July 2009

Talent Development & Excellence

Official Journal of the

Editors-in-Chief:

Albert Ziegler Jiannong Shi

This journal
Talent Development and Excellence is the official scholarly peer reviewed journal of the International Research Association for Talent Development and Excellence (IRATDE). The articles contain original research or theory on talent development, expertise, innovation, or excellence. The Journal is currently published twice annually. All published articles are assessed by a blind refereeing process and reviewed by at least two independent referees. Editors-in-Chief are Prof. Albert Ziegler, Ulm University, Germany, and Prof. Jiannong Shi of the Chinese Academy of Sciences, Bejing. Manuscripts can be submitted electronically to either of them or to Editor@iratde.org. Articles will be submitted for abstracting and indexing in Academic Search; Australian Education Index (AEI); British Education Index; Contents Pages in Education; EBSCO Online; EBSCO CD Rom Database; Education Journal; Educational Research Abstracts online (ERA); ERIC; e-psyche; ERIH (European Reference Index for the Humanities, Pedagogical and Educational Research); Gifted and Talented Abstracts; IBR (International Bibliography of Book Reviews of Scholarly Literature on the Humanities and Social Sciences); IBZ (International Bibliography of Periodical Literature on the Humanities and Social Sciences); ISI Social and Behavioural Sciences; National Database for Research into International Education (NDRI); psycINFO; PsychLit; Psychological Abstracts; Research into Higher Education Abstracts and Social Science Citation Index.

Editors-in-Chief:
Albert Ziegler, University of Ulm, Germany Jiannong Shi, Academy of Sciences, Beijng, China

Editorial Assistant:
Bettina Harder, University of Ulm, Germany

International Advisory Board:

Ai-Girl Tan, Nanyang Technological University, Singapore Barbara Schober, University of Vienna, Austria Carmen M. Cretu, University of IASI, Romania Elena Grigorenko, Yale University, United States Hans Gruber, University of Regensburg, Germany Mantak Yuen, The University of Hong Kong, P.R. China Marion Porath, University of British Columbia, Canada Osamah Ma'ajeeni, King Abdul Aziz University, Saudi-Arabia Petri Nokelainen, University of Tampere, Finland Robert Sternberg, Tufts University, United States Wilma Vialle, University of Wollongong, Australia Wolfgang Schneider, University of Wuerzburg, Germany

Impressum:
V.i.S.d.P.: Albert Ziegler, St.Veit-Str. 25, 81673 Mnchen, Germany

Welcome note

This issue represents the official inauguration of our scientific journal published under the umbrella of the International Research Association for Talent Development & Excellence (IRATDE). This issue also symbolizes launching of an academic venue with the highest possible intellectual quality where scientists, educators, and other professionals interested in the fields of gifted education, talent development, creativity, innovation and excellence can present their scientific research. Through IRATDE, we seek to encourage and support joint research initiatives among scholars of different nationalities. Our goal is to build bridges for inquisitive intercommunication that can deepen our understanding and enrich our research. Through our systematic partnership we aim at fortifying civilized dialogue among human societies concerning how to promote the global development of creative minds that care about constructing an affectionate, merciful, and wise humane society. My colleagues and I look forward to opportunities to work together with researchers and practitioners in the field of talent development and excellence to develop our association to serve our field better and to promote interdisciplinary and intercultural projects around the world.

Abdullah M. Aljughaiman President The International Research Association for Talent Development and Excellence

Talent Development & Excellence Volume 1 Number 1 2009

Contents

Editorial A. Ziegler and J. Shi The impact of smart fractions, cognitive ability of politicians and average competence of peoples on social development H. Rindermann, M. Sailer and J. Thompson Modeling mathematical Actiotopes: The potential role of CLARION S. N. Phillipson and R. Sun Relative age, talent identification and youth skill development: Do relatively younger athletes have superior technical skills? J. Schorer, J. Baker, D. Bsch, A. Wilhelm and J. Pabst Gifted Learners Epistemological Beliefs M. Porath, J. Lupart, J. Katz, C. Ngara and P. Richardson The Leonardo Laboratory: Developing Targeted Programs for Academic Underachievers with Visual-Spatial Gifts T. M. Newman, W. Brown, L. Hart, D. Macomber, N. Doyle, S. A. Kornilov, L. Jarvin, R. J. Sternberg and E. L. Grigorenko

27

45

57

67

Talent Development & Excellence Vol. 1, No. 1, 2009, 1-2

Editorial

Editorial
Launching a new scientific journal provokes the question of its raison dtre because of the now insurmountable flood of publications. As we know, less than 20% of researchers interested in a given field read more than even 1% of the newly appearing literature. Nevertheless we believe for several reasons that Talent Development and Excellence (TDE) will play an important role, and in four ways. As its title says, TDE shall first be a place of engagement for two important research currents that have been up to now largely separated: Giftedness research and Expertise research. The former researchers traditionally focus more on talented children, following the development of their achievement prospectively, while the latter traditionally start above all with persons who have already excelled, but of course with an interest in how their achievements developed. We think it is important and desirable for future research to unite these two branches. Our second wish in connection with TDE is for interdisciplinary access and a simultaneous broadening of traditional problems. Typically, talent development and excellence are investigated by psychologists and pedagogues. By contrast we would like to invite researchers in further disciplines (educational) economists, jurists, sociologists, political scientists, AI researchers, physiologists, neurologists, etc. to shed light from each of their specialist perspectives on the process of creativity and achievement development and their antecedents up to the level of expertise and innovation together. Moreover we explicitly exclude any limitation to individuals: Excellence can also be an attribute of a group of persons or a social community, and, in the age of Artificial Intelligence, even the attribute of a machine or machine network. While on the one hand publications in modern knowledge societies are more and more accessible, on the other the necessary expenses of libraries and universities have risen hugely in the past years. Our third wish is therefore to make TDE freely available, and so the journal can be read online at www.iratde.org. The print version is generously sponsored by the King Faisal University in Al-Hassa. Finally and fourthly, TDE is the official scholarly journal of the International Research Association for Talent Development and Excellence (IRATDE). We hope that TDE can provide IRATDE-associated researchers with a bit of intellectual community of their own. Heiner Rindermann, Michael Sailer and James Thompson present a pioneering study on the smart fraction theory. This says that the small share of high-achievers make an overproportional contribution to societal growth and prosperity. Such a finding would of course be the best legitimation for societys support of talent and excellence. The authors investigated whether there existed relations between the scores of 90 countries in international educational studies (TIMSS 1995-2007, PISA 2000-2006 and PIRLS 2001-2006) and a number of societal desiderata such as wealth (GDP), numbers of patents, scientists and Nobel Prizes, government effectiveness, democracy, but also AIDS and homicides. In particular they wanted to find out whether the scores of weaker achievers, average achievers, or of the smart fraction could better predict the factors mentioned. And in fact the authors find clear indications of the prognostic strength of the smart fraction theory: Our results emphasize the importance of nurturing the highly gifted. Could a new scientific journal start off with any more convincing proof than this of the societal significance of its field of research? Research areas are only as viable as their methods various. Shane N. Phillipson and Ron Sun propose a new method for giftedness research, which also expertise research has until now applied only sporadically: modeling. Their interest is particularly to examine potential benefits of and problems in modeling the Actiotope Model of giftedness using the cognitive architecture CLARION. We hope this innovative paper will help make giftedness researchers
ISSN 1869-0459 (print)/ ISSN 1869-2885 (online) 2009 International Research Association for Talent Development and Excellence http://www.iratde.org

A. Ziegler & J. Shi

aware of the advantages of modeling, as well as establish further bridges to expertise research. Especially promising is that modeling with CLARION as proposed by the authors offers a meta-theoretical unifying basis for vigorous scientific exchange. As circa two decades ago chance became accepted as an independent variable in various models of talent development, it met with the intense criticism of many traditionally oriented scholars. But quite soon several factors were found that, while randomly distributed in the population, do produce a systematic bias. The best studied of these is the birth date: Relatively older students within an age interval appear to be advantaged. The most significant explanations for this are held to be, first, maturational differences, and secondly, selection processes. J. Schorer, J. Baker, D. Bsch, A. Wilhelm, and J. Pabst broaden the state of our knowledge by pursuing three questions in two very interesting studies: a) Can the relative age effect be demonstrated in two samples of handball players being considered for national team selection? b) What role do variables play related to body-size and technical skill development? c) Can these findings be demonstrated also for the differences between those ultimately selected for the national team and those not selected? Marion Porath, Judy Lupart, Jennifer Katz, Constantine Ngara, and Pamela Richardson take up in their contribution what can well be considered a current hot topic in gifted education research: epistemic beliefs. These are concerned with making meaning of school, understanding ones own intentions with regard to learning, and interpreting ones learning in the context of knowledge traditions. Their significance is so great because most pedagogic measures do not act directly, but first must be filtered through the epistemic beliefs of their addressees. The authors are able to show in this exciting empirical study with pupils in grades 1-12 that their understanding of learning develops from a focus on the activities of school (e.g. reading, math), to interpretation of the meaningfulness of different learning activities, and finally to the meaning of knowledge itself. The study by Tina M. Newman, William Brown, Lesley Hart, Donna Macomber, Niamh Doyle, Sergey A. Kornilov, Linda Jarvin, Robert J. Sternberg, and Elena L. Grigorenko is concerned with a seldom investigated sample: twice exceptional children. The authors present with the Leonardo Laboratory a short-term program for children with coexisting learning disabilities and spatial gifts. In fact the phenomenon of underachievement, i.e. unexpected low achievement, is today among the greatest challenges to the promotion of talent. Precisely twice exceptional children are today considered by schools to be under-identified and under-served, and therefore to deserve particular attention. Although the program did not improve academic skills, children demonstrated gains in self-efficacy and improvements in organizational skills. Given the recognized difficulties in helping twice exceptional children, these results are definitely encouraging. Albert Ziegler Jiannong Shi

Talent Development & Excellence Vol. 1, No. 1, 2009, 3-25

Smart fraction

The impact of smart fractions, cognitive ability of politicians and average competence of peoples on social development
Heiner Rindermann1,, Michael Sailer1 and James Thompson2
Abstract: Smart fraction theory supposes that gifted and talented persons are especially relevant for societal development. Using results for the 95th percentile from TIMSS 19952007, PISA 2000-2006 and PIRLS 2001-2006 we calculated an ability sum value (N=90 countries) for the upper level group (equivalent to a within country IQ-threshold of 125 or a student assessment score of 667) and compared its influence with the mean ability and the 5th percentile ability on wealth (GDP), patent rates, Nobel Prizes, numbers of scientists, political variables (government effectiveness, democracy, rule of law, political liberty), HIV, AIDS and homicide. Additionally, using information on school and professional education, we estimated the cognitive competence of political leaders in N=90 countries. Results of correlations, regression and path analyses generally show a larger impact of the smart fractions ability on positively valued outcomes than of the mean result or the 5th percentile fraction. The influence of the 5th percentile fraction on HIV, AIDS and homicide, however, was stronger. The intelligence of politicians was less important, a longitudinal crosslagged analysis could show a positive influence on the cognitive development of nations. Keywords: high ability fraction, gifted portion, intelligence, patent rate, growth

A large amount of studies published in the last two decades has shown that cognitive ability levels of societies are relevant for the development of positively valued aspects of peoples and countries. Following an economic research tradition human capital is relevant for economic growth and wealth (Hanushek & Kimko, 2000; Lynn & Vanhanen, 2002, 2006; Jones & Schneider, 2006; Weede, 2006; Rindermann, 2008a). In addition, cognitive ability of nations has a positive impact on political development, in that it helps building up democracy, the rule of law and political liberty (Simpson, 1997; Rindermann, 2008b). Intelligence, knowledge and the intelligent use of knowledge also have beneficial effects on health, for instance they act as a brake on the spread of HIV (Oesterdiekhoff & Rindermann, 2007; Lakhanpal & Ram, 2008; Rindermann & Meisenberg, 2009). Finally, cognitive competence is relevant for the development of modernity as a societal and especially as a cultural phenomenon consisting of education, autonomy, liberty, morality and rationality (Habermas, 1985/1981; Meisenberg, 2004; Oesterdiekhoff, 2008; Lynn, Harvey & Nyborg, 2009). Societies at a higher ability level develop more complex, more evidence-based, more ethical and more rational world views. For some scholars like Georg Oesterdiekhoff (2000) or Michael Hart (2007) intelligence is the driving force of history. These broad effects at the cross-national data level are backed in different societies by results at the individual level for job performance and wealth (Bacharach & Baumeister, 1998; Schmidt & Hunter, 2004; Irwing & Lynn, 2006; Rindermann & Thompson, 2009), for tolerance, civic political attitudes and participation in elections (Herrnstein & Murray, 1994; Denny & Doyle, 2008; Deary, Batty & Gale, 2008), for health behavior and health (Goldman & Smith, 2002; Gottfredson, 2004), moral judgment (Piaget, 1997/1932; Kohlberg, 1987) and more rational world views (Oesterdiekhoff, 2000; Nyborg, 2009).
1

Institute of Psychology, Karl-Franzens-University Graz, Universitaetsplatz 2, A-8010 Graz, Austria Corresponding author. Email: heiner.rindermann@uni-graz.at 2 Department of Psychology, University College London, UK
ISSN 1869-0459 (print)/ ISSN 1869-2885 (online) 2009 International Research Association for Talent Development and Excellence http://www.iratde.org

H. Rindermann et al.

The mentioned positive and global effects of cognitive competences are also the reason why international economic or educational organizations like OECD and UNESCO are conducting studies on education and competence. But the terms used vary, ranging from human capital by economists (also including personality factors relevant for productive behavior like selfdiscipline; e.g. Becker, 1993/1964), literacy by educationists (OECD, 2003) and psychometricians and cognitive developmentalists focus on the concept of intelligence (Cattell, 1987/1971; Piaget, 2001/1947). The measurement of cognitive abilities in these diverse research traditions is based on mental tasks that can be solved with thinking and (within the different approaches) with varying amounts of specific knowledge. Students scores from all these different measurement approaches highly correlate at the individual level (Jensen, 1998; Ceci, 1991) and very highly at the national level (Rindermann, 2007a). Thus, the manifold of positive correlations among ostensibly dissimilar tests (e.g. verbal, math, science, figural and Piagetian tasks, psychometric and student assessments) is taken as an evidence that the same underlying latent factor is involved in all complex cognitive performance (Ceci, 1991). Formal education (years of schooling or degrees) often is used as a proxy or causal factor for cognitive ability. We use the term cognitive ability and cognitive competence interchangeably. They stand for intelligence (thinking ability), the disposal of true and important knowledge and the intelligent use of that knowledge. In spite of various terms, research methods, disciplines and paradigms different researchers came to the same result: Cognitive ability of individuals and societies is important for positive outcomes. This causal assumption is especially backed by the use of longitudinal designs controlling for the influence of other factors and of backward effects e.g. of wealth or democracy (Rindermann, 2008a, 2008b). Of course, intelligence and knowledge are not the only and single determinants of the positive attributes described. There are additional factors behind (like education, culture and genes) and beneath (like neighbor effects, personality or chance factors) cognitive ability and between cognitive ability and the positive outcomes (like the quality of institutions or meritoric principles). There are positively valued attributes which may not depend (positively) on cognitive competence (like happiness and mating success). Therefore, high competence is no guarantee for positive outcomes. Finally, intelligence not guided by ethics and rationality leads to biased, questionable and destructive results. A cognitive theory of action can explain the effect mechanism: An increased ability to understand information, causal relationships, and consequences of somebodys own and others behavior in everyday life improves ones own behavior and the behavior of important others like parents, friends, classmates, teachers, officials, managers, scientists and political leaders. Additionally, cognitive ability strengthens attitudes oriented towards a more ethical lifestyle including perspective taking, considerateness and general pensiveness. But cognitive ability is also connected meaning leading to, depending on and relying on common background factors to better environments (nurturing, training and gate keeper effects) and a greater appreciation of civic virtues and bourgeois values like appreciation of future lifes quality, of security, of education, work and individualized love. Thinking ability forms an integrated part of a global pattern oriented towards a rational, active, self-controlled and farsighted lifestyle. They all create positive feedback loops: own intelligence and the intelligence of others enhance environmental (the physical, social and cultural) quality and by this again cognitive development.

Smart fraction theory

Thus far we have solely mentioned mean cognitive ability effects. But different authors of the past and the present have stressed the importance of smart fractions. They place less emphasis on mean intelligence, but champion the importance of the cognitive ability level of a smart fraction, say a societys upper 10, 5, 1 or 0.1%, which should be mainly responsible for the progress of a society in technological, economic, political and cultural development (Thomson, 1937; La Griffe du Lion, 2004; Gelade, 2008; Murray, 2008; Weiss, 2009). Because of

Smart fraction

the assumed relevance of gifted and talented persons the Scottish pioneer of intelligence research, Godfrey H. Thomson (1881-1955), recommended the furthering of intelligent children (similarly Gelade, 2008; Murray, 2008). The smart fraction is regarded as responsible for progress in a utilitarian sense (wealth, health and power), but also for achievements in a non-utilitarian sense (music, literature, art, religion, ethics, philosophy and world-views). For many this kind of thesis has an elitist taste. But it is first of all an empirical question, which could be solved with a rigorous orientation towards scientific standards, based on philosophically justified epistemic principles: Is it true, that smart fractions are more relevant for important indicators of societal development like wealth and democracy or scientific and technological progress? Up to now there is no research on this subject using empirical data. Gelade (2008) has solely used an assumed ability level, he and the anonymous scholar La Griffe du Lion (2004) have estimated smart fractions from the mean value. But the student assessment studies (SAS) TIMSS, PISA and PIRLS give results for the 95th percentile (comparable to an IQ of 125 or higher, mean around IQ 130) enabling us to directly test a smart fraction theory. Conversely, SAS makes it possible to prove a non-smart fraction theory (5th percentile, comparable to an IQ of 75 or lower, mean around IQ 70): If the cognitive level of the lower groups is more relevant for outcomes standing for lower cognitive ability (and at least partly explainable by this) like HIV and AIDS (Oesterdiekhoff & Rindermann, 2007; Lakhanpal & Ram, 2008; Rindermann & Meisenberg, 2009) or criminality (Thomson, 1937; Herrnstein & Murray, 1994; Ellis & Walsh, 2003; Cunha, Heckman, Lochner & Masterov, 2006). A group in society especially important are statesmen (or rarely stateswomen). Simonton (1984, 1985, 2006) has done several studies on the measurement of political leaders cognitive ability and its impact on political success. Apart from politically (and for the media) interesting information (e.g. comparatively low cognitive ability of Bush junior, Simonton, 2006, or comparatively high cognitive ability of Obama, Murray, Cardoso & Mendes, 2008) Simonton and others (Deary et al., 2007) were able to show that cognitive traits of political leaders are important for their success. Political leadership is a cognitively demanding task (Suedfeld, Guttieri & Tetlock, 2003, p. 255), so the influence of politicians intelligence and knowledge on the development of societies (apart from their private success in being reelected or continuing in power) would be expected. The opponent hypothesis to the political leader thesis would be that it is not the attributes of politicians that are important but the ones of peoples themselves (e.g. Chomsky, 2009). Maybe there is no either-or: Both, the ability of peoples and their leaders are important. Empirical data will show us.

Method
To reduce problems of poor data quality and to ensure that countries are represented in the sample at all levels of cultural, social and economic development, an average score was formed from different measures and measurement points to create one construct for each country. Cognitive abilities Three cognitive competence measures were used: mean results in TIMS-, PISA and PIRLstudies, mean at 5th percentile in TIMS-, PISA and PIRL-studies and mean at 95th percentile in TIMS-, PISA and PIRL-studies. Sources were TIMSS 1995, 4th and 8th grade, math and science, TIMSS 1999, 8th grade, math and science, TIMSS 2003, 4th and 8th grade, math and science, TIMSS 2007, 4th and 8th grade, math and science; PISA (always around 15 year old students) 2000, 2003 and 2006, verbal, math and science literacy, 2003 also problem solving, PIRLS verbal literacy in 4th grade 2001 and 2006. All results were originally presented in student assessment scales (SAS M=500, SD=100).

H. Rindermann et al.

Resources had been for TIMSS 1995 (in the order of the grades and scales: Beaton, Mullis, Martin, Gonzalez, Kelly & Smith, 1996; Beaton, Martin, Mullis, Gonzalez, Smith & Kelly, 1996; Mullis, Martin, Beaton, Gonzalez, Kelly & Smith, 1997; Martin, Mullis, Beaton, Gonzalez, Smith & Kelly, 1997), TIMSS 1999 (Mullis, Martin, Gonzalez, Gregory, Garden, OConnor, Chrostowski & Smith, 2000; Martin, Mullis, Gonzalez, Gregory, Smith, Chrostowski, Garden & OConnor, 2000), TIMSS 2003 (Mullis, Martin, Gonzalez & Chrostowski, 2004; Martin, Mullis, Gonzalez & Chrostowski, 2004), TIMSS 2007 (Mullis, Martin & Foy, 2008; Martin, Mullis & Foy, 2008), PISA 2000 (OECD, 2003), PISA 2003 (OECD, 2004a, b), PISA 2006 (OECD, 2007a, 2007b), PIRLS 2001 (Mullis, Martin, Gonzales & Kennedy, 2003) and PIRLS 2006 (Mullis, Martin, Kennedy & Foy, 2007). A sum value of different scales, grades/age groups, studies and study approaches (grade vs. age level studies; studies trying to measure abilities defined by curriculum like TIMSS vs. studies trying to measure abilities defined by cognitive demands in modernity like PISA) is more convincing (say more representative, reliable and valid). High correlations between scales within and across studies allow to sum up scales to a general sum value (all factor loadings on an international G-factor were >.90; Rindermann, 2007a, 2007b). To form a common score the results were at first averaged within one grade, year and study between different scales (e.g. within TIMSS 1995, 4th grade, across math and science), secondly within one year and study between different grades (e.g. within TIMSS 1995, across 4th and 8th grade), thirdly within one study between different years (e.g. within TIMSS, across 1995, 1999, 2003 and 2007), fourthly within different grade vs. age study approaches across TIMSS and PIRLS (TIMSS and PIRLS are studies done in grades, PISA is a study done in a single age group), fifthly and finally between different study approaches (across grade and age approach studies: TIMSS-PIRLS-mean and PISA-mean). All averaging was done using ztransformations calculating means and standard deviations in countries which participated in all samples used for averaging (so z-formula are based on the same countries and over- or underestimation is avoided). Subsequently the z-results were re-normed using means and standard deviations obtained by simple arithmetical averaging of all three study results (SASscale with M=500 and SD=100, mean, 5th percentile, 95th percentile) as an orientation. At the end the values were transformed to the more usual IQ-scale, using Great Britain as reference country, SAS-SD were simply transformed to an IQ-scale (Greenwich-IQ, M=100, SD=15). Results are provided for N=90 countries. Means in SAS-scale are 453, 304 and 596, in UK-IQscale 90, 68 and 111. The results are not identical with the formally published cognitive ability values of Rindermann (2007a), because a) psychometric intelligence test results were not used here (Lynn & Vanhanen, 2006, provided no information on the tails), b) older student assessment studies like IEA-Reading and IAEP were not used (for IEA-Reading we could not find information on the tails, IAEP 1991 only published percentages of solved tasks, additionally samples were seriously biased), c) newer studies were included (PISA 2006, PIRLS 2006, TIMSS 2007), and d) the results were not corrected for age and grade or sample quality. Nevertheless the correlations are very high (with former corrected cognitive ability sum r=.92, with uncorrected r=.95, N=88). Normed values of all variables at international data level are somewhat arbitrary, e.g. the student assessment scale with M=500 and SD=100. The norms are estimated by the authors of the student assessment studies in orientation to results in OECD-countries (and sometimes in accordance with older results). OECD-membership, however, is no scientific criterion. IQnorms depend on the secular rise of intelligence and intelligence test results (Flynn-effect). Student assessment results are biased because only youth in school participated, in several countries pupils had been too old (especially in older studies and in developing countries), not all regions participated (especially in older studies and in developing countries) etc. (see Rindermann, 2007a; Wuttke, 2007). We have not applied corrections here because results at the higher (smart, gifted, talented) or lower (non-smart, imprudent or dull) levels are differently affected by the distortions, presumably the high level results are nearly not affected, but the low level results (and less the means). Corrections should be tried in further

Smart fraction

studies. But also in the other variables the norms are arbitrary, e.g. in GDP (inflation, Dollar or Euro), in democracy (which scale?). The competence levels of smart fractions, average and non-smart fractions are obtained through student assessment studies. But students do not work, do not win Nobel Prizes nor do they vote. We assume that the results of students could be generalized to adults, an assumption that is backed by high correlations with IQ measures (r=.87, N=86, Lynn & Vanhanen, 2006; often gained in adult samples), with an adult literacy study (r=.68, N=20; OECD, 2000) and the educational level of societies (r=.67, N=84, r=.75, N=85; measures see below). OECD is planning an adult literacy study for a larger country sample, by using their data on tails it would be possible to prove if our results are stable. Because in high ability societies there is a larger percentage of a cognitive elite at a higher level (and inverted for the lower level) the three values are correlated (see also Table 2): mean with high (r=.97), mean with low (r=.97), high with low (r=.90). As a consequence of these high correlations suppressor effects are expectable in regression analyses. For repeated measurements with cognitive abilities (Figure 5) old student assessment studies collected by Lee and Barro (1997) were used (here N=17; for further information see Rindermann, 2008b). Cognitive ability of leading politicians To estimate the cognitive ability of statesman (stateswomen) based on their formal education we selected the leading politicians of all the countries in the student assessment studies between 1960 and 2009. Leading politicians include presidents (33%), heads of government (64%) and the rest kings, emirs and sultans (3%). It was difficult for many countries to assess who has the real decisive power or who used or uses it (e.g. for Czech Republic, Iran, Poland, Russia and especially Switzerland). For these countries the two leading positions were used. A second problem was the modifications of countries (e.g. Yugoslavia, Soviet Union, Germany). In these cases the largest successor country represents the older one and vice versa, smaller successor countries start existing with their formal legalization. We also include (like SAS) some smaller territories which are not formal states (like Palestine or Taiwan). We always use the most usual names. A third problem often lies in not exactly knowing the leaders level of education. A fourth problem is the assessment of several educational vitae like those of clerical leaders in Iran (is it a university degree? we estimated it as a university degree in this case). We have not assessed the content or the quality of a university degree (e.g. in Science, Technology, Engineering and Mathematics/STEM or law, highly or lowly ranked institutions). The most serious problem lies in the low comparability of education and educational degrees across countries. There is considerable heterogeneity in educational standards across the world. According to the student assessment results, a secondary degree in OECD-countries is hardly comparable to a secondary degree in developing countries, because the former have much higher scores. Furthermore, corruption and forging at universities and especially in conferring and claiming to have degrees is not impossible. We tried to exclude all honorary doctorates. By using databanks (Munzinger-Biography: www.munzinger.de, MSN-Carta, Who is Who and only rarely Wikipedia) we could find information for N=896 leading politicians. Homepages of politicians were not used; from experiences in a study on Austrian and German politicians we found that they try to overestimate their educational record (e.g. university dropout given as a degree). We assessed education for school education (as highest level: no school; primary school; secondary school; high-school diploma; university degree; doctorate; doctorate plus further scientific degrees, Habilitation, Venia Legendi, or scientific achievement like publications), and education for professional training (as highest level: no vocational or professional training; vocational training; qualified training like technician; university degree; doctorate plus further scientific degrees, Habilitation, Venia Legendi, or scientific achievement like publications). For the last category (doctorate plus further scientific degrees or achievement) Fernando

