Bivariate Analysis with Contingency Tables Flashcards (36 cards)

All cards

1Question

What does a contingency table organize?

Answer

Counts for two categorical variables.

2Question

What does each cell in a contingency table count?

Answer

Observations with a particular combination of categories.

3Question

What does a marginal distribution show about a variable in a contingency table?

Answer

Its frequency distribution considered by itself.

4Question

Where is a marginal distribution shown in a contingency table?

Answer

In the corresponding margin.

5Question

What does a conditional distribution show in a contingency table?

Answer

One variable’s distribution within each category of the other variable.

6Question

How is an overall relative frequency table calculated from cell counts?

Answer

Divide every cell count by the total count and express the result as a percentage.

7Question

How are column percentages calculated from cell counts?

Answer

Divide each cell count by its column total.

8Question

How are row percentages calculated from cell counts?

Answer

Divide each cell count by its row total.

9Question

When are two categorical variables independent?

Answer

When one variable’s distribution is the same across every category of the other variable.

10Question

How does a 100% segmented bar chart display conditional distributions?

Answer

Each first-variable category gets a bar totaling 100%, divided into segments for second-variable categories.

11Question

What does the chi-squared statistic compare to measure categorical association?

Answer

Observed contingency-table counts and counts expected under independence.

12Question

How is the expected count in cell (i,j)\left(i,j\right) calculated under independence?

Answer

Eij=RiCjnE_{ij}=\frac{R_iC_j}{n}

13Question

What is the formula for the chi-squared statistic using observed and expected counts?

Answer

χ2=∑i∑j(Oij−Eij)2Eij\chi^2=\sum_i\sum_j\frac{(O_{ij}-E_{ij})^2}{E_{ij}}

14Question

What does Cramér’s V measure, and what are its endpoint values?

Answer

Association, from 0 for none to 1 for perfect association.

15Question

What is a lurking variable?

Answer

A third variable used to group observations when combining groups changes or removes a relationship between two variables.

16Question

What is Simpson’s paradox?

Answer

A trend or relationship within multiple groups disappears when the groups are combined according to a third variable.

17Question

Where does Variable 1 appear in the course’s contingency-table convention?

Answer

In the rows.

18Question

Where does Variable 2 appear in the course’s contingency-table convention?

Answer

In the columns.

19Question

In the sex-and-smoking example, what do the columns show?

Answer

Smoking-status distributions within each sex category.

20Question

In the sex-and-smoking example, what do the rows show?

Answer

Sex distributions within each smoking category.

21Question

What denominator does “percent of everyone” use for a table cell?

Answer

The table total.

22Question

What denominator does “percent of people with a specified characteristic” use?

Answer

The total for that characteristic.

23Question

In the sex-and-smoking example, what share of female students are non-smokers?

Answer

About three quarters.

24Question

In the sex-and-smoking example, what share of male students are non-smokers?

Answer

About three quarters.

25Question

What do similar-looking bars in a 100% segmented chart suggest?

Answer

They suggest similar conditional distributions and approximate independence.

26Question

What do similar separate pie charts suggest about conditional distributions?

Answer

They suggest approximate independence.

27Question

Among people tattooed at a commercial parlor, about what percentage had Hepatitis C?

Answer

About 33%.

28Question

Among people with no tattoo, about what percentage had Hepatitis C?

Answer

About 3.5%.

29Question

What do the Hepatitis C rates by tattoo status indicate about the variables?

Answer

Tattoo status and Hepatitis C are not independent.

30Question

What is the chi-squared statistic when variables are totally independent?

Answer

Zero.

31Question

What is the formula for Cramér’s V using the chi-squared statistic?

Answer

V=χ2nmin⁡(R−1,C−1)V=\sqrt{\frac{\chi^2}{n\min(R-1,C-1)}}

32Question

Which factors affect the magnitude of the chi-squared statistic?

Answer

Sample size and the numbers of rows and columns.

33Question

What Cramér’s V values describe sex and smoking, and tattoo origin and Hepatitis C?

Answer

0.0015 for sex and smoking; 0.3 for tattoo origin and Hepatitis C.

34Question

In the airline-arrivals example, what does the overall comparison suggest about airline and on-time arrival?

Answer

They appear almost independent.

35Question

In the airline-arrivals example, which variable reveals clearer differences between airlines?

Answer

Destination.

36Question

In the airline-arrivals example, which airline is more likely to arrive on time regardless of destination?

Answer

US Airways.

Test yourself with the quiz

Answer 15 questions on Bivariate Analysis with Contingency Tables.

1. A researcher records each customer's preferred payment method and membership tier. In a contingency table, what does the cell at the intersection of “card” and “premium” represent?

2. A table follows the convention that Variable 1 is shown in rows and Variable 2 in columns. If Variable 1 is region and Variable 2 is product type, where should product type appear?

Take the quiz →

Read the study guide

Go over the full study guide for Bivariate Analysis with Contingency Tables.

Read the study guide →

Related classes

Make your own flashcards

Upload your class notes and Revizly writes your flashcards in about a minute.

Flashcard maker