Confirmatory Factor Analysis and Profile Analysis via Multidimensional Scaling

This paper describes the Confirmatory Factor Analysis (CFA) parameterization of the Profile Analysis via Multidimensional Scaling (PAMS) model to demonstrate validation of profile pattern hypotheses derived from multidimensional scaling (MDS). Profile Analysis via Multidimensional Scaling (PAMS) is an exploratory method for identifying major profiles in a multi-subtest test battery. Major profile patterns are represented as dimensions extracted from a MDS analysis. PAMS represents an individual observed score as a linear combination of dimensions where the dimensions are the most typical profile patterns present in a population. While the PAMS approach was initially developed for exploratory purposes, its results can later be confirmed in a different sample by CFA. Since CFA is often used to verify results from an exploratory factor analysis, the present paper makes the connection between a factor model and the PAMS model, and then illustrates CFA with a simulated example (that was generated by the PAMS model) and at the same time with a real example. The real example demonstrates confirmation of PAMS exploratory results by using a different sample. Fit indexes can be used to indicate whether the CFA reparameterization as a confirmatory approach works for the PAMS exploratory results.     

Confirmatory Factor Analysis of Multitrait-Multimethod Data: The Construct Validation of Multidimensional Self-Concept Responses

ABSTRACT The multitrait-multimethod (MTMM) design is frequently used to test construct validity and is particularly appropriate for evaluating multidimensional instruments in personality research Despite its popularity there are important problems with both the traditional and confirmatory factor analysis (CFA) approaches to MTMM data In previous research (1989a, 1989b) I described a new CFA approach that I claimed to be relatively immune to many of these problems This approach is applied here to test the construct validity of preadolescent responses to three multidimensional self-concept instruments The substantive contribution of the present investigation is to demonstrate good support for the construct validity of two of the three instruments The methodological contributions are to further substantiate my claims about the new CFA approaches and to demonstrate this approach in a substantively meaningful context that has broad applicability to other personality research     

教师组织承诺结构的验证性因素分析

通过文献回顾、半结构访谈、问卷调查及理论分析,提出了教师组织承诺的结构。对278名教师进行测查以考察教师组织承诺的因素结构,验证了教师组织承诺的结构,发现教师组织承诺包含四个维度,即感情承诺、规范承诺、理想承诺和投入承诺。     

Determining the Dimensionality of Multidimensional Scaling Representations for Cognitive Modeling

Multidimensional scaling models of stimulus domains are widely used as a representational basis for cognitive modeling. These representations associate stimuli with points in a coordinate space that has some predetermined number of dimensions. Although the choice of dimensionality can significantly influence cognitive modeling, it is often made on the basis of unsatisfactory heuristics. To address this problem, a Bayesian approach to dimensionality determination, based on the Bayesian Information Criterion (BIC), is developed using a probabilistic formulation of multidimensional scaling. The BIC approach formalizes the trade-off between data-fit and model complexity implicit in the problem of dimensionality determination and allows for the explicit introduction of information regarding data precision. Monte Carlo simulations are presented that indicate, by using this approach, the determined dimensionality is likely to be accurate if either a significant number of stimuli are considered or a reasonable estimate of precision is available. The approach is demonstrated using an established data set involving the judged pairwise similarities between a set of geometric stimuli. Copyright 2001 Academic Press.     

Estimating Cognitive Profiles Using Profile Analysis via Multidimensional Scaling (PAMS)

Two of the most popular methods of profile analysis, cluster analysis and modal profile analysis, have limitations. First, neither technique is adequate when the sample size is large. Second, neither method will necessarily provide profile information in terms of both level and pattern. A new method of profile analysis, called Profile Analysis via Multidimensional Scaling (PAMS; Davison, 1996), is introduced to meet the challenge. PAMS extends the use of simple multidimensional scaling methods to identify latent profiles in a multi-test battery. Application of PAMS to profile analysis is described. The PAMS model is then used to identify latent profiles from a subgroup (N = 357) within the sample of the Woodcock-Johnson Psychoeducational Battery-Revised (WJ-R; McGrew, Werder, & Woodcock, 1991; Woodcock & Johnson, 1989), followed by a discussion of procedures for interpreting participants' observed score profiles from the latent PAMS profiles. Finally, advantages and limitations of the PAMS technique are discussed.     