H. Rindermann et al.

Henrique Cardoso from Brazil is an example (published several scientific books, professor of political science, member of or taught at Institute of Advanced Study, Princeton, Collge de France, Stanford, Brown, etc.). Both overlapping indicators correlate with r=.90 (Cronbach=.95). Finally we tried following Gottfredson (2004, 2005) to estimate IQ-levels: For no school IQ 80, for primary school IQ 90, for secondary school IQ 100, for high-school diploma IQ 113, for university degree 119, for doctorate 129 and for doctorate plus further scientific degree or achievement 138. For levels in between and different vocational or qualified trainings we gave values between 80 and 119. The mean of the two educational indicators and our IQ-estimate correlate with r=.96, IQ with school education r=.99, with professional training r=.88 (N=896). The mean IQ of politicians across countries is 118 (SD=7, N=90). The mean IQ is increasing from 1960 to 2000: in the 60s IQ 114, in the 90s IQ 118, in 00s IQ 119 and the variances are decreasing (from 12 to 7 and 7). Because we mainly include only countries participating also in student assessment studies the worldwide average could be lower. The sample for analyses was N=90 countries (for the map, Figure 2, N=95). Wealth and development indicators of societies Wealth: Gross domestic product 2003 (GDP per capita, purchasing power parity/ppp; Human Development Report/HDR, 2005, here for N=84 countries). GDP (ppp) 1998 per capita from Lynn and Vanhanen (2002), here for N=85 countries. GDP considers only goods and services produced within a country, not income received from abroad. GDP is an indicator for produced wealth. Their sources are UN data sets. Developmental level of society: Human development index (HDI) from HDR (2005). The HDI is a composite indicating three basic dimensions of human development: a long and healthy life (life expectancy at birth), knowledge (adult literacy rate and the combined gross enrolment ratio for primary, secondary and tertiary schools) and a decent standard of living, as measured by GDP per capita in ppp US dollars (here for N=85 countries). Educational level of societies and attributes of educational systems The general educational level of society was estimated in two ways: 1. Standardized sum of literate adults rate 1991, rate of persons between 12 and 19 years old from 1960 to 1985 having graduated from secondary school, and years of school attendance of persons 25 years or older 1990-2000 (Rindermann, 2007a). This indicator is used for analyses with data from a longer time interval (here for N=84 countries). 2. Education index 2003 (sum of adult literacy rate and combined gross enrollment ratio for primary, secondary and tertiary schools; HDR, 2005). This indicator is used for analyses with data around the turn of the century (here for N=82 countries). Young tracking age: Age of initial tracking between schools (reversed: young age), for some countries (e.g. Hong Kong and Switzerland) the information given by OECD/PISA was corrected. Class size: Large class sizes, many pupils per class. For the two educational policy variables see Rindermann and Ceci (2009). Attributes standing for high cognitive achievement of a society (excellence) Patent rate: Number of patents of a nation (sum of residents and nonresidents) related to population size, average annual patents per 1 million people 1960-2007 and 1991-2007 (the two decades, in which the majority of student assessment studies were carried out, here used for N=81 and 76 countries). Source is the World Intellectual Property Organization (WIPO, 2009), an agency of the United Nations. Nobel Prizes: Nobel Prizes in peace, literature and science 1901-2004 related to population size (sources: http://nobelprize.org/nobel_prizes/lists/all, http://en.wikipedia.org/wiki/ List_of_Nobel_laureates_by_country). Science sums up Nobel Prizes in physics, chemistry,

Smart fraction

medicine and economics. Mean correlations between those are around r=.90 (=.97, here for N=87 countries). Scientist rate: Scientists and engineers in research and development per million people, 19851995 (source: Kurian, 2001, p. 388, here for N=51 countries). High-technology exports: High-technology exports as percentage of manufacturing exports, 1997 (source: Kurian, 2001, p. 389-390, here for N=61 countries). Political indicators Government effectiveness: Government effectiveness 1996-2005 as the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the governments commitment to such policies (Kaufmann, Kraay & Mastruzzi, 2006, p. 2). The estimates are given here for N=88 countries. Democracy: Democracy 1950-2004 and 1996-2000 from Vanhanen (2005) and Marshall and Jaggers (2000), =.95. The estimates are used for N=84 countries. Rule of law: Rule of law 1970-2000 and 2000 with emphasis on ownership law from Gwartney and Lawson (2003) and Knack and Keefer (1995), =.90, here N=76 and 73. Political freedom: Political liberty or freedom was taken from Freedom House (2004) for the year 1999 (here N=86). For further description of political indicators see Rindermann (2008b). HIV, AIDS and criminality indicators HIV: HIV-infection rate 2001 and 2003 (mean r=.95, =.99) for adults and adults 15-49 years old are from UNAIDS/WHO (2003), here N=82. AIDS: Percentage (per 100.000) of people suffering on AIDS 1995-2000, source Statistisches Bundesamt (2002, p. 290ff.), here N=83. Homicide: Homicide rate (per 100.000 inhabitants) 1995 and 2002 (=.41, here N=80) from Interpol (2004). Problems in international data sets In all international data, problems and flaws are detectable. E.g. students are older than expected. How we define grade level if in some countries children start attending school by the age of four, in others by the age of seven? How can we compare wealth between First and Third World countries where much of the economy is based on subsistence farming and informal trading? Luxembourg, Ireland and the States seem to be too rich or at least richer than expected, Saudi Arabia is astonishingly constitutional (rule of law). How could literacy of students in Yemen be measured, where several regions are more or less out of state control? Kazakhstan seems to be too good in cognitive competence levels. There are some anomalies in data like decreasing rates of patents in UK and Switzerland or years without patents in Norway (changed into missing values). Nobel Prizes also seem to be given with political considerations, e.g. there is a serious decline of Nobel Prizes for Germany starting with 1933 (Nobel Prizes are given for lifes work achievement and not for one single achievement in a current year); related to the population size persons from Sweden seem to receive a Nobel Prize more easily, US-Americans less easily. And not all countries (e.g. Israel) existed during the whole 20th century. Finally, there is no Nobel Prize in social sciences and humanities (e.g. in psychology, sociology and philosophy) and mathematics (though there is the Fields Medal). For future analyses additional awards should be considered. The scientist rate is slightly misleading because many of the scientists and engineers in Western countries are of foreign origin (brain gain like in USA and Singapore). The quality of

10

H. Rindermann et al.

HIV data is sometimes doubted, especially for countries from Eastern Europe, for developing countries, Muslim countries, and countries from sub-Saharan Africa. We tried to solve or at least reduce these problems by aggregating data from different years and by using different indicators for the very construct, e.g. patent rates, Nobel Prizes, scientist rates and high-technology exports for cognitive excellence of a society (Figure 3). Statistical methods At first we calculated correlations between mean, non-smart and smart fraction competence measures on the one hand and different societal attributes on the other. Higher correlations for the non-smart or smart fraction indicate their larger relevance for these attributes. Secondly we present regression analyses showing relative impact of smart fraction ability or non-smart fraction ability compared to mean values. Finally we present results of crosssectional and longitudinal path-analyses. Because of suppressor-effects the smart and nonsmart fraction ability levels are not put together with mean ability in one regression analysis (only either smart or non-smart and average). Regression and path analyses are used to calculate direct, indirect and net effects of variables. In these analyses the standardized path coefficients () between different variables are to be interpreted. Correlations are always added in parentheses. Correlations help to quickly estimate the influence of other variables in the model (difference between correlation and path coefficient), they allow the model to be checked (r=R=1-error) and the proportion of the explained variance in each factor to be calculated (R=r). Good values for fit indices are SRMR.08 (Hu & Bentler, 1998, 1999) or SRMR.05 (Schermelleh-Engel, Moosbrugger & Mller, 2003) and CFI.95 (Hu & Bentler, 1998, 1999) or CFI.97 (Schermelleh-Engel et al., 2003), and acceptable fit is reached with SRMR.10 and CFI.95 (Schermelleh-Engel et al., 2003). For the analyses, SPSS 16.0 and LISREL 8.80 were used, while SAS 9.1 was used to produce the maps. Significance tests were not used for interpretation (see Rindermann, 2008a; for an in-depth justification e.g. Cohen, 1994; Falk & Greenbaum, 1995; Gigerenzer, 2004; Hunter, 1997). Especially in comparisons between countries, they are not appropriate for scientific reasoning. The results of significance tests depend on the number of observations. The observations here are for a limited number of countries (around 90), but each country-level observation is based on thousands of individual observations within each country. Possible causal relationships are not more or less true if they are significant or not (see Rindermann & Meisenberg, 2009; Rindermann & Ceci, 2009).

Results
Distributions The results for the most important variables are shown in Table 1. Figures (Maps) 1 and 2 show the distributions of the smart fractions and political leaders ability levels across the world. Because average, upper and lower levels are correlated there are at first sight no large differences: The highest values for the smart fractions are found in East Asia (1. Singapore IQ 127, 2. South Korea IQ 125, 3. Japan IQ 124, 5. Taiwan IQ 123, 9. Hong Kong IQ 122). A similar result was found in psychometric (average) intelligence or in student assessment studies (see Rindermann, 2007a). Different from the SAS, Scandinavia reaches in the cognitive elite not such a good rank (11. Finland IQ 121, 12. Estonia IQ 121 [the Baltics are added here], 16. Sweden IQ 120, 25. Denmark IQ 118, 34. Latvia IQ 117, 38. Lithuania IQ 116, 39. Iceland IQ 116, 41. Norway IQ 116). Maybe a homogenizing educational policy furthering weaker but disadvantaging high ability pupils leads to a smaller standard deviation and lower values for a gifted subgroup. Better are the traditional Commonwealth countries (5. New Zealand IQ 123, 7. Australia IQ 122 and 8. United Kingdom with IQ 122). They are followed by Western and Eastern European and North American countries, by South European countries, Arab or

Smart fraction

11

Muslim and Latin American countries and finally by sub-Saharan countries. The countries with the lowest results are 84. Botswana (IQ 96), 85. Saudi-Arabia (IQ 95), 86. Morocco (IQ 95), 87. Kyrgyzstan (IQ 94), 88. Belize (IQ 90), 89. Ghana (IQ 89) and 90. Yemen (IQ 84). Presumably many not participating countries would have lower values.

Figure 1. World map of cognitive ability level at 95th percentile (mean of TIMSS 1995-07, IGLU 2001-06, PISA 2000-06, N=90 nations, darker means higher competence, hachured: no data)

Figure 2. World map of leading politicians cognitive ability level 1960-2009, N=95 nations, darker means higher competence)

12

Table 1. Means of used variables (most important ones)

Country

CAmean

CA5%

CA95%

CApoliticians 60-09
120.20 106.63 117.83 125.33 115.55 128.20 119.00 107.00 121.31 119.00 121.14 116.00 116.50 115.92 120.11 122.75 119.67 119.67 122.33 118.25 115.00 90.00 118.42 120.11 120.24 119.67 123.00 123.25 118.60 123.07 119.00 119.77 120.08 119.33 115.38

CApoliticians 90-09
122.00 107.75 120.67 125.33 116.75 123.80 119.00 107.00 121.50 119.00 121.14 115.00 116.50 124.00 119.00 124.00 119.00 119.67 124.00 120.78 119.00 90.00 120.00 120.11 119.00 119.00 123.00 125.67 122.33 126.50 119.00 125.00 119.00 117.40 119.00

GDP 2003

HDI 2003

Patent rate 19602007

5.52 51.35 20.12 252.81 305.42 2.81 42.61 432.19 16.00 9.46 15.23 9.38 45.19 306.43 18.70 5.88 20.67 54.36 67.35 181.40 3.26 5.45 86.90 174.86 219.04 28.04 118.02 1.23 87.48 155.19 84.62 73.66 0.85 4.66

Patent rate 19912007

5.80 21.25 20.76 320.48 84.45 2.81 43.73 40.55 28.00 10.48 8.13 22.78 211.49 12.33 4.91 20.82 59.70 70.75 67.90 3.23 3.94 107.30 197.50 102.56 26.38 101.35 0.30 54.56 231.93 64.53 115.52 0.85 2.41

Nobel Peace 19012004

0 0 0.79 0 0 2.70 0 0 3.12 0 0 0 0 0.46 0 0 0 0 0 2.03 0.28 0 0 0 1.56 0 0.83 0 0 0 0 0 0 0.32

Nobel Literat. 19012004

per 10 mill 0 0 0 0 0.77 1.35 0 0 1.04 0 0 0 0 0 2.02 0.44 0 0 0.97 6.09 0.28 0 0 2.14 2.74 0 1.17 0 2.21 0 .99 47.66 0 0

Nobel Science 19012004

0 0 0.30 0 0.77 2.70 0 0 1.04 0 0 0 0 0.92 0 0 0 0 0.24 3.55 0.07 0 0 0.54 1.17 0 2.50 0 0 0 0.74 0 0 0

Scientists 198595
671 3166 1631 1814 168 2656 1978 1159 2647 458 19 2018 2812 2584 2843 774 98 1033 521

High technology 1997

1 22 15 39 24 23 18 25 19 20 19 13 27 7 16 24 26 31 26 12 29 39 20

Gov. eff. 19962005

92.6 90.7 100.4 93.9 129.7 127.4 86.4 108.7 125.9 97.9 89.8 111.3 99.0 96.7 130.9 119.6 98.3 103.1 118.0 111.4 132.1 97.4 95.6 113.0 131.7 124.7 90.5 126.4 98.2 112.4 123.3 111.2 131.3 93.9 93.4

Demo- Rule cracy of law 1950- 19702004 2000

IQ-scale (M=100) 90.0 87.2 103.8 107.3 128.3 129.9 89.5 83.8 132.9 105.5 104.6 101.1 93.6 123.1 101.4 104.3 104.8 119.6 102.8 133.0 85.6 96.4 116.7 127.3 122.5 107.3 128.8 89.7 117.1 95.9 132.2 89.1 86.4 96.3 82.7 95.2 125.8 127.7 103.6 128.5 103.2 110.1 103.5 107.8 127.0 93.6 87.7 107.5 100.8 104.7 128.0 91.1 84.1 107.4 119.7 114.2 131.0 84.9 102.6 117.7 105.9 122.9 91.0 86.7

Politi- HIV cal rate liberty 20011999 2003

% 92.8 85.2 108.0 96.6 119.4 119.4 89.0 77.6 115.6 119.4 111.8 100.4 108.0 119.4 111.8 96.6 96.6 119.4 115.6 119.4 85.2 108.0 115.6 119.4 115.6 100.4 115.6 104.2 111.8 96.6 115.6 119.4 96.6 81.4 0.05 0.06 0.61 0.12 0.09 0.21 0.04 0.17 0.18 1.93 33.99 0.59 0.04 0.27 0.27 0.51 0.04 0.03 0.05 0.17 0.04 0.59 0.84 0.08 0.33 0.05 0.09 2.73 0.17 0.09 0.05 0.19 0.09 0.09

H. Rindermann et al.

IQ-scale (M=100) Albania Algeria Argentina Armenia Australia Austria Azerbaijan Bahrain Belgium Belize Bosnia Botswana Brazil Bulgaria Canada Chile Colombia Croatia Cyprus Czech Republic Denmark Egypt El Salvador Estonia Finland France Georgia Germany Ghana Greece Hong Kong Hungary Iceland Indonesia Iran 81.10 80.56 81.50 93.06 101.12 99.65 84.62 84.24 99.13 63.55 90.60 73.93 81.59 93.46 101.75 83.62 80.61 95.96 91.59 99.96 98.46 81.14 77.53 102.26 102.91 98.17 87.62 99.08 61.25 94.37 103.66 99.37 96.45 81.75 82.83 55.84 63.23 54.72 69.34 79.06 78.16 73.01 61.99 75.02 40.93 69.88 50.79 58.43 67.92 79.59 60.95 58.15 77.23 68.65 78.92 76.86 53.73 59.36 84.40 84.96 77.01 66.16 75.71 32.86 71.45 83.32 78.07 75.34 62.00 60.64 103.56 97.94 105.79 116.94 121.94 119.34 98.84 105.80 116.53 89.95 110.07 96.15 104.65 117.22 120.32 105.97 101.38 115.06 112.63 119.96 118.17 107.28 96.19 120.75 120.92 117.77 107.50 119.72 89.38 115.46 121.54 119.77 116.00 100.93 104.46

ppp $ 4584 6107 12106 3671 29632 30094 3617 17479 28335 6950 5967 8714 7790 7731 30677 10274 6702 11080 18776 16357 31465 3950 4781 13539 27619 27677 2588 27756 2238 19954 27179 14584 31243 3361 6995

0-1 .780 .722 .863 .759 .955 .936 .729 .846 .945 .753 .786 .565 .792 .808 .949 .854 .785 .841 .891 .874 .941 .659 .722 .853 .941 .938 .732 .930 .520 .912 .916 .862 .956 .697 .736

per mill

per mill

Table 1. (continued)
Country

CAmean

CA5%

CA95%

CApoliticians 60-09

CApoliticians 90-09
117.80 118.25 123.00 119.00 113.91 113.00 119.80 105.00 119.00 121.73 119.00 123.00 123.00 124.00 119.00 122.33 122.33 129.00 124.00 120.43 119.00 123.86 122.33 116.00 120.67 113.00 122.33 121.50 118.50 121.53 121.00 101.50 123.77 124.00 90.00

GDP 2003

HDI 2003

Patent rate 19602007

Patent rate 19912007

Nobel Peace 19012004

3.19 6.24 0.18 0.09 4.03 0 0.30 0 0 0 0 0 0 0 0 0 0 0 0.17 0 0 0.76 0 5.12 0 0 0 0.52 0 0 0 0.10 0

Nobel Literat. 19012004

per 10 mill 9.58 3.12 1.11 0.19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.17 0 0 0 0 7.68 0 0 0 1.04 1.07 0 0 0.21 0

Nobel Science 19012004

0.80 0.78 0.32 0.19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 0 0 2.28 .87 2.56 0 0 0 0 0.27 0 0 0.13 0

Scientists 198595

High technology 1997

Gov. eff. 19962005

126.1 117.1 113.1 118.0 104.3 88.1 112.4 103.4 89.7 106.7 96.2 124.3 107.8 131.9 117.5 95.0 113.2 111.7 101.7 88.6 99.5 133.5 131.0 130.7 111.3 95.3 100.2 109.7 117.6 109.9 94.1 92.1 97.0

Demo- Rule cracy of law 1950- 19702004 2000

IQ-scale (M=100) 124.7 128.4 126.8 117.5 133.4 106.9 122.4 127.1 85.9 94.9 89.8 102.2 103.1 104.5 84.0 91.5 91.9 121.1 104.9 100.6 101.2 124.9 104.0 130.2 126.3 110.0 104.8 104.3 127.1 100.0 92.4 92.5 108.2 85.1 96.1 132.7 129.6 126.9 127.2 129.6 126.6 83.7 108.4 99.3 80.1 102.6 95.3 93.1 97.8 105.4 105.1 83.7 92.7 95.2 95.2 87.5 83.7 107.5

Politi- HIV cal rate liberty 20011999 2003

119.4 115.6 115.6 115.6 96.6 85.2 111.8 92.8 89.0 115.6 85.2 119.4 115.6 119.4 104.2 89.0 119.4 100.4 104.2 92.8 119.4 119.4 119.4 81.4 92.8 108.0 115.6 119.4 81.4 111.8 92.8 73.8 % 0.09 0.08 0.41 0.04 0.04 0.13 0.04 0.10 0.06 0.46 0.09 0.09 0.18 0.04 0.35 0.13 0.26 0.18 0.08 0.18 0.09 0.09 0.09 0.40 0.07 0.08 0.38 0.05 0.82 0.05

Ireland Israel Italy Japan Jordan Kazakhstan Korea (South) Kuwait Kyrgyzstan Latvia Lebanon Liechtenstein Lithuania Luxembourg Macau Macedonia Malaysia Malta Mexico Moldova Montenegro Morocco Netherlands New Zealand Norway Oman Palestine Peru Philippines Poland Portugal Qatar Romania Russia Saudi Arabia

99.92 92.57 96.57 104.55 86.08 101.93 106.37 75.72 69.93 97.47 83.61 100.93 96.96 98.31 101.11 84.58 95.54 92.41 85.37 92.29 84.22 71.02 101.89 100.11 95.80 80.64 79.96 74.03 73.55 96.95 92.12 72.11 89.00 97.27 74.40

IQ-scale (M=100) 78.55 119.95 117.75 64.65 117.52 117.00 74.09 117.45 125.22 82.85 124.30 117.25 61.37 108.47 113.35 79.52 122.11 113.00 86.11 125.25 118.11 53.10 97.77 105.00 45.78 93.36 119.00 77.07 116.96 121.73 61.50 105.99 119.22 78.79 121.21 123.00 76.70 116.41 123.00 76.19 118.84 127.00 84.43 117.94 119.00 60.08 107.12 122.33 74.74 115.92 119.67 63.40 116.70 126.14 64.97 105.47 121.22 70.06 112.71 120.43 63.95 104.36 119.00 47.48 95.36 118.65 82.74 119.96 124.82 75.94 122.65 114.40 73.73 115.83 112.47 55.43 104.12 114.00 52.23 106.04 122.33 49.77 97.00 118.27 46.61 101.02 121.57 74.99 117.89 121.31 70.89 112.14 121.53 49.37 96.20 97.67 65.77 110.77 117.00 75.66 118.09 116.18 53.11 95.40 90.00

ppp $ 37738 20033 27119 27967 4320 6671 17971 18047 1751 10270 5074 11702 62298 6794 9512 17633 9168 1510 4004 29371 22582 37670 13584 5260 4321 11379 18126 19844 7277 9230 13226

0-1 .946 .915 .934 .943 .753 .761 .901 .844 .702 .836 .759 .852 .949 .797 .796 .867 .814 .671 .631 .943 .933 .963 .781 .729 .762 .758 .858 .904 .849 .792 .795 .772

per mill 133.03 101.43 147.58 170.02 201.68 53.02 284.74 501.64 6.08 3.93 43.78 43.95 0.27 0.34 10.25 9.50 66.28 48.82 30.77 26.44 20.50 1489.94 87.96 13.82 19.30 21.82 21.19 24.94 46.22 70.59 249.93 20.83 32.43 26.39 27.86 6.88 9.10 94.30 74.07 432.51 556.67 264.38 260.45 7.17 7.53 5.98 7.08 45.51 32.25 62.28 20.47 39.63 24.90 137.30 70.60 3.24 4.21

per mill 1871 62 33 1325 15 6309 38 106 26 2636 39 4 703 24 1189 15 21 87 67 213 33 1539 9 27 2656 44 1778 11 3678 24 625 10 157 56 1299 12 1185 11 1382 7 3520 19 29

Smart fraction 13

14

Table 1. (continued)
Country

CAmean

CA5%

CA95%

CApoliticians 60-09

CApoliticians 90-09
119.00 125.67 121.00 110.40 119.00 116.60 119.00 121.00 125.67 118.11 122.33 100.00 124.00 124.00 91.67 116.00 121.50 124.00 100.00 122.27 118.51 7.50

GDP 2003

HDI 2003

Patent rate 19602007

Patent rate 19912007

Nobel Peace 19012004

0 0 0 1.24 0 4.99 4.93 0 0 0 0 0 0 0 0 1.63 0.90 0 0 0 0.53 1.31

Nobel Literat. 19012004

per 10 mill 0 0 0 0.83 1.44 7.48 3.28 0 0 0 10.03 0 0 0 0 1.45 0.48 0 0 0.48 1.37 5.40

Nobel Science 19012004

0 0 0 0.10 0.07 4.99 6.16 0 0 0 0 0 0 0 0 3.08 2.03 0 0 0 0.45 1.10

Scientists 198595

High technology 1997

Gov. eff. 19962005

135.7 107.2 112.9 108.8 124.9 130.1 135.3 85.8 118.9 104.9 107.3 109.9 100.1 89.3 110.3 131.6 127.3 108.7 88.4 91.1 109.4 14.4

Demo- Rule cracy of law 1950- 19702004 2000

IQ-scale (M=100) 95.4 125.8 124.7 108.7 127.4 113.0 103.7 116.7 106.4 114.0 129.9 121.1 130.0 134.9 86.0 83.7 91.6 116.0 93.7 104.5 113.9 102.4 84.7 97.8 112.1 96.1 118.6 89.9 84.3 91.5 126.7 126.6 124.4 134.1 115.1 102.4 85.4 88.9 106.9 16.6 107.0 14.5

