Você está na página 1de 27

 

The Effect of Demographic Factors and Internet 

Speeds on E-Commerce Growth 

Andrew Wang, Juneho Jang, Nick Flanigan 

Senior Project  

Friday December 15​th​, 2017 

 
 

 
 


Abstract  

We examine the effect of age, income, and broadband connection on 

e-commerce growth. In order to do this, we first compare overall mean expenditure 

shares to online retail growth rates and then overall mean expenditures with online mean 

expenditures across various demographic groups. Our results show that people with 

lower incomes seem to be moving more online. We also see that old people have 

greater online presence than young people. Interestingly, when we examine whether or 

not people have broadband, older and richer demographics seem to have greater online 

presence. 

Introduction 

Prior to the commercialization of the internet in the late 90’s, U.S. retail 

transactions typically involved an exchange between a vendor and a consumer in a 

physical location commonly referred to as a “brick-and-mortar store”. As the internet 

became more prominent, retail transactions began migrating online, giving birth to virtual 

marketplaces where consumers could shop from the comfort of their web-browser. 

According to the Census Bureau, the share of e-commerce sales in total U.S. retail 

sales has risen approximately 5% in the last 7 years (Figure 1). While it may be the case 

that faster speeds and a more accessible internet are the primary cause of this, we think 

that there are other reasons for this growth. It’s likely that not all industries are growing at 

this rate, and that some industries are going through a lot of change due to this steady 

growth. 


 
Figure 1: This graph shows the growth of e-commerce sales as a share of total retail sales in the US. Although this graph is from 
Statista, each quarterly share is from the Census Bureau.  
 
In tandem with the constant and steady growth of e-commerce over the past 7 

years, Figure 2 shows us that online retail revenue and physical retail revenue has 

steadily increased for automotive parts. Online retail also is a small part of total sales.  

 
Figure 2: This graph shows automotive parts industry’s physical retail revenue growth and online retail revenue growth. It seems that 
online retail is still a small part of all sales. 
 
However, in figure 3, we see a totally different picture for men’s clothing. Although 

there is almost no physical retail growth in this industry, the online retail growth is 


substantial. As opposed to just 5 years ago, online retail is larger than brick and mortar 

for men’s clothing. These two different trends show that in different industries, the impact 

of e-commerce can be great. 

 
Figure 3: This graph shows men’s clothing industry’s physical retail revenue growth and online retail revenue growth. It seems that 
online retail has overtaken physical retail. 
 
We believe that e-commerce has varying impacts across industries because 

different demographic groups have different shopping preferences. We go one step 

further to say that whether or not a person has broadband internet has an impact on their 

shopping behavior. 

We see that connection speeds are growing in the US and believe that a person 

with broadband will be more likely than a person without it to spend money online. It may 

be the case that having broadband allows for faster speeds which makes shopping more 

convenient or that people who have broadband spend more of their time online and end 

up shopping online as well. Unfortunately, we could not find any data to directly support 

these arguments. 


 
Figure 4: Average internet connection speed in the US from 2007 to 2017 
 

Research Question:  

What demographic factors are driving e-commerce growth for different products? Are 

people with broadband more likely to spend money online? Do people with broadband 

spend more than those without it? 

Hypothesis:  

We expect that age and income will have an effect on online spending behavior. 

Specifically, we expect the spendings of people of younger age and lower income to 

have a positive relationship with the growth of e-commerce. We also expect to see that 

younger and poorer people with stronger internet connections are the groups that spend 

the most online. 


Data Collection 

We gather data from three sources: IBISWorld (IBIS), the Consumer Expenditure 

Survey (CE) from the Bureau of Labor Statistics, and comScore Web Behavior Database.  

IBIS provides Industry Market Research with different specifications and multiple 

datasets. Specifically for our purposes, we use the Online Retail dataset under US 

Specialized Industry Reports. From this data set, we extract the annual online retail 

revenues of various industries across different years. 

CE provides data on average expenditures and aggregate expenditures for 

different demographic characteristics of US consumers.  