Confirmatory Factor Structure and Psychometric Properties of the Multidimensional Peer Victimization Scale

The Multidimensional Peer Victimization Scale (MPVS; Mynard and Joseph, British Journal of Educational Psychology, 67, 51-54 2000) is a 16-item self-report scale that captures peer victimization across four dimensions: physical victimization, verbal victimization, social manipulation, and attacks on property. Performance of the scale has not been evaluated among older adolescents. We examined the factor structure, internal consistency reliability, and performance of the scale in two separate epidemiological U.S. samples representing different age groups: 9–14 year olds (N?=?610) and 15–17 year olds (N?=?524). The four-factor structure of the scale was affirmed in both samples, however; there was not metric invariance by gender in the younger age group. The scale and its subscales were found to have good internal consistency. Expected relationships between the MPVS and measures of irritability, anxiety, and depression were affirmed. Results support continued use of the MPVS among child and adolescent samples.     

工作倦怠结构的验证性因素分析

用所编制的工作倦怠问卷,先后对541名被试的调查,对工作倦怠的结构进行探讨。结果表明,工作倦怠由耗竭、人格解体和成就感降低三个因素构成。工作倦怠的三因素结构模型在三个职业样本的验证性因素分析中都得到了支持。工作倦怠问卷的信度与效度指标也都达到了心理测量学的要求。     

The Comprehensive Alcohol Expectancy Questionnaire: Confirmatory Factor Analysis,Scale Refinement,and Further Validation

The Comprehensive Alcohol Expectancy Questionnaire (CAEQ; Demmel & Hagen, 2003a, 2003b) is a self-report measure designed to assess an individual’s alcohol expectancies. In this study, we examined the CAEQ in a student sample (N= 932) and in a clinical sample of alcohol-dependent inpatients (N= 744). The Five-factor structure was confirmed by means of confirmatory factor analysis. Convergent validity of the revised CAEQ was supported by showing significant relationships to quantity and frequency of drinking. The results of this study suggest that the revised CAEQ appears to be a psychometrically sound tool for the assessment of alcohol expectancies among both students and alcohol-dependent inpatients.     

《学习适应量表》的验证性因素分析

在北京多所学校对425名学生进行了《学习适应量表》的测量,验证性因素分析结果表明,总量表和各个分量表的模型拟合都未违反基本适令标准,总量表的整体模型拟合性良好,内在结构适合性也比较好。学习方法、学习习惯、学习态度、学习环境和身心适应五个分量表的整体模型拟合性较好,但内在结构不佳。     

Validation of the Multidimensional Blood/Injury Phobia Inventory: Evidence for a Unitary Construct

The Multidimensional Blood/Injury Phobia Inventory (MBPI) was developed from a theoretical framework to characterize a broad range of feared stimuli and phobic reactions associated with this pathology. The MBPI consists of 40 items that cross 4 types of stimulus content (injections, hospitals, blood, injury), 5 types of phobic responses (fear, avoidance, worry, fainting, disgust), and a self versus other focus. This study reports on administration of the MBPI to 558 undergraduates, 9 of whom had blood/injury phobia. The instrument had a Cronbach's alpha of .91 and demonstrated good concurrent validity, convergent validity, and discriminant validity. One large factor emerged in an unrotated principal components analysis, suggesting that blood/injury phobia is a unitary psychometric construct. Exploratory factor analyses revealed a 6-factor solution defined by stimulus content domains and fainting, each of which may be important to consider clinically when assessing the unique concerns of treatment-seeking individuals.     

Some Theory and Applications of Confirmatory Second-Order Factor Analysis

This paper shows how confirmatory factor analysis can be used to test second- (and higher-) order factor models in the areas of the structure of abilities, allometry, and the separation of specific and error variance estimates. In the latter area, an idea of Joreskog's is extended to include a new conceptualization of the estimation of validity. Second-order models are placed in a hierarchy of factor analysis models, which shows how the fit of various models can be compared. The concept of discriminability is introduced to describe a situation in which two models may both be identified, and yet the goodness-of-fit for both will be the same. This problem can usually be avoided by careful design of a study. Several examples are discussed.     

Multigroup Confirmatory Factor Analysis of the Cognitive Dysfunction Questionnaire: Instrument refinement and measurement invariance across age and sex

Vestergren, P., Rönnlund, M., Nyberg, L. & Nilsson, L.‐G. (2012). Multigroup confirmatory factor analysis of the Cognitive Dysfunction Questionnaire: Instrument refinement and measurement invariance across age and sex. Scandinavian Journal of Psychology 53, 390–400. The study adopted Confirmatory Factor Analysis (CFA) to investigate the factorial structure and reduce the number of items of the Cognitive Dysfunction Questionnaire (CDQ). The analyses were based on data for a total of 1,115 participants from population based samples (mean age: 63.0 ± 14.5 years, range: 25–95) randomly split into a refinement (N = 569) and a cross‐validation (N = 546) sample. Equivalence of the measurement and structural portions of the refined model was demonstrated across the refinement and cross‐validation samples. Among competing models the best fitting and parsimonious model had a hierarchical factor structure with five first‐order and one second‐order general factor. For the final version of the CDQ, 20 items within five domains were selected (Procedural actions, Semantic word knowledge, Face recognition, Temporal orientation, and Spatial navigation). Internal consistency reliabilities were adequate for the total scale and for the subscales. Multigroup CFAs indicated measurement invariance across age and sex up to the scalar level. Finally, higher levels of cognitive dysfunction as reflected by CDQ scores were predicted by advancing age, fewer years of education, and with deficits in general cognitive functioning as reflected by scores on the Mini‐Mental State Examination. In conclusion, the CDQ appears to be psychometrically sound and shows the expected relationships with variables known to be associated with cognitive dysfunction and dementia. Future studies should apply it among clinical groups to further test its usefulness.     