Political liberty 1999

89.0 111.8 115.6 115.6 115.6 119.4 119.4 73.8 111.8 108.0 115.6 85.2 92.8 100.4 85.2 115.6 119.4 115.6 85.2 89.0 105.0 13.8

HIV rate 20012003

% 0.18 0.04 0.05 18.64 0.53 0.09 0.37 0.04 1.07 2.17 0.04 0.05 0.68 0.13 0.40 0.20 0.09 0.11 0.91 4.24

H. Rindermann et al.

Singapore Slovakia Slovenia South Africa Spain Sweden Switzerland Syria Taiwan. RoCh Thailand Trinidad/Tobago Tunisia Turkey Ukraine United A. Emirates United Kingdom United States Uruguay Yemen Yugoslavia/Serbia M SD

104.56 97.59 98.57 63.26 95.65 100.14 99.83 80.57 102.93 90.11 84.55 80.81 87.06 92.99 91.91 100.00 98.41 87.99 63.52 90.20 89.83 (453) 10.95 (73)

IQ-scale (M=100) 78.86 127.22 119.00 75.61 117.83 125.67 78.13 118.27 121.00 35.69 100.06 114.17 75.36 115.19 122.43 79.21 119.98 117.73 77.25 120.07 122.33 64.91 104.84 120.71 80.92 122.57 121.33 71.12 109.99 117.12 57.61 110.05 121.50 60.33 100.63 109.50 65.69 110.17 120.30 70.91 113.33 124.00 67.76 115.05 92.50 76.14 121.92 115.10 74.90 120.30 119.36 61.08 112.19 118.87 43.27 84.10 100.00 67.81 111.03 117.58 67.53 111.21 117.72 (304) (596) 12.00 9.54 7.14 (80) (64)

ppp $ 24481 13494 19150 10346 22391 26750 30552 3576 7595 10766 7161 6772 5491 22420 27147 37562 8280 889 15331 11399

0-1 .907 .849 .904 .658 .928 .949 .947 .721 .778 .801 .753 .750 .766 .849 .939 .944 .840 .489 .824 .104

per mill 433.42 910.35 327.11 58.58 87.20 78.02 115.57 98.04 25.06 327.75 125.43 559.78 72.71 4.56 3.37 5.45 46.10 26.89 10.57 7.85 5.25 8.27 51.74 63.53 239.84 72.43 168.81 272.74 27.96 13.48 0.28 30.04 10.74 113.11 197.59 83.24 141.55

per mill 2728 71 1821 15 2544 16 938 1210 17 3714 34 28 1 119 43 388 11 261 9 3173 2417 41 3732 44 688 8 0 1667 1301 24.08 15.46

Note: As country names the normally used names; all standardized and unstandardized scales are somewhat arbitrary (e.g. GDP and inflation, IQ and reference groups); CA-mean: cognitive ability mean normed according to UK (Greenwich-norm) M=100, as SD used students assessment results (SD=100) transformed into IQ-scale (SD=15), means calculated across countries, for student assessment results are also reported original mean in SAS-scale (M=500, SD=100); CA-5%: cognitive ability mean at 5th percentile; CA-95%: cognitive ability mean at 95th percentile; CA-politicians 60-09: estimated cognitive ability of leading politicians (1960-2009) of a country according to their formal education; CA-politicians 90-09: estimated cognitive ability of leading politicians (1990-2009); GDP 2003: GDP per capita in purchasing power parity (ppp) US Dollars; HDI 2003: Human Development Index 2003, scale 0 (low) to 1 (high); Patent rate 1960-2007: average annual patents per 1 million people 1960-2007; Patent rate 1991-2007: average annual patents per 1 million people 1991-2007; Nobel Peace, Literature and Science 19012004: sum of Nobel Prizes in Peace, Literature and Science per 10 million people 1901-2004; Scientists 1985-95: scientists and engineers in research and development per million people, 1985-1995; High technology 1997: high-technology exports as percentage of manufacturing exports, 1997; Gov. eff. 1996-2005, Democracy 1950-2004, Rule of law 1970-2000 and Political liberty 1999: government effectiveness 1996-2005, democracy value across 1950-2004, rule of law value across 1970-2000 and political liberty in 1999 in IQ-scale with M=100 and SD=15 (in world samples, here positively selected by having data in student assessment results); HIV rate 2001-2003: HIV rate 2001-2003 of adults in percent.

Smart fraction

15

Some astonishing results are observable like the high level of Kazakhstan (6., IQ 122) and the comparatively low for Israel (31., IQ 118, mean 93). For Kazakhstan we have only results from TIMSS 2007 (4th grade); Mullis et al. (2008, p. 34) describe sample anomalies, a correction would be necessary. Israel has participated in several studies, compared to older studies and Jews in the Western World the results are deteriorating (e.g. Lynn & Longley, 2006). Most probably multiple reasons are responsible and not only the 20% fraction of Arabs (a thorough analysis would be necessary). There are also characteristic differences between mean, upper and lower levels. For instance between Canada and USA there is no difference in the upper level (IQ 120 and 120), but in the lower level (IQ 80 and 75). The past history of slavery and a different immigration policy (or different success of migration policies and geographical distance to societies with lower mean abilities) may be reflected into this difference. A similar pattern could be found for Finland and Germany: The difference in the upper level is only 1.20 IQ-points (IQ 121 and 120), but at the lower level 9.60 IQ-points (IQ 85 and 76). Most likely different immigration histories are reflected here, furthermore differences in educational policy (age of tracking, in Germany between age 10 and 12, in Finland at age 16). Early tracking increases ability variance. Using regression analysis (as predictors mean and lower level) the largest residual (standing for difference between upper level and the rest) is found in South Africa (with its heterogeneous population of European, Asian and African descent), inverted the largest residual (standing for difference between lower level and the rest) is found in Belgium (probably a result of immigration and educational policy). The highest cognitive ability values (measured by education and educational degrees) of leading politicians are found in 1. Austria (IQ 128), 2. Luxembourg (IQ 127) and 3. Malta (IQ 126), the lowest in Arab countries: 82. Tunisia (IQ 110), 83. Bahrain (IQ 107), 84. Algeria (IQ 107), 85. Kuwait (IQ 105), 86. Yemen (IQ 100), 87. Qatar (IQ 98), 88. Emirates (IQ 93), 89. Egypt (IQ 90) and 90. Saudi-Arabia (IQ 90). In all these countries the leaders are not elected by their population. Austria has the highest statesman cognitive ability level because up to the early 70s it was common (and almost universal) to complete studies with a doctorate (as it is still the case with medicine in Germany and Austria). Cross-sectional analyses The upper and lower levels are highly correlated with the means (r=.97), among each other with r=.90 (see Table 2). The correlation of the 95th percentile with cognitive ability of political leaders is slightly higher (r=.364 and .365) than of the average competence values (r=.361 and .354) or of the 5th percentile (r=.345 and .324), indicating that the politicians ability depend more on the smart fraction value. Political leaders most likely stem from this fraction (and considering the complexity of political tasks this is a reasonable result; s. Suedfeld et al., 2003). A regression analysis (criterion: IQ of politicians) increases this pattern (for politicians 1960-2009, average ability aAPIQ=.13, 95%PIQ=.24; for politicians 1990-2009, same time ability measures aAPIQ=.00, 95%PIQ=.37, N=90).
Table 2. Correlations within cognitive ability indicators CA-mean CA-mean (average) CA-5% CA-95% CA-politicians 60-09 CA-politicians 90-09 N Note: See Table 1. .97 .97 .36 .35 90 CA-5% CA-95% CApoliticians 60-09 CApoliticians 90-09

.90 .35 .32 90

.36 .37 90

.93 90

90

16

Table 3: Correlations between cognitive ability indicators and indicators of societal development GDP 2003 HDI 2003 Patent Patent rate rate 1960- 19912007 2007 .40 .45 .35 .42 .22 .13 81 .37 .50 .03 .04 76 Nobel Peace 19012004 .21 .15 .23 .02 .01 87 Nobel Literat. 19012004 .13 .13 .13 .06 .04 87 Nobel Science 19012004 .34 .31 .34 .10 .07 87 Scientists 198595 .61 .57 .64 .10 .16 51 High Gov. Demo Demotechno- eff. cracy cracy logy 1996- 1950- 19961997 2005 2004 2000 .38 .61 .60 .62 .35 .40 .15 .12 61 .55 .63 .19 .12 88 .55 .62 .52 .45 84 .55 .65 .58 .57 84 Rule of law 19702000 .62 .56 .66 .22 .13 76 Rule of law 2000 .59 .53 .63 .12 .02 73 Political liberty 1999 .49 .42 .53 .55 .52 86 HIV rate 20012003 -.30 -.31 -.24 -.05 -.10 82 AIDS 19952000 -.21 -.24 -.20 .08 .05 83

H. Rindermann et al.

CA-mean (average) CA-5% CA-95% CA-politicians 60-09 CA-politicians 90-09 N

.61 .56 .61 .13 .08 84

.79 .74 .78 .27 .25 85

Note: See Table 1; AIDS 1995-2000: percentage of people suffering on AIDS 1995-2000.

Table 4: Regression analyses, predictors: cognitive ability indicators, criteria: indicators of the development of societies GDP 2003 HDI 2003 Patent rate 19602007 -.13 Patent rate 19912007 (-.75) Nobel Peace 19012004 -.35 Nobel Literat. 19012004 .09 Nobel Science 19012004 .13 Scientists 198595 -.12 High Gov. Demo techno- eff. cracy logy 1996- 19501997 2005 2004 -.29 -.12 .02 Democracy 19962000 -.13 Rule of law 19702000 -.30 Rule of law 2000 -.35 Political liberty 1999 -.41 HIV rate 20012003 .03 -.34 .40 84 .33 85 .55 81 (1.23) 76 .57 87 .04 87 .22 87 .76 51 .68 61 .75 88 .59 84 .54 .27 83 .77 84 .56 .32 83 .95 76 .61 -.05 73 .97 73 .58 -.22 71 .93 86 .47 .32 84 82 83 82 83 AIDS 19952000 .15 -.36

CA-mean (average) CA-5% CA-95% N Educational level CA-politicians N

.22

.47

Note: See Table 1 and 3. For educational level is used for long term criteria (democracy 1950-2004, rule of law 1970-2000) a general long-term educational index, for more recent criteria (democracy 1996-2004, rule of law 2000 and political liberty1999) a current education index from HDR, the same was done with cognitive ability estimations of politicians.

Smart fraction

17

The smart fraction (or gifted or high ability fraction) is not only more relevant than the average cognitive ability level or the non-smart fractions level for the cognitive competence levels of politicians, but also for wealth (GDP; see Tables 3 and 4). For the general human development level of a society, the 95th is less important than the average, but more than the 5th level. Even stronger are the effects on all indicators of high intellectual achievement from patent rates, over Nobel Prizes to the rates of scientists and high technology exports. Only for Nobel Prizes in Literature the impact is lower, presumably because of the low reliability and validity of this judgement process (if it should rate literature quality) and because of political considerations in giving this award (sometimes it seems to be an award given to nations not receiving science Nobel Prizes, aiming at equal distribution across different cultures and continents). Also impressive is the high impact of the smart fraction (exactly: variance in indicators between different countries statistically explained by differences in cognitive levels of the 95th percentile groups) on all political variables: on government effectiveness, democracy, rule of law and political liberty. The political development of nations seems to depend heavily on the cognitive ability of their smart fraction. On the other hand the development in indicators standing for undesired outcomes (HIV and AIDS) depends on the cognitive level of the lower fraction. This is a strong support for all studies demonstrating that health and especially HIV depends on cognitive ability, leading to more or less risky behavior resulting in the long run in health or death (e.g. Goldman & Smith, 2002; Gottfredson, 2004; Rindermann & Meisenberg, 2009). Additionally, the opposite effects of the high ability fraction on intellectual outcomes and of the low ability fraction and their abilities on anti-intellectual outcomes strongly support the theoretical framework: Cognitive ability specifically matters through a cognitive effect mechanism. Finally, similar results for crime (homicide) substantiate the pattern: Homicide is higher (negatively) correlated with the ability level of the 5th percentile than with the average or even with the ability of the 95th percentile (5th r=-.34, average r=-.33, 95th r=-.23; N=80). The regression analysis strengthens the result: 5%Hi=-.35, aAHi=-.01, N=80). Because of this opposite effect pattern an objection, that the results are a consequence of different reliability (high for 95th percentile ability, mean for the average value, low for 5th percentile), could be ruled out. The results for political leaders and their ability levels across the countries are less impressive: Correlations are always low except for democracy and political freedom. For both variables political leaders competence could explain the variance beneath the general educational level of society (democracy: EduDem=.54, PIQDem=.27, N=83; political freedom: EduPoF=.47, PIQPoF=.32, N=84).

.38

.60

.45

.86

Patent rate 1960-2007

Nobel Prize in Science rate 1901-2004

.78 (.78)

Scientist rate 1985-1995

.74 (.74)

High tech export rate 1997

.37 (.37)

.64 (.64)

Cognitive ability of 95%

1 (1) .43 (.75)

GDP 1998

.01

Intellectual excellence (STEM) Cognitive ability mean

.32 (.74)

.94 (.94)

Wealth

.98 (.98) .44 .12

GDP 2003

.04

Figure 3. Standardized path coefficients (and correlations in parentheses) between cognitive ability of nations, high intellectual achievement and wealth (error term as unexplained variance), N=48 nations

18

H. Rindermann et al.

For a path analysis four indicators of intellectual excellence in STEM subjects were used (patent rate, Nobel Prizes in science rate, scientist rate and high technology export rate), wealth was measured by GDP 1998 and 2003 (latent variables in circles, measured/manifest in boxes; see Figure 3). Only countries with data in all measures were used (N=48), the pattern of results is similar to the larger samples (see Tables 3 and 4). The fit indices are good (SRMR=.04, CFI=.99). For high STEM achievement the cognitive competence of the smart fraction rate is more important (95%IE=.43, aAIE=.32). The impact of cognitive ability on wealth goes completely through intellectual excellence (IEGDP=.94), a further direct effect would be small and even negative (aAGDP=-.05 and 95%GDP=-.16), indicating, that cognitive ability not going through STEM-like achievement has no positive impact on wealth (similar: Schofer, Ramirez & Meyer, 2000). Longitudinal analyses Student assessment studies with information on the tails do not allow longitudinal analyses, the time interval between the 90s and the first decade of 21st century is too short. But we can do these analyses with political leaders cognitive ability from the 60s to the 00s. In the Tables 3 and 4 high correlations and a cross-sectional impact of politicians ability on democracy were observable. Maybe this relationship does not include an effect of ability on democracy, but of democracy on politicians ability. This could be tested by a cross-lagged panel analysis.

Cognitive ability of political leaders 1960-64

Cognitive ability of political leaders 2000-09

Democracy 1960-64

Democracy 1996-2000

Figure 4. Cross-lagged coefficients (and correlations in parentheses) between cognitive ability of political leaders and democracy (error term as unexplained variance, correlated error), N=64 nations

Cognitive ability of political leaders 1960-69

Cognitive ability of political leaders 2000-09

Cognitive ability mean

1964-1972

Cognitive ability mean

1995-2007

Figure 5. Cross-lagged coefficients (and correlations in parentheses) between cognitive ability of political leaders and society (error term as unexplained variance), N=17 nations

Smart fraction

19

The relation between politicians ability and democracy could be analyzed between the early 60s and the 00s (see Figure 4; the fit of the model is perfect). The years for political leaders are somewhat misleading because the majority of leaders in 1960-64 had ruled the country also for several years before, the same is true for 2000-09. The cognitive ability of politicians has only a small impact on the development of democracy (PIQ1Dem2=.04), but democracy leads to more educated and competent politicians (Dem1PIQ2=.18). The low competence levels of Arab leaders described above could at least be partly explained by their nondemocratic (self-)appointment (and only under rare circumstances non-democratic leaders like Mohammed VI in Morocco and bin Said in Oman could push their societies forward). But politicians ability has a positive impact on cognitive development of a society (PIQ1CA2=.21; see Figure 5; SRMR=.01, CFI=1.00). The reverse effect is smaller (CA1PIQ2=.03). Unfortunately the sample with 17 countries is small, but after wealth and rule of law it is the third factor that longitudinally seems to influence cognitive development at the country data level. In the long run smart politicians increase their nations abilities. Cognitive ability and educational policy Large class sizes affect low ability fractions stronger than high ability fractions, the opposite is true for early tracking, which shows larger benefits for high ability fractions and smaller for low ability fractions (see Table 5). Both results are in accordance with assumptions and studies at the national and within-national data level (Rindermann & Ceci, 2009): Less able students need smaller classes (Krueger, 1999). Early tracking is more beneficial for highly capable students because they will be together with comparable high able students leading to a more stimulating instruction, interaction and school climate, the ability of others nurtures ones own ability (Rindermann & Heller, 2005), but less able students lose their possible models. Both patterns of results strongly support former analyses at the country data level. Rather astonishing is the high correlation between early tracking and cognitive ability of political leaders (r=.41-.44, N=66; compare them with correlations in Table 3). Maybe in educational systems with early tracking the political elite stems from the cognitive development stimulating higher tracks and higher education is seen as something prestigious and valuable for a political career within parties, society and voters (furthering and gate keeper effect of early tracking). The other way around it may be that more intelligent and competent politicians are in favor of tracking systems (which seem to benefit cognitive development). Future longitudinal analyses should prove this.

Table 5: Correlations between cognitive ability indicators and attributes of educational policy

Young tracking age CA-mean CA-5% CA-95% CA-politicians 60-09 CA-politicians 90-09 N .31 .25 .33 .41 .44 66

Class size -.45 -.47 -.38 -.19 -.22 74

Note: See Table 1; Young tracking age: tracking age inverted; Class size: large class size and pupilteacher-ratio.

20

H. Rindermann et al.

Discussion
As other studies before have shown, there are large cognitive ability differences between societies. They are reflected in the upper and lower levels, but East Asian and Commonwealth countries seem to have comparatively stronger smart fractions than Scandinavian or other First-World-countries. Maybe these countries not only benefit from their educational policies and, especially in East-Asia, from harsh educational efforts with very extensive studying times per day, week, year and youth, but also from a more successful migration policy. Beneath cultural background factors relevant for education in family and school genetic factors could be important, but genes for cognitive ability are not known up to now. The ability value (intelligence, knowledge and the intelligent use of knowledge) of the smart fraction (95th percentile, comparable to an IQ 125 or higher in within-country norms) is more important for country differences in wealth, nations intellectual excellence (in STEM fields: patents, Nobel Prices in science, scientists, high technology exports) and political attributes of societies (government effectiveness, democracy, rule of law and political liberty) than the average ability or the ability level of a non-smart fraction (5th percentile, comparable to an IQ 75 or lower in within-country norms). But the cognitive ability level of the non-smart fraction is more important for country differences in HIV AIDS and homicide. Wealth differences , between countries could be completely explained through differences in high intellectual achievement in STEM fields, which itself largely depends on differences in smart fractions ability. The smart fraction is essentially relevant for beneficial societal development. The cognitive ability of political leaders is far less important. We could only find higher correlations to democracy and political liberty, in a longitudinal analysis democracy has a positive impact on cognitive ability of political leaders. People, if they have the chance to elect their leaders, prefer more educated ones. Political leaders have, in the long run, a positive influence on countries cognitive ability, presumedly by creating better educational and social environments increasing cognitive ability. This study could show how in former studies discovered mean cognitive ability effects on growth could work: In societies with a higher cognitive average the smart fraction reaches a higher cognitive level (resp. is also larger from a real threshold of high ability on see below). This smart fraction pushes growth through excellence in areas relevant for economic affluence, like in technology and science. We did not expect such a high impact of the smart fraction on the destiny of societies. The current data do not allow us to present a historical analysis of modernization processes from the ancient past up to 1960. We can only use our findings on present day comparisons between countries in terms of cognitive ability as an analogy to surmise that the same cognitive effects occurred during historical periods. It seems very likely that the achievements of cognitively eminent persons coming from the smart fraction, and stimulated by their peers, was decisive for the betterment of their societies. But the results are somewhat contradictory: On one hand, international differences in important attributes of societies (and, we would argue, in the historical development of nations) depend on the cognitive ability of an elite. On the other hand these differences do not depend on the cognitive ability levels of politicians. Peoples and nations themselves seem to be important for the destiny of their countries and societies, more so than politicians. Leaders stem by a majority from smart fractions of their societies and they seem to reflect social attributes and especially their cognitive ability levels, rather than influencing them. One possibility is that in developed societies with a large smart fraction, political leadership is drawn from somewhat less intelligent elite members (see Cox, 1926, p. 84), willing to earn less intellectual merits or money in return for the possibility of fame. Another possibility is that the cognitive ability of politicians is less important because they have competent consultants and experts who give advice (the first author discussed this in 2006 at an ISIRmeeting in San Francisco with Dean Keith Simonton, his reply was, that politicians also need to understand them). Maybe formal education is not the best estimate, better would be real life criteria like used by Simonton. But we should not use verbal ability here. Rhetorical brilliance

Smart fraction

21

is not solving problems. Verbal fluency could be a dangerous competence, a seduction to talk people into believing or doing something that they would not do by the use of thinking and rationality. In the Greek tradition, Plato (Gorgias) similarly assessed rhetorical competence in a very critical manner. Finally, in German analyses political and weltanschauliche orientations are showing more explanative strength than education of politicians. One remark on Singapore: Its long-term Prime Minister Lee Kuan Yew has in our data set only as highest degree university degree, no doctorate or an additional scientific degree. But he has studied at London School of Economics and in Cambridge and finished his studies with exceptional Double Starred First Class Honours. Our assessment procedure seems to underestimate his cognitive ability level. Furthermore, Singapore has reached the highest rank in the smart fraction ability (IQ 127) in our list and the second highest rank in average ability (IQ 105), but only the 14th rank in the lower non-smart fraction ability ranking (IQ 79). In patents (1991-2007) Singapore has the first place. And, that is especially remarkable, Singapore has reached the first place in government effectiveness. Singapore seems to have the best government in the world. Lee Kuan Yews ability indicated by his success seems to be underestimated again by the solely use of formal education. Of course, Singapore was and still is no standard-bearer of liberty and democracy and Lee Kuan Yew has attracted criticism because of this. But, he stands apart from other leaders in terms of his exceptional success for Singapore in growth, modernization, technology and since several years also in science (up to now only STEM including biotechnology). Lee is also apparently the only politician who has read and used the results of intelligence research in his politics. In speeches he has cited Thomas Bouchard and Richard Lynn (Chan & Chee, 1984), and he is the only statesman, who has seen that intelligence enhancement not only needs an improvement in the environment (like in educational policy) but also in demographic policies, because parents transfer cognitive ability to their children by creating a stimulating environment (especially by education and modeling) and by transmission of their genes. Further research should not only use as an indicator for smart fractions a value like the 95th percentile and the ability at this level, but also a defined threshold of cognitive ability like IQ 130 or SAS 700 and the percentage of population above this threshold. The present editing of TIMSS, PISA and PIRLS results complicates this. Research using different ability levels seems to be a fruitful approach. Also further indicators of high achievement like Fields Medal or of low achievement like traffic accidents should be used as indicators to stand for high or low intelligence (Dagona, 1994; OToole & Stankov, 1992). Migrations and their impact on countries cognitive ability and development should be analyzed (so Singapore benefits from a smart fraction import, similarly to Switzerland, Australia and Canada, but Eastern Europe and less so Germany suffer from a brain flight, others from a low brain immigration; e.g. Levels, Dronkers & Kraaykamp, 2008; te Nijenhuis, de Jong, Evers & van der Flier, 2004). The possibilities of education for cognitive enhancement are still not sufficiently explored (e.g. Heckman, 2000; Nisbett, 2009; Rindermann & Ceci, 2009). Our results emphasize the importance of nurturing the highly gifted. Their support will be beneficial in the long run not only for themselves but even more through general effects on societal development including wealth, health, politics, science, ethics and culture for the less smart and non-smart fractions. Finally, at the level of societies, there is an integrative theoretical framework on causes and mechanisms of cognitive competence still missing. By showing the influence of ability on wealth through high intellectual STEM achievement we hope to provide some further building block in development of such a theory.

