Discrepancies Between Factor Analysis and Multivariate Discrimination Among Groups As Applied to Personality Theory

1958 ◽  
Vol 104 (436) ◽  
pp. 713-721 ◽  
Author(s):  
Lowell H. Storms
Keyword(s):  
Factor Analysis ◽  
Principal Components ◽  
Group Differences ◽  
Personality Theory ◽  
Factor Analyses ◽  
Independent Evidence ◽  
Loss Of Information ◽  
Response Measures ◽  
Intervening Variables

It is clearly advantageous to the personality theorist to know which of his measures covary and to what extent. The results of factor analyses can be useful in summarizing some of the patterns of covariation and classifying response measures. An indefinitely large number of sets of factors may equally well describe the same set of data, however, and, before any choice of one particular set of factors, dimensions, principal components, etc., can be considered as revealing basic intervening variables or constructs in a personality theory, independent evidence must be provided. It is the purpose of this paper to reveal some of the pitfalls in making such a choice by showing how concentration on one set of factors obtained by analysing a multivariate scatter can lead to a serious loss of information when applied to the analysis of group differences in the same set of data and can suggest interpretations which are at odds with the full evidence.

Profile Factor Analysis and Variable Factor Analysis

1964 ◽  
Vol 15 (1) ◽  
pp. 119-127 ◽  
Author(s):  
Robert G. Ryder
Keyword(s):  
Factor Analysis ◽  
Principal Components ◽  
The Other ◽  
Factor Analyses ◽  
Variable Factor ◽  
Equivalent Factor

Nunnally (1962) mentions three types of profile factor analyses: correlational, covariance, and raw score sums of crossproducts. When principal components factors are used, each of these profile analyses corresponds to a precisely equivalent factor analysis of variables. In particular, if variables have previously been standardized over Ss, factor analysis of raw sums of profile crossproducts yields exactly the same results as factor analysis of correlations among variables. An example is worked out in which it is shown that factor analysis of raw sums of profile crossproducts yields exactly the same factors as factor analysis of correlations between variables. It is concluded that, of the procedures considered, factor analysis of correlations between variables is generally the most useful, because findings are easiest to communicate, and because results can be used to estimate results of the other possible factor analyses.

Factor Analysis of the Motor Steadiness Battery

1973 ◽  
Vol 36 (3) ◽  
pp. 779-783
Author(s):  
Stewart Meikle ◽  
Brenda Lonsbury ◽  
Richard Gerritse
Keyword(s):  
Factor Analysis ◽  
Elementary School ◽  
School Children ◽  
Principal Components ◽  
Elementary School Children ◽  
Loss Of Information

50 elementary school children between the ages of 8 and 12 yr. were tested on the Motor Steadiness Battery. The 20 derived raw scores were intercorrelated and factored by the principal components technique. Seven of the 10 resulting factors were interpreted as supporting the view that attempts to shorten the battery could only be undertaken at the expense of some loss of information.

Factor analysis and personality theory

1987 ◽  
Vol 1 (1) ◽  
pp. 21-36 ◽  
Author(s):  
Paul Kline
Keyword(s):  
Factor Analysis ◽  
Analytic Approach ◽  
Analytic Method ◽  
Personality Theory ◽  
Factor Analyses ◽  
Factor Analytic Approach ◽  
Factor Analytic ◽  
Personality Questionnaires ◽  
Factor Analytic Method

This article deals with the factor analytic approach to personality. More specifically, it deals with problems in factor analyses of personality questionnaires which contribute to factorial confusion. It is stated that in fact the factorial results make better sense than is usually admitted. The apparent disparity of results can be accounted for by technical defects in the chosen factor analytic method. Furthermore, it is shown that another source of disagreement lies in the interpretation of what are essentially the same factors.

Towards a Validation of Multiple Features in the Assessment of Emotions

1998 ◽  
Vol 14 (3) ◽  
pp. 202-210 ◽  
Author(s):  
Suzanne Skiffington ◽  
Ephrem Fernandez ◽  
Ken McFarland
Keyword(s):  
Factor Analysis ◽  
Pilot Study ◽  
Cognitive Appraisal ◽  
Negative Emotions ◽  
The Self ◽  
Factor Analyses ◽  
Data Set ◽  
Multiple Features ◽  
Action Tendency ◽  
Intensity Range

This study extends previous attempts to assess emotion with single adjective descriptors, by examining semantic as well as cognitive, motivational, and intensity features of emotions. The focus was on seven negative emotions common to several emotion typologies: anger, fear, sadness, shame, pity, jealousy, and contempt. For each of these emotions, seven items were generated corresponding to cognitive appraisal about the self, cognitive appraisal about the environment, action tendency, action fantasy, synonym, antonym, and intensity range of the emotion, respectively. A pilot study established that 48 of the 49 items were linked predominantly to the specific emotions as predicted. The main data set comprising 700 subjects' ratings of relatedness between items and emotions was subjected to a series of factor analyses, which revealed that 44 of the 49 items loaded on the emotion constructs as predicted. A final factor analysis of these items uncovered seven factors accounting for 39% of the variance. These emergent factors corresponded to the hypothesized emotion constructs, with the exception of anger and fear, which were somewhat confounded. These findings lay the groundwork for the construction of an instrument to assess emotions multicomponentially.

