Higher-order Factor Structure Research Articles

A Higher Order Analysis of the Factor Structure of the Myers-Briggs Type Indicator

This study examined the higher order factor structure of Form G of the Myers-Briggs Type Indicator (MBTI; K. C. Briggs & I B. Myers, 1987). A third-order component analysis of a sample (N = 926) found 2 higher order components. This higher order analysis contributes to the research literature pertaining to the generalized structure of the personality measure.

Evidence of a Single Second-Order Factor in Student Ratings of Teaching Effectiveness

In student-evaluation research, there has been considerable debate on the number of higher order factors in student ratings of university teaching. Using a distance learning program as the context for research, the study described in this article applied structural equation modeling to investigate the 2nd-order factor structure in student ratings of teaching effectiveness. An a priori model was hypothesized, which consisted of four 1st-order factors and one 2nd-order factor. Data were collected from 2,021 distance students using a multifactor questionnaire. In contrast to all past studies, the presence of a single 2nd-order factor in student ratings was empirically demonstrated through hierarchical confirmatory factor analysis. Implications for future research on the higher order factor structure in student ratings of conventional on-campus courses are also discussed.

The Dimensionality of the Maslach Burnout Inventory across Small Business Owners and Educators

This study tested the dimensionality of the Maslach Burnout Inventory (MBI) by comparing three factor structures (i.e., a one-factor structure, a three-factor structure, and a higher order factor structure) in two diverse samples. The comparison of the LISREL measurement models was extended by a series of measurement invariance tests. Additionally, constructs related to burnout had a pattern of correlations to the three MBI dimensions that was similar across the two samples. In aggregate, the analyses suggested that the three-factor structure of the MBI is the most plausible model. By using a sample of small business owners, the current research contributed to existing knowledge on the MBI by establishing the dimensionality and generalizability of the MBI beyond human service occupations.

Gender differences in dimensions of anxiety sensitivity

Anxiety sensitivity (AS) is the fear of anxiety-related sensations arising from beliefs that these sensations have harmful physical, psychological, or social consequences. AS is measured using the Anxiety Sensitivity Index (ASI), a 16-item self-report questionnaire. Little is known about the origins of AS, although social learning experiences (including sex-role socialization experiences) may be important. The present study examined whether there were gender differences in: (a) the lower- or higher-order factor structure of the ASI; and/or (b) pattern of ASI factor scores. The ASI was completed by 818 university students (290 males; 528 females). Separate principal components analyses on the ASI items of the total sample, males, and females revealed nearly identical lower-order three-factor structures for all groups, with factors pertaining to fears about the anticipated (a) physical, (b) psychological, and (c) social consequences of anxiety. Separate principal components analyses on the lower-order factor scores of the three samples revealed similar unidimensional higher-order solutions for all groups. Gender × AS dimension analyses on ASI lower-order factor scores showed that: females scored higher than males only on the physical concerns factor; females scored higher on the physical concerns factor relative to their scores on the social and psychological concerns factors; and males scored higher on the social and psychological concerns factors relative to their scores on the physical concerns factor. Finally, females scored higher than males on the higher-order factor representing the global AS construct. The present study provides further support for the empirical distinction of the three lower-order dimensions of AS, and additional evidence for the theoretical hierarchical structure of the ASI. Results also suggest that males and females differ on these various AS dimensions in ways consistent with sex role socialization practices.

Personality Assessment Inventory: A Confirmatory Factor Analysis

The Personality Assessment Inventory is a recently constructed multidimensional self-report measure of personality traits. Morey reported the results of a scale factoring, claiming that the instrument measures four separate higher-order dimensions; however, in an independent Australian study of the psychometric properties of the inventory, Boyle and Lennon found five higher-order dimensions, using factor analytic procedures intended to maximize simple structure. The present paper reports the results of a confirmatory factor analysis for the proposed model based on the Australian data. The results indicate that the model does not provide a satisfactory fit, raising questions about the higher-order factor structure.