From CE’s 2016 data, we use three datasets all under the Average expenditure, 

share, and standard error tables section: Age of reference person, Deciles of income 

before taxes, and Population size of area of residence. Each of these datasets provided 

us with the amount a particular consumer group spent on average on a specific product 

category. For example, in the Age of reference person dataset, we have the average 

amount a person under the age of 25 spent on Medical Supplies in 2016 to be 45 dollars. 

In particular, we used the share each of these values were of the demographic group’s 

total income. So in this case, we were interested in the data telling us that this spending 

on Medical supplies was 0.1 % of the total income of an average person under the age of 

25. 

From CE’s 2011 and 2012 data, we looked at Aggregate expenditure share tables 

for two datasets: Age of reference person and Income before taxes. Each table we use 

the proportion a demographic group spent on a specific product category. For example, 


in the Age of reference person dataset for 2012, we see that in 2011, approximately $17 

billion was spent on Medical Supplies and that people under 25 were responsible for 

2.2% of this. 

We also use expenditure mean data for Population size of area of residence for 

both 2011 and 2012.  

ComScore Web Behavior Database (comScore) provides detailed browsing and 

online buying behavior of US households, which are represented by unique machine IDs. 

We looked at comScore’s 2011 and 2012 Transactions data to use product total price and 

product category. From Demographics data, we get age of oldest head of household, 

household income, zip code and connection speed. We match these two datasets using 

machine IDs included in each year. It should be noted that only approximately 42% and 

47% of the machine IDs listed actually had any transaction in 2011 and 2012, respectively.  

Methods & Hypotheses 

Overall Expenditure Shares (CE) & Online Retail Growth Rates (IBIS) 
 
Before examining actual online buying behavior potentially described by the 

comScore data, we use CE’s average expenditure shares and IBIS from 2016 to see how 

the consumer buying behavior and the growth of certain online retail industries are 

related. First, we matched the industries from IBIS to the product categories in CE, as 

seen in Table 1. We then decided to examine three demographic groups - Young vs. Old, 

Poor vs. Rich, and Rural vs. Urban people - using the relevant Average expenditure, 

share, and standard error tables. 


Categories 
IBISWorld (IBIS)  Consumer Expenditure Survey (CE) 
Automotive Parts & Accessories  Maintenance and repairs 
Baby & Infant Apparel  Children under 2 
Beer, Wine, & Liquor  Alcoholic beverages 
Camera & Camcorder  Other entertainment supplies, equipment, and services 
Children's Toys  Toys, hobbies, and playground equipment 
Eyeglasses & Contact Lens 
Medical Supplies 
Medical Supplies 
Event Ticket  Fees and admissions 
Computer & Tablet 
Flower  Miscellaneous household equipment 
Hardware & Tools 
Grocery  Food at home 
Home Furnishing  Household furnishings and equipment 
Household Furniture  Furniture 
Jewelry & Watch  Other apparel products and services 
Large Kitchen Appliance  Major Appliances 
Men's Clothes  Men, 16 and over 
Perfume & Cosmetic  Personal care products and services 
Pet food & Pet Supplies  Pets 
Shoes  Footwear 
Vitamin & Supplements  Drugs 
Table 1: We match possible overlapping categories across both using the glossary in the CE and descriptions from IBISWorld. Some 
categories in CE are a collection of miscellaneous goods to a certain big investment. For example, ‘Miscellaneous household 
equipment’ includes multiple products that one will buy for their home, but because it is under a general ‘Housing’ category, it is 
treated as miscellaneous. We bolded these categories that were defined by a collection of categories from IBISWorld. 
 
For Young vs. Old, CE divides its consumers into different age thresholds (e.g 

under 25 years, 25 to 34 years, 65 years and older). We defined being under 35 years 

old as ‘Young’ and being over 54 as ‘Old.’ In order to calculate the expenditure shares 

for Young and Old people, we take their weighted averages. For example, for Young 

people, we first add the number of consumers under 25 and between 25 to 34: 

U nitsyoung = U nitsunder 25 + U nits25 to 34  

Then we proceed with the weighted average calculation. We’ll use alcoholic beverages 

as our example: 
U nitsunder 25 U nits25 to 34
S hareyoung, alcohol = U nitsyoung · S hareunder 25, alcohol + U nitsyoung · S hare25 to 34, alcohol  


