The factor analysis procedure for exploration: a short guide with examples / El análisis factorial exploratorio: una guía breve con ejemplos

Several methods of factor extraction have recently gained popularity as a procedure for dealing with estimation problems associated with small sample sizes, which can be found in the various behavioral science disciplines, such as comparative psychology and behavior genetics. Two popular approaches for particularly small samples (below 50) include unweighted least squares factor analysis (ULS-FA) and regularized exploratory factor analysis (REFA). However, it is unclear how well each of the approaches performs with small samples in the context of exploratory bifactor modeling. In the current study, a comprehensive simulation study was conducted to evaluate the small sample behavior of the two approaches in terms of bifactor structure recovery under different sample size, factor loading, number of variables per factor, number of factors, and factor correlation experimental conditions. The results show that REFA is recommended for use over ULS-FA, particularly in the conditions involving low factor loadings, few group factors, or a small number of variables per factor.

Neural Topologies of Reinforcement Sensitivity Theory: A Latent Variable Approach to Magnetic Resonance Imaging Data.

IDENTIFICATION OF LATENT RESERVE RISK FACTORS FOR DEMENTIA IN TWO LARGE PROSPECTIVE STUDIES: CONFIRMATORY FACTOR ANALYSIS

The Problem Exploratory factor analysis (EFA) serves many useful purposes in human resource development (HRD) research. The most frequent applications of EFA among researchers consists of reducing relatively large sets of variables into more manageable ones, developing and refining a new instrument’s scales, and exploring relations among variables to build theory. Because researchers face a number of decisions when conducting EFA that can involve some subjectivity (e.g., factor extraction method, rotation), poor analytic decisions regarding how the EFA should be conducted (e.g., number of factors to extract) can produce misleading findings to the detriment of these efforts, especially theory building. The Solution Steps must be taken to improve the quality of the decision making associated with conducting EFAs if sound theory building and research related to this statistical method is to continue. Higher quality EFAs facilitate higher quality theory building and research. The Stakeholders HRD theorists, researchers, and scholar-practitioners are the intended audience of this article. In particular, those interested in refining measures and theory building would benefit most from being exposed to best EFA decision-making practices.

Background:Chronic obstructive pulmonary disease (COPD) is a common disease with considerable consequences for patients’ daily lives. The Clinical COPD Questionnaire (CCQ) was designed to measure these consequences in daily practice. Although the CCQ is widely used, its original structure has never been tested.Aims:This study examines the psychometric properties of the CCQ with regard to its latent structure in a sample of primary care patients with COPD.Methods:Two cross-sectional studies were conducted; in study 1 (N=243) exploratory analyses, including exploratory factor analysis (EFA) and Mokken scale analysis, were performed to explore the latent structure of the CCQ. In study 2 (N=244), confirmatory factor analysis (CFA) was conducted to evaluate the model fit of the structure found in study 1.Results:Both EFA and Mokken scale analysis revealed a structure of two dimensions (‘general impact’ α=0.91 and ‘cough’ α=0.84). This structure, however, was not confirmed in study 2, nor was the original structure. However, subsequently removing items that violated the assumption of a normal response distribution did result in an excellent model fit with two dimensions measuring ‘dyspnoea’ and ‘cough’ (CFA: comparative fit index (CFI) 0.98; normed fit index (NFI) 0.97; root mean squared error of approximation (RMSEA) 0.08 (0.04)).Conclusions:In primary care, factor analyses on the CCQ revealed a two-component structure measuring ‘general impact’, and ‘cough’. A shortened and more specific version of the CCQ could be regarded as a useful instrument to screen for exacerbations by measuring dyspnoea, coughing and producing phlegm.

