Você está na página 1de 21

Shades of Gray: Ambiguity Tolerance & Statistical Thinking

Robert H. Carver
Stonehill College/Brandeis University Session 385 JSM 2007 Salt Lake City

Outline
Brief review of JSM 2006 paper Modifications in current work Methods Results Invitation to participate

1 August 2007

Ambiguity Tolerance
Frenkel-Brunswik, Else (1948)

Ambiguity Tolerance Construct:

Some are stimulated by ambiguity, some are threatened Personality trait vs. preferred process Enduring personality attribute vs. contextdependent Relationship to rigidity, uncertainty tolerance, openness

1 August 2007 3

Very low A.T.

Never, ever, think outside the box

1 August 2007

JSM 2006 paper


Ambiguity tolerance construct Focus on inferential thinkingskill of

drawing actionable conclusions based on incomplete information Hypothesized that people with Low AT would have difficulty becoming facile with inferential thinking tasks Mixed findings
1 August 2007 5

Research Questions
Is ambiguity tolerance (AT) a

predictor of success in a students development of statistical thinking skills? Does AT interact with other success factors?

1 August 2007

Sample
Sample: 85 undergraduates enrolled over 2 semesters Differences among sections
Technology:

Minitab vs. SAS (Learning Ed.) Normal, Learning Community, Honors

1 August 2007

Sample
Informed consent
Illustration

of research design Modeling ethical research practice Illustration of some methods


Credit & incentives
Course-embedded data collection

1 August 2007

Methods
Dependent variable:
Score on Comprehensive Assessment of Outcomes

for a first course in Statistics (CAOS) post-test

Developed by Web ARTIST Project (U.Minnesota and Cal Poly) team Pre- and Post-test (40 items each)

URL:

https://data.gen.umn.edu/artist//tests/index.html

1 August 2007

CAOS post-test
Post vs. Pre-test Scores
90 80 70
Gender Male Female

CAOSPost

60 50 40 30 30 40

1 August 2007

50 CAOSPre

60

70

80
10

Questions/Methods

Independent Measures & variables:


McLains AT scale:

22 question instrument 7-point Likert Scales


Max

score for extreme tolerance = 74 Min score for extreme intolerance = - 58

Reliability: Cronbachs alpha = 0.897


In

this sample a = 0.872

1 August 2007

11

Typical Scale Items


I dont tolerate ambiguous

situations well. Im drawn to situations which can be interpreted in more than one way. I enjoy tackling problems which are complex enough to be ambiguous. I find it hard to make a choice when the outcome is uncertain.
1 August 2007 12

Distribution of AT
AT Scores for Sample
25

20

Frequency

15

10

-16

16 AT

32

48

64

1 August 2007

13

Covariates investigated
Score on CAOS Pre-test Prior Stat Education (37% had some) Section dummy variables (Honors, L.C., etc.) Course Performance variables Attendance Gender dummy (49% female; 51% male) First-year student dummy (61% 1st year) Math SAT

1 August 2007

14

Findings: CAOS Pre-test


Variable Constant Female dummy AT scale Coeff 9.07 -1.13 0.048 Signif 0.438 0.638 0.537

First year dummy Prior course dummy

-5.581 5.256

0.028 0.032

Math SAT score


F

0.063
4.89

0.001
0.001

Adj R2

21.3%

A.T. did not have a significant main effect on Pre-test scores


1 August 2007 15

Findings:CAOS Post-Test
Variable Coeff Signif

Constant CAOS Pre-test score AT scale First Year dummy Prior course dummy F Adj R2

33.374 0.559 0.110 -3.726 -3.406 12.29 37.0%

0.000 0.000 0.079 0.072 0.099 0.000

AT score has an effect (p < 0.10) on Post-Test reasoning score


1 August 2007 16

Findings:CAOS Post-Test
Variable Coeff Signif

Constant

-2.529

0.751

CAOS Pre-test score


AT scale Course Cumulative Avg Prior course dummy F Adj R2

0.437
0.117 0.473 -3.946 19.46 48.9%

0.000
0.039 0.000 0.035 0.000

AT score has a significant (p < 0.05) effect on Post-Test reasoning score


1 August 2007 17

Discussion
Main Findings:

AT showed a positive main effect AT was not predictive of course performance Concerns: CAOS measure several aspects of statistical thinking AT scale may measure several factors Small sample Substantial unexplained variance

1 August 2007 18

Discussion & Questions


An individuals orientation toward ambiguity can affect his/her success with statistical reasoning. AT construct may provide a metaphor for statistical thinking Relationship between AT and Learning Styles? Can these results be replicated, especially in larger samples?

1 August 2007

19

Discussion & Questions


Would the results hold up with different measures of statistical reasoning? Do other personality or personal style variables shape success in statistical reasoning? How can we structure pedagogy to address personality variation among learners? Does A.T. affect application of statistical reasoning in practice?

1 August 2007 20

Replication?
Contact me
rcarver@stonehill.edu rcarver@brandeis.edu

http://faculty.stonehill.edu/rcarver/

1 August 2007

21

Você também pode gostar