22

H. Rindermann et al.

References
Bacharach, V R., & Baumeister, A. A. (1998). Direct and . indirect effects of maternal intelligence, maternal age, income, and home environment on intelligence of preterm, low-birth-weight children. Journal of Applied Developmental Psychology, 19, 349-363. Beaton, A. E., Mullis, I. V S., Martin, M. O., Gonzalez, E. . J., Kelly, D. L., & Smith, T. (1996). Mathematics achievement in the middle school years. Chestnut Hill: TIMSS Study Center. Beaton, A. E., Martin, M. O., Mullis, I. V S., Gonzalez, E. . J., Smith, T., & Kelly, D. L. (1996). Science achievement in the middle school years. Chestnut Hill: TIMSS Study Center. Becker, G. S. (1993/1964). Human capital: A theoretical and empirical analysis with special reference to education. Chicago: University of Chicago Press. Cattell, R. B. (1987/1971). Intelligence: Its structure, growth and action. Amsterdam: Elsevier. Ceci, S. J. (1991). How much does schooling influence general intelligence and its cognitive components? Developmental Psychology, 27, 703-722. Chan, Ch. K. & Chee, H. L. (1984). Singapore 1984: Breeding for Big Brother. In Ch. K. Chan & H. L. Chee (Eds.), Designer genes: IQ, ideology and biology (pp. 4-13). Kuala Lumpur: Selangor/Institute for Social Analysis (INSAN). Chomsky, N. (2009). Politics, society, and Obama: Change we can believe in? The Owl, 15, 16-21. Cohen, J. (1994). The earth is round (p<.05). American Psychologist, 49, 997-1003. Cox, C. M. (1926). The early mental traits of three hundred geniuses. Stanford: Stanford University Press. Cunha, F., Heckman, J. J., Lochner, L., & Masterov, D. V . (2006). Interpreting the evidence on life cycle skill formation. In E. A. Hanushek, & F. Welch (Eds.), Handbook of the economics of education (I, pp. 697-812). Amsterdam: North-Holland. Dagona, Z. K. (1994). Substance use and road traffic accidents among Nigerian commercial motor cyclists. Ife PsychologIA, 2, 81-93. Deary, I. J., Batty, G. D., & Gale, C. R. (2008). Childhood intelligence predicts voter turnout, voting preferences, and political involvement in adulthood. Intelligence, 36, 548-555. Deary, I. J., Ferguson, K. J., Bastin, M. E., Barrow, G. W. S., Reid, L. M., Seckl, J. R., Wardlaw, J. M., & Maclullich, A. M. J. (2007). Skull size and intelligence, and King Robert Bruces IQ. Intelligence, 35, 519-525. Denny, K., & Doyle, O. (2008). Political interest, cognitive ability and personality: Determinants of voter turnout in Britain. British Journal of Political Science, 38, 291-310. Ellis, L., & Walsh, A. (2003). Crime, delinquency and intelligence: A review of the worldwide literature. In H. Nyborg (Ed.), The scientific study of general intelligence (pp. 343-365). Oxford: Pergamon. Falk, R., & Greenbaum, Ch. W. (1995). Significance tests die hard. The amazing persistence of a probabilistic misconception. Theory & Psychology, 5, 75-98. Freedom House (2004). Freedom in the world. Washington: www.freedomhouse.org. Gelade, G. A. (2008). IQ, cultural values, and the technological achievement of nations. Intelligence, 36, 711-718. Gigerenzer, G. (2004). Mindless statistics. Journal of Socio-Economics, 33, 587-606. Goldman, D. P., & Smith, J. P. (2002). Can patient selfmanagement help explain the SES health gradient? Proceeding of the National Academy of Sciences, 99, 10929-10934. Gottfredson, L. S. (2004). Life, death, and intelligence. Journal of Cognitive Education and Psychology, 4, 23-46. Gottfredson, L. S. (2005). Implications of cognitive differences for schooling within diverse societies. In C. L. Frisby & C. R. Reynolds (Eds.), Comprehensive Handbook of Multicultural School Psychology (pp. 517-554). New York: Wiley. Gwartney, J., & Lawson, R. (2003). Economic freedom of the world. 2003 annual report. Vancouver: Fraser Institute. Habermas, J. (1985/1981). The theory of communicative action. Boston: Beacon. Hanushek, E. A., & Kimko, D. D. (2000). Schooling, labor-force quality, and the growth of nations. American Economic Review, 90, 1184-1208. Hart, M. H. (2007). Understanding human history. Augusta: Washington Summit. Heckman, J. J. (2000). Policies to foster human capital. Research in Economics, 54, 3-56. Herrnstein, R. J., & Murray, Ch. (1994). The bell curve. Intelligence and class structure in American life. New York: Free Press. Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure modeling. Psychological Methods, 3, 424-453. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis. Structural Equation Modeling, 6, 1-55. Human Development Report of the United Nations (HDR) (2005). Human Development Report 2005. New York. Retrieved May 25, 2007 from http://hdr.undp.org/reports/global/2005/pdf/h dr05_HDI.pdf. Hunter, J. E. (1997). Needed: A ban on the significance test. Psychological Science, 8, 3-7. Interpol (2004). International Crime Statistics. Lyon. Retrieved October 22, 2004 from www.interpol.int/Public/Statistics/ICS/ downloadList.asp.

Smart fraction

23

Irwing, P., & Lynn, R. (2006). The relation between childhood IQ and income in middle age. Journal of Social, Political and Economic Studies, 31, 191196. Jensen, A. R. (1998). The g factor. The science of mental ability. Westport: Praeger. Jones, G., & Schneider, W. J. (2006). Intelligence, human capital, and economic growth. Journal of Economic Growth, 11, 71-93. Kaufmann, D., Kraay, A., & Mastruzzi, M. (2006). Governance matters V: Aggregate and individual governance indicators for 1996-2005. Worldbank. Retrieved November 28, 2006 from www.worldbank.org/wbi/governance. Knack, S., & Keefer, Ph. (1995). Institutions and economic performance. Economics and Politics, 7, 207-227. Kohlberg, L. (1987). Child psychology and childhood education. A cognitive-developmental view. New York: Longman. Krueger, A. B. (1999). Experimental estimates of education production functions. Quarterly Journal of Economics, 114, 497-532. Kurian, G. Th. (2001). The illustrated book of world rankings. Armonk: Sharpe. La Griffe du Lion (2004). Smart fraction theory II. 6(2). Retrieved July 13, 2009 from www.lagriffedulion.f2s.com/sft2.htm. Lakhanpal, M., & Ram, R. (2008). Educational attainment and HIV/AIDS prevalence: A crosscountry study. Economics of Education Review, 27, 14-21. Lee, J.-W., & Barro, R. J. (1997). Schooling quality in a cross-section of countries. New York. Retrieved November 18, 2004 from www.nber.org/papers/W6198. Levels, M., Dronkers, J., & Kraaykamp, G. (2008). Immigrant childrens educational achievement in Western countries. American Sociological Review, 73, 835-853. Lynn, R., Harvey, J., & Nyborg, H. (2009). Average intelligence predicts atheism rates across 137 nations. Intelligence, 37, 11-15. Lynn, R., & Longley, D. (2006). On the high intelligence and cognitive achievements of Jews in Britain. Intelligence, 34, 541-547. Lynn, R., & Vanhanen, T. (2002). IQ and the wealth of nations. Westport: Praeger. Lynn, R., & Vanhanen, T. (2006). IQ and global inequality. Athens: Washington Summit. Marshall, M. G., & Jaggers, K. (2000). Polity IV Project. Political regime characteristics and transitions, 1800-1999. University of Maryland: www.bsos.umd.edu/cidcm/inscr/polity. Martin, M. O., Mullis, I. V S., Beaton, A. E., Gonzalez, E. . J., Smith, T., & Kelly, D. L. (1997). Science achievement in the primary school years. Chestnut Hill: TIMSS Study Center. Martin, M. O., Mullis, I. V S., & Foy, P. (2008). TIMSS . 2007 international science report. Chestnut Hill:

TIMSS & PIRLS Study Center. Martin, M. O., Mullis, I. V S., Gonzalez, E. J., & . Chrostowski, S. J. (2004). TIMSS 2003 international science report. Chestnut Hill: TIMSS & PIRLS International Study Center. Martin, M. O., Mullis, I. V S., Gonzalez, E. J., Gregory, K. . D., Smith, T. A., Chrostowski, S. J., Garden, R. A., & OConnor, K. M. (2000). TIMSS 1999 international science report. Chestnut Hill: TIMSS Study Center. Meisenberg, G. (2004). Talent, character, and the dimensions of national culture. Mankind Quarterly, 45, 123-168. Mullis, I. V S., Martin, M. O., Beaton, A. E., Gonzalez, E. . J., Kelly, D. L., & Smith, T. (1997). Mathematics achievement in the primary school years. Chestnut Hill: TIMSS Study Center. Mullis, I. V S., Martin, M. O., & Foy, P. (2008). TIMSS . 2007 international mathematics report. Chestnut Hill: TIMSS & PIRLS Study Center. Mullis, I. V S., Martin, M. O., Gonzalez, E. J., Gregory, K. . D., Garden, R. A., OConnor, K. M., Chrostowski, S. J., & Smith, T. A. (2000). TIMSS 1999 international mathematics report. Chestnut Hill: TIMSS Study Center. Mullis, I. V S., Martin, M. O., Gonzalez, E. J., & . Chrostowski, S. J. (2004). TIMSS 2003 international mathematics report. Chestnut Hill: TIMSS & PIRLS International Study Center. Mullis, I., Martin, M. O., Gonzales, E. J., & Kennedy, A. M. (2003). PIRLS 2001 international report. Chestnut Hill: IEA. Mullis, I., Martin, M. O., Kennedy, A. M., & Foy, P. (2007). PIRLS 2006 international report. Chestnut Hill: IEA/TIMSS & PIRLS International Study Center. Murray, Ch. (2008). Real education: Four simple truths for bringing Americas schools back to reality. New York: Crown Forum. Murray, Ch., Cardoso, R., & Mendes, D. (2008). Miscigenao diminui o QI dos brasileiros. Isto, 2032(15-10-08), 9-11. Nisbett, R. E. (2009). Intelligence and how to get it. Why schools and cultures count. New York: Norton. Nyborg, H. (2009). The intelligence-religiosity nexus. Intelligence, 37, 81-93. OECD (2000). Literacy in the information age. Final report of the International Adult Literacy Survey. Paris: OECD. OECD (2003). Literacy skills for the world of tomorrow. Further results from PISA 2000. Paris: OECD. OECD (2004a). Learning for tomorrows world. First results from PISA 2003. Paris: OECD. OECD (2004b). Problem solving for tomorrows world. Paris: OECD. OECD (2007a). PISA 2006. Science competencies for tomorrows world. Volume 1: Analysis. Paris: OECD.

24

H. Rindermann et al.

OECD (2007b). PISA 2006. Volume 2: Data/Donnes. Paris: OECD. Oesterdiekhoff, G. W. (2000). Zivilisation Strukturgenese. Frankfurt: Suhrkamp. und

effects of science on national economic development, 1970-1990. American Sociological Review, 65, 866-887. Simonton, D. K. (1984). Genius, creativity and leadership. Cambridge: Harvard University Press. Simonton, D. K. (1985). Intelligence and personal influence in groups: Four nonlinear models. Psychological Review, 92, 532-547. Simonton, D. K. (2006). Presidential IQ, openness, intellectual brilliance, and leadership: Estimates and correlations for 42 US chief executives. Political Psychology, 27, 511-526. Simpson, M. (1997). Informational inequality and democracy in the new world order. In M. Midlarsky (Ed.), Inequality, democracy, and economic development (pp. 156-176). Cambridge: Cambridge University Press. Statistisches Bundesamt (2002). Statistisches Jahrbuch 2002 fr das Ausland. Suttgart: Metzler-Poeschel. Suedfeld, P., Guttieri, K., & Tetlock, Ph. E. (2003). Assessing integrative complexity at a distance. In J. M. Post (Ed.), The Psychological assessment of political leaders (pp. 246-270). Ann Arbor: The University of Michigan Press. te Nijenhuis, J., de Jong, M.-J., Evers, A., & van der Flier, H. (2004). Are cognitive differences between immigrant and majority groups diminishing? European Journal of Personality, 18, 405-434. Thomson, G. H. (1937/1936). Intelligence and civilisation: A Ludwig Mond lecture delivered at the University of Manchester on October 23rd, 1936. Journal of the University of Manchester, 1, 18-38. Reprinted in: Deary, I. J., Lawn, M., Brett, C. E., & Bartholomew, D. (2009). Intelligence and Civilisation. Intelligence, 37, 48-61. UNAIDS/WHO (2003). AIDS epidemic Geneva: WHO/www.unaids.org. update.

Oesterdiekhoff, G. W. (2008). Magic and animism in old religions: The relevance of sun cults in the world-view of traditional societies. Croatian Journal of Ethnology and Folklore Research, 45, 43-66. Oesterdiekhoff, G. W., & Rindermann, H. (2007). The spread of AIDS in developing countries: A psycho-cultural approach. Journal of Social, Political and Economic Studies, 32, 201-222. OToole, B. I., & Stankov, L. (1992). Ultimate validity of psychological tests. Personality and Individual Differences, 13, 699-716. Piaget, J. (1997/1932). Moral judgement of the child. New York: Free Press. Piaget, J. (2001/1947). The psychology of intelligence. London: Routledge. Rindermann, H. (2007a). The g-factor of international cognitive ability comparisons: The homogeneity of results in PISA, TIMSS, PIRLS and IQ-tests across nations. European Journal of Personality, 21, 667-706. Rindermann, H. (2007b). The big G-factor of national cognitive ability. European Journal of Personality, 21, 767-787. Rindermann, H. (2008a). Relevance of education and intelligence at the national level for the economic welfare of people. Intelligence, 36, 127-142. Rindermann, H. (2008b). Relevance of education and intelligence for the political development of nations: Democracy, rule of law and political liberty. Intelligence, 36, 306-322. Rindermann, H., & Ceci, S. J. (in press). Educational policy and country outcomes in international cognitive competence studies. Perspectives on Psychological Science. Rindermann, H., & Heller, K. A. (2005). The benefit of gifted classes and talent schools for developing students competences and enhancing academic self-concept. Zeitschrift fr Pdagogische Psychologie, 19, 133-136. Rindermann, H., & Meisenberg, G. (2009). Relevance of education and intelligence at the national level for health: The case of HIV and AIDS. Intelligence, 37, 383-395. Rindermann, H., & Thompson, J. (2009). Parents education, and not their money, nurtures the intelligence of their children. Manuscript submitted for review. Schermelleh-Engel, K., Moosbrugger, H., & Mller, H. (2003). Evaluating the fit of structural equation models. Methods of Psychological Research, 8, 23-74. Schmidt, F. L., & Hunter, J. E. (2004). General mental ability in the world of work. Journal of Personality and Social Psychology, 86, 162-173. Schofer, E., Ramirez, F. O., & Meyer, J. W. (2000). The

Vanhanen, T. (1997). Prospects of democracy. London: Routledge. Vanhanen, T. (2005). Measures of democracy 18102004. Finnish Social Science Data Archive (FSD). Retrieved January 17, 2006 from www.fsd.uta.fi/ english/data/catalogue/FSD1289. Weede, E. (2006). Economic freedom development. Cato Journal, 26, 511-524. and

Weiss, V (2009). National IQ means transformed from . Programme for International Student Assessment (PISA) scores, and their underlying gene frequencies. Journal of Social, Political and Economic Studies, 34, 71-94. World Intellectual Property Organization (WIPO) (2009). Patent applications by patent office (18832007), by resident and non-resident. Geneva. Retrieved June 22, 2009 from www.wipo.int/ipstats/en/statistics/patents. Wuttke, J. (2007). Uncertainties and bias in PISA. In S. Th. Hopmann, G. Brinek & M. Retzl (Eds.), PISA zufolge PISA - PISA according to PISA (pp. 241263). Wien: LIT.

Smart fraction

25

The authors
Heiner Rindermann (born 1966) is Professor for Developmental Psychology at University of Graz (Austria). He studied psychology and earned his Ph.D. in Heidelberg (1986-1995). 1994-1999 he worked together with Kurt A. Heller (LMU Munich) in two research projects on evaluation of programs for the gifted. 1999-2008 he taught as assistant and associate professor psychological methods (Magdeburg) and educational science (Saarbrucken, Paderborn). His research topics are development, furthering and consequences of intelligence at the individual, school and society level, emotional competence, teaching quality and its evaluation, cross-cultural psychology and educational policy.

Michael Sailer (born 1984) graduated from Secondary School in Leoben and studied psychology in Graz 2003-2009. He is currently working on his thesis about the Impact of politicians and the cognitive ability of smart fractions on social development. Since 2007 he is a member for the KFU-Graz students association.

Dr James Thompson is Honorary Senior Lecturer in Psychology in the Department of Psychology, University College London. Currently he teaches medical students psychology applied to medicine. His clinical work is in the field of post-traumatic stress disorder. His original thesis was on the intelligence and scholastic abilities of people who had sustained localized cortical injuries in childhood. His current interests are on the ways in which general intelligence can be used as an explanatory variable in a range of social behaviors, particularly health behaviors.

26

H. Rindermann et al.

Talent Development & Excellence Vol. 1, No. 1, 2009, 27-43

Modelling mathematical Actiotopes with CLARION

27

Modeling mathematical Actiotopes: The potential role of CLARION

Shane N. Phillipson1,* and Ron Sun2
Abstract: The challenges facing research into giftedness are considerable, including the lack of a clear meta-theoretical framework and agreement on terminology. If giftedness, in broad terms, refers to the interactions between complex patterns of actions, the psychological aspects of self and the social environment toward the attainment of achievement excellence, then it is not surprising that the study of giftedness presents a number of conceptual and practical difficulties. Using mathematical achievement as an example, this article explores the potential benefits and problems of modeling the Actiotope Model of giftedness using the cognitive architecture CLARION. In using simulation studies to explore giftedness, the meta-theoretical framework, including terminologies, is clearly defined, helping to direct future field studies using the more traditional methods of research. Keywords: achievement excellence, Actiotope Model of Giftedness, CLARION, cognitive architecture, mathematics, modeling

Introduction
The current state of research into giftedness has been described by a number of researchers as fragmented and contradictory. The aim of this article is to argue that the Actiotope Model of Giftedness (AMG) (Ziegler, 2005) provides a much needed sense of coherence and unity to this research. The AMG focuses on the development of action repertoires necessary for the attainment of excellence. In broad terms, the development of action repertoires is dependent on physiological, psychological and environmental (or social) variables, and excellence is viewed within social contexts. Although the AMG is difficult to test experimentally using traditional research methods because of its reliance on these different conceptual levels, we also argue that it is possible to simulate the AMG using techniques developed by researchers currently working in the area of cognitive modeling. Using excellence in the domain of mathematics as an example, we describe the CLARION cognitive architecture as one possible way to model the development of higher forms of mathematical thinking from its antecedents of innate number sense. In describing the general principles of both the AMG and CLARION, this preliminary discussion then shows how the two can be mapped onto each other and the immediate challenges in using this approach and thereby providing some suggestions for future research.

Conceptions of giftedness
In reviewing the status of research into giftedness, Eyre (2009) pointed out that, despite the many decades of work, conceptions of giftedness amongst psychologists are fragmented and often contradictory. Recent volumes that specifically address different conceptions of giftedness (Phillipson & McCann, 2007; Shavinina, in press; Sternberg, 2004b; Sternberg & Davidson, 2005), the related topics of intelligence (Sternberg, 2000), creativity (Kaufman & Baer, 2006; Kaufman & Sternberg, 2006) and terms such as high ability, talent and
1

Shane N. Phillipson, Department of Educational Psychology, Counselling and Learning Needs, The Hong Kong Institute of Education, 10 Lo Ping Road, Tai Po, NT Hong Kong * Corresponding author. Email: shane@ied.edu.hk 2 Ron Sun, Cognitive Science Department, Rensselaer Polytechnic Institute, Troy, NY, USA.
ISSN 1869-0459 (print)/ ISSN 1869-2885 (online) 2009 International Research Association for Talent Development and Excellence http://www.iratde.org

28

S. N. Phillipson & R. Sun

precocity, for example, would tend to confirm this impression. Given this status, it is not surprising that the broad field appears confusing to researchers working either within or outside the field. It is this confusion that provides detractors with ample opportunity to criticize and, in some cases, deride the research into giftedness altogether (Marsh, 1998; Seaton et al., 2008). If conceptions of giftedness amongst psychologists are numerous and fragmented, it is not surprising that the practice of gifted education is fragmented. The reasons for the development of the three broad educational paradigms (Eyre, 2009) in gifted education lie partly in the different agendas of the various educational systems around the world, and partly in the different conceptions of giftedness that are adopted by the policy makers working within these systems. In Hong Kong, for example, policy documents supporting gifted education, either explicitly or implicitly, simultaneously refer to multiple intelligences, creativity, the Marland report, IQ and Gagns Differentiated Model of Giftedness and Talent as conceptions of giftedness. The documents also mention disabilities and underachievement as characteristics of some gifted individuals (Phillipson et al., in press). Using the Hong Kong context again, the policy of including as many conceptions of giftedness is politically attractive. Being inclusive means that the cohort of gifted students can be as wide as possible, reducing the likelihood of alienating certain members of the community (Eyre, 2009). As Eyre pointed out, the cohort approach depends on the prior identification of students who are gifted in some way and depend, therefore, on psychological and/or personality commonalities amongst these students. Perhaps the best known examples of research involving the cohort paradigm are those that centre on mathematical precocity (Lubinski & Benbow, 2006). Significantly, this research has identified some of the important environmental and personal components that work together to enable the attainment of mathematical excellence. Eyre (2009) has identified definitions as one of the key questions for future research in gifted education. It is timely that psychologists researching in the field of giftedness begin to think about how to bring coherence to the diversity of conceptions of giftedness. This coherence needs to take into account cultural diversity as well as personality variables such as differences in measured intelligence, creativity and motivation, for example, and differences in conceptions of achievement. As Eyre concluded, the field must reorganize itself if it is to move beyond the educational margins and take its rightful place in identifying the conditions that enable and encourage exceptional ability to flourish (pp. 17, 18). Eyres (2009) concerns are not new. Ziegler (2005) has argued that most of the theoretical conceptions of giftedness do not have a sufficient empirical basis, that the relationship between gifts and achievement is often unclear, and that research methods and definitions are often characterized by tautologies. In discussing the relationship between intelligence and giftedness, for example, Callahan (2000) observed that the commonly cited conceptions of giftedness such as Renzullis three-ring definition, Tannenbaums psychosocial definition, Sternbergs triarchic and WICS models, and Gardners multiple intelligences model rely on different explanatory levels, such as cognitive skills, expressions of appropriate sets of behaviors, production of knowledge, bio-physical potential, personality, luck and culture. Furthermore, combinations of these different explanatory levels sometimes exist in the one model. Of course, models that rely on different explanatory levels are not easily falsifiable. Last, Ziegler and Heller (2000) pointed out that much of the reported research to date lacks the rigor normally expected of scientific research, such as appropriate control groups. Heller and Schofield (2000) also concluded from their content analysis of international trends on research on giftedness that the field is overwhelmingly dominated by applied rather than basic research, and that there was an urgent need for improvements in the quality of research designs and measurement techniques [and] most importantly, basic research [requiring] the intensification of cooperation with researchers (pp. 136-137). Although there is an urgent need for a systematic analysis of research since 1999, Eyre (2009) concluded that the situation has probably not changed very much since that time.

Modelling mathematical Actiotopes with CLARION

29

Generally, research on giftedness centers on exceptional performance of some kind, and on the antecedents and processes that lead to exceptional performance. Despite the problems in research approaches and methods, there is broad consensus that giftedness involves interactions between aspects of self (such as intelligence, cognitive processes, self-efficacy and motivation, for example) and the environment (parents, education, and culture, for example) that help propel the individual or the group to exceptionality (Sternberg, 2004a). Last, exceptional performance is contextual, meaning that it refers to different things at different times and for different sociocultural groups (Phillipson & McCann, 2007).

The critical state view of giftedness

Each of the psychologists working within the field of giftedness brings to their research, of course, their individual prejudices and preferences. This should be encouraged since it is the source of debate and the engine of further research. In order to bring a much needed unity to this debate, Ziegler and Heller (2000) described a meta-theoretical framework for research into giftedness. Termed the critical state view of giftedness (CSG), the framework describes a probabilistic rather than deterministic relationship between the critical state (CS) and achievement excellence (AE). Boundary conditions operate directly on the CS, to either promote or decrease the probability of an individual reaching AE (Ziegler & Heller). The term talented refers to someone who has the necessary psychological components, but not yet fulfilled the conditions of the CS. Gifted refers to someone who has fulfilled the conditions of the CS but, because of unfavorable boundary conditions, not yet reached AE. Expert refers to persons who have reached AE. The two stages of the CSG refer to the temporal relationship between the critical state (CS) and achievement excellence (AE) (Figure 1). Gifts are the psychological components within the CS but there are other aspects of personality such as motivational states with all being necessary and sufficient for AE. Individuals achieve AE after fulfilling the conditions within the CS. Within any domain there can be many types of AE and there may be many ways to achieve the any one AE. Achievement excellence (AE) is at first glance the most contentious area of the CSG. Ziegler and Heller (2000) proposed that AE needs to be understood using a sequential series of problems beginning with a clarification of domain, focus, frame of reference and significance respectively. Ultimately, however, solutions to these problems are more likely to be found by persons working within rather than outside the domain (Phillipson & Callingham, in press).

Development, learning processes CS a Critical State (CS) a CS b Personality characteristics inducive to gifts (unnecessary, sufficient) Gift (insufficient, nonredundant) Environmental conditions CS n Achievement eminence in Domain X

CS c

Figure 1. Achievement excellence in Domain X originates from one of Critical State (CS) a , CS b CS n under optimal boundary conditions for each critical state. Each CS operates independently of the others and the gift, a psychological process, fulfills the inus condition of Postulate 2. (Adapted from Ziegler & Heller, 2000, p. 15).

Modelling mathematical Actiotopes with CLARION

43

Cognition and multi-agent interaction: from cognitive modeling to social simulation (pp. 326). Cambridge: Cambridge University Press. Sun, R. (2006b). The CLARION cognitive architecture: Extending cognitive modeling to social simulation. In R. Sun (Ed.), Cognition and multiagent interaction: from cognitive modeling to social simulation (pp. 79-99). Cambridge: Cambridge University Press. Sun, R. (2007). The motivational and metacognitive control in CLARION. In W. Gray (Ed.), Modeling integrated cognitive systems (pp. 63-75). Oxford University Press: New York. Sun, R., & Naveh, I. (2004). Simulating organizational decision-making using a cognitively realistic agent model. [Electronic Version]. Journal of Artificial Societies and Social Simulation, 7, from http://jasss.soc.surrey.ac.uk/7/3/5.html Sun, R., & Naveh, I. (2007). Social institution, cognition, and survival: a cognitive-social simulation Mind & Society, 6(2), 115-142. Sun, R., & Zhang, X. (2004). Top-down versus bottomup learning in cognitive skill acquisition. Cognitive Systems Research, 5, 63-89. Sun, R., & Zhang, X. (2006). Accounting for a variety of reasoning data within a cognitive architecture Journal of Experimental & Theoretical Artificial Intelligence, 18, 169-191. Sun, R., Zhang, X., & Mathews, R. (2006). Modeling meta-cognition in a cognitive architecture. Cognitive Systems Research, 7, 327-338.