After calculating all the expenditure shares of each product category for Young 

and Old people, we now calculate the percent change in revenue from 2016-2017 for the 

above categories. For example, online baby & infant apparel sales grew 7.64 % from 

2016 to 2017 using the following equation: 


Rev 2017 −Rev 2016 2190.7−2035.3
%ΔRevenue2016 = 100 · Rev 2016
= 100 · 2035.3
= 7.64%  

After calculating revenue changes from 2016-2017 for all our product categories, 

we then want to compare mean expenditure shares for Young and Old people. First we 

take the differences between the weighted averages for the expenditure shares of 

Young and Old for each product category. Then we standardize the differences by 

finding a z-score. From this, we create a scatterplot of online retail revenue growth rates 

as the dependent variable and the z-scores as our independent variable. Finally, we 

regress the two and find the relationship. 

We repeat the aforementioned steps for Poor vs. Rich people, with ‘Poor’ being 

those who are in the lowest three deciles and ‘Rich’ being those who are in the highest 

three deciles. Once again, we repeat the same process for people in urban and rural 

areas. ‘Urban’ is defined as someone who lives in a Metropolitan Statistical Area, or an 

area with a population above 2,500. ‘Rural’ is defined as someone who lives in an area 

that is not a Metropolitan Statistical Area and has a population below 2,500.  

Online Spending (comScore) & Overall Spending (CE) 

Now we use the comScore transaction data to calculate the average spending of 

Young vs. Old and Poor vs. Rich on the relevant product categories in 2011 and 2012.  


First we match the comScore product categories with the CE ones as seen below 

in Table 2. 

comScore CE Matched Categories 


Consumer Expenditure Survey (CE)  comScore 
Maintenance and repairs  Automotive Accessories 
Apparel and services  Apparel 
Alcoholic beverages 
Food & Beverage 
Food 
Cameras & Equipment 
Other entertainment supplies, equipment, and services  PC Video Games 
Console Video Games 
Toys & Games 
Pets, toys, hobbies, and playground equipment 
Pet Supplies 
Tools & Equipment 
Miscellaneous household equipment**  Desktop Computers 
Laptop Computers 
Computer Software 
Miscellaneous household equipment**  Other Computer Supplies 
Flowers 
Movie Tickets 
Fees and admissions 
Event Tickets 
Furniture  Home Furniture 
Other apparel products and services  Jewelry & Watches 
Major Appliances  Home Appliances 
Personal care products and services 
Health & Beauty 
Drugs 
Footwear  Shoes 
Vehicle rental, leases, licenses, and other charges  Car Rental 
Household textiles  Bed & Bath 
Small appliances, miscellaneous housewares  Kitchen & Dining 
Audio & Video Equipment 
Audio and visual equipment and services 
Video Game Consoles & Accessories 
Reading  Books & Magazines 
Housekeeping supplies  Garden & Patio 
Table 2: This table is the categories we matched up between CE and comScore. Because overall expenditure has expenditure on 
everything (such as housing, transportation, and education) the relatively small expenditures for goods that can be purchased online 
were grouped together. Because comScore has over sixty product categories, there was more grouping of product categories across 
both datasets than for IBISWorld and CE. For these groupings, we bolded the category names. 
**Miscellaneous household equipment is repeated for clarity. 
 
One thing to note before we explain our process is that our ‘Poor’ and ‘Rich’ 

thresholds are slightly different across CE and comScore datasets. ‘Poor’ in CE is defined 

by someone with an income less than $20,000 per year, but in comScore it is defined by 

10 
someone with an income of less than $25,000 per year. ‘Rich’ is defined in CE by 

someone with an income of more than $70,000 per year, but in comScore it is defined by 

someone with an income of more than $75,000 per year. We believe the differences are 

small enough for a reasonable approximation. 

To calculate average spending, we put each person in his relevant group (i.e 

Young or Old, Poor or Rich). We then sum the product total prices of every transaction in 

the group for each product category and divide the sum by the number of unique 

machine IDs in the group. 