PurposeAs consumers interact with various small businesses, they develop a mental image, called a prototype, to represent what small businesses are as a generalized, conceptual category. However, prior research has said little about what this small business prototype entails. Thus, the aim of this study is to explore consumers’ perceptions of the prototypical small business by identifying common attributes among small businesses that differentiate them from large businesses.Design/methodology/approachThis study undertakes a thorough review of the relevant consumer research literature for the attributes that consumers use to evaluate small businesses. Then, using a contemporary parallel analysis approach, it conducts an exploratory factor analysis (EFA) on a sample of 266 university students who were asked to evaluate how common those attributes are of small businesses. A second comparative EFA for large businesses is also conducted.FindingsThe EFA reveals two dimensions on which consumers evaluate small businesses: a sincere–authentic dimension and a disruptive–innovative dimension. Specifically, consumers view the prototypical small business to be relatively high on sincere–authentic and moderate on disruptive–innovative dimensions.Originality/valueThrough a comprehensive literature review and exploratory analysis, this study provides a novel understanding of consumers’ conceptualizations of small businesses. In studying the mental image consumers associate with the prototypical small business, this research fills a significant gap in the existing literature and provides important insights for practitioners and researchers alike.

Extended Abstract Purpose: Quality of college life (QCL) concept refers to overall feelings of satisfaction a student experiences with life at the college and based on bottom-up spillover theory. Quality of College Life (QCL) Measure which was developed by Sirgy et al. (2007) has been found valid and reliable across many different countries such as Korea (Yu and Lee, 2008), Germany, Italy, Lebanon and Mexico (Sirgy et al. 2010) with different student and university profiles. The purpose of this study is to adapt the Quality of College Life (QCL) to Turkish university students and to present the validity and reliability of the scale. Method: The Turkish form of the QCL Measure was administered to university students for the validity and reliability analysis after examining equivalency between English and Turkish versions. The universe of the study is Akdeniz University students. Applicable data has been obtained from 1.297 students from 13 different education units (e.g. faculty, vocational school, etc.) in campus area. Survey instrument consist of 3 dimensions, 17 sub dimensions and 70 items. Respondents answer the questions on a 5-point Likert type scale varying from “very dissatisfied 1â€? to “very satisfied 5â€?. In order to examine the validity and reliability properties of the measure exploratory factor analysis, Cronbach alpha correlation coefficients, item-total correlation, convergent – discriminant, and concurrent validity were used. Findings: The Cronbach's alpha coefficients are calculated to find the reliability value of the measure. All the three dimensions have a Cronbach alfa value between 0.701 and 0.947. These are above the acceptable lower limit (0.70). Only academic reputation sub dimension's Cronbach alfa value is 0.625 which consist of two items. All of the item-total corelation values are between 0.283 and 0.903. These are above the acceptable lower limit (0.25). Reliability findings indicates the QCL Measure is a reliable construct. KMO and Bartlett test are done to see if the Quality of College Life Measure is suitable for exploratory factor analysis. The result of KMO test is found to be significant at 0.853, 0.868, 0.870 levels for three dimensions. Barlett sphericity is found to be significant at (p<0.01) level, and it is concluded that exploratory factory analyisis can be applied to the data. In the exploratory factor analysis principle component analysis method and varimax technique which is one of the orthogonal rotations is used. Three factors are obtained with the factor analysis. The first explaines the 56.62%, the second explains the 71.60%, and the third explains 63.21% of the scale. Convergent – discriminant, and concurrent validity findings are also indicate that QCL Measure is a valid construct. Results and Suggestions: Only Og6 item deleted from the scale because of low values of exploratory factor analysis. Also security services and classroom environment sub dimensions split half , eventually there are 19 sub dimensions. It is thought that different education circumstances and environment are the causes. It is conculed that Turkish version of the Quality of College Life (QCL) of Students Measure is valid and reliable for Turkish university students. Key words: Quality of College Life (QCL), Scale Adaptation, College Students.

Lack of perceived support and control during labor and childbirth is known as an important predictor of post-traumatic stress disorder following childbirth. However, there is no standard scale to measure perceived support and control for Iranian women. This study determined the validity and reliability of the support and control in birth questionnaire for Iranian women. Support and control in the birth questionnaire were translated into Persian by the forward and backward translation method in 2019. Among a total of 102 healthcare centers in Tabriz, 50 urban and 10 rural centers were selected randomly. Then, 660 women with vaginal childbirth during the postpartum period were extracted by each center and selected randomly. The validity of the Persian version was evaluated in terms of face, content, and construct validity. Internal consistency and reliability were assessed by Cronbach's alpha coefficient and intraclass correlation coefficient, respectively. Exploratory and confirmatory factor analyses were used for evaluating the construct validity of the tool. Cronbach's alpha coefficient (0.95) and intraclass correlation coefficient (0.99) were acceptable. In an exploratory analysis, three factors were extracted and these three factors explained 63.1% of the total variance. Items 14 and 17 were removed from the Persian version due to low factor loading and impact factor values. Confirmatory factor analysis supported the three factors extracted in the exploratory analysis. Confirmatory factor analysis showed suitable indexes of fitness for 31 items. Persian version of the support and control in the birth questionnaire is a valid and reliable tool for the Iranian women population.