Upal, M. A. (2005). Simulating the emergence of new religious movements [Electronic Version]. Journal of Artificial Societies and Social Simulation, 8. Retrieved June 12, 2009 from http://ideas.repec.org/a/jas/jasssj/2004-352.html. Wilensky, U., & Rand, W. (2007). Making models match: Replicating an agent-based model [Electronic Version]. Journal of Artificial Societies and Social Simulation, 10 from http://ideas.repec.org/a/jas/jasssj/2007-72.html. Wood, J. N., & Spelke, E. S. (2005). Infants' enumeration of actions: Numerical discrimination and its signature limits. Developmental Science, 8(2), 173-181. Ziegler, A. (2005). The Actiotope Model of Giftedness. In R. J. Sternberg & J. Davidson (Eds.), Conceptions of Giftedness (pp. 411-436). New York: Cambridge University Press. Ziegler, A., & Heller, K. A. (2000). Conceptions of giftedness from a meta-theoretical perspective. In K. A. Heller, F. J. Mnks, R. J. Sternberg & R. F. Subotnik (Eds.), International handbook of giftedness and talent (2 ed., pp. 3-21). Oxford, UK: Elsevier Science. Ziegler, A., & Stger, H. (2004). Identification based on ENTER within the conceptual frame of the actiotope model of giftedness Psychology Science 46, 324-341.

The authors
Dr Shane N. Phillipson is an Associate Professor in the Department of Educational Psychology Counselling and Learning Needs, in the Hong Kong Institute of Education. His research interests include cultural conceptions of giftedness and models of achievement. He has been awarded a number of research grants, resulting in research publications in many international peer reviewed journals, including High Ability Studies and Educational Psychology. His books include Phillipson, S. N. (2007). Learning diversity in the Chinese classroom: Contexts and practice for students with special needs. Hong Kong: The Hong Kong University Press, and Phillipson, S. N., & McCann, M. (2007). Conceptions of giftedness: Sociocultural perspectives. Marwah, NJ: Lawrence Erlbaum Associates. Ron Sun is Professor of Cognitive Science at Rensselaer Polytechnic Institute, and formerly the James C. Dowell Professor of Engineering and Professor of Computer Science at University of Missouri-Columbia. His research interest centers around the study of cognition, especially in the areas of cognitive architectures, human reasoning and learning, cognitive social simulation, and hybrid connectionist-symbolic models. He is the founding co-editor-in-chief of the journal Cognitive Systems Research, and also serves on the editorial boards of many other journals. He is the general chair and the program chair of CogSci 2006, and the program chair of IJCNN 2007. He is a member of the Governing Boards of Cognitive Science Society and International Neural Networks Society.

44

S. N. Phillipson & R. Sun

Talent Development & Excellence Vol. 1, No. 1, 2009, 45-56

Relative age effects and technical skills

45

Relative age, talent identification and youth skill development: Do relatively younger athletes have superior technical skills?
Jrg Schorer1,2,*, Joe Baker3, Dirk Bsch2,4, Andreas Wilhelm2,5 and Jan Pabst2,4
Abstract: Relative age effects (RAEs) refer to differences among individuals in age-based cohorts typically used in sport. These effects usually favour relatively older members of the cohort and are thought to result from differences in maturation and experience among athletes of different chronological age. Recently, researchers suggested that relatively younger participants may not be as disadvantaged as previously thought. In two studies, we examined whether relatively younger athletes who were able to survive in a system that advantages their relatively older counterparts would develop superior technical skills. In study one, participants aged 13-15 years (n=140) drawn from a regional handball talent selection camp in Germany demonstrated a general relative age effect but no differences between relatively older and relatively younger athletes in physical body size (i.e., height/weight) or technical skills. In study two, similar tests were considered with a larger sample (n=478) and revealed similar results. Furthermore, there were no differences between those selected for the national youth team and those not selected. Differences in RAEs do not seem to be due to technical skills or body size variables. Moreover, the homogeneity of these results suggests causes of the relative age effect occur early in development.

Keywords: birth-date, maturation, expertise, skill acquisition

There are a range of primary and secondary factors influencingg the development of expertise in sport (Baker & Horton, 2004). One secondary factor believed to affect access to high quality training and coaching is relative age. Relative age effects (RAEs) refer to chronological age differences between individuals within annually age-grouped cohorts (Barnsley, Thompson, & Barnsley, 1985). In most youth sports, annual or biannual agegroupings, which are thought to create homogenous groups within the competition system of each sport, unfortunately create a sport structure that perpetuates RAEs. For example, consider a youth competing in a talent-development program for athletes 10 years of age. If the youth is born very early in the competition year (i.e., is relatively older), he would be ten percent older than relatively younger opponents born at the end of the competition year (cf. Helsen, Starkes, & van Winckel, 2000). First to report RAEs in sport were Grondin, Deschaies, and Nault (1984), who presented data for Canadian ice-hockey and volleyball demonstrating an unequal distribution across birth quartiles. At recreational, competitive, and senior professional levels more players were born shortly after the cut-off date for their respective sports. Similar findings have been noted in Canadian ice-hockey within developmental leagues (Barnsley & Thompson, 1988; Barnsley et al., 1985). In many sports, RAEs are not new phenomena. Wattie, Baker, Cobley, and Montelpare (2007) found RAEs in Canadian ice-hockey hall of fame athletes going back over
1

Institute for Sport Science, Westflische Wilhelms-University Mnster, Horstmarer Landweg 62b,48149 Mnster, Germany 2 Research Group Handball * Corresponding author. Email: jschorer@wwu.de 3 School of Kinesiology and Health Science, York University, Toronto, Canada 4 Institute for Applied Training Science, Leipzig, Germany 5 Institute for Sport Science, Christian-Albrechts-University Kiel, Germany
ISSN 1869-0459 (print)/ ISSN 1869-2885 (online) 2009 International Research Association for Talent Development and Excellence http://www.iratde.org

46

J. Schorer et al.

four decades. For professional soccer in Germany, Cobley, Schorer, and Baker (2008) demonstrated RAEs for players and coaches (but not for referees) since the beginning of the Bundesliga (Highest Federal league) in the early 1960s. In more recent years, RAEs have been confirmed internationally for male athletes on many different levels of competition and in a variety of sports, including Australian Rules football (Abernethy & Farrow, 2005), baseball (Thompson, Barnsley, & Stebelsky, 1991), basketball (Delorme & Raspaud, 2009), soccer (Helsen, Starkes, & Van Winckel, 1998, 2000; Helsen, Van Winckel, & Williams, 2005; Jimenez & Pain, 2008), handball (Schorer, Cobley, Bsch, Brutigam, & Baker, in press; Schorer, Baker, Lotz, & Bsch, in press), ice-hockey (Wattie et al., 2007), rugby (Till et al., in press), and volleyball (Barnsley et al., 1985). Findings for female athletes are less consistent. Vincent and Glamser (2006) found marginal RAEs at the National level for female soccer players but not at the state level. Schorer, Cobley, et al. (in press) observed a significant over-representation of relatively older athletes (i.e., from the first quartile) in young female handball players, but these effects were smaller than for male athletes. In general, RAEs are confirmed with few exceptions for a variety of sports for both genders over a range of competition and development levels (Cobley, Baker, Wattie, & McKenna, 2009). Our understanding of the mechanisms behind RAEs is much less developed. To date, two hypotheses have been forwarded to explain relative age differences. Most often cited are maturational differences (e.g., Barnsley & Thompson, 1988), which assumes that greater height and mass provide an advantage for relatively older athletes in sports where size is important (Malina, 1994; Malina, Bouchard, & Bar-Or, 2004; Norikazu, Atsushi, & Toru, 2007). The relevance of height and weight in youth sport is supported by a study of Brewer, Balsom, Davis, and Ekblom (1992), who found that elite junior soccer players were above the 95th percentile of normative data for their population. A second, related, hypothesis considers the role of selection processes. The above mentioned maturational differences, which are stronger in the early years of athlete development, lead to an increased probability of selection by coaches for higher tiers of competition (Schorer, Cobley, et al., in press; Sherar, Baxter-Jones, Faulkner, & Russell, 2007), which perpetuates a vicious cycle for the relatively younger. Because they are less likely to make it into the selection, relatively younger athletes have less access to training with highly qualified coaches and better medical resources (Helsen et al., 1998). Additionally, they are less likely to gain match experience at higher levels of competition (considered essential to developing expertise, Baker & Horton, 2004), thereby increasing the differences between relatively older and relatively younger athletes. Recently, researchers have discovered that relatively younger participants may not be as disadvantaged as previous research suggested. One study by Ashworth and Heyndels (2007) found that relatively younger soccer players had higher salaries than their relatively older counterparts. Further, Baker and Logan (2007) found that relatively younger ice-hockey players were more highly sought after during the National Hockey League draft. In the study by Schorer, Cobley, et al. (in press) relatively younger athletes had a higher probability of becoming members of the adult national team. More specifically, early selections (i.e., the first regional and national selections) showed clear RAEs with significant over representations of relatively older athletes; however, the adult national team had approximately the same percentage of players born in the first and fourth quartiles indicating an over representation of relatively younger players relative to the population from which players were drawn. Although the specific mechanisms of these effects are not known, Schorer, Cobley, et al. (in press) proposed that systems that perpetuate relative age effects may turn out to be beneficial for some relatively younger players who develop superior technical or tactical skills in order to compete successfully against their older, generally bigger opponents. Ultimately, they hypothesize, relative age effects result in those relatively younger players who survive the system having a larger repertoire of skills and therefore demonstrating superior performance. We explore this idea in the present study. Below we summarize two studies examining these relationships in German handball. Overall, our objectives were to a) establish relative age effects in two samples of handball players being considered for national team selection, b) consider differences between relatively

Relative age effects and technical skills

47

older and relatively younger players on variables related to body size and technical skill development, and c) consider differences between those selected for the National team versus those not-selected.

Study 1
In this first study, we consider the above relationships with a sample of participants being considered for national handball team selection in Germany. During this preparation camp for national selection, five regional D-squads tested their teams (Wohlrab, Landgraf, & Feldmann, 1998). By the end of the week, an All-Star-team was selected by the coaches based on their impressions over the week. Based on the preliminary research done to date (Baker & Logan, 2007; Schorer, Cobley, et al., in press), we hypothesized that the sample would show an overall relative age effect and that the relatively youngest players would have higher scores in technical skills while relatively older ones would have larger body sizes, although these hypotheses were largely exploratory. Methods Athlete data were acquired during the Sdcamp 2008 of the Southern German Handball Federation in Tailfingen. Participants were between 13-15 years of age and most of them passed two selection levels (district and region/state). For this study, 69 female and 71 male athletes provided their birth-date, height (in cm), and weight (in kg). Their height ranged from 1.54-1.81m for the female and from 1.66-1.96m for the male players. Their body mass was between 46-73kg for females and 52-88kg for males. To test for RAEs, birth-months of the players were re-coded to reflect the athletes birth quartile (Q). Because the handball annual age-grouping starts with the 1st of January, quartiles were calculated accordingly: Q1: JanuaryMarch, Q2: April-June, Q3: July-September; and Q4: October-December. As with most of the previous research, comparisons were drawn on the assumptions of an equal distribution (see Cobley et al., 2008). For Germany, a recent study by Schorer, Baker, et al. (in press) demonstrated equal distributions between quartiles. Chi-square analyses were conducted to test for differences in quartile distributions among the overall sample as well as between players selected for the All-Star-team and those not selected. In addition to these tests we conducted a technical talent assessment where respondents were required to throw ten balls as precisely and as quickly as possible alternating between the upper two corners of the goal. This task was chosen by the national coaches because it represents one of the main skills athletes at this age need to have acquired since it is the basis for most throwing skills in handball. The players had to pick up the ball from a box positioned twelve meters away from the goal in the middle of the field and then throw the balls after three approaching steps from approximately nine metres. The duration of these ten throws was measured by an electronic watch. Due to time limitations during the camp only 43 male and 48 female players were able to perform the test. During these ten throws two independent raters evaluated the over arm throw (German: Schlagwurf) in four different categories (run up, shooting position, throwing movement, and final position). Scores demonstrated a high degree of inter-rater-reliability ((overall)=.97) and the ratings for both judges were amalgamated into a general technical rating (Pabst, Bsch, Wilhelm, & Schorer, 2009). Additionally, the speed of throws was measured by Speed-Trac (Sport Thieme, Germany) and the percentage of correct hits to the goal corners provided a measure of throw accuracy. This task resulted in four dependent variables (1) test duration, (2) expert ratings, (3) throw speed, and (4) throw accuracy. Instead of using a descriptive statistical approach, a more exploratory data analysis was conducted for physiognomic data and technical skill data, because cell sizes were small and varied among the quartiles (cf. Sedlmeier, 1996). Figures are presented as boxplots, based on a rank scale, which are less influenced by outliers. The line in the middle of the box is the median. The box shows 25% and 75% quartiles and the lines above and below show the

48

J. Schorer et al.

highest and lowest values that are not outliers. Circles (1.5 to 3.0 standard deviations) and asterisks (more than 3.0 standard deviations) above or below these lines indicate outliers. The width of the boxes allows a relative comparison of cell sizes per group (cf. Benjamini, 1988; Sedlmeier, 1996). For the inferential statistical analysis, a conservative Exact Test by Kruskal-Wallis for independent samples was calculated, which works well with small samples and rank scales. For the Monte Carlo sampling the default values corresponding to a sample size of 10,000 and confidence levels of 99% were used. Due to the small sample size, the exact or asymptotic pvalue seemed inappropriate; therefore, Monte Carlo p-values are reported. When applicable, post-hoc comparisons of independent samples were calculated using Mann-Whitney-U-Tests (alpha criterion set on .05 and for technical skills alpha was Bonferroni adjusted). For calculating effect sizes and power, G*Power 3.0.10 was used (Faul, Erdfelder, Lang, & Buchner, 2007). Results The results are presented in three sections. First, we provide the analyses of overall relative age effect. Second, we test birth quartile differences in physiognomic variables (i.e., weight and height). In the final section, we investigate differences between quartiles in the technical skill test. For all three sections, we differentiate those selected for the all-star game and those not selected, to test whether the selection affected the strength of the RAE (as hypothesized by Musch & Grondin, 2001) The analysis of the overall sample revealed a significant relative age effect, (3, n=140) = 24.29, p<.01, w=.42. Differentiating between selected and not-selected players, significant differences from the normal distribution were found for selected players, (3, n=43) = 6.76, p=.04, w=.40, and for not-selected players, (3, n=97) = 18.59, p<.01, w=.44. Comparing distributions of not-selected players to selected players, no significant differences were found, (3, n=97) = 3.51, p=.33, w=.19, 1 =.45 (see Figure 1).

Quartile 1

Quartile 2

Quartile 3

Quartile 4

45 40 Player per quartile (in %) 35 30 25 20 15 10 5 0

Overall (n = 140) Selected (n = 43) Not selected (n = 97)

Figure 1. Distribution of players birth quartiles overall (n=140) as well as differentiated between selected (n=43) and non-selected (n=97) for study 1.

Relative age effects and technical skills

49

200

90

70 58

190

80

170

Weight (in kg)

Quartile 1 Quartile 2 Quartile 3 Quartile 4

Height in cm

180

70

60

160

50

150

40 Quartile 1 Quartile 2 Quartile 3 Quartile 4

Birth-quartiles

Birth-quartiles

35

58 72 62 57

30

28 30

Ratings by experts (in points)

Duration of test (in ms)

26

25

24

22

20

20

26

18 15 Quartile 1 Quartile 2 Quartile 3 Quartile 4

101

Quartile 1

Quartile 2

Quartile 3

Quartile 4

Birth-quartiles

Birth-quartiles

100

90

80

Accuracy of throws (in n of hits)

Quartile 1 Quartile 2 Quartile 3 Quartile 4

Speed of throws (in km/h)

70

60

50

0 Quartile 1 Quartile 2 Quartile 3 Quartile 4

Birth-quartiles

Birth-quartiles

Figure 2. Boxplots for height (in cm/upper left), weight (in kg/upper right), test duration (in secs/middle left), expert ratings (in points/middle right), throw speed (in km/h/lower left), and throw accuracy (in n of hits/lower right) differentiated by birth quartile.

50

J. Schorer et al.

As can be seen in Figure 2 (upper row), there were no differences among the birth quartiles for either height, H(3)=1.20, p=.76, or weight, H(3)=0.55, p=.91. Further, differentiating between selected and not-selected players revealed no differences between the groups; height, selected: H(3)=6.31, p=.09, not-selected: H(3)=2.87, p=.41, and weight, selected: H(3)=2.14, p=.56, not-selected: H(3)=0.59, p=.90. Finally, we considered differences between players from different quartiles concerning their technical skills. As can be seen in Figure 2 (middle and lower rows), there were no differences among the groups for any of the four technical variables. Inferential statistics confirmed no significant differences for (1) expert ratings, H(3)=3.10, p=.38, (2) throw accuracy, H(3)=1.77, p=.64, (3) throw speed, H(3)=1.46, p=.70, and (4) test duration, H(3)=0.90, p=.83. Similarly, no differences were found between selected and not-selected players, (1) expert ratings, selected: H(3)=0.95, p=.83, not-selected: H(3)=4.39, p=.22, (2) throw accuracy, selected: H(3)=0.22, p=.98, not-selected: H(3)=3.35, p=.35, (3) throw speed, selected: H(3)=1.48, p=.71, not-selected: H(3)=1.51, p=.68, and (4) test duration, selected: H(3)=2.94, p=.42, not-selected: H(3)=4.46, p=.22. Discussion Analyses indicated very little difference among the birth quartiles for the study outcomes (height, weight, or technical skills). Furthermore, these variables did not differentiate who was selected versus who was not-selected. While these results are intriguing from a theoretical point of view, one main methodological limitation in this study needs to be considered that warrants caution in our discussion of the results. The sample size of this study was rather small, which resulted in cell sizes of five or less especially for the relatively younger and selected athletes. This small number is reached even without differentiating between genders, a known moderator of RAEs (Cobley et al., 2009; Schorer, Cobley, et al., in press). For the comparisons of birth quartile distributions to normal distributions these limitations are not as important as for testing the difference hypothesis on physiognomic factors (Malina, 1994; Malina et al., 2004; Norikazu et al., 2007) and technical skills (Schorer, Cobley, et al., in press). Based on the limitations noted above, a replication of this study was needed with a larger number of participants. While replications are often not seen as an attractive alternative for researchers, their necessity for the consolidation and improvement of a scientific disciplines is unquestionable (Amir & Sharon, 1990). We undertake this task in Study Two.

Study 2
The national handball try-outs in Germany provide the opportunity to collect data from up to 480 young athletes. Out of this sample, up to 80 players are chosen for a second selection camp out of which the German national youth team is selected. This larger sample should allow for hypothesis testing inferential statistics, at least for the combined data set. The aims of Study Two were the same as in Study One. We investigated whether a sample of female and male players considered for the youth national team would show an overall RAE and whether body seize and technical skill measures would differ between relatively older and younger athletes or predict who was selected for the National team. In addition to these objectives from the first study, we examined differences between male and female subgroups. Based on the preliminary research done to date (i.e., Baker & Logan, 2007; Schorer, Cobley, et al., in press) and similar to Study 1, we hypothesized that the sample would show an overall relative age effect and that the relatively youngest players would have higher scores in technical skills and the relatively older ones would have larger body seizes. We also hypothesized that these effects would be smaller in female than in male players. Methods Athlete data were acquired during the 2009 try-outs for the female and male German national youth teams. Participants in these camps were between 13-15 years of age and most of them

Relative age effects and technical skills

51

passed through two previous selection levels (district and region/state). Their height ranged from 1.53-1.86m for the female players and from 1.62-1.98m for the males. Their body mass ranged from 43-85kg and from 55-106kg for the females and males respectively. By the end of these try-outs, lists of female and male candidates for the national youth teams were aggregated. For this study, 238 female and 240 male athletes provided their birth-date, height (in cm), and weight (in kg). To test for RAEs, birth-months of the players were re-coded to reflect the athletes birth quartile (Q). Similar to Study One, quartiles were calculated as Q1: January-March, Q2: April-June, Q3: July-September; and Q4: October-December and comparisons were drawn on the assumption of an equal distribution (Cobley et al., 2008; Schorer, Baker, et al., in press). Chi-square analyses were conducted to test for differences in quartile distributions among the overall sample as well as between players selected for the second national selection level and those not selected. Additionally, the technical skill test described in Study One was performed by 469 out of the 478 try-outs participants. Nine participants were not able to fulfil this test due to injuries. Due to the amount of time necessary to evaluate the technique of the throws it was impossible to conduct this rating during the try-outs. Therefore, only accuracy and mean speed of throwing were utilized as dependent variables for the technical test. Results As in Study One, results are presented in three main sections: overall RAEs, differences between the birth quartiles for height and weight, and differences for technical skills (i.e., throwing speed, throwing accuracy, and test duration). In each section we also consider males and females separately. As expected, significant overall RAEs were revealed, (3, n=478) = 41.72, p<.01, w=.29. As can be seen in Figure 3, RAEs were also found in the female, (3, n=238) = 20.49, p<.01, w=.29, and male sub-groups, (3, n=240) = 24.23, p<.01, w=.32.

60 50 40 30 20 1 0 0

Quartile 1

Quartile 2

Quartile 3

Quartile 4

Player per quartile (in %)

Overall (n = 240)

Selected (n = 50) Male

Not selected (n = 1 90)

Overall (n = 238)

Selected (n = 39) Female

Not selected (n = 1 98)

Figure 3. Distribution of male players birth quartiles (left side) overall (n=240) as well as differentiated between selected (n=50) and non-selected (n=190) and female players birth quartiles (right side) overall (n=238) as well as differentiated between selected (n=39) and non-selected (n=198) for study 2.

Differentiating between selected and not-selected female players, significant differences from the equal distribution were found for both selected, (3, n=39) = 13.82, p<.01, w=.59, and not-selected players, (3, n=199) = 11.99, p<.01, w=.24. Comparing the distributions of the not-selected players against the selected players, significant differences were also found, (3, n=199) = 65.81, p<.01, w=.41. Considering male athletes, significant differences from the

52

J. Schorer et al.

normal distribution were revealed for selected players, (3, n=50) = 18.00, p<.01, w=.60, and not-selected players, (3, n=190) = 12.53, p<.01, w=.26; further, significant differences were also found between these distributions, (3, n=190) = 36.36, p<.01, w=.39. Because there were few players in the third and fourth quartile for the selected players (as small as only 3 players), only differences between quartiles in the combined group were considered for height, weight and technical skills. For the physiognomic variables KruskalWallis-tests revealed no significant differences between birth quartiles for height, H(3)=1.00, p=.80, or for weight, H(3)=0.72, p=.87, for males. As can be seen in Table 1, similar results were found for the females. There were no significant differences between birth quartiles for height, H(3)=1.32, p=.72, or for weight, H(3)=0.57, p=.90. For males, no significant differences between groups could be revealed for mean speed of throws, H(3)=3.72, p=.29, and accuracy of throwing, H(3)=1.55, p=.67. For females, no significant differences between birth quartiles were revealed for mean speed of throws, H(3)=2.08, p=.56, or for accuracy of throwing, H(3)=2.79, p=.43.

Table 1. Comparison of means (standard deviations) per birth quartiles concerning height (in m), weight (in kg), accuracy of throws (n out of ten), and mean speed of throw (in km/h).

Gender

Dependent variable
Height Weight

Quartile 1
1.82 (0.07) 75.3 (8.6) 3.81 (1.96) 63.7 (6.8)

Quartile 2
1.81 (0.07) 74.5 (7.8) 3.66 (1.71) 64.8 (7.3)

Quartile 3
1.82 (0.07) 74.7 (9.9) 3.47 (1.81) 62.3 (6.4)

Quartile 4
1.81 (0.06) 74.3 (7.7) 3.38 (1.54) 64.8 (8.4)

Male Accuracy of throwing Mean speed of throws

Height Weight Female Accuracy of throwing Mean speed of throws

1.68 (0.06) 61.3 (7.4) 3.97 (1.78) 54.3 (5.5)

1.67 (0.07) 60.7 (7.1) 3.66 (1.82) 54.3 (5.7)

1.67 (0.06) 61.1 (7.7) 3.38 (1.67) 53.8 (5.5)

1.67 (0.07) 60.8 (8.1) 3.71 (1.66) 52.8 (5.1)

Discussion As expected (Schorer, Cobley, et al., in press), young handball talents showed a general relative age effect; however, contrary to our hypotheses, there were no relative age differences in any of the player-specific outcomes examined in this study. Players across the birth quartiles were similar in height, weight, and technical skills (i.e., test duration, expert rating, throw speed and accuracy). Moreover, there were no differences between players selected for the national youth team and those not selected. While we were able to overcome one of Study Ones limitations by differentiating between genders, the small number of selected athletes limited the sophistication of our analyses. As in Study One, the cell sizes for the third and fourth quarter within the selected male and female groups became too small to conducted differentiated parametric analyses. An attempt to use data from the previous selection year was not possible, because the technical test was altered for the selection of 2009. This analysis may be possible in the future, provided the test remains constant for the next two to three years. By choosing random comparison samples, null-hypothesis testing as suggested by our results should be possible.