We want to calculate the e-commerce share of product category expenditures (i.e 

the percentage of expenditure accounted for by online purchases). First we need to 

obtain average product category expenditures for our demographic groups above from 

the relevant CE Aggregate expenditure share tables. Since we are using data from 

aggregate rather than average expenditure share tables, we do the following: for ‘Young’ 

people, we sum the expenditure shares of a product category for those under 25 years 

and between 25 to 34; we also sum the consumer units for those categories. Then we 

multiply the combined share by the aggregate annual expenditures for the product 

category. This gives us the amount ‘Young’ people spent on the product category.  

Finally we divide this amount by the number of consumer units who are ‘Young’ to obtain 

the average amount spent on the product category by a ‘Young’ person. We repeat this 

procedure for Poor vs. Rich people. 

Finally we divide our comScore averages by our CE averages. This yields the 

e-commerce share of product category expenditures. Then for each of the 

11 
demographics, we graph e-commerce share of product category expenditures according 

to years and contrasting demographics. With these proportions, we want to see visually 

how e-commerce has grown over the two years, with the demographics split up by color, 

and also how each demographic spends, with the years split up by color. The first graph 

can show us how e-commerce has grown for each product category, and perhaps which 

demographic spends more on the ones that are growing. The second graph is more 

telling as it shows us which demographic is spending more of their total expenditure on a 

certain product online. We use a 45 degree line in each graph to see which side the 

points are leaning towards to determine which demographic or year has more spending. 

Examining Broadband Connection 

In order to address the issue of broadband connection, we examine the impact of 

broadband connection on online spending from two angles: 1) whether or not having 

broadband makes a consumer more likely to spend online and 2) how having broadband 

affects online consumer expenditure. 

We first want to see how broadband affects the likelihood a person spends online. 

To start, we determine whether or not a person spent a positive amount online by 

creating an expenditure dummy variable ​T​ where ​T ​= 1 if a person spent more than $0 on 

a given item and ​T ​= 0 otherwise. We then regress the expenditure dummy on a 

combination of independent dummy variables, including ​connection_speed ​(1 if a person 

has broadband, 0 if not), ​young​ (1 if a person is younger than 35, 0 if older than 64), and 

poor ​(1 if a person makes less than 25k per year, 0 if he makes at least 75k). Since our 

12 
dependent variable is a dummy variable, we utilize a​ ​logit model​ ​to estimate the 

probability that a person of certain demographic characteristics spends money online. 

The logit model will give us parameters that maximize the probability of getting the data 

we observed.  

We run the following regressions: 

T i = β 0 + β 1 connection_speed

T i = β 0 + β 1 connection_speed + β 2 young

T i = β 0 + β 1 connection_speed + β 3 poor

T i = β 0 + β 1 connection_speed + β 2 young + β 3 poor

T i = β 0 + β 1 connection_speed + β 2 young + β 3 poor + β 4 young * poor

To help us form a hypothesis for Part 2 about whether or not having a broadband 

connection causes people to spend more online, we first identify the people that made 

transactions in both 2011 and 2012. Among those that spent money in both years, we 

identify a total of 34 people that changed connection speeds from 2011 to 2012, whether 

it be switching to or losing broadband, and assign them to the relevant group. Within the 

two groups, we aggregate the group’s spending for each month and divide by the 

number of people in the group. For example, we sum the January 2011 spending of every 

person who switched to broadband and divide that sum by the number of people who 

switched to broadband. We then plot average spending for each month, with t = 1 and 

t = 24 representing January 2011 and December 2012, respectively. 

13 
 
Figure 5: Average monthly expenses for those that changed connection speeds between 2011 and 2012. ​Vertical dotted line 
represents t=13 (January 2012). No person changed connection speed within a year. 
 

There does not appear to be any pattern before and after changing connection 

speeds. Although informed by a small sample size, we hypothesize that having 

broadband does affect the amount a person spends online. 

To test our hypothesis, we calculate each person’s average expenses in each 

month. Each person has at most 24 observations: his average expenses in January 2011, 

February 2011, March 2011, up to and including his average expenses in December 2012. 

We then regress such expenses on a combination of the independent variables 

connection_speed​, ​young​, and ​poor​.  