This research attempts to measure the impact of service-learning on community recipients – at-risk high school students in urban Southern California. The service-learning project, an integrative, six-week assignment, involves upper-division business majors delivering the Options: Business Education and Life Skills curriculum to at-risk students in two local alternative education high schools. In addition to delivering business education and life skills, a critical design component of the curriculum is the opportunity for college students to be role models and provide mentoring guidance to at-risk high school students. This study used surveys to gather data on student perceptions of four constructs: (1) strengths and values, (2) school and work-related skills, (3) business etiquette and resume building, and (4) future life and career planning. Pre-tests and post-tests were administered to gauge differences in perception during the six-week service-learning project. Results indicated positive effects of the service-learning curriculum overall. Further, the data revealed statistically significant results with particularly noteworthy outcomes in the planning for the future and preparing for the world of work responses.

Our objective was to examine the use and quality of exploratory factor analysis (EFA) in articles published in Rehabilitation Psychology. Trained raters examined 66 separate exploratory factor analyses in 47 articles published between 1999 and April 2014. The raters recorded the aim of the EFAs, the distributional statistics, sample size, factor retention method(s), extraction and rotation method(s), and whether the pattern coefficients, structure coefficients, and the matrix of association were reported. The primary use of the EFAs was scale development, but the most widely used extraction and rotation method was principle component analysis, with varimax rotation. When determining how many factors to retain, multiple methods (e.g., scree plot, parallel analysis) were used most often. Many articles did not report enough information to allow for the duplication of their results. EFA relies on authors' choices (e.g., factor retention rules extraction, rotation methods), and few articles adhered to all of the best practices. The current findings are compared to other empirical investigations into the use of EFA in published research. Recommendations for improving EFA reporting practices in rehabilitation psychology research are provided.

Grounded in the Systemic–Transactional Model, this study developed the Perceived Stress Scale in Romantic Relationships and evaluated its psychometric properties in adults currently involved in a romantic relationship. Using a scale development design, data were collected from two independent samples of individuals who had been in a romantic relationship. The exploratory analyses were conducted with 571 participants (age range 20–63 years), and the confirmatory analysis was conducted with 246 participants (age range 21–64 years). Construct validity was examined using exploratory factor analysis and confirmatory factor analysis. Item functioning and precision were further evaluated using a graded response model within the item response theory framework. Discriminative power was tested by comparing lower and upper score groups, and criterion-related validity was examined through correlations with couple burnout, relationship satisfaction, and relationship consensus. Results supported a unidimensional five-item short form. Confirmatory analyses indicated good model fit, and multi-group confirmatory factor analysis supported strict measurement invariance across gender. Internal consistency was high in both samples. Criterion-related validity evidence showed strong positive associations with couple burnout and negative associations with relationship satisfaction and relationship consensus. Item-level results indicated adequate discrimination and that the scale provided the most precise measurement at moderate to high levels of perceived stress. Overall, the Perceived Stress Scale in Romantic Relationships appears to be a brief, valid, and reliable measure for assessing relationship-specific perceived stress.