Relative age effects and technical skills

53

General Discussion
Taken together, results from these studies support previous work (e.g., Cobley et al., 2008; Schorer, Cobley, et al., in press) suggesting that the primary mechanism driving relative age effects acts early in the athlete development process and that latter levels of selection, such as being selected for the national youth team, do not add to the effect. If this result bears out in further research, it provides important information for reducing or eliminating relative age effects. Specifically, that the earliest levels of talent identification and selection need to be targeted and modified to provide a more balanced selection process. On the other hand, these results do little to explain the preliminary work by Baker and Logan (2007) as well as by Ashworth and Heyndels (2007) indicating that relatively younger athletes are more highly valued than their relatively older counterparts. The clear similarities among the birth quartiles in this study indicates either that relatively younger athletes are superior on some other outcome not measured in this study (e.g., tactical or decision-making skill) or that a similar advantage for relatively younger athletes is not present in handball (which would be contrary to Schorer, Cobley et al., in press). Both explanations might be influenced by missing moderator variables. Playing position, for instance, has been seen as important for understanding the RAE in handball (Schorer, Cobley, et al., in press), soccer (Ashworth and Heyndels, 2007) and ice hockey (Edwards, 1994; Grondin & Koren, 2000). As a result, the expected difference between relatively younger and relatively older would need to be considered relative to playing position, which, unfortunately, was not possible with the sample investigated here. It is also possible that a very homogenous group of players is being considered as talented right from the first regional level of selection (which in handball occurs around 10 years of age). As the present results indicate, they have similar stature and weight (similar results have been noted in ice-hockey by Baker, Cobley, Montelpare, Wattie, & Faught, in press; Sherar et al., 2007). It is possible that compensation for the relatively younger occurs during earlier selection levels since the developmental differences are typically much larger pre-puberty than post-puberty. Additionally, it would be helpful to expand the measures of growth and maturation beyond simple measures of height and weight to more objective measures of biological maturity. These measures would rule out whether relatively younger athletes are simply maturing more quickly than their relatively older peers. Another reason for the homogeneity of the groups in technical skills might be the consistency in types and amounts of training after the first selection. As Helsen et al. (1998) hypothesized more qualified training on technical skills might happen in their selection team training resulting in similar movement patterns for the players, although the homogeneity of training hypothesis does not explain the results by Baker and Logan (2007) or Ashworth and Heyndels (2007). Furthermore, observation of individual players at most levels of play clearly suggests a degree of individual difference even among players with the same coach (reflecting a given coachs signature). It is possible that the technical skill examined in the present study (throwing) was too simple and as a result a large degree of homogeneity across the birth quartiles is not surprising. Future work may consider more sophisticated measures of technical skills to consider whether these differences are affected by RAEs. Overall, future studies considering differences between relatively younger and older athletes should investigate the very first selection level, to investigate why relatively younger athletes are selected for further talent developmental programmes as well as later levels of selection to determine why specific selection decisions are being made. Especially helpful would be longitudinal data showing the development of technical as well as perceptual and tactical skills in players across the selection process. From a broader perspective, recent studies (including the data presented here) reinforce the need for a more elaborate theory of RAEs, which considers the complex interaction among moderator variables and main effect predictors. A comprehensive understanding of this phenomenon would facilitate behavioural and policy changes to eliminate this inequality.

54

J. Schorer et al.

Acknowledgements
We would like to thank the German Handball Federation for their cooperation. Additionally, we would like to express our gratitude to Sebastian Bagats, Christina Janning, Rebecca Rienhoff, Lennart Fischer and all the students from the seminar talent detection in handball for their helping hand with the data collection. This study was in part funded by the Federal Institute of Sport Science, Germany, IIA1-070704/09.

References
Abernethy, B., & Farrow, D. (2005). Contextual factors influencing the development of expertise in Australian athletes. Proceedings of the 11th World Congress of Sport Psychology, Sydney, Australia. Amir, Y., & Sharon, I. (1990). Replication Research: A "must" for the scientific advancement of psychology. In J. W. Neuliep (Ed.), Handbook of replication research in the behavioral and social sciences (pp. 51-69). Corte Medera, CA: Select. Ashworth, J., & Heyndels, B. (2007). Selection bias and peer effects in team sports: The Effect of Age Grouping on Earnings of German Soccer Players. Journal of Sports Economics, 8, 355-377. Baker, J., Cobley, S., Montelpare, W. J., Wattie, N., & Faught, B. (in press). Exploring proposed mechanisms of the relative age effect in Canadian minor hockey. International Journal of Sport Psychology. Baker, J., & Horton, S. (2004). A review of primary and secondary influences on sport expertise. High Ability Studies, 15, 211-228. Baker, J., & Logan, A. J. (2007). Developmental contexts and sporting success: birth date and birthplace effects in national hockey league draftees 2000-2005. British Journal of Sports Medicine, 41, 515-517. Barnsley, R. H., & Thompson, A. H. (1988). Birthdate and success in minor hockey: the key to the NHL. Canadian Journal of Behavioural Science, 20, 167-176. Barnsley, R. H., Thompson, A. H., & Barnsley, P. E. (1985). Hockey success and birthdate: The RAE. Canadian Association for Health, Physical Education, and Recreation, 51, 23-28. Benjamini, Y. (1988). Opening the box of a boxplot. The American Statistician, 42, 257-262. Brewer, J., Balsom, P. D., Davis, J. K., & Ekblom, B. (1992). The influence of birth date and physical development on the selection of a male junior international soccer squad. Journal of Sport Science, 10, 561-562. Cobley, S., Baker, J., Wattie, N., & McKenna, J. M. (2009). Annual age-grouping and athlete development: A meta-analytical review of relative age effects in sport. Sports Medicine, 39, 235-256. Cobley, S., Schorer, J., & Baker, J. (2008). Relative age effects in elite German soccer: a historical analysis. Journal of Sports Sciences, 26, 15311538. Delorme, N., & Raspaud, M. (2009). The relative age effect in young French basketball players: a study on the whole population. Scandinavian Journal of Medicine and Science in Sport, 19, 235242. Edwards, S. (1994). Born too late to win? Nature, 370, 186. Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Reseach Methods, 39, 175-191. Grondin, S., Deschaies, P., & Nault, L. P. (1984). Trimestres de naissance et rendement scolaire. Apprentissage et Socialisation, 16, 169-174. Grondin, S., & Koren, S. (2000). The relative age effect in professional baseball: A look at the history of major league baseball and at current status in Japan. Avante, 6, 64-74. Helsen, W. F., Starkes, J. L., & Van Winckel, J. (1998). The influence of relative age on success and dropout in male soccer players. American Journal of Human Biology, 10, 791-798. Helsen, W. F., Starkes, J. L., & Van Winckel, J. (2000). Effect of a change in selection year on success in male soccer players. American Journal of Human Biology, 12, 729-735. Helsen, W. F., Van Winckel, J., & Williams, A. M. (2005). The relative age effect in youth soccer across Europe. Journal of Sports Sciences, 23, 629-636. Jimenez, I. P., & Pain, M. T. G. (2008). Relative age effect in Spanish association football: Its extent and implications for wasted potential. Journal of Sports Sciences, 26, 995-1003. Malina, R. M. (1994). Physical growth and biological maturation of young athletes. Exercise and Sport Sciences Reviews, 22, 389-434. Malina, R. M., Bouchard, & Bar-Or, O. (2004). Growth, maturation, and physical activity Champaign Illinois: Human Kinetics. Musch, J., & Grondin, S. (2001). Unequal competition as an impediment to personal development: A review of the relative age effect in sport. Developmental Review, 21, 147-167. Norikazu, H., Atsushi, H., & Toru, F. (2007). Relationship between date of birth, skeletal age, and anthropometric characteristics in adolescent elite soccer players. Paper presented at the 12th Annual Congress of the ECSS 2007.

Relative age effects and technical skills

55

Pabst, J., Bsch, D., Wilhelm, A., & Schorer, J. (2009). Haben es gute Handballer (selbst) in der Hand? [Can good handball players be identified by hand?], Manuscript in preparation. Schorer, J., Baker, J., Lotz, S., & Bsch, D. (in press). Influence of early environmental constraints on achievement motivation in talented young handball players. International Journal of Sport Psychology. Schorer, J., Cobley, S., Bsch, D., Brutigam, H., & Baker, J. (in press). Influences of competition level, gender, player nationality, career stage and playing position on relative age effects. Scandinavian Journal of Medicine and Science in Sports. Sedlmeier, P. (1996). Jenseits des SignifikanztestRituals: Ergnzungen und Alternativen. Methods of Psychological Research Online, 1, 41-63. Sherar, L. B., Baxter-Jones, A. D. G., Faulkner, R. A., & Russell, K. W. (2007). Do physical maturity and birth date predict talent in male youth ice hockey players? Journal of Sports Sciences, 25,

879-886. Thompson, A. H., Barnsley, R. H., & Stebelsky, G. (1991). Born to Play Ball - the Relative Age Effect and Major-League Baseball. Sociology of Sport Journal, 8, 146-151. Till, K., Cobley, S., Wattie, N., O'Hara, J., Cooke, C., & Chapman, C. (in press). The Prevalence, Influential Factors and Mechanisms of Relative Age Effects in UK Rugby League. Scandinavian Journal of Medicine and Science in Sport. Vincent, J., & Glamser, F. D. (2006). Gender differences in the relative age effect among US Olympic Development Program youth soccer players. Journal of Sports Sciences, 24, 405-413. Wattie, N., Baker, J., Cobley, S., & Montelpare, W. J. (2007). Tracking relative age effects over time in Canadian NHL players. International Journal of Sport Psychology. Wohlrab, S., Landgraf, L., & Feldmann, K. (1998). Talentcamp - Erlebnis-Event und AusbildungsCheck. Handballtraining, 20, 27-31.

The authors

Dr. Jrg Schorer is a post-doctoral fellow at the University of Mnster in the Department of Sport Psychology and is currently finishing his habilitation. His research considers the development and maintenance of expertise across the life-span from multiple perspectives.

Dr. Joe Baker is an associate professor in the School of Kinesiology and Health Science at York University, Canada and a visiting researcher at Leeds Metropolitan University in the UK. His research examines the factors affecting athlete development across the lifespan. Dr. Baker is the current president of the Canadian Society for Psychomotor Learning and Sport Psychology.

PD Dr. Dirk Bsch is head of the Department of Game and Combat Sports and has responsibility for elite youth sports at the Institute for Applied Training Science University of Leipzig, Germany.

56

J. Schorer et al.

Prof. Dr. Andreas Wilhelm is currently working at the Institute for Sport and Sport Science at the Christian-Albrechts-University Kiel, Germany. His research interests include topics like groups, leadership, coaching, and sport games.

Dr. Jan Pabst is currently employed as the Head of the Research Group Handball at the Institute for Applied Training Science in Leipzig, Germany.

Talent Development & Excellence Vol. 1, No. 1, 2009, 57-66

Gifted Learners Epistemological Beliefs

57

Gifted Learners Epistemological Beliefs

Marion Porath1,*, Judy Lupart2, Jennifer Katz1, Constantine Ngara1 and Pamela Richardson1
Abstract: This paper focuses on gifted children's and adolescents' narrative interpretations of learning using a neo-Piagetian theory of conceptual development (Case, 1992; Case & Okamoto, 1996) as a framework. Children's narratives develop from action based reasoning to intentional reasoning that incorporates understanding of their own and others' mental states. In adolescence, thinking becomes more interpretive, with a focus on the psychology of individuals involved. Similarly, understanding of learning develops from a focus on the activities of school (e.g., reading, math) to interpretation of the meaningfulness of different learning activities and the meaning of knowledge itself. Students narrative accounts of learning are used to articulate the development of epistemological beliefs across childhood and adolescence. Keywords: Learning, conceptual development, giftedness, epistemological beliefs

There are significant gaps in our knowledge of how children understand the educational enterprise, and of how this understanding may impact their adjustment to school and realization of their talents. In particular, it is imperative that we understand how children with significantly developmentally advanced ability profiles understand their own learning. Familiar norms do not apply to these children. Significant adaptations need to be made to curricula and educational programming to meet their needs (Robinson, Zigler, & Gallagher, 2000). Listening to children and providing environments and educational strategies that engage them in understanding and directing their own learning are critical to the notion of gift-creation advocated by Hymer (2009) and Huxtable (2009) and consonant with the vision of contemporary classrooms where growth, enquiry, and personally relevant education are the focus (Hymer, Whitehead, & Huxtable, 2009). Harter (1996) pointed out how important it is to know how children perceive different aspects of school life, in order that we can take these perceptions into account in supporting them through the school years. However, education typically disregards learners' perspectives on knowledge and their understanding of their capacity for learning (Bruner, 1996). Children are thinkers (Bruner) with competent and legitimate ways of making sense of the world. Children's understanding of learning is a necessary starting point for pedagogy. Children's understandings are rarely engaged in this way, but when they are, the result for children is deeper and more meaningful learning (Bransford, Brown, & Cocking, 2000; Fosnot, 1996; Griffin & Case, 1996; McKeough, 1992). Epistemic beliefs, or beliefs about knowledge and knowing, are related to self-regulated learning and achievement (Muis, 2007) and intellectual performance (Kuhn, Cheney, & Weinstock (2000). What do children understand about learning and knowledge? This paper investigates the epistemological beliefs of gifted learners, concentrating on how they understand the relationship between themselves as knowers/learners and knowledge (Burr & Hofer, 2002). Following Burr and Hofer, personal epistemology is the focus rather than epistemology in the philosophical sense.
1

Faculty of Education, The University of British Columbia, 2125 Main Mall, Vancouver, BC, Canada V6T 1Z4 * Corresponding author. Email: marion.porath@ubc.ca 2 University of Alberta
ISSN 1869-0459 (print)/ ISSN 1869-2885 (online) 2009 International Research Association for Talent Development and Excellence http://www.iratde.org

58

M. Porath et al.

Research on epistemological beliefs has been conducted primarily with adults and some hypothesize that epistemological reasoning is not possible before the acquisition of formal thinking (Hofer & Pintrich, 1997). There may, however, be developmental precursors to epistemological thought, with abstract conceptions building on first thoughts about what it means to learn, as suggested by Burr and Hofer (2002), Kitchener (2002), and Kuhn et al. (2000). Very little research has been done on childrens epistemological beliefs; the origins of epistemological awareness (Burr & Hofer, p. 200) are not well understood. Studies on children include Burr and Hofers study of 3- to 5-yearolds focused on the relationship of theory of mind and epistemological beliefs and Kuhn et al.s studies of 7- and 8-year-olds and 10-year-olds. The former suggested a relationship between ability to predict and justify what others will think and do and theory of mind ability characterized by development from inability to do either set of tasks, to the ability to predict and justify in a relevant way, to the ability to coordinate both abilities successfully. Burr and Hofer saw their study as only beginning to address a critical developmental connection and call for more fine-grained, developmental studies that investigate process. Similarly, they note the need to connect such studies to education, highlighting that another gap in our understanding is that learners beliefs and the school context have not been linked in any practical way (Ziegler, Stoeger, & Mundi, 2004). Kuhn et al.s work, while tangentially related to education in its articulation of development of conceptions of knowledge from childhood to adulthood, also does not address the educational implications of the influence of different views of knowledge on the process of learning. Mansfield and Clinchys (2002) study focused on children aged 10, 13, and 16, finding an increasing ability to coordinate events with the inner worlds of individual knowers (p. 225). While they included a vignette featuring a teacher, the focus was on justifications for why the child thought the teacher was nice or mean. Their identification of the age-related increase in complexity and nature of epistemological position is important in understanding the general nature of epistemological development but needs practical translation to be meaningful in schools and the education of gifted learners.

A Neo-Piagetian Perspective on the Understanding of Learning Neo-Piagetian theory is the framework for the work presented in this paper. NeoPiagetian theories allow for the mapping of "precise changes in the development of academic skills and understandings" (Mascolo, Li, Fink, & Fischer, 2002, p. 120). Because of its articulation of the structure and content of childrens representations of the academic and social dimensions of school, and the developmental course of these representations, Cases (1992; Case & Okamoto, 1996) neo-Piagetian theory provides an appropriate framework for this study. Case (1992) described childrens central conceptual structures in a number of domains (mathematical, spatial, social narrative; see Case & Okamoto, 1996). These structures are blueprints of childrens understanding, the internal mental entities that consist of the relations among a number of concepts. Because the relations are semantic, articulating the meanings children assign to concepts, structures are conceptual. Structures are considered central to a domain because they form the basis for understanding a wide range of related tasks within that domain. Studies of gifted learners framed within Cases (1992) neo-Piagetian theory have found development to be relevant to the level of conceptual understanding they demonstrate (see, for example, McKeough, Genereux, & Jeary, 2006; Okamoto, Curtis, Jabagchourian, & Weckbacher, 2006; Porath, 2006). That is, maturation is a factor in the level of conceptual complexity demonstrated, even among highly able learners, when responses are considered from the perspective of central conceptual structures. For this reason, work with children who represented a typical range of abilities was used to inform this study of gifted learners conceptions of learning.

Gifted Learners Epistemological Beliefs

59

Childrens conceptions of learning. Bickerton (1994) articulated a developmental progression in how children understand learning, using Cases (1992) theoretical model. This research provides a basis for determining the sort of understanding that may be central to children's conceptions of academic tasks. Bickerton asked elementary school children what learning means, and what happens when they learn. She found that childrens understanding of the meaning of learning develops in the following manner. Four-year-olds think of learning as behavioural events (e.g., playing, going to school, building a tower). They also think of learning as involving the presence of a learning agent, such as Mom or Dad. Six-year-olds define learning as a relationship between a behavioural event and an internal state. Internal states include thoughts, feelings, and judgments. For example, If I do good work, I get happy. Sometimes learning gets me frustrated because I make so many mistakes (Bickerton, pp. 6-7). Eight-year-olds are able to relate a behavioral event to two internal states when defining learning. You get better at it if you try your hardest (Bickerton, p. 7). Ten-year-olds add a personal element to their definitions of learning. To me, learning means knowing how to do something without having any problems doing it (Bickerton, p. 7). Twelve-year-olds responses are more psychological in nature, recognizing states of mind as important in learning. Learning is developing a smarter mind. Learning is knowing and understanding things you didnt know before (Bickerton, p. 8). Bickerton (1994) conceptualized childrens understanding of learning as related to narrative thought, that is, as one of the understandings related to a central social narrative structure (Case & Okamoto, 1996; McKeough, 1992). Childrens responses clearly reflected human action and motivations related to the social context and traditions of schooling and education more broadly defined; they were making meaning of their learning experiences. As children enter the world of formal schooling, they face exceptional demands to make meaning of their experiences, including understanding their own learning and why others behave the way they do. Bruner (1986) described meaning making as the simultaneous construction of two landscapes.
One is the landscape of action, where the constituents are the arguments of action: agent, intention, or goal, situation, instrument. The other landscape is the landscape of consciousness: what those involved in the action know, think, and feel, or do not know, think or feel. (p. 12).

In a similar vein, Perner (1991) stated, "Mental states need to be systematically linked to externally observable events" (p. 101).

Method
This study focused on gifted children's perspectives on learning. The task and analyses reported are from a larger study focused on gifted childrens understanding of learning and teaching, their learning identities, and social aspects of education (Porath & Lupart, 2007). Eighty-one students from Grades 1 through 12 (43 boys; 38 girls) identified as gifted participated in the larger study. Seventeen of the participants were enrolled in a 2year program that facilitated early entrance to university (11 in Year 1; 6 in Year 2). All grades except 8, 9, and 11 were represented in the sample. (Canadas school systems comprise twelve grades, preceded by kindergarten.) The study included public and parochial schools in two large western Canadian cities. The students were identified in different ways, as is characteristic across school districts in western Canada. Identification strategies included combinations of teacher nomination, superior academic achievement, and/or superior intelligence or cognitive ability test scores. The participants attended a variety of programs for gifted learners, including segregated classes and pullout programs.

60

M. Porath et al.

Table 1. Development of Thinking about Learning

Level 1 Nature of thought about learning Action-based Examples You do a lot of new things; doing stuff; brain is remembering all that stuff To do no, not to do but to think about what something means. Learning can mean different things like my one example is probably you can learn about yourself, like whatyourself means to you, what is really inside youor you can learn knowledge based like math, and reading, and drawing, PE, all sorts of learning to do with knowledge well actually PE isnt to do with knowledge but whatever andI guess learning can also mean learning about relationships like sometimes you might have problems with your friends like you might get into a fight but you always learn from your mistakesso I think thats what learning means. Im finding out new things, researching things, and basically passing on what others have learned. It doesnt always happen to a kid. Learning doesnt stop when you become an adult, so being a scientist its always learning new stuff. Learning to me means to find new things, new understandings of things. You absorb information from others and you try to use it in your own life. Changing behaviour through experience and applying the knowledge. Its a function of mind. There is a spark when you are happy about what you are learning, asking questions, and making your own connections. Its interesting to learn new things and open your point of view. Putting information together to see if it can be combined to come up with an idea.

Gaining knowledge through a variety of modes and learning in different ways Understanding/meaning is key Rudimentary notion of knowledge Future-oriented General idea of brains role in learning

Acquisition of competence increasing knowledge; learning new things Awareness of interests, weaknesses, boredom, curiosity Multiple modes of learning Learning as continuous Links you to knowledge traditions Transformative Awareness of the breadth of learning (variety of sources, vicarious learning) Views of learning and knowledge related to personal preferences and goals Recognition of ideas and different perspectives on knowledge Characterization of learning as abstract Recognition of neuropsychological factors in learning (perception, associative learning)

Conceptions of Learning This task was the first of a series of research tasks completed by the participants. Each child participated in an individual semi-structured interview, using the questions below as a framework for discussing their perspectives on learning. Their responses were recorded and transcribed for analysis. What does learning mean? What is happening when you are learning? Where does learning come from? (Bickerton, 1994) Coding guidelines for responses to the first two questions were devised based on inductive analyses of childrens responses and Cases (1992) theoretical framework. The

Gifted Learners Epistemological Beliefs

61

scheme also drew on McKeoughs (1992; McKeough & Genereux, 2003) work on development in the narrative domain, viewing learning, generally and specifically, as an intentional act focused primarily on knowledge acquisition during middle childhood and an interpretive epistemological act in adolescence (Table 1). Coding guidelines were the result of an iterative process in which data were examined in light of Cases (1992) neoPiagetian theory of development; prototypical characteristics of stages in child and adolescent development defined; and childrens best responses assigned to a level of development. Coding was done first by four individual coders who were research assistants on the project; codes were then compared and discussed in pairs; the final stage involved comparing and discussing codes in the group of four to reach consensus. The process used was a constructivist revision of grounded theory (Henwood & Pidgeon, 2003, p. 134); the data guided interpretation within the conceptual framework of developmental theory using the steps of open coding, constant comparison, and theoretical integration (Henwood & Pidgeon). Analyses were primarily qualitative in nature. However, following other neo-Piagetian studies, a numerical score was assigned to levels of development and used in a statistical analysis of developmental trend. Responses to the question regarding the source of learning were content analyzed, focusing on whether the source was external, internal, or a combination of the two and the degree of sophistication evident in the response. This analysis was also informed by Cases (1992) theory of development; however, no numerical scores were assigned.

Results
Development of Conceptions of Learning Participants were grouped by grades; these groupings approximated the age ranges specified by Case (1992) for each of his hypothesized developmental stages (Grades 1 and 2, n=6; Grades 3 and 4, n=12; Grades 5 and 6, n=24; Grade 7, n=12; Early Entrance 1 and 2 with Grade 10, n=23; and Grade 12, n=4). Mean (SD) learning scores by group are reported in Table 2. Learning scores were analyzed for age effect using a oneway ANOVA. A significant between groups effect was found, F(5, 76)=6.73, p=.000. The linear trend was significant, F(1,80)=26.02, p=.000, with no significant deviation from a linear trend, F(4, 77)=1.90, p=.119. Post hoc comparisons were not done due to the unequal group sizes.
Table 2. Mean (SD) learning level by age group Group Grades 1 and 2 Grades 3 and 4 Grades 5 and 6 Grade 7 Early entrance 1 and 2; Grade 10 Grade 12
a

Mean age (SD) in months 93.67 (7.66) 109.42 (8.10) 124.75 (6.16) 145.17 (3.07) 176.35 (8.49) 204a

Learning level 1.50 (.45) 1.83 (.62) 2.40 (.64) 2.54 (.72) 3.15 (1.07) 2.38 (.85)

Average age is an estimate based on the usual age of Grade 12 students. Age for these students was not noted during data collection.

62

M. Porath et al.

Childrens responses form a developmental progression generally related to age, as illustrated in Table 1. The youngest children understand learning as doing stuff and learning new things. Children in the middle grades of elementary school understand learning as acquisition of knowledge that is important to them in the future and as more than just academic. Learning makes one conscious of ones own interests. Children of this age also have a rudimentary understanding that meaning is important in learning. Without meaning, learning is boring. In pre-adolescence, there is recognition of the competence that comes with learning, the link to established knowledge, and the human drive to learn. Early adolescent participants talked about learning as a transformative experience (You think in ways you didnt before) and their emergent place in a knowledge tradition. Some of the oldest participants, in particular students from the program designed to support early entrance to university, demonstrated sophisticated epistemological beliefs epistemic inclination (one must be interested and willing), epistemic acquisition (the availability of knowledge and its perceivability), and attributes of the perception (its certainty and the uniqueness of the perception) (Ziegler et al., 2004). They began to play with notions of the relativity of knowledge and unique ideas that emerge when individual perceptions interact with established knowledge domains, although these ideas were not yet well formed. Sources of Learning Childrens attributions for the sources of learning developed from external sources, often with little relationship to learning, in young children to a blend of internal and external sources later in middle childhood. Through adolescence a trend was observed that saw movement from primarily internal to primarily external to a blend of the two, all in more sophisticated statements that reflected growing recognition of the complexity of learning (Table 3).

Table 3. Developmental Progression in Attributions for the Source of Learning

Developmental Stage

Source Examples
External Scientists, teachers government

Middle childhood - Early primary school (Grades 1 and 2) Middle childhood - Elementary school (Grades 3 to 6) Early adolescence (Grade 7)

and

other

smart

people;

the

Internal and external Some learning comes from your brain and some learning comes from experiences. Internal I think learning comes from our instincts. We just have to keep exploring new things. Thats what drives evolution. External (more sophisticated notions) Life itself provides a huge learning opportunity. I think it can come anywhere. Internal (more sophisticated notions) and external; interaction between brain/mind and environment Your brain senses because you need to understand. Biologically, there are associations of stuff what we see, perceive, think.

