No abstract

No abstract

EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT 1974, 34, 537-545. The effect of the number of scale intervals of a continuous variable on the results of principal components factor analysis was investigated. Analyses were performed for seven different numbers of scale intervals. The general effect was a decrease in the size of the eigenvalues, communalities, and factor loadings as the number of scale divisions was reduced. The magnitude of the effect was, however, not large and the pattern of the rotated factor loadings was not appreciably affected.IN recent years, there has been widespread interest in, and application of, factor analysis in a number of fields. One consideration that confronts the researcher is that the model underlying factor analysis seldom matches precisely the characteristics of the data being analyzed. One problem that has concerned investigators is the effect of the number of intervals along the measurement scale of a, continuous variable on the results of the analysis. In its extreme form it is desired to know what effect reducing a continuous measurement scale to a dichotomy has on the analysis where the Pearson product moment correlation is used to represent the relationships between dichotomous variables (Carroll, 1961;Henrysson and Thunberg, 1965).

No abstract

How can people best able to learn Morse code be identified? To find out, a battery of 14 ability tests was given 310 airmen. Their rate of progress in training was related to this test performance. "A factor analysis of the ability measures was carried out and the factor structure of the criterion measures determined." Tests work best for predicting who will do best in the initial periods of Morse code training. After that it depends more on specific habits acquired in training.