14 
We run the following regressions: 

mo_avg_expi = β 0 + β 1 connection_speed  

mo_avg_expi = β 0 + β 1 connection_speed + β 2 young  

mo_avg_expi = β 0 + β 1 connection_speed + β 3 poor  

mo_avg_expi = β 0 + β 1 connection_speed + β 2 young + β 3 poor  

mo_avg_expi = β 0 + β 1 connection_speed + β 2 young + β 3 poor + β 4 young * poor  

Empirical Results 

Overall Expenditure Shares & Online Growth Rates 

The first part of our research was looking at the relationship between overall 

spending and online growth. After matching the categories above, we ran a regression 

across the growth rates of the product categories to the z-scores of differences in 

demographic groups. The following tables tell us the strength of the relationships 

between different demographics and growth rates. 

Each point represents the relationship between the differences in urban and rural 

(as the x-variable) and growth rates (as the y-variable) for each of the products. To 

interpret this result, we can take a look at the graph. If it were true that spending of 

people living in rural areas had a relationship with e-commerce growth, we would see a 

positive trend line with rural people spending more on the products that have the higher 

online retail growth rates. A negative trend line will tell us that people in urban areas are 

spending more on those products. From this regression, however, it is clear spending 

depending on location has little to no relationship with e-commerce growth.  

15 
 
Figure 6: This is the result of our regression on growth rates with differences in spending by location (rural - urban). Each point on the 
graph is defined by a product category.  
 
 

  Coefficients  Standard Error  P-value 

Intercept  0.07039  0.00884  2.62E-07 

Rural - Urban 
-0.0005  0.00907  0.96027 
z-score  
Table 3: Regression coefficients with SE for difference in location.  
Our Rural-Urban z-score coefficient is not statistically significant. 
 

Overall spending based on location has no relationship with growth rates. With a 

very high p-value and a horizontal line of best fit there is clearly no correlation here. This 

result still does not mean that location has no relevance for the growth of e-commerce, 

but only that there is no relevance to how overall expenditure of rural or urban 

demographics are affecting the growth of e-commerce. Overall, neither of rural or urban 

are spending more than the other on products that are showing relatively stronger 

growth online. 

16 
 

 
Figure 7: This is the result of our regression on growth rates with differences in spending by age (young - old). Each point on the 
graph is defined by a product category. 
 

  Coefficients  Standard Error  P-value 

Intercept  0.06899  0.00904  6.94E-07 

Young - Old 
0.00735  0.00929  0.4397 
z-score 

Table 4: Regression coefficients with SE for difference in spending by age group.  


Our Young-Old coefficient is not statistically significant. 
 
Here it seems that there almost no significant correlation as our p-value is very 

high. So how young or old consumers spend their income on products overall has little or 

no relationship with how much those products are growing online. It may be the case still 

that age has relevance to how people spend online. It is also important to point out that 

here we are comparing shares of income. Although we do calculate weighted averages, 

this means that we are comparing how young people spend different portions of their 

income on certain products to how old people spend different portions of their income 

on those same products. Later on we use shares of aggregate expenditure, and the 

results may differ. 

17 
 
Figure 8: This is the result of our regression on growth rates with differences in spending by income (poor - rich) populations’ 
spendings. Each point on the graph is defined by a product category. 
 

  Coefficients  Standard Error  P-value 

Intercept  0.07039  0.00721  1.30E-08 

Poor - Rich  0.02224  0.00740  0.007597 


z-score 
Table 5: Regression coefficients with SE for difference in spending by income group.  
Our Poor-Rich coefficient is statistically significant. 
 
Income seems to be the one demographic with a clear correlation. From this we 

can say that people of lower income brackets spend more on the products that are 

growing in online retail than people of higher income brackets do. While it is no causal 

relationship, we can see if this is reflected in a relationship between the aggregate 

spendings online and overall. 

Looking at growth rates with overall average expenditure shares only gives us a 

vague idea for the relationship between consumer spending and the growth of 

e-commerce. So the age and location effects may have stronger relationships with online 

growth of certain products when analyzing them through different methods. The same 

can be said for the strong relationship we found between income and e-commerce 

18 
growth, in that there may not be a strong relationship after all. For stronger evidence, we 

took a look at the online expenditure and compare with overall expenditure. 