Higher-Order Abilities Conceptualized with in Guilford's Structure-Of-Intellect (Soi) Model for a Sample of United States Coast Guard Academy Cadets: A Reanalysis of an Soi Data Base

This study was undertaken to provide empirical validation of the first-order and higher-order factor structures of the structure-of-intellect (SOI) model conceptualized by Guilford. Both confirmatory maximum likelihood and exploratory (varimax and promax) factor analytic methodologies were employed to reanalyze an existing correlational data base (N - 178) involving 20 SOI experimental tests, 9 Coast Guard aptitude tests, and 10 Coast Guard criterion measures originating from the University of Southern California Aptitudes Research Project Report Number 13 entitled "The Relation of Certain Thinking Factors to Training Criteria in the U.S. Coast Guard Academy" (Guilford, Kettner, and Christensen, 1955). Although nine of the originally identified 11 first-order orthogonal factors could be replicated, oblique solutions involving higher-order factors accounted for a larger proportion of covariance among the test and criterion measures than did orthogonal solutions. The presence of identifiable first-order and higher-order correlated factors as well as a general factor suggested that the SOI model could be reconceptualized as a pyramid-like hierarchical theory of intelligence.

Psychometric evaluation of a short form of the inventory of drinking situations.

The Inventory of Drinking Situations (IDS) was developed to assess the likelihood of alcoholic relapse in eight high-risk situations. Previous research questioned the factor stability of the instrument; therefore, an alternative, short form of the IDS (IDS42) was developed and subjected to preliminary examination. This project reports on the further examination of the psychometric characteristics of the IDS42. The instrument's primary factor stability and higher-order factor structure were examined, and subscale means, standard deviations, alpha coefficients, interscale correlations and independence (unique variance) were determined. Also, the relationships between the IDS42 subscale scores and alcohol dependence, social desirability and demographic characteristics were explored. Results of the principal components analysis support the instrument's primary factor stability and indicate that the scales load onto one higher order dimension. The subscale internal consistency estimates were very high. Significantly high positive correlations were found between the IDS42 subscales and alcohol dependence. The IDS42 is influenced by a socially desirable response set that may result in spuriously low IDS42 scale elevations. The subscales initially demonstrated high interscale correlations; however, estimates of unique variance show that the IDS42 scales have an acceptable level of independence. Except for a significant negative correlation with age for three subscales, demographic characteristics do not appear to influence IDS42 results. These findings support the use of the IDS42 in assessing the extent to which heavy alcohol use has occurred in five situations. With this information, it may be possible to prepare to avoid or cope with these situations in the future and, consequently, to reduce the potential for relapse.(ABSTRACT TRUNCATED AT 250 WORDS)

Factor structure within and across three family-assessment procedures.

Higher order factor structures within and across the Family Environment Scale (FES), the Family Assessment Measure (FAM), and Family Adaptability and Cohesion Evaluation Scales III were examined. A sample of 138 families was obtained; separate analyses were conducted for mothers, facthers, and their adolescent children.

Higher-Order Factors in Structure-of-Intellect (SOI) Aptitude Tests Hypothesized to Portray Constructs of Military Leadership: A Re-Analysis of an SOI Data Base

Taken from a 1959 research report by Marks, Guilford, and Merrifield, a correlation matrix of 21 structure-of-intellect (SOI) tests intended to reflect aptitudes required in military leadership was re-analyzed through use of relatively objective exploratory factor analytic procedures (varimax and promax solutions for orthogonal and oblique structures, respectively) and through employment of confirmatory maximum likelihood factor analyses. For the sample of 204 Marine officers, only modest correspondence was found in the exploratory analyses between the solution by Marks et al. based on the orthogonal rotation of factor axes from a centroid matrix and the factor solutions obtained from application of the varimax and promax methods of rotation. An evaluation of selected alternative models comprising first-order and higher-order factors involving use of LISREL methodology revealed that a two-factor oblique model affording a separation of semantic and figural contents yielded the greatest promise in accounting for the covariation among the 21 aptitude tests. A higher-order oblique factor model furnishing a differentiation among three psychological operations of cognition, divergent production, and evaluation ranked second in its capability of reproducing the original correlation matrix. The conclusion was reached that the covariance among these aptitude tests can be meaningfully conceptualized within higher-order factor structures. A short discussion highlights possible difficulties and limitations likely to be realized in establishing the construct validity of the hypothesized first-order dimensions within the structure-of-intellect model.

The Structure of Masculinity/Femininity: An Application of Confirmatory Factor Analysis to Higher-Order Factor Structures and Factorial Invariance.

