312 Chapter 10 Alternative Correlational Techniques
a. Plot these data and fit a regression line.
b. Calculate and test it for significance.
c. Interpret the results.
10.3 Compare the results you obtained in Exercises 10.1 and 10.2. What can you conclude?
10.4 Why would it not make sense to calculate a biserial correlation on the data in Exercises 10.1
and 10.2?
10.5 Perform a t test on the data in Exercise 10.1 and show the relationship between this value
of t and.
10.6 A graduate-school admissions committee is concerned about the relationship between an
applicant’s GPA in college and whether or not the individual eventually completes the
requirements for a doctoral degree. They first looked at the data on 25 randomly selected
students who entered the program 7 years ago, assigning a score of 1 to those who
completed the Ph.D. program, and of 0 to those who did not. The data follow.GPA: 2.0 3.5 2.75 3.0 3.5 2.75 2.0 2.5 3.0 2.5
Ph.D.: 0000000011
GPA: 3.5 3.25 3.0 3.0 2.75 3.25 3.0 3.33 2.5 2.75
Ph.D.: 1111111111
GPA: 2.0 4.0 3.0 3.25 2.5
Ph.D.: 11111a. Plot these data.
b. Calculate.
c. Calculate.
d. Is it reasonable to look at in this situation? Why or why not?
10.7 Compute the regression equation for the data in Exercise 10.6. Show that the line defined
by this equation passes through the means of the two groups.
10.8 What do the slope and the intercept obtained in Exercise 10.7 represent?
10.9 Assume that the committee in Exercise 10.6 decided that a GPA-score cutoff of 3.00
would be appropriate. In other words, they classed everyone with a GPA of 3.00 or higher
as acceptable and those with a GPA below 3.00 as unacceptable. They then correlated this
with completion of the Ph.D. program.
a. Rescore the data in Exercise 10.6 as indicated.
b. Run the correlation.
c. Test this correlation for significance.
10.10 Visualize the data in Exercise 10.9 as fitting into a contingency table.
a. Compute the chi-square on this table.
b. Show the relationship between chi-square and .frbrbrpbrpbrpb