Online Expenditure & Overall Expenditure 

After looking at the shares of overall expenditure, we moved on to aggregate 

expenditures for both online (comScore) and overall (CE). With this we found values that 

tell us the share of average overall spending that average online spending has. We can 

look at two things here: 1. The contrast between years to see which demographics are 

driving the growth more and 2. The contrast between demographics for both years to 

see which demographics are buying more of the same goods online. We do this by 

drawing a 45 degree line to see which side each point is leaning towards. The following 

graphs are grouped by the demographics we looked at. For age and income we were 

able to show the differences across years, but for census region we did not group 

census regions to have a two-dimensional graph for comparing demographics. 

 
Figure 9: This shows a comparison between years. The (x, y) is (old for 2011, old for 2012) and (young for 2011, young for 2012).  

19 
 
Figure 10: This shows a comparison between demographics. (x, y) is (young for 2011, old for 2011) and (young for 2012, old for 2012). 
 
From Figure 9, we can tell that there has been growth for most of the categories 

and that there is no real significant difference in whether or not the older or younger 

people are driving the growth. In Figure 10, however, it is clear that older people spend 

more online on almost all the categories. Other than books, older demographics spend a 

greater proportion of their total expenditure online.

Figure 11: This shows a comparison between years. The (x, y) is (poor for 2011, poor for 2012) and (rich for 2011, rich for 2012). 
 

20 
Figure 12: This shows a comparison between demographics. (x, y) is (poor for 2011, rich for 2011) and (poor for 2012, rich for 2012). 

We see that in Figure 11, all products are growing for rich people from 2011 to 

2012. Almost all products are growing for poor people as well, but what is interesting is 

that even from this graph we can tell poorer people are spending more online because 

of how all the red (rich) points are gathering near the origin and how the green (poor) 

points are clustered in the bottom corner. As we see in the second graph, this is seen 

clearly as nearly all product categories are leaning towards the poor side. 

From this analysis, we see again that poor people have a clear relationship with 

online spending. With this, we can say that poor people definitely spend more of their 

income on the categories that are growing online stronger and also spend more of their 

total expenditures online. We also see that older people are spending more of their total 

expenditures online. While this is by all means just a simple graphical analysis, we have 

enough evidence from the first analysis of growth rates to overall expenditures and this 

one to believe that people of lower incomes have an effect on e-commerce growth. 

21 
Examining Broadband Connection 
 
To examine broadband connection, we want to measure two effects: 1. If having 

broadband made people more likely to spend online and 2. How having broadband 

affects online consumer expenditure. To see the first effect, we incorporate a logit model 

to find the probability a person may purchase online. Table 6 shows our result for each of 

our five regressions.  

 
Table 6: Regression of expenditure_dummy on connection_speed, young, poor, and interaction between young and poor. 
 
It seems that having broadband is significantly associated (p-value ≤ 0.001) with 

making an online transaction. Here we regress to capture the effect of broadband, and 

change add variables such as age and income to see how broadband affects these both 

mutually and together. It seems that with broadband, older people were more likely to 

spend online, which is in line with our result in our previous analysis of comparing online 

22 
to overall expenditure. However, it seems here that richer people, not poorer people, are 

more likely to spend online with broadband. Until this point in our research we have seen 

how heavily in favor of poor people spending more online our analyses were. We will 

touch on this point later. 

To calculate the probability of a person spending money online, we plug the 

binary state of a person’s demographic characteristics into the following equation: 

1
P r(T i = 1) = 1+e−(Xβ)
 

where X β = β 0 + β 1 x1 + β 2 x2 + ... + β k xk   

Table 7 shows our results. 

  Probability of transacting 

Young and poor with broadband  0.6876 

Young and poor without broadband  0.3428 

Old and poor with broadband  0.7039 

Old and poor without broadband  0.3603 

Young and rich with broadband  0.6955 

Young and rich without broadband  0.3511 

Old and rich with broadband  0.7847 

Old and rich without broadband  0.4633 

Table 7: Probability of spending money online for person with specific demographic characteristics..  

Here we see that an older, richer person with broadband has nearly an 80% 

probability that they will spend online. Each kind of consumer seems to be more likely to 

spend money if they have broadband. Next we want to see how broadband changes 

how people spend money online. 

