STATISTICS FOR MANAGERS

Answers, Ch. 1

1.19

(a) 12-47-83-76-22-skip 99-65-93-10-skip 65-skip 83-61-36-skip 98-89-58-86-92

(b) 12-47-83-76-22-skip 99-65-93-10-65-83-61-36-skip 98-89-58

.

1.23

(a) Either a simple random sample or a systematic sample could be used.  In a simple random sample, 200 numbers between 1 and 4,000 should be generated and those students should be included in the sample.  In a systematic sample, a random number between 1 and 20 is generated as the starting point, and then every 20th student after that should be included in the sample.

(b) The simple random sample is the most elementary of the sampling techniques.

(c) A systematic sample is easier and faster to use than a simple random sample.

(d) A stratified sample should be taken in this case.  The sample of 200 should reflect the gender and class status of the university as a whole:  17.5% freshman females (35 of 200), 14% freshman males (28 of 200), and so on, to 9.5% senior males (19 of 200).

(e) A cluster sample should be taken.  Each floor or dorm would be a cluster.  The floors and dorms are homogeneous across each other, and heterogeneous within each cluster.  Four floors or one dorm could be chosen at random, and the residents sampled.

.

1.27

a.  Some categories of employees may have been oversampled and some may have been undersampled.

b.  Some of the employees may choose not to participate in the survey.

c.  A nonrepresentative sample of the employees may have been chosen just by chance.

d.  There are three possible measurement errors:

• The questions asked may have been confusing or leading

• The respondent may have been trying to please the interviewer, i.e. management

• The employees may have exaggerated or not given full effort to their response