When a pilot study variance is used to estimate σ2 in the sample size formula, the resulting is a random variable. The authors investigate the theoretical behavior of . Though is more likely to underachieve than overachieve the unbiased n, correction factors to balance the bias are provided.
Get full access to this article
View all access options for this article.
References
1.
AbramowitzMilton and StegunIrene A., eds. (1965), Handbook of Mathematical Functions.New York: Dover Publications.
2.
GroeneveldRichard A. and MeedenGlen (1977), “The Mode, Median, and Mean Inequality,” American Statistician, 31 (August), 120–1.
3.
GuentherWilliam C. and ThomasP. O. (1965), “Some Graphs Useful for Statistical Inference,” Journal of the American Statistical Association, 60 (March), 334–43.
4.
LehmannDonald R. (1985), Marketing Research and Analysis.Homewood, IL: Richard D. Irwin, Inc.
5.
RohatgiV. K. (1976), An Introduction to Probability Theory and Mathematical Statistics.New York: John Wiley & Sons, Inc.
6.
ShifflerRonald E. and HarwoodGordon B. (1985), “An Empirical Assessment of Realized a-Risk When Testing Hypotheses,” Educational and Psychological Measurement, 45 (Winter), 811–23.
