British Journal of Mathematical and Statistical Psychology

Dunsdon, M. I.; Roberts, J. A. Fraser

doi: 10.1111/j.2044-8317.1957.tb00001.xpmid: N/A

Bernyer, G.

doi: 10.1111/j.2044-8317.1957.tb00002.xpmid: N/A

The following paper presents a brief survey of current factorial procedures, with a discussion of their chief advantages and limitations. It is intended primarily for research workers in general psychology who desire to analyse the factors involved in their data without making a detailed study of all available procedures. An attempt is made to determine the commoner faults, either of theory or of practice, which have rendered so many recent factorial investigations contradictory or inconclusive; and a number of defects are noted in existing techniques which appear to call for further investigation by the mathematical theorist. The main conclusion is that, like other statistical procedures, factor analysis is merely an adjunct to psychological research, not a method sufficient in itself.

Stuart, Alan

doi: 10.1111/j.2044-8317.1957.tb00003.xpmid: N/A

For purposes of statistical comparison, matched samples are in general more accurate than unmatched. In testing the statistical significance of the results, it is essential to take into account the gain in accuracy, due to the matching. The following paper indicates how this can be done in the case of attributes.

Burt, Cyril; Howard, Margaret

doi: 10.1111/j.2044-8317.1957.tb00004.xpmid: N/A

An attempt is made to remove current misconceptions about the way in which individual differences in general intelligence appear to be inherited. Further support for a multifactorial form of the Mendelian theory is found in an examination (a) of the commoner objections, both methodological and statistical, and (b) of the chief alternative hypotheses, viz., (i) the theory of blending inheritance and (ii) the theory of exclusively or predominantly environmental determination.

doi: 10.1111/j.2044-8317.1957.tb00005.xpmid: N/A

Book reviewed in this article: