QA 4

Flashcards covering organizing and presenting categorical data, including frequency tables, cross-tabulations, proportions, percentages, ratios, pie charts, and bar charts.

What is the primary purpose of organizing data?

To pinpoint where data values tend to concentrate, which is its distribution.

What does the distribution of a variable tell us?

It tells us what values the variable takes and how often it takes these values.

What is a frequency distribution table for a categorical variable?

It lists the categories and gives either the count, relative frequency, or percent of individuals who fall in each category.

What are cross-tabulations?

Tables that display the distribution of data across two categorical variables.

How is relative frequency calculated?

Relative frequency is calculated as the frequency (f) divided by the sum of all frequencies (sigma f), which is equal to the sample size (n).

What are descriptive statistics?

Numbers that describe certain characteristics of a sample, highlighting salient features of a data distribution.

How do you calculate a proportion?

By dividing the portion you are interested in (frequency) by the whole (sample size).

How do you convert a proportion to a percentage?

Multiply the proportion by 100.

From the social media preference example with n=50, what proportion of respondents prefer Instagram if 9 people preferred it?

0.18 (9 divided by 50).

From the social media preference example, what is the ratio of respondents who preferred Snapchat to those who prefer Twitter if 8 preferred Snapchat and 4 preferred Twitter?

2 to 1 (8 divided by 4).

What is a pie chart used for?

To show the distribution of a categorical variable as a 'pie' whose slices are sized by the percentage for the categories, emphasizing each category
's relation to the whole.

When should pie charts be avoided?

If there are too many categories, if the percentages do not sum to 100%, or to display distribution across two categorical variables.

What is a bar chart used for?

To represent each category of a variable as a bar, where bar heights show category counts or percentages.

What is a cluster bar chart?

A bar chart that displays and compares two or more groups along the same variable.

What is a segment bar chart?

A chart that displays the distribution of a categorical variable as portions (segments) of a rectangle, with the area of each segment proportional to the percentage of individuals in the corresponding category.

When is a bar chart generally preferred over a pie chart?

When comparing the magnitude of differences between categories, when there is a larger number of categories, or when emphasizing the distribution of data.

When is a pie chart generally preferred over a bar chart?

When emphasizing the relationship of parts to a whole, with a smaller number of categories, or for a simple comparison of proportions or percentages, provided all categories
' percentages sum to 100%.

What are the key elements of a good graph?

Title, plot, source, legend, and axis titles.