Chi Square (1) Comm 215
Chi Square (1) Comm 215
PRACTICE PROBLEMS
Chi Square Tests
1. Brook is in charge of the accounts receivable department of COREX, Inc. An accountant wishes to
check on the department and asks Brook for his ideas about outstanding accounts. He replies that he
believes that accounts should be about as follows:
The accountant, knowing that it is impractical to examine all accounts, chooses a random sample of
60 accounts and finds 27, 19, 11 and 3 accounts in these categories. Can the accountant conclude
that Brook’s ideas are accurate? Set up appropriate hypotheses and report your conclusion using a
5% level of significance.
2. The theory that changes in employment and wage rates for various occupations are not independent
has been proposed. The following data were collected for furniture workers. Do the data support
the theory? Test the appropriate hypotheses at a .01 level of significance.
3. As a business major, did you study foreign languages? If so, will your foreign language skills make
you more marketable in the business community? To answer these questions, researchers mailed
questionnaires to personnel directors of both foreign-based and domestic businesses. The 215
responses to the question of whether a firm would give hiring preference to business majors
knowledgeable in foreign languages are summarized in the table:
YES NEUTRAL NO
U.S. FIRMS 50 57 19
FOREIGN FIRMS 60 22 7
a. Conduct an analysis to determine whether the percentages in the response categories for the
question depend on the type of firm. Use a 10% level of significance.
b. Construct a 90% confidence interval estimate for the percentage of U.S. firms that give hiring
preferences to business majors with foreign language skills.
5. An experimental project was undertaken to verify the belief that novice computer users prefer a
menu-oriented interface and that experienced users prefer a command-oriented interface. A software
product that can be used with both types of interfaces was selected to test this experiment. Novice,
trained and experienced computer users were given various tasks to perform and a tally was kept on
whether the user selected a menu-oriented or a command-oriented interface. The following table
provides the results of the experiment.
Do the data given provide sufficient evidence to indicate a relationship between experience level and
interface preference at the 5% level of significance?
6. The scores on a national achievement test are of considerable interest. The scores from a random
sample of 100 students selected from several schools are given in the table below.
a. Do these data provide sufficient evidence to indicate that scores on a national achievement
test are related to gender? Use a 5% level of significance.
b. Assume that the population mean of scores is 550 and the standard deviation is 75. From the
sample of males given above, what is the probability that the sample mean will not exceed a
score of 570?
a. Is the current distribution of reasons why employees were fired different from the results
obtained by ABC Inc.? Test the hypothesis using a 5% level of significance.
b. Of the 400, 350, 150 and 100 employees interviewed, respectively, 350, 320, 140 and 90 had
received a warning from their employer prior to the firing. At a 10% level of significance, can it
be concluded that the reason for firing is related to previous warnings given by the employer?
8. A job-training program offers training in advance computer programming. It is suspected that the
probability that an individual will be able to find a programming job within six months (after the
completion of the program) depends on previous related working experience.A random sample of 80
individuals trained in the program yields the following frequency table:
Zero 52
One year 16
More than one year 12
Of the 52, 16 and 12 individuals in the three experience categories listed in the table, respectively,
20, 8 and 10 were able to find a programming job within six months.
a. At a 5% level of significance, is there any evidence to support the hypothesis that the
probability of finding a job (within six months) depends on prior related working experience?
b. At a 10% level of significance, is there any evidence against the hypothesis that the
percentage of individuals in the program with “zero”, “one year” and “more than one year”
of experience are respectively 70%, 15%, and 15%?