1

Case study (American family)

Your Name

Pima Community College

ABC 101: Course Name

Professor (or Dr.) Firstname Lastname

Date
 2

Case study (American family)

Introduction:

The main goal of this in-depth case study is to look into the complicated mechanisms that

control how people spend their money by looking into the link between annual income,

household size, and yearly credit card charges. The given dataset gives information about the

personal finances of 50 people, including how much they earn each year, who lives in their

home, and how much they spend on credit cards. As they try to figure out what causes people to

get into debt, financial companies and governments would do well to learn about these trends.

Additionally, people who want to do smart financial planning would also find these trends

helpful in their efforts. The goal of this study is to look at the given dataset for trends, find out

how income and family size affect credit card charges, and then decide which variable is a better

predictor in this case.

For example, we used regression analysis and other complex statistical methods to find

hidden relationships in the dataset. We expect that using regression models will help us better

understand the complex relationships between things like income, family size, and annual credit

card costs. We also look into the possibility of making our models better at making predictions

by adding factors that are more independent. For individuals, businesses, and governments alike,

this in-depth research is useful because it helps them make smarter financial choices and learn

more about how people behave as consumers. Our study's goal is to shed light on the basic

factors that affect the complicated balance between income, household size, and credit card

expenses. This will make a major academic addition to the field of consumer finance.
 3

Descriptive Statistics

Table 1: Descriptive statistics

Income, family size, and credit card spending are just some of the key indicators that can

be better understood with the help of descriptive statistics. Let us look at the average pay of

$43.48 to get a sense of the typical earnings in this population. This number can be used as a

proxy for the median income of the people under consideration. The large standard deviation of

$14.55 indicates substantial variation around the mean. Despite the comparatively modest mean

income of $43.48 per month, the incomes observed in the sample display considerable variation

around the sample mean, indicating a varied range of socioeconomic situations among the

persons polled.

The average family size, as measured by the statistic known as the "household size," is

3.42 people. The aforementioned numerical value is of the highest relevance in the process of

inferring the racial and ethnic composition of the provided sample. The standard deviation of

1.74 indicates the wide diversity in family sizes, ranging from single-person households to those

with as many as seven members. The inclusion of a wide range of household sizes is crucial

because it improves the accuracy of the portrayal of real-world circumstances, where the number

of people living in a household has a major impact on spending habits.

We look at the costs of using credit cards each year and find that, on average, people in

our sample rack up $3964.06 in fees and interest over the course of a year. The presented chart

summarizes, at a single glance, the most prominent spending trends reflected in the collected

information. The $933.49 standard deviation from the mean indicates how much individual

spending habits might wander from the average. Some people's credit card expenses will have a

large standard deviation, meaning they will be very different from the average. The wide range
 4

of yearly credit card payments, from a low of $1864 to a high of $5678, reflects the wide variety

of consumers' spending habits. The wide variation seen is likely attributable to the fact that

people have widely varying perspectives on how they should manage their money, ranging from

extreme frugality to lavish spending (Xu et al., 2022).

The descriptive statistics provided paint a detailed picture of the dataset under

consideration. The authors laid the groundwork for future studies by directing our investigation

into the connections between income, household size, and credit card spending. Understanding

the major tendencies and variations inherent to these variables is crucial for correctly

comprehending the results of our regression analysis and for fully appreciating the complex

dynamics that influence consumer-purchasing behavior.

Regression analysis (Annual income as the independent variable)

Table 2: Regression analysis

Source SS df MS Number of obs = 50

F(2, 47) = 36.65
Model 6321.54435 2 3160.77218 Prob > F = 0.0000
Residual 4052.93565 47 86.2326734 R-squared = 0.6093
Adj R-squared = 0.5927
Total 10374.48 49 211.724082 Root MSE = 9.2862

income Coefficient Std. err. t P>|t| [95% conf. interval]

householdsize -5.842319 1.159 -5.04 0.000 -8.173926 -3.510713

amountcharged .0180288 .0021591 8.35 0.000 .0136853 .0223724
_cons -8.006706 6.29346 -1.27 0.210 -20.66751 4.654103

The objective of the regression analysis employing income level as the independent

variable is to investigate the impact of income and family size on annual credit card expenditure.

The results present compelling empirical support for the correlation between the variables under
 5

investigation, so offering novel perspectives on the underlying determinants that influence

consumers' purchasing behaviors.

The regression model reveals a significant association between the three factors and

annual credit card costs. In general, the model has statistical significance, as evidenced by an F-

statistic of 36.65 and a p-value of 0.0000. Hence, the impact of income and household size might

be employed to elucidate the disparity in yearly credit card expenses. Based on the coefficient of

determination (R2) of 0.6093, it can be inferred that approximately 60.93% of the variability in

annual credit card spending can be accounted for by characteristics such as income and family

size. The R-squared value indicates a strong match between the model and the data,

demonstrating that a significant percentage of the variability in credit card spending can be

accounted for by individual and family income and size.

 6

Graphical representation:

Figure 1: Regression analysis (Annual income as the independent variable)

Considering the quantity of predictors incorporated in the model, the adjusted R-squared

coefficient of determination, which amounts to 0.5927, signifies that approximately 59.27% of

the variance in credit card charges can be accounted for by the variables of income and

household size. A high value of adjusted R2 signifies that the model effectively captures the

associations among the variables. The coefficient pertaining to income has been shown to be

0.0180. This corresponds to an increase of around $0.0180 in yearly credit card expenses for

each unit of disposable income. When holding income constant, the coefficient for household

size is -5.8423, indicating that an increase of one more member in the household is associated

with a decrease of around $5.84 in yearly credit card costs. The presence of a negative sign in the
 7

analysis suggests the existence of an inverse relationship between household size and credit card

charges. Specifically, when considering income as a factor, it can be shown that bigger

households generally exhibit lower levels of credit card charges. This regression analysis

underscores the significance of income and family size as explanatory factors for the variability

observed in annual credit card expenditures (Xu et al., 2022). There is a positive correlation

between individuals' income levels and the magnitude of credit card charges they incur.