Adolescence I (Early entrance and Grades 10 and 12)

Adolescence II (Early entrance and Grades 10 and 12)

Gifted Learners Epistemological Beliefs

63

Discussion
While the research was limited by unequal sample sizes, with few participants at the youngest and oldest points in the age range, and the cross-sectional design, it provides a beginning to articulating how gifted learners think about learning and knowledge. The youngest childrens responses suggest that there are developmental precursors to epistemological beliefs and that these earliest understandings lay the foundations for more sophisticated understandings of learning and knowledge. As the research was conducted in school settings, a bridge was made to what learning and knowledge mean to children in schools, contributing to our knowledge of how personal epistemologies are relevant to school learning. The childrens responses also offer insight into how some of our brightest students conceive of learning. Understanding of Learning: A Social Narrative or Epistemological Structure? Childrens understanding of learning can be conceptualized as part of a central social narrative structure. It is concerned with making meaning of actions, understanding ones own intentions with regard to learning, and in adolescence, interpreting ones learning in the context of knowledge traditions. However, understanding of learning may be broader than the domains conceived by Case as having unique conceptual and operational structures (Case, 1992; Case, Demetriou, Platsidou, & Kazi, 2001). In Cases system, central conceptual structures have been proposed in the domains of number, spatial representation, and social narrative.
These structures (1) represent the core content in a domain of knowledge, (2) help children to think about the problems that the domain presents, and (3) serve as a tool for the acquisition of higher order insights into the domain in question. (Case et al., p. 328)

Childrens understandings of learning, while they can be explained by a central social narrative structure, appear to be relevant to other central conceptual structures as well. A central epistemological structure might be proposed. However, given the domainspecific nature of central conceptual structures, a central epistemological structure, to be theoretically consistent, would need to explain understandings in a delimited domain of understanding. Understanding of learning appears to entail a more general conceptual understanding. It may be that beliefs about learning are processes that underlie or interconnect (Case et al., 2001) central conceptual structures. Cases (1992) executive processes or Demetriou, Efklides, and Platsidous (1993) hypercognitive system, for example, entail self-understanding and self-regulation. These enable the thinker to select and organize the particular computations and concepts required by a task addressed to any of the specific domains or to any combination of them (Case et al., p. 322). These constructs resemble the reflective abstraction that was, for Piaget, a central developmental process (Campbell, 1993). Analysis of the relationships of these beliefs to childrens conceptions of themselves as learners in different academic disciplines (reading, writing, and mathematics) suggests strong parallels with the development of understanding of learning (Porath & Lupart, in press), providing some support for understanding of learning and oneself as a learner as a central developmental process, an underlying mechanism for childrens understandings of learning in general and academic subjects in particular. Epistemological beliefs are concerned with making meaning of school, understanding ones own intentions with regard to learning, and interpreting ones learning in the context of knowledge traditions. These beliefs develop from action-based understanding of the educational endeavour to the awareness of personal preferences and needs and, ultimately, to sophisticated reflections on the nature of knowledge and ones role in knowledge development. Educational Implications Understanding the epistemological underpinnings of students approaches to learning and academic subjects can help educators to support students to consolidate their

64

M. Porath et al.

current constructions of learning and move them forward to more sophisticated constructions (McKeough, Okamoto, & Porath, 2002). Appreciating and supporting the foundations of learning in this way may contribute to students achieving richer and deeper knowledge of academic disciplines. Mansfield and Clinchy (2002) support this approach, arguing that teaching should start from students perspectives on what is to be studied, allowing for personal meaning to be considered in education. Through helping students to consolidate their understanding and then supporting them to build bridges to more sophisticated understanding, teachers also provide students the support necessary to achieve optimal levels of understanding (Kitchener, 2002). Without this support, their understandings, even those of the brightest students, can remain at a functional level (Kitchener). Directions for Research This study provides a starting point for continued exploration of gifted learners epistemological beliefs and the role those beliefs play in achieving excellence. Further research could take a longitudinal approach, tracking students growing sophistication in beliefs about learning and knowledge. Fine-grained analyses of process, such as can be accomplished with microgenetic analysis that examines changes while they occur during instruction, could provide knowledge about what instructional strategies and supports facilitate development. The fine-grained and intensive examination of processing that microgenetic analysis allows can provide suggestions about cognitive change mechanisms relevant to personal epistemology and indicate directions for extension of instructional models (McKeough & Sanderson, 1996; Pellegrino, Chudowsky, & Glaser, 2001). It facilitates the tracking of developmental processes from specifiable beginnings (as described by central conceptual understanding) through increasingly competent approximations to a more sophisticated understanding (Catan, 1986). Importantly, this tracking of process would take place in the context of learning more about what children have to tell us about their perspectives on themselves as learners and their views of knowledge.

Acknowledgements
A version of this paper was presented in the symposium, Developmental Trajectories in Narrative Thought: Contributing Factors and Underlying Mechanisms, XIIth European Conference on Developmental Psychology, University of La Laguna, Tenerife. This research was made possible by a grant from the Social Sciences and Humanities Research Council of Canada.

References
Bickerton, G. (1994, June). Narrative knowledge as revealed in childrens perceptions of learning. Paper presented at the Canadian Society for the Study of Education, The University of Calgary. Bransford, J. D, Brown, A. L. & Cocking, R. C. (Eds). (2000). How people learn: Brain, mind, experience, and school. Washington, DC: National Academy Press. Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press. Bruner, J. (1996). The culture of education. Cambridge, MA: Harvard University Press. Burr, J. E., & Hofer, B. K. (2002). Personal epistemology and theory of mind: Deciphering young childrens beliefs about knowledge and knowing. New Ideas in Psychology, 20, 199-224. Campbell, R. L. (1993). Commentary. Epistemological problems for neo-Piagetians. In A. Demetriou, A. Efklides, & M. Platsidou, The architecture and dynamics of developing mind (pp. 168-191). Monographs of the Society for Research in Child Development, 58(5-6, Serial No. 234). Case, R. (1992). The minds staircase: Investigating the conceptual underpinnings of childrens thought and knowledge. Hillsdale, NJ: Erlbaum. Case, R., Demetriou, A., Platsidou, M., & Kazi, S. (2001). Integrating concepts and tests of intelligence from the differential and developmental traditions. Intelligence, 29, 307-326. Case, R., & Okamoto, Y. (1996). The role of central conceptual structures in the development of

Gifted Learners Epistemological Beliefs

65

childrens thought. Monographs of the Society for Research in Child Development, 61(1-2, Serial No. 246). Catan, L. (1986). The dynamic display of process: Historical development and contemporary uses of the microgenetic method. Human Development, 29, 252-263. Demetriou, A., Efklides, A., & Platsidou, M. (1993). The architecture and dynamics of developing mind. Monographs of the Society for Research in Child Development, 58(5-6, Serial No. 234). Fosnot, C. T. (Ed.) (1996).Constructivism. Theory, perspectives, and practice. New York: Teachers College Press. Griffin, S., & Case, R. (1996). Evaluating the breadth and depth of training effects when central conceptual structures are taught. In R. Case & Y. Okamoto (Eds.), The role of central conceptual structures in the development of childrens thought (pp. 83-102). Monographs of the Society for Research in Child Development,61(1-2, Serial No. 246). Harter, S. (1996). Teacher and classmate influences on scholastic motivation, selfesteem, and level of voice in adolescents. In J. Juvonen & K. R. Wentzel (Eds.), Social motivation: Understanding children's school adjustment (pp. 11-42). New York: Cambridge University Press. Henwood, K., & Pidgeon, N. (2003). Grounded theory in psychological research. In P. M. Camic, J. E. Rhodes, & L. Yardley (Eds.), Qualitative research in psychology: Expanding perspectives in methodology and design (pp. 131-155). Washington, DC: American Psychological Society. Hofer, B. K., & Pintrich, P. R. (1997). The development of epistemological theories: Beliefs about knowledge and knowing and their relation to learning. Review of Educational Research, 67, 88-140. Huxtable, M. (2009). Creating inclusive and inclusional understandings of gifts and talents through living educational theory research. In T. Balchin, B. Hymer, & D. J. Matthews (Eds.), The Routledge international companion to gifted education (pp. 292-298). London: Routledge. Hymer, B. J. (2009). Beyond compare? Thoughts towards an inclusional, fluid and nonnormative understanding of giftedness. In T. Balchin, B. Hymer, & D. J. Matthews (Eds.), The Routledge international companion to gifted education (pp. 299-307). London: Routledge. Hymer, B., Whitehead, J., & Huxtable, M. (2009). Gifts, talents and education: A living theory approach. Chichester, UK: Wiley-Blackwell. Jackson, N. E. (2000). Strategies for modeling the development of giftedness in children. In R. C. Friedman & B. M. Shore (Eds.), Talents unfolding: Cognition and development (pp. 2754). Washington, DC: American Psychological Association. Kitchener, K. S. (2002). Skills, tasks, and definitions: Discrepancies in the understanding and data on the development

of folk epistemology. New Ideas in Psychology, 20, 309-328. Kuhn, D., Cheney, R., & Weinstock, M. (2000). The development of epistemological understanding. Cognitive Development, 15, 309-328. Mansfield, A. F., & Clinchy, B. McV. (2002). Toward the integration of objectivity and subjectivity: Epistemological development from 10 to 16. New Ideas in Psychology, 20, 225-262. Mascolo, M. F., Li, J., Fink, R., & Fischer, K. W. (2002). Pathways to excellence: Value presuppositions and the development of academic and affective skills in educational contexts. In M. Ferrari (Ed.), The pursuit of excellence through education (pp. 113-146). Mahwah, NJ: Erlbaum. McKeough, A. (1992). A neo-structural analysis of children's narrative and its development. In R. Case, The mind's staircase: Exploring the conceptual underpinnings of children's thought and knowledge (pp. 171-188). Hillsdale, NJ: Lawrence Erlbaum. McKeough, A., & Genereux, R. (2003). Transformation in narrative thought during adolescence: The structure and content of story compositions. Journal of Educational Psychology, 95, 537-552. McKeough, A., Genereux, R., & Jeary, J. (2006). Structure, content, and language usage: How do exceptional and average storywriters differ? High Ability Studies, 17, 203-223. McKeough, A., Okamoto, Y., & Porath, M. (2002, April). A design for development: The legacy of Robbie Case. Paper presented at the American Educational Research Association, New Orleans. McKeough, A., & Sanderson, A. (1996). Teaching storytelling: A microgenetic analysis of developing narrative competency. Journal of Narrative and Life History, 6, 157-192. Muis, K. R. (2007). The role of epistemic beliefs in self-regulated learning. Educational Psychologist, 42, 173-190. Okamoto, Y., Curtis, R., Jabagchourian, J. J., & Weckbacher, L. M. (2006). Mathematical precocity in young children: A neo-Piagetian perspective. High Ability Studies, 17, 183-202. Pellegrino, J. W., Chudowsky, N., & Glaser, R. (Eds.) (2001). Knowing what students know. The science and design of educational assessment. Washington, DC: National Academy Press. Perner, J. (1991). Understanding the representational mind. Cambridge, MA: MIT Press. Porath, M. (2006). The conceptual underpinnings of giftedness: Developmental and educational implications. High Ability Studies, 17, 145-157. Porath, M., & Lupart, J. (2007, May). The child as psychologist: Gifted students making meaning of learning and knowledge. Paper presented at the Canadian Society for the Study of Education, University of Saskatchewan. Porath, M., & Lupart, J. (in press). Gifted childrens representations of learner identities. Exceptionality Education International.

66

M. Porath et al.

Robinson, N. M., Zigler, E., & Gallagher, J. J. (2000). Two tails of the normal curve: Similarities and differences in the study of mental retardation and giftedness. American Psychologist, 55, 1413-1424.

Ziegler, A., Stoeger, H., & Mundi, M. (2004). The epistemic learner model. In Tourn, J. (Ed.), Educational technology for gifted education: From information age to knowledge era (pp. 13-20). European Council for High Ability: Pamplona, Spain.

The authors
Marion Porath is a Professor in the Department of Educational and Counselling Psychology, and Special Education in the Faculty of Education at the University of British Columbia, Vancouver, Canada. Her research interests include different forms of giftedness, young childrens social development, instructional applications of developmental theory, and problem-based learning. Her current research focuses on gifted and gifted/learning disabled childrens and adolescents' views of themselves as learners and their understanding of teaching and learning and the roles of intrapersonal and interpersonal competencies in pedagogy.

Judy L. Lupart is a Professor and Canada Research Chair in Special Education, beginning January 2003, in the Department of Educational Psychology at the University of Alberta. She has served as the Founding Director of the University of Calgary Centre for Gifted Education, and the Founding Editor of the journal Exceptionality Education Canada. Her research, publication and teaching interests include inclusive education and school transformation; cognition, transfer and educational applications; learning disabilities, giftedness and at-risk learners; and girls, women and achievement.

Dr. Jennifer Katz has recently joined the Department of Educational Administration, Foundations and Psychology in the Faculty of Education at the University of Manitoba, Canada. Dr. Katz received her PhD in 2008 from the University of British Columbia and is currently writing a book on Teaching to Diversity: Universal Design for Learning and Inclusive Learning Communities for Pearson Education Press. Dr. Katz has been a successful special education teacher, sessional lecturer, educational consultant, classroom teacher, editor, and guidance counselor. Following her early years as an elementary teacher, her work as an advocate of inclusive education is evident by her written works and conference presentations. Dr. Katz brings to the faculty broad research interests in the areas of: Inclusive special education, teaching to diversity, assessment, and social and emotional learning development. Constantine Ngara (PhD, M. Ed, PGDE, B. Ed.) is an Assistant Professor at University of Bahrain Teachers College. In his doctoral thesis (at University of British Columbia), Constantine studied Shona stone sculptors of Zimbabwes talent attributions and proposed a dynamic interactive process model (DIPM) that explains talent development from an African perspective espoused in Shona culture. His research agenda is to understand giftedness and creativity from a cultural perspective with a view to inform gifted education from a sociocultural perspective. His collaborative work and publications focus on conceptions of giftedness and creativity.

Pamela Richardson is a Doctoral Candidate in Special Education (High Ability) at the University of British Columbia in Vancouver, British Columbia, Canada, and an Instructor in the Faculty of Education at Vancouver Island University in Nanaimo, British Columbia, Canada. In her research, she employs arts-based, narrative and poetic inquiry methods to explore experiences and representations of giftedness within actual and imaginary contexts.

Talent Development & Excellence Vol. 1, No. 1, 2009, 67-78

The Leonardo Lab

67

The Leonardo Laboratory: Developing Targeted Programs for Academic Underachievers with Visual-Spatial Gifts
Tina M. Newman1, William Brown2, Lesley Hart1, Donna Macomber1, Niamh Doyle3,#, Sergey A. Kornilov4, Linda Jarvin5,#, Robert J. Sternberg5,# and Elena L. Grigorenko1,4,6*
Abstract: This article presents a short-term program, The Leonardo Laboratory, for children with coexisting learning disabilities and spatial gifts. The program was developed in collaboration with a museum and administered as an after-school program to a group of elementary-school children (20 boys and 5 girls) so that its effectiveness could be evaluated preliminarily. Although the students in the program did not demonstrate statistically significant gains in academic skills, they demonstrated gains in self-efficacy and improvements in organizational skills. Thus, the program we developed has promise for certain kinds of skills and attitudinal development. Our work shows the importance of addressing the needs of children with disabilities and gifts (twice exceptional children). Kewords: Twice exceptional children, learning disabilities, spatial gifts, after-school program

Students with coexisting learning disabilities and special talents (twice exceptional children) are an often under-identified and under-served segment of the school-age population (Baum, 2004; Dix & Schafer, 2005; Kalbfleisch & Iguchi, 2008). Although some of these students are provided services for either their gifts or their learning difficulties, very few are identified for services that both develop their areas of weakness and allow them to explore their areas of strength (Bianco, 2005; Brody & Mills, 1997). This oversight may have significant consequences, both direct and indirect, on students opportunities to succeed in careers that utilize their areas of strength (Johnsen & Kendrick, 2005). Directly, the students have little or no opportunity to develop their abilities and learn to compensate for their disabilities. Indirectly, this lack of services may create a lessened sense of self-efficacy. Ultimately, successfully intelligent individuals both capitalize on strengths and compensate for or correct weaknesses, so instruction should address both strengths and weaknesses (Sternberg & Grigorenko, 2007). Given that these students require services both for their gifts and for their learning disabilities, there have been a number of documented interventions aimed at developing services specifically for this group of students. Predominantly, these programs are described in literature that covers case studies and anecdotal reports (e.g., Hua, 2002; Thrailkill, 2005; Turk & Campbell, 2005; Volker, Lopata, & Cook-Cottone, 2006); yet, there are a few illustrations of small-scale group-administered programs (see below). In terms of their objectives, these programs can be classified into three groups. First, some are designed to address specific weaknesses these students possess. For example,
1

Yale University, Child Study Center, 230 South Frontage Road, New Haven, CT 065191124,USA 2 The Eli Whitney Museum, USA 3 Syracuse University, USA # At Yale University during the completion of the Leonardo Lab project 4 Moscow State University, Russia 5 Tufts University, USA 6 Columbia University, USA * Corresponding author. Email: elena.grigorenko@yale.edu
ISSN 1869-0459 (print)/ ISSN 1869-2885 (online) 2009 International Research Association for Talent Development and Excellence http://www.iratde.org

68

T. M. Newman et al.

one program utilizes a process-based remediation strategy to improve the reading skills of three students who have both gifts and learning disabilities (Crawford & Snart, 1994). Second, other programs address areas of weakness but are primarily designed to develop areas of strength through enrichment activities (e.g., Baum, 1988; Cooper, Baum, & Neu, 2004). One such intervention, Project High Hopes, sought to identify and develop scientific talents in students with special needs (Cooper et al. 2004). Finally, a number of programs are designed to develop areas of strength while remediating weaknesses (e.g., Bees, 1998; Weinfeld, Barnes-Robinson, Jeweler, & Shevitz, 2002). For example, the Wings program (Weinfeld et al., 2002) was developed to ensure that students with both gifts and learning disabilities have access to an enriched and accelerated curriculum, while also receiving instruction, adaptations, and accommodations in their areas of weakness through a cascade of services from integrated to self-contained classes, depending on each students needs. In addition to focusing on academic and cognitive skills, many programs acknowledge an additional goal of increasing student self-efficacy and self-regulation. Both the literature on students with gifts and the literature on students with learning disabilities describe children who are at risk for socio-emotional difficulties, such as poor self-concept and high levels of frustration and anxiety (Baldwin, 1999; Dole, 2000). Thus, a significant benefit can be provided by programs designed to increase students self-efficacy. While limited in number and scale, the literature on interventions for twice exceptional students provides some generalizations regarding advisable approaches to educating such students. One focal finding has been the importance of providing these students with a curriculum relevant to their gifts and talents (Baum, Cooper, & Neu, 2001; Neilsen & Mortoff-Albert, 1989; Weinfeld et al., 2002). The benefits of gifted-and-talented programming have been seen in both academic achievement and self-concept. For example, in a study of 76 grade 3 to 5 twice exceptional students, researchers (Neilsen & Mortoff-Albert, 1989) found that those receiving a combination of gifted-and-learningdisability services or only gifted services reported higher self-concept than those students receiving interventions exclusively focused on remediating weaknesses. Another finding in the literature has highlighted the importance of having an individualized education plan that sets both strength-building and weakness-correcting goals (Baum et al., 2001; Cooper et al., 2004; Shevitz, Weinfeld, Jeweler, & BarnesRobinson, 2003; Weinfeld et al., 2002). For example, the Wings program ensures that students have individualized and appropriate instruction and accommodations in their area of disability, while securing access to gifted programs or mentorships depending upon need (Weinfeld et al., 2002). Pedagogical programs for students with both gifts and learning disabilities have not been rigorously evaluated. Yet, the preliminary evidence suggests that approaches addressing both strengths and weaknesses appear to have a positive impact on students attitudes toward school and their commitment to academic work. Prompted by the success of programs for students with learning disabilities whose gifts are in the analytical domain, and recognizing that strengths in other areas can promote success in life, we developed and implemented the Leonardo Laboratory program. This program was designed to address the creative and practical gifts of children who experience difficulties in academic subjects, but who have strong spatial abilities. Gifts that allow people to be successful in life can be found in a variety of domains, including strengths in visual-spatial skills or creativity and innovation (Aaron, Joshi, & Ocker, 2004; Mann, 2006; VonKarolyi & Winner, 2004). One goal of the Leonardo Laboratory program was to develop an enrichment curriculum that allows students to develop their identified strengths while at the same time identifying and remediating areas of weakness (Sternberg & Grigorenko, 2007). A second goal was to increase self-efficacy and enhance the self-regulation of these students, whose combination of gifts and disability has often not led to the development of adequate related skills.

The Leonardo Lab

69

The present project was designed to (1) identify students who have academic difficulties in reading, spelling, or math, while also displaying exceptional abilities in the creative arts (e.g., drawing, painting, or design) or in practical/spatial skills (e.g., model building); (2) develop an after-school program designed to address both the strengths and weaknesses of these students; and to (3) preliminarily evaluate the effectiveness of this program. Thus, we wanted to appraise findings from prior case-study and small-group research suggesting that when children with both gifts and learning disabilities are provided with a program that addresses their strengths, they demonstrate growth in areas of weakness that are targeted within the program (e.g., task approach and follow-through), even though they might not demonstrate improvement in areas of weakness that are not targeted (e.g., reading-related and math skills). We also wanted to validate a claim from prior research that students with both gifts and learning difficulties demonstrate growth in self-efficacy and self-regulation when provided with an intervention program that highlights their strengths and gives them opportunities to meet other students with similar learning profiles. To summarize, this project was conceived to develop an intervention for children with coexisting spatial gifts and difficulties in reading/writing and/or math. In this article, we describe the intervention, comment on the feasibility of its implementation in after-school settings, and present some pilot data reflecting the promise of the intervention. Of note is that, although there are no formal estimates of the prevalence of double exceptionalities, the literature suggests that the number of twice exceptional children is low. Given this fact, recruiting a second group of twice exceptional children to serve as controls was not feasible, while a group of more typical students would not serve as adequate controls, being drawn from too different a population. Hence our study was done as a preliminary study only to assess whether there were any effects apparently associated with the program, with plans for a future controlled study if these minimal conditions showed any effects.

Method
Participants The participating students were recruited through a brochure sent out to local elementary schools and families who had previously participated in programs at the Eli Whitney Museum (http://www.eliwhitney.org/) asking teachers and parents to nominate students in grades 4 through 6 who demonstrated, simultaneously, both learning difficulties and talents in drawing or building activities. A total of 33 students were nominated by their teachers and parents; of those, 30 students attended an evaluation session. Of these students, 11 were in grade 4, 14 in grade 5, 3 in grade 6, and 2 were being homeschooled. There were 24 male students and 6 female students, and the average age of the participants was 10 years 2 months (SD=.81). The program inclusion criteria were (1) reading, writing, or mathematics difficulties and (2) talent in at least one other area of endeavor (building, practical problem-solving, drawing). Of the 30 students assessed, one was reported as not having any learning difficulties at school, and one was not interested in the initial building project, and therefore deemed not well matched to the program. A total of 28 students began the program; 25 students completed the program. Table 1 presents the group by age, gender, and ethnicity.
Table 1. Means and Standard Deviations of Age, Grade, Gender, and Ethnicity of Program Participants

Age
M 10.12 SD 0.67 M

Grade
SD 0.71 4.60

Gender
Male 20 Female 5 Asian 4%

Ethnicity
African-Amer 12% Caucasian 84%

70

T. M. Newman et al.

Project Site The intervention program was designed in collaboration with and implemented by the Eli Whitney Museum (http://www.eliwhitney.org/). The Museum is founded on the ideas of industrial revolutionist Eli Whitney and seeks to encourage the resourcefulness, innovation, and entrepreneurship exhibited by its namesake through a unique tradition of learning via self-guided, hands-on problem-solving and experiences beyond the conventional classroom. The Museum offers projects that begin in play and grow into art, architecture, engineering and science, and they include Japanese Toys and Technology, Rubber-Band Powered Cars, Aeromodeling, and Boat Building. The intervention program at the Eli Whitney Museum featured here was based on the works of Leonardo DaVinci and was led by the director of the museum, William Brown. Procedure To collect pilot data on the outcomes of the intervention, participants were assessed before beginning the Leonardo Laboratory program (Time 1, Pretest), then re-evaluated again at the completion of the program (Time 2, Posttest). The tests used were groupadministered measures of nonverbal reasoning, math calculation, phonological and orthographical knowledge, and perception of task approach and follow-through skills. For performance assessment, students were asked to build a rubber-band powered car and their projects were assessed on two dimensions: (1) practical problem-solving and (2) creativity. In addition, parents completed a rating scale assessing their childs task approach and follow-through skills. Targeted Skills In designing the intervention program, the primary goal was to focus on students strengths while developing areas of weakness. In a recent surge of research in developing self-regulated learning skills for children with learning disabilities (Harris, Reid, & Graham, 2004), self-regulation and the related concept of metacognition are described as encapsulating a number of skills, including (1) setting goals; (2) developing strategies to meet those goals (planning) (Butler, 1998); (3) consciously monitoring progress (Butler & Winne, 1995); (4) adjusting approaches as required (Butler & Winne, 1995); and (5) using motivation to cope with setbacks or obstacles (Corno, 1993). There is evidence that students with learning disabilities in postsecondary education continue to have difficulties with these task-approach and follow-through strategies, and that those college students with learning disabilities who employ self-regulation strategies are more successful (Butler, 1998; Trainin & Swanson, 2005). The purpose of this intervention was to help students develop these self-regulation strategies specifically task approach and follow through while they completed projects that capitalized on their strengths and helped them practice compensating for weaknesses, including elements of reading/writing and doing math. The program also intended to increase childrens self-efficacy. Materials Pretest measures included (1) a measure of nonverbal cognitive skills, (2) baseline measures of academic skills that were areas of difficulty for students as specified in the recruitment inclusion/exclusion criteria, and (3) a performance-based measure of practical problem-solving and creativity. For the academic skills, specifically, we assessed: (1) reading and reading-related skills (phonological processing, orthographical processing, and print exposure); (2) math calculation skills; and (3) achievement as indicated by a parental rating. We also collected measures on target areas that are often challenging for students with learning difficulties, including parent and child ratings of task approach and follow-through skills. We also administered a hands-on performance