The Structure Among Measures of Personality, Social Attitudes, Values, and Social Norms

2007 ◽  
Vol 28 (4) ◽  
pp. 240-251 ◽  
Author(s):  
Lazar Stankov
Keyword(s):  
Factor Analysis ◽  
Individual Differences ◽  
Social Norms ◽  
Social Interactions ◽  
Correlation Matrix ◽  
Social Attitudes ◽  
Group Differences ◽  
Factor Scores ◽  
Ethnic Group Differences ◽  
Inside Out

Abstract. This paper presents the results of a study that employed measures of personality, social attitudes, values, and social norms that have been the focus of recent research in individual differences. These measures were given to a sample of participants (N = 1,255) who were enrolled at 25 US colleges and universities. Factor analysis of the correlation matrix produced four factors. Three of these factors corresponded to the domains of Personality/Amoral Social Attitudes, Values, and Social Norms; one factor, Conservatism, cut across the domains. Cognitive ability showed negative correlation with conservatism and amoral social attitudes. The study also examined gender and ethnic group differences on factor scores. The overall interpretation of the findings is consistent with the inside-out view of human social interactions.

Factor Analysis of the Stability of Self Scale

1980 ◽  
Vol 47 (3_suppl) ◽  
pp. 1160-1162 ◽  
Author(s):  
Stephen L. Franzoi ◽  
Benjamin J. Reddish
Keyword(s):  
Factor Analysis ◽  
Factor Structure ◽  
Principal Components ◽  
Analysis Data ◽  
Factor Solution ◽  
The Stability

The factor structure of Rosenberg's Stability of Self Scale (1965) was investigated via principal components factor analysis. Data from 92 male and 171 female undergraduates yielded a one-factor solution, supporting Rosenberg's contention that the scale is unidimensional.

Preliminary Examination of Psychometric Properties of the Korean Version of the Specific Interpersonal Trust Scale

2007 ◽  
Vol 100 (2) ◽  
pp. 355-364 ◽  
Author(s):  
Sook-Jeong Lee
Keyword(s):  
Factor Analysis ◽  
Confirmatory Factor Analysis ◽  
Psychometric Properties ◽  
Principal Components Analysis ◽  
Principal Components ◽  
Interpersonal Trust ◽  
Confirmatory Factor ◽  
Components Analysis ◽  
Original Scale ◽  
Trust Scale

The purpose of this study was to examine the psychometric properties of the Specific Interpersonal Trust scale of Johnson-George and Swap in Korean samples as a part of the process of providing an exemplary tool for intercultural studies of trust. A translated version of the original scale was administered to 337 university students (157 males, 180 females) in Seoul, Korea. Data were subjected to a principal components analysis and a confirmatory factor analysis. In principal components analysis for the Korean sample ( n = 167), three factors were identified and labeled: Overall Trust (Cronbach α=.89), Emotional Trust (Cronbach α = .88), and Reliableness (Cronbach α=.84). A confirmatory factor analysis ( n=170) showed that the three-factor model was valid for the sample (χ2/ df= 1.78, RMR=.06, RMSEA = .07, TLI=.92, CFI=.93). Internal consistency reliability and factorial validity were satisfactory in the case of the Korean sample. The Korean version of the Specific Interpersonal Trust Scale made good use of three factors of trust and appeared to be valid without sex differences, while the original scale distinguished the Males subscale from the Females subscale. Implications and limitations of this study were discussed.

Research Note: Some Problems with Measurement of Macro-organizational Structure

1981 ◽  
Vol 2 (3) ◽  
pp. 287-296 ◽  
Author(s):  
Peter H. Grinyer ◽  
Masoud Yasai-Ardekani
Keyword(s):  
Factor Analysis ◽  
Empirical Study ◽  
Organizational Structure ◽  
Electrical Engineering ◽  
Original Study ◽  
Research Note ◽  
Loss Of Information ◽  
The Uk

Problems associated with the use of Aston psychometrically based measures are evalu ated in the light of experience gained in and the findings of an empirical study of 45 electrical engineering companies in the UK in which the Aston methodology was used. It is shown that (a) the unidimensionality of multi-item measures must be clearly established if loss of information is to be avoided, (b) scales constructed by aggregation of a number of subscales suggested by factor analysis should not be given general labels beyond the description of subscales included in the final scale, and (c) abbreviated scales based on the original study may only reflect sample-specific relationships and may not be used as proxies of original scales in the study of other samples. The objectivity of factor analysis is also addressed.

Sign in / Sign up

Export Citation Format

Share Document