Two-Way Tables (2 of 5)

Example

Community College Enrollment

The following table summarizes the full-time enrollment at a community college located in a West Coast city. There are a total of 12,000 full-time students enrolled at the college. The two categorical variables here are gender and program. The programs include academic and vocational programs at the college. Assume that a student can enroll in only one program.

Let’s consider a few preliminary questions to get familiar with this new data set.

• 1. What proportion of the total number of students are male students?

Answer:

[latex]\frac{\text{number of male students}}{\text{total number of students}} = \frac{5,802}{12,000} = 0.4835\text{(or 48.35%)}[/latex]

• 2. What proportion of the total number of students are Bus-Econ students?

Answer:

[latex]\frac{\text{number of Bus - Econ students}}{\text{total number of students}} = \frac{925}{12,000} = 0.077\text{(or 7.7%)}[/latex]

Note that to calculate this proportion, we used two numbers in the margin that relate to just one of the categorical variables (program). This calculation is therefore called a marginal proportion.

Note: This proportion does not help us determine if gender is related to program because it involves only one of the variables.

Now consider the following question:

• If we choose one student at random from among all 12,000 students at the college, how likely is it that this student will be in the Bus-Econ program?

From our previous calculation, we know that only about 8% (7.7%) of the students at the college are in the Bus-Econ program. That’s a fairly low number, so it is not very likely that our random student will be a Bus-Econ student.

One way to state our conclusion is to say:

• There is about an 8% chance of picking a Bus-Econ major.

This means that if we selected 100 students at random, we would expect on average that 8 of them would be in the Bus-Econ program.

Here is another way to state this conclusion:

• There is about an 0.08 probability of picking a Bus-Econ major.

Because this probability is exactly the same as the marginal proportion we calculated earlier, we call it a marginal probability.

Example

Conditional Probability

Here is the same community college enrollment data.

Here is our first question:

• If we select a female student at random, what is the probability that she is in the Health Sciences program?

Answer: Of the 6,198 female students at the college, 421 are enrolled in Health Sciences. (Find these numbers in the table.) The probability we are looking for is:

[latex]\frac{421}{6,198} \approx 0.07[/latex]

Therefore, the probability that a female student is in the Health Sciences program is approximately 0.07.

Focus on Language

We need to pause here and be very careful about the language we use in describing this situation.

Note that we start with a female student and then ask what is the probability that this female student is in the Health Sciences department.

In this case, our starting point is that the student is a female. This information sets the conditions for calculating the probability. Once the condition (student is female) is set, we focus on the female student population. In terms of the two-way table, it means that the only numbers we will be using are in the Female row: 421 and 6,198.

What Is a Conditional Probability?

The probability we calculated earlier is an example of a conditional probability. In general, a conditional probability is one that is based on a given condition. Here the given condition is that the student is female.

Here is the notation we use for a conditional probability:

• Original question: If we select a female student at random, what is the probability that she is in the Health Sciences program?
• Notation: P(student is in Health Sciences given that student is female).
• We also write this as P(Health Sciences given female).

An even shorter way of writing this is to use a vertical bar | in place of given:P(Health Sciences | female).