170 Chapter 6 Categorical Data and Chi-Square
6.15 Darley and Latané (1968) asked subjects to participate in a discussion carried on over an in-
tercom. Aside from the experimenter to whom they were speaking, subjects thought that
there were zero, one, or four other people (bystanders) also listening over intercoms. Part-
way through the discussion, the experimenter feigned serious illness and asked for help.
Darley and Latané noted how often the subject sought help for the experimenter as a func-
tion of the number of supposed bystanders. The data follow:What could Darley and Latané conclude from the results?
6.16 In a study similar to the one in Exercise 6.15, Latané and Dabbs (1975) had a confederate
enter an elevator and then “accidentally” drop a handful of pencils. They then noted whether
bystanders helped pick them up. The data tabulate helping behavior by the gender of the by-
stander:
Gender of Bystander
Female Male
Help 300 370 670
No Help 1003 950 1953
1303 1320 2623
What could Latané and Dabbs conclude from the data? (Note that when we collapse over
gender, only about one-quarter of the bystanders helped. That is not relevant to the ques-
tion, but it is an interesting finding that could easily be missed by routine computer-based
analyses.)
6.17 In a study of eating disorders in adolescents, Gross (1985) asked each of her subjects
whether they would prefer to gain weight, lose weight, or maintain their present weight.
(Note: Only 12% of the girls in Gross’s sample were actually more than 15% above their
normative weight—a common cutoff for a label of “overweight.”) When she broke down
the data for girls by race (African-American versus white), she obtained the following re-
sults (other races have been omitted because of small sample sizes):
Reducers Maintainers Gainers
White 352 152 31 535
African-American 47 28 24 99
399 180 55 634
a. What conclusions can you draw from these data?
b. Ignoring race, what conclusion can you draw about adolescent girls’ attitudes toward
their own weight?
6.18 Use the likelihood ratio approach to analyze the data in Exercise 6.10.
6.19 Use the likelihood ratio approach to analyze the data in Exercise 6.12.
6.20 It would be possible to calculate a one-way chi-square test on the data in row 2 of the table
in Exercise 6.12. What hypothesis would you be testing if you did that? How would that
hypothesis differ from the one you tested in Exercise 6.12?Sought Assistance
Ye s N o
0 11 2 13
Number of
1 16 10 26
Bystanders
4 4 9 13
31 21 52