Basic Statistics

REFERENCES 139

to either control (n = 142) or the new educational program (n = 140), and

their prescriptions of the analgesic were monitored from medical records. After

the education program, 58 of the physicians in the control group prescribed the

analgesic and 42 in the education group. Test for no difference in the proportion

prescribing the analgesic at the o = .05 level.

10.7 In high schools A and B, the proportion of teenagers smoking has been estimated

to be .20. The same antismoking educational program will be used again in

high school A but a new program is planned for high school B. There is concern

whether the new program will be better or worse than the old one or if the

results will be about the same. A study is planned to compare the proportion

smoking in the two high schools in June after the programs have been run.

The investigators want to take an adequate sample of students and ask them

questions concerning their smoking. They want to make sure that their sample

size is large enough so they can detect a difference of .10 in the proportion

smoking with Q: = .05 and p = .lo. What size sample should they take in each

high school?

REFERENCES

Cleveland, W. S. [1985]. The Elements of Graphing Data, Monterey, CA: Wadsworth, 264.

Dixon, W. J. andMassey, F. J. [ 19831. Introduction to StatisticalAnalysis, NewYork: McGraw-

Fleiss, J.L, Levin, B. and Paik, M. C. [2003]. Design andAnalysis of Clinical Experiments,

Good, P. J. and Hardin, J. W. [2003]. Common Errors in Statistics (and How to Avoid Them),

Mould, R. F. [ 19981. Introductory Medical Statistics. 3rd ed., Philadelphia: Institute of Physics

van Belle, G., Fisher, L. D., Heagerty, P. J. and Lumley, T. [2004]. Biostatistics: A Methodology

Hill, 281-284.

Hoboken, NJ: Wiley.

Hoboken, NJ: Wiley, 117-1 18.

Publishing, 61-14.

for the Health Sciences, 2nd ed., Hoboken, NJ: Wiley-Interscience, 15 1-163.