Conversely, individuals residing in larger households tend to have lower credit card costs. These

findings have the potential to provide valuable insights into the spending behaviors of

consumers, which can be of great interest to financial advisors, business strategists, and

legislators.

Regression analysis (household size as the independent variable)

Table 3: Regression analysis

Source SS df MS Number of obs = 50

F(2, 47) = 60.07
Model 106.511793 2 53.2558966 Prob > F = 0.0000
Residual 41.6682068 47 .886557591 R-squared = 0.7188
Adj R-squared = 0.7068
Total 148.18 49 3.02408163 Root MSE = .94157

householdsize Coefficient Std. err. t P>|t| [95% conf. interval]

income -.0600649 .0119157 -5.04 0.000 -.0840361 -.0360936

amountcharged .0019932 .0001857 10.73 0.000 .0016196 .0023669
_cons -1.869584 .5889482 -3.17 0.003 -3.054395 -.6847731

This regression analysis lets us examine household size and annual credit card bills with

income as a covariate. The relationship between family size, income, and credit card spending is

examined. An F-statistic of 60.07 and a p-value of 0.0000 indicate model significance. This
 8

shows that the model is very reliable, as household size and income explain the variance in

annual credit card fees. R-squared = 0.7188, so household size and income explain 71.88% of

annual credit card bills. This high R-squared value shows that the model adequately accounts for

most of the variance in credit card expenses due to household size and income. A model with

degrees of freedom equal to household size and income explains 70.68 percent of credit card

charges, according to an adjusted R-squared value of 0.7068. Even with many predictors, this

modified R-squared value confirms the model's ability to explain credit card transactions. The

coefficient for household size is -0.0601. Keeping income constant, each additional household

member reduces annual credit card charges by $0.0601. When income is considered, larger

families have lower credit card charges. The income coefficient -0.0019 is negative. All else

being equal, credit card costs would decrease by $0.0019 per year for each additional dollar in

annual income. This negative coefficient indicates a slight decrease in credit card expenses as

income rises, controlling for household size. The intercept, constants, is -1.8696. This fixed term

represents annual credit card costs assuming zero family size and income. Real households and

incomes never reach zero, so these numbers are useless.

 9

Figure 2: Regression analysis (Household size as the independent variable)

After controlling for income, this regression analysis shows a significant inverse

relationship between household size and annual credit card charges, suggesting that larger

households have lower credit card charges. Even after controlling for family size, income has a

small negative relationship with credit card charges (Zhang, 2023). These findings illuminate the

complex relationship between household demographics, income, and consumer spending habits,

which aids financial planning, market analysis, and policymaking.

Better predictor of annual credit card charges?

Credit card spending is more strongly correlated with annual income than with

household size. Using income as the independent variable in a regression analysis, it is clear that
 10

income (0.0180) is higher than household size (-5.8423). Assuming a fixed income, the annual

credit card bill increases by $0.0180 per $1 and decreases by $5.84 for each additional person in

the household. The adjusted R-squared value of the model where annual income is the

independent variable is 0.5927, which is higher than the value obtained when using household

size, which is 0.7068. This indicates that the income model provides a better explanation for

variance in credit card charges when both factors are included. This means that both variables are

significant, but annual income is slightly more predictive.

Predicted annual credit card charge for a three-person household with an annual

income of $40,000?

By plugging the relevant numbers into the multiple regression equation, we can

calculate the expected yearly credit card bill for a family of three earning $40,000. The estimated

coefficients from the multiple regression model will be used for this purpose. Coefficients

created from the data would be required for an exact prediction, however, these are not included

in the data set you have provided (Tunc et al., 2023). The multiple regression analysis

coefficients and intercept will determine the likely fee.

Discuss the need for other independent variables that could be added to the model.

What additional variables might be helpful?

Annual income and household size are good predictors, but the model may utilize more

independent variables to improve its accuracy. Additional variables may include, higher credit

scores indicating prudent financial behavior, which can affect credit card usage. Financial

literacy and decision-making are affected by education. Employment because employed people

may spend differently than jobless or self-employed people (Xiao et al., 2022).
 11

Discuss the summary of the average consumer debt of the American family

This case study solely covers the data provided and does not represent the typical

American family's finances. A more complete and representative data set is needed to summarize

American household consumer debt. Mortgages, school loans, car loans, and credit card debt

must be considered to fully understand American household finances. Several macro-level

factors affect US consumer debt. Economic conditions, interest rates, government laws, and

cultural norms. Income, expenses, and economic growth affect consumer debt. Checking

trustworthy sources including government economic reports, banking institutions and research

groups can provide thorough and up-to-date statistics on average consumer debt in the US.
 12

References

Tunc, C., & Kilinc, M. (2023). Household debt and economic growth: Debt service matters.

Open Economies Review, 34(1), 71-92.

Xiao, J. J., & Yao, R. (2022). Good debt, bad debt: family debt portfolios and financial burdens.

International Journal of Bank Marketing, 40(4), 659-678.

Xu, Y., & Yao, R. (2022). US household financial vulnerability: Prediction analyses in the

COVID-19 pandemic. Journal of Financial Counseling and Planning, 33(2), 228-242.

Zhang, Z. (2023). Check and Balance Overspending in Average American Household

Consumption. Highlights in Business, Economics and Management, 7, 1-7.