Unique problems are considered with the analysis of proportional frequencies in the 2 by 2 design due to the possibility of interchanging effects by rearranging cells. It is noted that the researcher must precisely state the hypotheses of interest, make a decision about population proportions, and determine the validity of the test of significance. Each of these decisions is considered in derail and an example presented by Gerjuoy and Myers (1964) is used to illustrate the different choices.
Get full access to this article
View all access options for this article.
References
1.
BancroftT. A.Topics in intermediate statistical methods. Vol. 1. Ames, Iowa: Iowa State Univer. Press, 1968.
2.
CohenJ.Multiple regression as a general data-analytic system. Psychological Bulletin, 1968, 70, 426–443.
3.
GerjuoyH.MyersA. E.Analysis of variance: effects of interchanging main effects and interactions by rearranging data. Perceptual and Motor Skills, 1964, 18, 833–838.
4.
HockingR. R.SpeedF. M.Unbalanced data. American Statistician, 1976, 30, 205–206.
5.
KutnerM. H.Hypothesis testing in linear models (Eisenhart Model I). American Statistician, 1974, 28, 98–100.
6.
KutnerM. H.Hypothesis tests in linear models. American Statistician, 1975, 29, 133–134.
7.
ShafferJ. P.Reorganization of variables in analysis of variance and multidimensional contingency tables. Psychological Bulletin, 1911, 84, 220–228.
8.
SpeedF. M.HockingR. R.The use of the R()-notation with unbalanced data. American Statistician, 1976, 30, 30–33.
9.
WinerB. J.Statistical principles in experimental design. (2nd ed.) New York: McGraw-Hill, 1971.
10.
YatesF.The analysis of multiple classifications with unequal numbers in the different classes. Journal of the American Statistical Association, 1934, 29, 51–66.