The Leonardo Lab

71

measure to assess students potential areas of strengths, including visual/spatial building skills, practical problem solving, and creative skills. At posttest, we collected academic skill measures, including the reading and readingrelated skills, math calculation, and student and parent ratings of task-approach and follow-through skills. A qualitative feedback measure was also obtained. Nonverbal ability. The Cattell Test of g: Culture Fair Scale 2, Form B (Cattell & Cattell, 2002) is a measure of nonverbal reasoning in which the student analyzes visual patterns, for example, determining missing parts or finding similar patterns. This test was administered at Time 1 to provide a benchmark level of cognitive functioning and to ensure that all students were of average to above average ability in nonverbal reasoning. Scores on this measure indicated that all students had average to superior ability: the average score for the group was 114.84 (SD=14.30). Academic skills measures. The Title Recognition Task (Cunningham & Stanovich, 1990) gauges students' exposure to print. This test, consisting of a list of real and false book titles, was originally designed to predict reading experience in third- and fourth-grade children, but has been correlated with other reading-related skills in students in grades 5 through 9 (McBride-Chang, Manis, Seidenberg, Custodio, & Doi, 1993). Participants were instructed to put a mark next to the titles that they knew were real books. In addition, they were told that the list contained both real and false titles and authors, and therefore they were not to guess. This test has two versions and administration was randomly assigned at pretest, with the other version administered at posttest. Both versions have 25 real books; version 1 has 15 false titles, version 2 has 16 false titles. Scoring of this measure consisted of the percentage of real book titles selected minus the percentage of false titles selected. The same measure was administered at both pretest and posttest. The Phonological Choice Task (Gayan & Olson, 2001) is a measure of phonological-processing skill consisting of forced-choice decisions between pairs of nonwords that sound like common words if read aloud and two nonwords that do not sound like known words. Students were asked to circle the word in each row that sounded like a real word if read aloud. There are five practice items and 60 test items that were randomly divided into 2 sets of 30 items to create A and B versions of the test. Students randomly received either the A or B version at pretest and were given the alternate version at posttest. Scoring of this measure consisted of the percentage of correct responses. The same measure was administered at both pretest and posttest. The Orthographic Choice Task (Olson, Forsberg, Wise, & Rack, 1994) is a measure of orthographic-processing skill consisting of a forced-choice decision between a word and a phonetically identical pseudohomophonic nonword (e.g., take taik). The task requires the participant to recognize the correct orthographic pattern for the word independent of its phonology. There are five practice items and 78 test items that were randomly divided into 2 sets of 39 items to create an A and B version of the test. Students randomly received either the A or B version at pretest and were given the alternate version at posttest. Scoring of this measure consists of the percentage of correct responses. The same measure was administered at both pretest and posttest. The Woodcock-Johnson (WJ-III) Math Calculation Subtest (Woodcock, McGrew, & Mather, 2001) provides a measure of a students mathematical-computation ability. Questions progress through single-, double-, and triple-digit addition, subtraction, multiplication, and division. Harder items require students to perform calculations with fractions, decimals, and negative numbers. Students in the study were asked to complete as many questions as they could. Standard scores were obtained. The same measure was administered at both pretest and posttest. Self-regulated learning measures. The Organizational Rating Scale (Parent-Report) is a measure developed for this project that asks parents to rate their childs behavior on a 5point scale. The measure consists of 33 items assessing task-approach and follow-through skills such as planning, preparing, overcoming obstacles, and independence in completing home chores, school tasks, and extra-curricular tasks. This measure was administered at both pretest and posttest. The Organizational Rating Scale (Child-Report) is

72

T. M. Newman et al.

a measure developed for this project that asks students to rate their own behavior on a 5point scale. The scale consists of 36 items measuring self-perception of task approach and follow-through skills such as planning, preparing, and completing tasks at home and school. This measure was administered at both pretest and posttest. Performance measures. The Rubber-band-powered car task, developed by the Eli Whitney Museum, gives students the parts and instructions for building a basic car powered by rubber-bands. The task was assessed on two factors (1) function (practical) and (2) form (creative). For the functional assessment, the student had to compete to see how far their car could travel compared with the cars of other students. The students were given trial and revision opportunities for testing their car. These trial and revision opportunities were limited in time but not in number, so students could try their car and revise as many times as they chose during a time period that was set for everyone. For the form assessment, students were provided with resources to make their car unique. Qualitative measures. Qualitative feedback was solicited from both parents and students following the completion of the program. Parents were given a feedback form asking three questions: (1) please describe any benefits of the program for your child; (2) please describe any negative outcomes of the program for your child; and (3) please provide any additional information that you feel may be important for us in evaluating this program. Students were also given a feedback form with three questions: (1) please describe three things that made this program good for you; (2) please describe anything about the program that was not so good for you; and (3) please describe anything else about the program that you think was great or not so great. Intervention General overview. The program included a series of ten workshops that invited students to complete experimental building projects derived from Leonardo DaVincis notebooks. Using activities initially sketched by DaVinci and further developed for the purposes of this intervention, the program integrated exercises developing important learning skills such as: (1) following the demands of the instructor while making projects on their own; (2) coping with environmental constraints, such as availability of materials and tools; (3) progressing through steps of the projects, i.e., planning, preparation, time scheduling, exploring, and follow-through; and (4) specific skills in following directions, overcoming obstacles, and recognizing their own successes. Specific application of the goals. Each assignment allowed students to take responsibility and realize the choices available in directing or steering the project. It also acknowledged the restrictions that can be placed on an assignment by facilitators (i.e., museum staff members), reminding the student that in most learning situations, there are boundaries that must be considered. Each project was also subject to environmental constraints, such as resources and time constraints that can push an endeavor forward, backward or in a different direction. At the start of each project, students were provided with the basic materials and instructions for completing the project. A demonstration of the steps for completing the basic model and examples of different ways of individualizing the project were provided. Additional resources were provided (e.g., bins of markers, scraps of wood, pieces of cloth, pipe cleaners, pieces of wire), and students were reminded that they could use any other resources found at the museum. During project building, facilitators circulated among the students, asking them about their plans, how they were going to reach their goal, and what they were going to need. The facilitators also reminded the students to consider the time available. No specific ideas were provided to the students, although they were free to talk amongst themselves and were encouraged to make the project their own. Each obstacle was presented as an opportunity for choice, a time for decision-making, or a lesson to be learned. In deciding how to overcome obstacles, students were supported by choices such as back-tracking and finding alternate ideas, methods, and materials. At the end of each session, students were taken aside individually and encouraged to consider the obstacles they had

The Leonardo Lab

73

encountered and overcome, the opportunities that had arisen, the opportunities captured, the set-backs faced, the discoveries or successes, the disappointments, and the lessons learned in their projects.

Results
Descriptive Statistics To determine any significant effects of age and gender on the results, correlations were calculated between these three variables and the rating scales and the academic-skills measures. No significant correlations were found for age or grade with the Child-Report measure, the Parent-Report measure, or any of the academic-skills measures (phonological, orthographical, title recognition, or calculation) at either pretest or posttest. Intervention Outcomes The Parent-Report Organizational Rating Scale and The Child-Report Organizational Rating Scale were two new measures developed for this study. As these rating scales were new, we wanted to ensure their reliability across time. The child and parent scales showed a high degree of consistency with each other, indicating that children and parents tended to agree on the abilities of the children. With respect to the internal consistency of the items, the Child Rating Scale had a Cronbachs alpha of 0.88 and the Parent Rating Scale had a Cronbachs alpha of 0.91, demonstrating a high degree of internal consistency. Given the pre/post design of this pilot study and the small sample size, paired-sample ttests were conducted to determine any significant differences between the pre- to posttest academic skills and the pre/post child and parent-ratings on task approach and follow-through skills. Academic skills assessed by group-administered tests of reading-related skills and math calculation skills showed no significant differences from pre to post test, neither on the omnibus, nor on individual tests. As these were not areas of intervention, this finding was expected (see Table 2 for descriptive and inferential statistics).

Table 2. Means, Standard Deviations, and T-test Data of Pretest and Posttest Measures of Readingand Math-Related Skills

N
M

Pretest
SD 0.22 0.25 0.12 14.69

Posttest
M 0.27 0.65 0.85 95.39 SD 0.18 0.27 0.13 15.64

Difference
t-score -1.00 -0.19 -0.17 -0.85 p-value 0.33 0.85 0.86 0.40

Title recognition task Phonological choice task Orthographic choice task Math calculation task

21 18 22 22

0.21 0.64 0.84 93.06

The parent-rating scale and child-rating scale assessed skills that were targeted in the intervention, and therefore, differences between pre and post ratings of parents and childrens perceptions of task approach and follow-through skills were expected. The parent rating scale showed a significant difference between pre and post ratings, with

74

T. M. Newman et al.

scores rising following the intervention, indicating that, overall, parents saw an improvement in their childrens task-approach and follow-through skills. In particular, the following items showed a significant positive change or approached a significant change in rating from pretest to posttest: (1) overcoming obstacles in completing a homework task; (2) leaving the house on time for school with minimal prompting; (3) following multiple steps to complete a household chore; (4) overcoming obstacles in completing a household chore; and (5) following instructions to complete a one-step household chore. The child-rating scale demonstrated no comparable rise (see Table 3), however, given that parents are the ones who provide all the support and/or reminders, they are likely the ones who notice change first. Of note also are the dynamics of the correlations between parent and child ratings at pre- and posttests (see Table 4).

Table 3. Means, Standard Deviations, and T-test Data of Pretest and Posttest Parent and Child Rating Scales of Task Approach and Follow-Through Skills

N
M

Pretest
SD 12.78 19.09

Posttest
M 59.84 126.15 SD 2.37 16.20

Difference
t-score -2.41 -0.62 p-value 0.03 .55

Parent Rating Scale Child Rating Scale

17 13

56.42 122.62

Table 4. Correlations of the Pre and Post Administrations of the Parent and Child Rating Skills

Pre Parent Pre Parent Post Parent Pre Child Post Child 0.89*** 0.72** 0.30

Post Parent

Pre Child

0.65* 0.32

0.67*

*** p < .0001, ** p < .01, * p < .05

Qualitative feedback from parents converged on three themes: (1) the development of areas of weakness, for example, he is more willing to sit down and do homework, less resistant, more persevering; (2) the development of areas of strength, for example, he continued to work on each project when he got home; and (3) the development of selfefficacy, for example, he doesnt shrug off his accomplishments. In general, parent qualitative feedback confirmed the findings of the parent rating scale that children showed improvement in areas of weakness. In addition, parents also indicated that children experienced benefits to their self-efficacy and developed further their areas of strength. Qualitative feedback from the students reflected primarily themes of selfefficacy, for example, showed me I could do good things that people love and want.

The Leonardo Lab

75

Discussion
The results of the current study are consistent with the literature that suggests that students who have both gifts and learning disabilities benefit from programs targeted to developing their areas of strength. Although very preliminary and based only on informants reports, this study provides some quantification of positive effects on the improvement of targeted skills by the intervention. The students who participated in this study diverged from those in previous studies in the fact that the focus was not on students who demonstrated cognitive gifts, but on those whose gifts were in the creative and visual-spatial domains. The intervention offered also differed in that students were delivered a program out of the typical school curriculum, focused on developing not only their strengths but also on areas of weakness that impact school performance. In particular, the intervention targeted skills in task initiation and follow-through, areas often found to be weak in students with learning difficulties. To quantify these skills, new measures were devised to obtain ratings from parents and students on the application of these skills from the intervention setting to outside settings, specifically, the home environment and especially homework tasks. The new measures offered a way to capture parents perceptions of students strengths and weaknesses in task approach and follow-through. Parents reported significant improvement on a number of items. Finally, parent and student qualitative statements corroborated these findings. One of the most important findings reported by both parents and students was the growth in self-efficacy and self-recognition of an area of strength. At pretest, many of these students were reported as experiencing social and emotional difficulties, including anxiety and sadness. Parents reported concerns that their children were exhibiting indications of learned helplessness in their school environments and were demonstrating less effort in their schoolwork. At the time of the posttest, both parents and students reported a re-engagement in goal-directed tasks and a sense that the student could accomplish something worthwhile. One indicator of this programs positive benefit to these students is the number of students who have continued to engage in programs at the Eli Whitney museum as both students and, later, facilitators, after first being introduced to the Leonardo Laboratory. Yet, although this work resulted in the development of a new and engaging program, the data we present here are only preliminary and, though encouraging, should be interpreted with caution. To validate and strengthen our claims about the potential of the Leonardo Laboratory, a number of steps need to be taken. First, a formal effectiveness study with a large sample of participants and a control group needs to be carried out. Second, a battery of performance-based assessments capturing the skills targeted by the program needs to be administered using the pre-/posttest design. Here we used convergent information, provided by parents and the students themselves, but, as indicated above, this is only the first step in carrying out an evaluation of the program. Clearly, other adults, such as students teachers and museum personnel can be included as additional informants. The positive feedback from parents and students might be indicative of wishful thinking for change, rather than change itself. Finally, the relationships that a number of the students subsequently built with the museum after their participation in the Leonardo Laboratory, although encouraging, should also be interpreted with caution: their comfort and satisfaction with the atmosphere of the museum might not be indicative of learning and change. In summary, this report is only an initial presentation of a promising intervention program for children who have coexisting gifts and disabilities. With the program developed, the next step is to formally evaluate it. Hopefully, there will be future opportunities for that step to be made.

76

T. M. Newman et al.

Acknowledgements
This research was supported by Grant R206R950001 under the Javits Act program as administered by the Office of Educational Research and Improvement, U.S. Department of Education. Grantees undertaking such projects are encouraged to express freely their professional judgment. This article, therefore, does not necessarily represent the position or policies of the U.S. Department of Education, and no official endorsement should be inferred. We express our gratitude to Ms. Mei Tan for her editorial assistance.

References
Aaron, P. G., Joshi, R. M., & Ocker, E. S. (2004). Dyslexia and visual-spatial talents: Are they connected? In T. M. Newman & R. J. Sternberg (Eds.), Students with both gifts and learning disabilities: Identification, assessment, and outcomes (pp. 199-234). New York, NY: Kluwer Academic/ Plenum Publishers. Baldwin, A. Y. (1999). Learning disability: The mysterious mask - The USA perspective. In A. Y. Baldwin & W. Vialle (Eds.), The many faces of giftedness (pp. 103-134). Belmont, CA: Wadsworth Publishing. Baum, S. M. (1988). An enrichment program for gifted learning disabled students. Gifted Child Quarterly, 32, 226-230. Baum, S. M. (Ed.). (2004). Twice-exceptional and special populations of gifted students. Thousand Oaks, CA: Corwin Press. Baum, S. M., Cooper, C. R., & Neu, T. W. (2001). Dual differentiation: An approach for meeting the curricular needs of gifted students with learning disabilities. Psychology in the Schools, 38, 477-490. Bees, C. (1998). The GOLD program: A program for gifted learning disabled adolescents. Roeper Review, 21, 155-161. Bianco, M. (2005). The effects of disability labels on special education and general education teachers referrals for gifted programs. Learning Disability Quarterly, 28, 285-293. Brody, L. E., & Mills, C. J. (1997). Gifted children with learning disabilities: A review of the issues. Journal of Learning Disabilities, 30, 282-296. Butler, D. L. (1998). The strategic content learning approach to promoting self-regulated learning. In B. J. Zimmerman & D. Schunk (Eds.), Developing self-regulated learning: From teaching to self-reflective practice (pp. 160-183). New York, NY: Guilford Press. Butler, D. L., & Winne, P. H. (1995). Feedback and self-regulated learning: A theoretical synthesis. Review of Educational Research(65), 245-281. Cattell, R. B., & Cattell, A. K. S. (2002). Culture Fair Intelligence Tests. Champaign, IL: Institute for Personality and Ability Testing, Inc. Cooper, C. R., Baum, S. M., & Neu, T. W. (2004). Developing scientific talent in students with students with special needs: An alternative model for identification, curriculum, and assessment. The Journal of Secondary Gifted Education, 15, 162-169. Corno, L. (1993). The best laid plans: Modern conception of volition and educational research. Educational Researcher, 22, 14-22. Crawford, S., & Snart, F. (1994). Process-based remediation of decoding in gifted LD students: Three case studies. Roeper Review, 16, 247-252. Cunningham, A. E., & Stanovich, K. E. (1990). Assessing print exposure and orthographic processing skill in children: A quick measure of reading experience. Journal of Educational Psychology, 82, 733-740. Dix, J., & Schafer, S. (2005). From paradox to performance: Practical strategies for identifying and teaching Gifted/LD students. In S. K. Johnsen & J. Kendrick (Eds.), Teaching gifted students with disabilities (pp. 153-159). Waco, TX: Prufrock Press. Dole, S. (2000). The implications of the risk and resilience literature for gifted students with learning disabilities. Roeper Review, 23, 9196. Gayan, J., & Olson, R. K. (2001). Genetic and environmental influences on orthographic and phonological skills in children with reading disabilities. Developmental Neuropsychology, 20, 483-507. Harris, K. R., Reid, R. R., & Graham, S. (2004). Selfregulation among students with LD and ADHD. In B. Wong (Ed.), Learning about learning disabilities (3rd ed., pp. 167-195). San Diego, CA: Elsevier Academic Press. Hua, C. B. (2002). Career self-efficacy of the student who is gifted/learning disabled: A case study. Journal for the Education of the Gifted, 25, 375-404. Johnsen, S. K., & Kendrick, J. (Eds.). (2005). Teaching gifted students with disabilities. Waco, TX: Prufrock Press. Kalbfleisch, M. L., & Iguchi, C. M. (2008). Twiceexceptional learners. In J. A. Plucker & C. M. Callahan (Eds.), Critical issues and practices in gifted education: What the research says (Vol. 707-719). Waco, TX: Prufrock Press. Mann, R. L. (2006). Effective teaching strategies for gifted/learning-disabled students with spatial strengths. The Journal of Secondary Gifted Education, 17, 112-121. McBride-Chang, C., Manis, F. R., Seidenberg, M. S., Custodio, R. G., & Doi, L. M. (1993). Print

The Leonardo Lab

77

exposure as a predictor of word reading and reading comprehension in disabled and nondisabled readers. Journal of Educational Psychology(85), 230-238. Neilsen, M. E., & Mortoff-Albert, S. (1989). The effects of special education service on the self-concept and school attitude of learning disabled/gifted students. Roeper Review, 12, 29-36. Olson, R. K., Forsberg, H., Wise, B., & Rack, J. (1994). Measurement of word recognition, orthographic, and phonological skills. In G. R. Lyon (Ed.), Frames of reference for the assessment of learning disabilities: New views on measurement issues (pp. 243-277). Baltimore, MD: Paul H. Brookes. Shevitz, B., Weinfeld, R., Jeweler, S., & BarnesRobinson, L. (2003). Mentoring empowers gifted/learning disabled students to soar! Roeper Review, 26, 37-40. Sternberg, R. J., & Grigorenko, E. L. (2007). Teaching for successful intelligence (2nd ed.). Thousand Oaks, CA: Corwin Press. Thrailkill, C. (2005). Patrick's story: A gifted/learning-disabled child. In S. K. Johnsen & J. Kendrick (Eds.), Teaching gifted students with disabilities (pp. 99-103). Waco, TX: Prufrock Press. Trainin, G., & Swanson, H. L. (2005). Cognition, metacognition, and achievement of college

students with learning disabilities. Learning Disability Quarterly, 28, 261-262. Turk, T. N., & Campbell, D. A. (2005). What's wrong with Doug? The academic struggles and triumphs of a gifted student with ADHD from preschool to college In S. K. Johnsen & J. Kendrick (Eds.), Teaching gifted students with disabilities (pp. 105-141). Waco, TX: Prufrock Press Volker, M. A., Lopata, C., & Cook-Cottone, C. (2006). Assessment of children with intellectual giftedness and reading disabilities. Psychology in the Schools, 43, 855-869. VonKarolyi, C., & Winner, E. (2004). Dyslexia and visual-spatial talents: Are they connected? In T. M. Newman & R. J. Sternberg (Eds.), Students with both gifts and learning disabilities: Identification, assessment, and outcomes (pp. 95-118). New York, NY: Kluwer Academic/ Plenum Publishers. Weinfeld, R., Barnes-Robinson, L., Jeweler, S., & Shevitz, B. (2002). Academic programs for gifted and talented/learning disabled students. Roeper Review, 24, 226-233. Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock-Johnson III Tests of Achievement. Itasca, IL: Riverside Publishing.

The authors
Tina Newman, Ph.D. received her doctorate in School/Applied Child Psychology from McGill University in Canada. Dr. Newman is currently an Associate Research Scientist at the Yale University Child Study Center. Her clinical and research interests are in assessment and intervention with children who have individual learning needs, including learning disabilities, ADHD, and giftedness. Dr. Newman is particularly interested in children with double exceptionalities (e.g., giftedness and learning disabilities) and circumscribed interests in autism spectrum disorders.

William Brown, MSW, is the Director of the Eli Whitney Museum and Workshop in Hamden, Connecticut. He is a designer, educator, and social worker who trained in child development at Columbia University in New York. In addition to directing all aspects of the Eli Whitney Museums programs, Mr. Brown runs an apprenticeship program and designs and teaches core summer workshops.

Dr. Hart received her Ph.D. in cognitive psychology from the University of Pittsburgh, with a concentration in neuroscience through the Center for the Neural Basis of Cognition. Currently, Dr. Hart is an Associate Research Scientist at the Yale University Child Study Center. Her research interests focus on reading development, skilled reading performance, and impairments in language and reading processes.

78

Tina M. Newman et al.

Donna Macomber has a Masters Degree in Psychology and is a Research Assistant at the Child Study Center of Yale University. Her primary interests include childhood development, at-risk youth, educational psychology, and learning/cognition. She is currently working with at-risk preadolescent and adolescent children toward improving social problem solving skills. She also works closely on other projects within her department including a project committed to improving the delivery of educational services to court involved youth while in detention.
Niamh Doyle is originally from Dublin, Ireland where she received a B. Ed in Elementary Education from St. Patrick's College (Ireland) and a B.Sc in Psychology from Open University (U.K.). Niamh received her MA in Psychology and is currently in her 5th year of a doctoral program in School Psychology at Syracuse University (USA). Prior to pursuing graduate studies, she was an elementary school teacher in Ireland and Australia and a research assistant at the Yale University Child Study Center and PACE center. Research interests include school and family integration, parent-teacher relationships, early detection of developmental disabilities, and transitions from pre-k to kindergarten for children with disabilities. Niamhs master's thesis examined parents perspectives on education and teacher quality and levels of involvement in school life. Niamh hopes to extend that line of research examining the validity of parents perspectives on teacher quality and its relationship to classroom quality.

Sergey Kornilov received his MSc/MA in Educational Psychology from Moscow State University and is currently a graduate student at the University of Connecticut. His research interests include giftedness and gifted education, creativity, spoken language comprehension and learning disabilities. His masters thesis examined longitudinal changes in academic achievement of Russian college students as predicted by ability measures developed within R. Sternbergs theory of successful intelligence and was awarded the Russian National Contest for Best Student Scientific Paper Award in 2009. Linda Jarvin, Ph.D., is an Associate Research Professor in the Department of Education at Tufts University, and director of its Center for Enhancing Learning and Teaching (CELT). She received her PhD in Cognitive Psychology from the University of Paris V (France) and her postdoctoral training at Yale University. She has extensive experience with curriculum planning and development, designing and implementing professional development opportunities for k-12 and college teachers focusing on teaching and assessment, and facilitating programmatic evaluation plans. Robert J. Sternberg is Dean of the School of Arts and Sciences, Professor of Psychology, and Adjunct Professor of Education at Tufts University (formerly-IBM Professor of Psychology and Education, Professor of Management, and Director of the Center for the Psychology of Abilities, Competencies, and Expertise at Yale). Sternberg was the 2003 President of the American Psychological Association and is President of the International Association for Cognitive Education and Psychology. He is the author of about 1200 journal articles, book chapters, and books, and has received over $20 million in government and other grants and contracts for his research. Dr. Elena L. Grigorenko is Associate Professor of Child Studies and Psychology at Yale and Adjunct Professor of Psychology at Columbia University (USA) and Moscow State University (Russia). She has published more than 250 peerreviewed articles, book chapters, and books. She has received awards for her work from five different divisions of the American Psychological Association. In 2004, Dr. Grigorenko won the APA Distinguished Award for an Early Career Contribution to Developmental Psychology. Dr. Grigorenkos research has been funded by NIH, NSF, DOE, Cure Autism Now, the Foundation for Child Development, the American Psychological Foundation, and other federal and private sponsoring organizations.

Instructions to authors
The Journal Talent Development and Excellence is an international scholarly journal. Manuscripts should be written in accordance with the publication manual of the American Psychological Association. See http://apastyle.apa.org/ Manuscripts must be written in English. It is recommended that non-native English speakers have their papers checked in regard to language accuracy prior to submission. Submission and Reviewing Manuscripts may be submitted electronically to one of the Editors-in-Chief. All personal data will be removed by the Editor prior to the refereeing process to allow for a masked review. Manuscript Length: A paper submitted should not exceed 6000 words including abstract, references, illustrations and appendixes. Abstract: An abstract should consist of a maximum 125 words. The abstract must, if it is the result of an empirical research, briefly outline theoretical basis, research question/s, methodology and instrumentation, sample/s and important characteristics (e.g. number, gender and age) as well as the main findings of the study. Also, it should include main conclusions. An abstract for a theoretical article or a review should describe the topic, the objective or thesis and the scope of the article. It should also outline the main conclusions. Footnotes should be numbered consecutively with superscript Arabic numerals. If reporting statistics, sufficient information must be included according to the APA Manual. Figures and Tables Figures and Tables must be placed on separate pages; not included in the text. They should be submitted as separate file/s. Figures and Tables must have an Arabic number, an explaining text and a title. Their approximate place in the text should be clearly indicated. Authors should follow APA format in designing tables and figures and consider the fact that illustrations supplements - not duplicates - the text. In the text, refer to every table and figure and tell the reader what to look for. Discuss only the table's highlights. Presentation of the Authors For the presentation of the Author(s) please hand in a digital photograph (min. 300 dpi, 4x5 cm/ 1.6x2 inch) and a description of yourself (max. 100 words). If you have any questions about references or formatting your article, please contact the Editors-in-Chief. Copyright It is a condition of publication that authors assign copyright or licence the publication rights in their articles, including abstracts, to the International Research Association for Talent Development and Excellence. Authors are themselves responsible for obtaining permission to reproduce copyright material from other sources.

ISSN 1869-0459 (print)/ ISSN 1869-2885 (online)