AimTo explore the construct validity of the ‘Experience in Close Relationships – Relationship Structures’-questionnaire (ECR-RS) in a population of out-of-hospital cardiac arrest (OHCA) survivors. Objectives were to (i) describe item- and scale-level response patterns, and (ii) evaluate the preliminary construct validity of the ECR-RS, including its dimensional (structural), known-groups, and convergent validity.MethodsAn exploratory cross-sectional validation study, with OHCA survivors completing the ECR-RS, Hospital Anxiety and Depression Scale (HADS), and the mental health component from the Short Form-12 (SF-12 MCS) three months post- arrest. Descriptive statistics and floor/ceiling analyses were performed. Dimensional validity was assessed using response distribution patterns and exploratory factor analysis (EFA), followed by reliability using Cronbach’s άs. Known-group validity was tested using a priori hypotheses, Spearman’s correlations, and Mann-Whitney U tests. Convergent validity was evaluated by correlating ECR-RS total scores with HADS and SF-12 MCS.ResultsAmong 123 survivors (median age 59.9 years, 84 % male), ECR-RS total scores were low on both subscales and floor effects were observed at scale level (31 % for avoidant and 72 % for anxious attachment). EFA supported the expected two-factor structure, though item 4–6 showed poor loadings/cross-loadings. Internal consistency was acceptable (total scale Cronbach’s α = 0.88) and improved when problematic items were excluded. Known-group hypotheses were not supported. Anxious attachment correlated moderately with symptoms of anxiety and depression and was inversely correlated with mental health scores.ConclusionThe ECR-RS demonstrated partial construct validity among OHCA survivors, but item-level inconsistencies and pronounced floor effects limit its utility. Findings highlight the need for a revised instrument better suited to post-cardiac arrest relational and psychological dynamics.

The Alcohol Quality of Life Scale (AQoLS) is a new patient-reported outcome 34-item questionnaire measuring health-related quality of life (HRQOL), specific to patients with an alcohol use disorder, developed from the patients' perspective. This is the first report establishing evidence in support of measurement reliability and validity of the AQoLS. A total of 285 randomly selected patients receiving interventions for alcohol use disorder in addiction specialised care settings in France were included in the study (response rate 80.1%). Exploratory factor analysis was conducted to evaluate the hypothesised-during-development-stage dimensional structure of the AQoLS. Internal consistency of the total score and the dimensions subscores were assessed through Cronbach's alpha coefficients. Construct validity was tested through correlations with the Short-Form 36 Health Survey (SF-36) and EuroQol 5 dimensions (EQ-5D). Exploratory analysis indicated seven observed dimensions which differed slightly from the 7 dimensions defined a priori in the framework hypothesised during the scale development: activities, relationships, living conditions, negative emotions, self-esteem, control and sleep. A major common factor allows the summing of the 34 items to obtain a total score. All the 34 items were acceptable. Cronbach's alpha for the AQoLS total score was 0.96 and ranged from 0.8 to 0.9 for the dimensions subscores. Negative correlations between AQoLS and all dimensions of the SF-36, but general health and positive correlations between AQoLS and all items of the EQ-5D were shown. As expected, the correlations were mostly moderate in magnitude, low with scores referring to physical areas and the highest with the SF-36 MSC. This study provides evidence of the measure's psychometric properties in terms of construct validity and internal consistency. The "control" and "self-esteem" dimensions are of particular interest as these concepts are not captured in existing HRQOL. Further longitudinal validation of the scale is necessary to assess sensitivity to change.

A library for comparing factorial solutions obtained in exploratory analyses is presented. The procedure allows both orthogonal and oblique solutions to be compared by computing several indices. The library is controlled with interactive menus or programs.

The extended psychosis phenotype, or the expression of nonclinical positive psychotic experiences, is already prevalent in adolescence and has a dose-response risk relationship with later psychotic disorder. In 2 large adolescent general population samples (n = 5422 and n = 2230), prevalence and structure of the extended psychosis phenotype was investigated. Positive psychotic experiences, broadly defined, were reported by the majority of adolescents. Exploratory analysis with Structural Equation Modelling (Exploratory Factor Analysis followed by Confirmatory Factor Analysis [CFA]) in sample 1 suggested that psychotic experiences were best represented by 5 underlying dimensions; CFA in sample 2 provided a replication of this model. Dimensions were labeled Hallucinations, Delusions, Paranoia, Grandiosity, and Paranormal beliefs. Prevalences differed strongly, Hallucinations having the lowest and Paranoia having the highest rates. Girls reported more experiences on all dimensions, except Grandiosity, and from age 12 to 16 years rates increased. Hallucinations, Delusions, and Paranoia, but not Grandiosity and Paranormal beliefs, were associated with distress and general measures of psychopathology. Thus, only some of the dimensions of the extended psychosis phenotype in young people may represent a continuum with more severe psychopathology and predict later psychiatric disorder.