23 
To measure the effect of broadband on online expenditure, we use a linear 

regression model with the same variables as our analysis measuring the probability of 

spending. Table 8 shows our results. 

Table 8: Regression of monthly average expenses on connection_speed, young, poor, and interaction between young and poor. 

 
It seems that having broadband is significantly associated (p-value ≤ 0.001) with 

monthly average expenditures. While it is definitely interesting that having broadband 

itself on average causes a person to spend $27.24 more than someone without, it is 

quite alarming how negative of an effect being poor has. We see that older and richer 

people with broadband spend the most online again, but from our previous analyses we 

thought that poor people would drive growth. 

24 
It is most likely that because the way we analyzed these effects of broadband, on 

both the probability of spending and monthly spending, we see that poor people are 

spending less. In our previous analyses we calculated averages and found shares. We 

captured only the expenditures relative to each demographic’s income, but not relative 

to total spending. However, in this analysis we measure both parts in reference to total 

spending. It is likely that the average rich person spends more than the average poor 

person, so it makes sense that here the rich has stronger spending characteristics. 

_____________________________________________________________________ 

Conclusion 

In terms of products, we saw no significant differences. Our analyses showed that 

it was much more interesting to examine the comprehensive differences and effects of 

each measure as the products different demographics were spending on seemed to tell 

us little about how e-commerce was growing.  

While we have no concrete evidence to say that e-commerce is being driven by 

certain demographics, we have some data to show that it may be the case. In our first 

analysis, we saw that poorer people were spending more of their income on the goods 

that were growing in online retail. With this result, we also found that poorer people were 

spending more online as a share of their total expenditure on nearly all products that we 

analyzed. We also found that older people were spending more online from their total 

expenditure. With this we were able to say that it is highly likely that poor people spend 

more online and are to a certain extent driving the growth of e-commerce. In our last 

analysis we saw that with broadband, older people were even more likely to spend 

25 
online and spent more on average than younger people. We also saw that rich people 

were more likely to spend online and spent more on average than poor people. This 

effect may be because older people are generally richer, but we still saw that all 

demographics were more likely to spend online with broadband connection. While we 

have no further evidence to show that broadband definitely brings about more spending, 

we can say that there is a strong correlation between online expenditure and broadband 

connection. 

Limitations and Future Research 

What we wanted to include in our final analysis was location. From the start we 

believed that location may have an impact on online spending for multiple reasons. It 

may be that rural people buy products online because stores are too faraway or that 

urban people shop online because distribution centers are closer to urban areas. Given 

that comScore provides zip codes for each machine ID, we attempted to include this in 

our analysis. CE provides 2011 and 2012 annual expenditure means tables for urban and 

rural spending on different product categories. We were also able to find a list of urban 

areas and their five-digit zip codes from the US Census Bureau. However, we could not 

match them with comScore due to possible privacy restrictions that resulted in them 

producing many four-digit zip codes. As we could not just exclude the ones without zip 

codes for accuracy of the research, we could not produce anything else from our urban 

and rural demographics. Analysis on this may be enlightening for future research. 

26 
Works Cited 

● Consumer Expenditure Survey (Bureau of Labor Statistics), 

https://www.bls.gov/cex/tables.htm​, accessed September 30, 2017. 

● comScore (Wharton Research Data Services), 

https://wrds-web.wharton.upenn.edu/wrds/​ (needs username), accessed October 

1, 2017.  

● US Specialized Industry Reports (IBISWorld), 

http://clients1.ibisworld.com/reports/us/specializedreports/home.aspx​, accessed 

September 10, 2017.  

● Quarterly E-Commerce Report Historical Data (U.S. Census Bureau), 

https://www.census.gov/retail/ecommerce/historic_releases.html​, accessed 

September 7, 2017. 

● Quarterly share of e-commerce sales of total U.S. retail sales from 1st quarter 2010 

to 3rd quarter 2017 (Statista)​, 

https://www.statista.com/statistics/187439/share-of-e-commerce-sales-in-total-us-r

etail-sales-in-2010/Statista​, accessed September 10, 2017. 

27 

Você também pode gostar