验证性因素分析中模型拟合的判断

采用与Hu ＆ Bender不同的设计进行模拟研究,发现在他们推荐的8个指数中,NNFI、CFI、IFI在所有样本量条件下几乎都能得到最低的α与β错误率,而且最佳界值都在0．95左右,属于优良的指数。但Mc与GammaHat的表现却不好,因此不建议使用。他们建议的2指标策略被证明能降低α与β错误率,在样本量较小时尤其值得尝试。     

The Acceptance of Cosmetic Surgery Scale: Confirmatory Factor Analyses and Validation among Serbian Adults

We examined the validity of the Serbian version of the Acceptance of Cosmetic Surgery Scale (ACSS; Henderson-King and Henderson-King 2005). A total of 622 Serbian adults completed the ACSS, along with Serbian translations of measures for the discrepancy between actual body weight and ideal body weight, body appreciation, sociocultural attitudes toward appearance, and demographics. Confirmatory factor analyses were conducted to compare how different ACSS models fitted the collected data. A three-factor model provided the best fit to the data relative to two- and one-factor models. The three-factor model had good internal consistency, convergent and discriminant validity, and nomological validity. The ACSS seems to be a valid instrument for use in Serbian populations. Our study will contribute towards better understanding of the acceptance of cosmetic surgery from a cross-cultural perspective.     

Confirmatory Factor Analysis of the Multi-Attitude Suicide Tendency Scale

This paper reports an examination of the factor structure of Orbach's Multi-Attitude Suicide Tendency Scale (MAST) utilizing confirmatory factor analytic techniques. Participants were 215 undergraduate students ranging in age from 18 to 20 who completed the MAST, the Reynolds Adolescent Depression Scale, the Suicide Ideation Questionnaire, and a self-harmful behavior scale. Support was found for the original factor structure, reliability, and validity of the MAST. A correlated model fit the data significantly better than an uncorrelated model. The results of this study indicate that the MAST is a valid and reliable measure appropriate for use with college-aged individuals for research purposes. Because the correlated model fit the data better than the uncorrelated, we suggest that future studies should attempt to replicate this finding, which would lead to a reexamination of the theory which underlies the MAST.     

Careless Responding to Reverse-Worded Items: Implications for Confirmatory Factor Analysis

Many self-report measures include some items worded in the direction opposite to that of other items. These so-called reverse-worded (RW) items can reduce the reliability and validity of a scale, and frequently form a separate method factor that does not appear to be substantively meaningful. One possible explanation for factors defined by RW items is respondent carelessness. The purpose of the present study was to evaluate whether relatively few careless responders to RW items can influence confirmatory-factor-analysis model fit enough that researchers would likely reject a one-factor model for a unidimensional scale. Results based on simulations indicated that if at least about 10% of participants respond to RW items carelessly, researchers are likely to reject a one-factor model for a unidimensional scale.     

Das-Naglieri认知评估系统（中文版）在学前儿童样本中的结构验证

本研究以246名5-6岁幼儿园儿童（男101名,女145名）为研究对象,考察基于智力PASS理论开发的Das-Naglieri 认知评估系统（5-7岁版）在中国学前儿童中的适用性,对评估系统中的12个任务进行了内在一致性信度评估,以幼儿早期数学能力测试作为参考评估其校标效度,并采用验证性因素分析对评估系统的结构效度进行评估。结果发现,该版本评估系统具有较好的信度和校标效度,并且很好地拟合了PASS模型。     

同事评价教师绩效的结构验证性因素分析

考察了教师绩效同事评价结果的因素结构。通过文献回顾、开放式调查、关键事件访谈和理论分析,提出了教师绩效的结构,通过验证性因素分析验证了同事评价教师绩效的结构。发现教师绩效包含6个维度,即职业道德、职务奉献、助人合作、教学效能、教学价值与师生互动,前3个维度组成关系绩效,其余维度组成任务绩效。关系绩效与任务绩效有较高的相关。