The purpose of this study is to examine the factor structure of responses to the masculinity/femininity (MF) scale of the Comrey Personality Scales for males (N = 366) and females (N = 384). The study also illustrates the use of confirmatory factor analysis for testing hierarchical factor structures and factorial invariance. The MF scale consists of M and F items designed to measure five distinguishable components of MF. Neither a one-factor solution with a single bipolar MF factor, nor a two-factor solution with separate M and F factors, was able to fit the data. However, a five-factor solution derived from the design of the MF scale provided an adequate fit, and the relationships among the five first-order factors could be explained by a single higher-order, bipolar MF factor. Contrary to the findings and proposals from androgyny research, when separate masculinity and femininity constructs were defined, the correlation between the constructs was nearly - 1.0. The findings argue that MF is a multifaceted, hierarchical construct instead of either a single bipolar MF construct, or separate M and F factors.

A reanalysis of the higher-order factor structure of the motivation analysis test and the eight state questionnaire

In an exploratory study of the higher-order factor structure of the Motivation Analysis Test (MAT) and the Eight State Questionnaire (8SQ), Boyle (1983c) obtained an 11-factor solution comprising 9 second-order MAT factors and 2 higher-order 8SQ factors. However, application of more conservative criteria regarding the size of significant factor pattern loadings, significance of derived factors, together with reinterpretation of the appropriate Scree ‘break’, suggest that in Boyle's earlier analysis 2 factors too many were extracted. Recalculation of the factor pattern for the MAT (using 8SQ data as ‘hyperplane stuff’) supports the view that 7 rather than 9 second-order MAT factors were appropriate. A separate dR-factoring of the intercorrelations of the subscale difference scores indicated 3 higher-order 8SQ factors, indicative of change dimensions, and not 2 factors as found in the static single-occasion combined factoring of both the MAT and 8SQ data.

Higher-order factor structure of Cattell's MAT and 8SQ

Previous research has failed to delineate adequately the higher-order factor structure of the Eight State Questionnaire (8SQ), and of the Motivation Analysis Test (MAT). Kline (1979, p.185) concluded that, "... there is considerable need for further investigaion... utilizing factorial procedures in accord with the technical demands which are listed... so that one can have confidence in the attainment of simple structure and the consequent psychological import of the results." In response to Kline's suggestion, the present study indicated two third-order 8SQ factors, and nine second-order MAT factors, on an Australian sample of 258 college students. Use of Cattell's Rotoplot program was shown to be a valid and useful analytic procedure. Findings are discussed and tentative labels are assigned to each higher-order factor.

The Construct Validity of Higher Order Structure-of-Intellect Abilities in a Battery of Tests Emphasizing the Product of Transformations: A Confirmatory Maximum Likelihood Factor Analysis

Within the context of a causal modeling system involving use of Joreskog's confirmatory maximum likelihood factor analysis in conjunction with the LISREL IV computer program devised by Joreskog and Sorbom an evaluation was made of the construct validity underlying the higher order factor structure of a given correlation matrix of 46 structure-of-intellect (SOI) tests emphasizing the product of transformations. The tenability of four alternative factor models was examined relative to their accounting for the intercorrelations among the 46 tests that had been hypothesized to represent first-order abilities involving one kind of operation, one type of content, and one form of product. The major elements of each of the three postulated dimensions of the SOI model (the five operations, the five contents, and the six products) were defined as the 16 possible third-order factors. The following conclusions resulted: (a) the five third-order SOI factors comprising operations of cognition, convergent production, divergent production, memory, and evaluation do not appear to be highly differentiated (with the cognition and convergent production factors being substantially correlated); (b) the two third-order factors of symbolic and semantic content are relatively discrete; (c) the third-order product factor involving the ability to complete tasks requiring hypothesized transformations is an identifiable construct; and (d) evidence of the construct validity of the third-order factors in the SOI test battery analyzed appears to be at best only equivocal. The predominance of tests intended to reflect the product of transformations (32 out of 46) very possibly created inflated levels of intercorrelations among SOI tests that prevented the emergence of distinct higher order dimensions. New confirmatory maximum likelihood factor analyses of correlational matrices derived from SOI tests in which a careful balance has been achieved in the representation of hypothesized third-order factors across the three basic dimensions of the SOI model are expected to demonstrate differentiability among the higher order constructs.