Independently Fisher and K. Pearson suggested similar methods for combining probability integrals from independent tests of significance. These methods (a) can lead to anomalous results, (b) are not sensitive to hypotheses alternative to the hypothesis under test, and (c) are biased. An alternative method, free of these limitations, is proposed. Combined probabilities from the Fisher, Pearson, and proposed methods are compared against results obtained from traditional analyses, and the proposed method is found to be superior.
Get full access to this article
View all access options for this article.
References
1.
AlexanderH. W.Elements of mathematical statistics. New York: Wiley, 1961.
2.
BakerP. C.Combining tests of significance in cross-validation. Educational and Psychological Measurement, 1952, 12, 300–306.
3.
BirnbaumA.Combining independent tests of significance. Journal of the American Statistical Association, 1954, 49, 559–574.
4.
DavidF. N.On the Pλn test for randomness: remarks, further illustration, and table of Pλn for given values of —log10 λn. Biometrika, 1934, 26, 1–11.
5.
EdwardsA. L.Statistical methods for the behavioral sciences. New York: Rinehart, 1954.
6.
FisherR. A.Statistical methods for research workers. (4th ed.) London: Oliver & Boyd, 1932.
7.
FisherR. A.Combining independent tests of significance. American Statistician, 1948, 2, 30.
8.
GordonM. H.LovelandE. H., & CuretonE. E.An extended table of chi-square for two degrees of freedom for use in combining probabilities from independent samples. Psychometrika, 1952, 17, 311–316.
9.
GuilfordJ. P.Fundamental statistics in psychology and education. (4th ed.) New York: McGraw-Hill, 1965.
10.
JanisI. L., & FieldP. B.Sex differences and personality factors related to persuasibility. In HovlandC. I. & JanisI. L. (Eds.), Personality and persuasibility. New Haven: Yale Univer. Press, 1959. Pp. 55–68.
11.
JohnsonP. O., & JacksonR. W. B.Modern statistical methods: descriptive and inductive. Chicago: Rand-McNally, 1959.
12.
JonesL. V., & FiskeD. W.Models for testing the significance of combined results. Psychological Bulletin, 1953, 50, 375–382.
13.
KatzD.McClintockC., & SarnoffI.The measurement of ego defense as related to attitude change. Journal of Personality, 1957, 25, 465–474.
14.
KendallM. G.The advanced theory of statistics. Vol. 2. (3rd ed.) London: Charles Griffin, 1951.
15.
KozelkaR. M.Elements of statistical inference. Reading, MA: Addison-Wesley, 1961.
16.
LancasterH. O.The combination of probabilities arising from data in discrete distributions. Biometrika, 1949, 36, 370–382.
MostellerF., & BushR. R.Selected quantitative techniques. In LindzeyG. (Ed.), Handbook of social psychology. Vol. 1. Cambridge, MA: Addison-Wesley, 1954. Pp. 289–334.
20.
NairK. R.A note on the exact distribution of λn. Sankhyā: The Indian Journal of Statistics, 1937, 3, 171–174.
21.
NairK. R.On Tippett's “Random sampling numbers.”Sankhyā: The Indian Journal of Statistics, 1938, 4, 65–72.
22.
NeymanJ.“Smooth” test for goodness of fit. Skandinavisk Aktuarietidskrift, 1937, 20, 149–199.
23.
PearsonE. S.The probability integral transformation for testing goodness of fit and combining independent tests of significance. Biometrika, 1938, 30, 134–148.
24.
PearsonK.On a method of determining whether a sample of size n supposed to have been drawn from a parent population having a known probability integral has probably been drawn at random. Biometrika, 1933, 25, 379–410.
25.
PearsonK.On a new method of determining “goodness of fit.”Biometrika, 1934, 26, 425–442.
26.
RAND Corp. A million random digits with 100,000 normal deviates. Glencoe, IL: Free Press, 1955.
27.
RaoC. R.Advanced statistical methods in biometric research. New York: Wiley, 1952.
28.
ThorndikeE. L.Sex differences in status and gain in intelligence from thirteen to eighteen. Pedagogical Seminary, 1926, 33, 167–181. (a)
29.
ThorndikeE. L.On the variability of boys and girls from thirteen to eighteen. Pedagogical Seminary, 1926, 33, 182–184. (b)
30.
TippettL. H. C.Random sampling numbers. (Tracts for Computers No. 15) Cambridge, Eng.: University Press, 1927.
31.
TippettL. H. C.The methods of statistics. (4th ed.) New York: Wiley, 1952.
