Ninety mice from two emotionally divergent strains, SJL/J and SWR/J, were obtained from the Jackson Laboratory. The mice belonged to litters of three, five, or seven. At 33 days of age they were placed in separate cages and at 60 days of age they were run through a battery of tests consisting of 19 measures of emotionality. Scores were factor analyzed by alpha factoring followed by varimax and promax rotations. Factor scores were computed for all subjects on the six factors obtained and analyzed on a :2 X 3 factorial design with main effects for strain and litter size. Strain differences were found on five factors and litter-size differences on three. The differential effect of litter size on factors was discussed.One of the alleged advantages of factor analysis is that experimental results can be reported more precisely. Thus a factor analytic study in "emotionality" can distinguish several equally valid meanings of that term and make separate statements on each. The present study was directed toward effects of litter size during infancy on adult emotionality in mice. Studies on this and related topics of population density and infantile stimulation show a wide diversity of tests and consequently the terms "emotionality," "stress," and "arousal" have to be understood in a general sense. Using measures ranging from the open field to ratings on an "easy-to-handle" scale, Seitz (1954) found rats from larger litters (12) more emotional than rats from smaller litters. He suggested an optimal litter size for eliciting maternal behavior reductive of adult emotionality in the offspring. Mothers of small litters were rated "more maternal" than mothers of large litters. With the help of automated cages and with litter sizes of 4, 8, and 12, Ader and Grota (1970) showed that this is correct, at least as far as total time spent with litters is concerned. How-1 This research is reported from a thesis submitted by the first author in partial fulfillment of the master's degree at the University of Alberta.
Teachers’ rating of academic ability in mathematics and English were regressed on test scores in a sample of second grade students (n = 694 in mathematics, n = 736 in English). Fitted equations were interpreted as judgmental models of the rating process, and normative subjective probabilities were assigned to individual ratings, showing the extent to which they were defensible in light of the test scores. Within the terms of the model, teachers’ ratings were only marginally less defensible than ratings based on a previous administration of the same tests. Since systematic error in teacher ratings cannot be assumed, some new directions for research are suggested.
This paper examines the use of cluster analysis in educational research. Its first conclusion is that the important issues are not technical but interpretative. While we must, of course, reject clusters which fail to meet minimum technical requirements, it does not follow that those which are technically admissable are therefore meaningful explanatory constructs. In general, I will argue, they are not.The principal difficulty with cluster analysis is that it introduces a concept of 'type' which is unscientific. As a result, we become embroiled with character-sketches and pen-pictures more appropriate to literature than to research. This point has been touched on by critics who find cluster analysis to be 'gross' in certain respects: in its "failure to pinpoint significant teaching variables", in its tendency to "intensify" what are described as "unproductive polarizations" (Elliott, 1978, p. 77), in its propagation of "emotionally laden catch-all terms" (Wragg, 1976, p. 285). The second conclusion of the paper is that when the 'grossness' of cluster analysis is understood and accepted, the technique can nonetheless serve an important heuristic function, provided certain conditions are met. The paper refers to the Lancaster and ORACLE studies to illustrate the problems and possibilities for cluster analysis in educational research. It ends with some recommendations. Cluster AnalysisAn informal description of cluster analysis is adequate for our present purposes. Everitt (1974) is still the most accessible introduction. For an account of recent work, see Hand (1981, chapter 7). CLUSTAN (Wishart, 1978) is the most convenient computer package.In cluster analysis profiles of scores for individuals are transformed into a matrix of distances or similarities between all pairs of observations. So, if we have scores of teachers (or pupils) on a vector of variables, we can use cluster analysis to create groups which are similar in the sense that their score-profiles fall within some predetermined measure of closeness to each other. Educational research has typically used a clustering method which minimises the within-cluster sum of squares. This may be likened to the creation a posteriori of maximally different treatment groups in the ANOVA sense.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.