Eighth Grade Mathematics: Constructing Two-Way Frequency Tables
While scatter plots are the premier statistical tool for analyzing quantitative measurements like height, time, and temperature, much of the data collected in modern society is categorical: gender, grade level, voting preferences, favorite subjects, or transportation modes. When researchers investigate whether two categorical characteristics collected from the same group of individuals are related, they organize the data into a grid of rows and columns known as a two-way frequency table. In this chapter, you will master the structure of two-way frequency tables, learn how to tally raw bivariate survey data, and calculate row, column, and grand totals.
The Structure of a Two-Way Frequency Table
A two-way frequency table organizes bivariate categorical data into a rectangular matrix where the rows represent the categories of one variable and the columns represent the categories of the second variable:
The Anatomy of a Two-Way Table:
Column Variable (Category 1) Column Variable (Category 2) Row Total
Row Variable (Cat A) Joint Frequency (Cell A1) Joint Frequency (Cell A2) Total for Cat A
Row Variable (Cat B) Joint Frequency (Cell B1) Joint Frequency (Cell B2) Total for Cat B
Column Total Total for Category 1 Total for Category 2 GRAND TOTAL (N)
The word frequency simply means the count of how many individuals fall into a particular category.
The internal cells of the table contain joint frequencies: counts of individuals that belong to both the specified row category AND the specified column category simultaneously.
The outer edges of the table contain marginal frequencies: the row totals and column totals. The bottom-right corner cell contains the grand total N, which represents the total sample size of all surveyed individuals.
Constructing a Table from Raw Survey Data
To construct a two-way frequency table, researchers tally responses across both categories simultaneously:
Survey Scenario: 100 Eighth-Grade Students were asked:
1. Do you participate in school athletics? (Yes or No)
2. Do you play a musical instrument in the school band? (Yes or No)
Raw Survey Results:
- 35 students play in band and participate in athletics.
- 15 students play in band but do NOT participate in athletics.
- 25 students do not play in band but DO participate in athletics.
- 25 students neither play in band nor participate in athletics.
Step-by-Step Table Construction:
Athletics: Yes Athletics: No Row Total
Band: Yes 35 15 35 + 15 = 50
Band: No 25 25 25 + 25 = 50
Column Total 35 + 25 = 60 15 + 25 = 40 50 + 50 = 100 (Grand Total)
Notice that the sum of the row totals (50 + 50 = 100) must identically equal the sum of the column totals (60 + 40 = 100). This double-check confirms that no individual was double-counted or omitted.
Solving for Missing Values in Incomplete Tables
In many statistical and standardized test problems, tables are presented with several blank cells. You can solve for the missing values using basic subtraction and addition, because each row must sum to its row total, and each column must sum to its column total.
Incomplete Table Example:
Bus Riders Walkers Car Riders Total
7th Graders 45 [ ? ] 20 90
8th Graders [ ? ] 35 15 [ ? ]
Total 85 60 [ ? ] 180
Solving the Missing Cells:
1. 7th Grade Walkers: 90 - 45 - 20 = 25 walkers.
2. 8th Grade Bus Riders: 85 - 45 = 40 bus riders.
3. 8th Grade Total: 40 + 35 + 15 = 90 students.
4. Total Car Riders: 20 + 15 = 35 car riders.
5. Verification: Walkers: 25 + 35 = 60.
Grand Total Check: 85 + 60 + 35 = 180. (Verified!)
By systematically finding rows or columns with only one missing value, an entire complex table can be decoded effortlessly.
Chapter Exercises
Exercise 1: A survey of 120 middle school students asked whether they prefer studying in the morning or evening, and whether they own a personal computer. The data shows: 42 students who own a computer prefer morning study; 48 students who own a computer prefer evening study; 8 students who do not own a computer prefer morning study; and 22 students who do not own a computer prefer evening study. Construct a complete two-way frequency table including all row totals, column totals, and the grand total.
Exercise 2: An incomplete two-way table tracks lunch choices among 80 students: 30 boys chose pizza; 15 boys chose salad; 25 girls chose pizza; and the grand total is 80 students. Determine how many girls chose salad, the total number of boys, the total number of girls, and the total count of pizza and salad choices.
Exercise 3: Explain in paragraph form why the sum of the row totals in a two-way frequency table must always equal the sum of the column totals, discussing how every surveyed individual is counted.
Solutions and Explanations
Solution 1: To construct the two-way frequency table, assign the rows to Computer Ownership (Owns Computer, No Computer) and the columns to Study Time Preference (Morning, Evening). Place the given counts into their corresponding joint cells: for Owns Computer and Morning, enter 42; for Owns Computer and Evening, enter 48; for No Computer and Morning, enter 8; and for No Computer and Evening, enter 22. Now calculate the marginal totals: the row total for Owns Computer is 42 + 48 = 90 students; the row total for No Computer is 8 + 22 = 30 students. The column total for Morning is 42 + 8 = 50 students; the column total for Evening is 48 + 22 = 70 students. Finally, verify the grand total: summing the row totals yields 90 + 30 = 120 students; summing the column totals yields 50 + 70 = 120 students. Both directions match the total sample size of 120 students.
Solution 2: To solve the incomplete lunch table: first, calculate the total number of boys by adding the boys who chose pizza to the boys who chose salad: 30 + 15 = 45 boys. Next, find the total number of girls by subtracting the total number of boys from the grand total of 80 students: 80 - 45 = 35 girls. Now determine how many girls chose salad: of the 35 total girls, 25 chose pizza, so 35 - 25 = 10 girls chose salad. Finally, calculate the column totals: total students who chose pizza is 30 + 25 = 55 students; total students who chose salad is 15 + 10 = 25 students. Verifying: the sum of pizza and salad choices is 55 + 25 = 80, which perfectly matches the grand total.
Solution 3: The sum of the row totals must always equal the sum of the column totals because both calculations represent two different ways of partitioning the exact same set of surveyed individuals. When you calculate the row totals, you are sorting every person in the entire sample into mutually exclusive row categories (such as 7th graders and 8th graders) and adding up the sizes of those groups; their sum encompasses every person surveyed once and only once. When you calculate the column totals, you are taking those exact same people and sorting them into mutually exclusive column categories (such as bus riders, walkers, and car riders) and adding up those groups; their sum likewise encompasses every person surveyed once and only once. Because every single individual in the study corresponds to exactly one internal cell in the table, summing across rows counts every person once, and summing across columns counts every person once. Both sums must identically equal the grand total sample size N.