scholarly journals Approximated Penalized Maximum Likelihood for Exploratory Factor Analysis: An Orthogonal Case

Psychometrika ◽  
2018 ◽  
Vol 83 (3) ◽  
pp. 628-649 ◽  
Author(s):  
Shaobo Jin ◽  
Irini Moustaki ◽  
Fan Yang-Wallentin
Keyword(s):  
Factor Analysis ◽  
Maximum Likelihood ◽  
Exploratory Factor Analysis ◽  
Penalized Maximum Likelihood ◽  
Orthogonal Case

scholarly journals The Importance of Service Attributes between Low-cost and Full-service Carrier Customers: A case of Airline Type Hybridisation

2021 ◽  
Vol 10 (10(6)) ◽  
pp. 1741-1757
Author(s):  
Nkululeko Funyane
Keyword(s):  
Factor Analysis ◽  
Maximum Likelihood ◽  
Exploratory Factor Analysis ◽  
Low Cost ◽  
Latent Factors ◽  
Significant Difference ◽  
The Difference

This study sought to assess if the importance attached by customers to the airline service attributes differed across low-cost and full-service airline models. A Mann-Whitney U Test was used to assess the difference between the two models. However, before subjecting the data to differential tests, an exploratory factor analysis (maximum likelihood) was performed on the fifty-five items of service attributes, reducing them into forty-two items retained into ten latent factors (airline service attributes). The results of the revealed a significant difference in the importance attached to staff competence, courtesy and responsiveness only. Such findings suggest that the positioning of airlines into binary (FSC - LCC) models could be a waste of effort and resources since airlines seem to be converging.

An Exploratory Factor Analysis of Hartmann's Boundary Questionnaire and an Empirically-Derived Short Version

2001 ◽  
Vol 21 (2) ◽  
pp. 131-144 ◽  
Author(s):  
David Rawlings
Keyword(s):  
Factor Analysis ◽  
Individual Differences ◽  
Maximum Likelihood ◽  
Exploratory Factor Analysis ◽  
Perceived Competence ◽  
Full Scale ◽  
Short Version ◽  
Maximum Likelihood Factor Analysis ◽  
Scale Reliability ◽  
Promax Rotation

The Boundary Questionnaire (BQ) of Ernest Hartmann measures individual differences in the “thinness” of the mental boundaries presumed to separate the contents of consciousness. We report a study in which 300 undergraduates completed the 145-item BQ. Their scores were factor analyzed using Maximum Likelihood factor analysis followed by Promax rotation. Seven substantially uncorrelated factors emerged from the rotation. Subsequent analyses aimed at developing an empirically-derived short version of the questionnaire (abbreviated BQ-Sh) produced six subscales labeled Unusual Experiences, Need for Order, Trust, Perceived Competence, Childlikeness, and Sensitivity, with the Trust subscale omitted from calculation of the total score. Subscales of the 46-item BQ-Sh varied in reliability (alpha) from .80 to .65. It had a full-scale reliability of .74 and correlated strongly with the original BQ ( r = .88). It is compared with an alternative, rationally-derived short version with which it showed a correlation of .77.

scholarly journals A Comparison of Penalized Maximum Likelihood Estimation and Markov Chain Monte Carlo Techniques for Estimating Confirmatory Factor Analysis Models With Small Sample Sizes

2021 ◽  
Vol 12 ◽  
Author(s):  
Oliver Lüdtke ◽  
Esther Ulitzsch ◽  
Alexander Robitzsch
Keyword(s):  
Monte Carlo ◽  
Factor Analysis ◽  
Markov Chain ◽  
Markov Chain Monte Carlo ◽  
Maximum Likelihood ◽  
Confirmatory Factor Analysis ◽  
Parameter Estimates ◽  
Penalized Maximum Likelihood ◽  
Confirmatory Factor ◽  
Bayesian Estimators

With small to modest sample sizes and complex models, maximum likelihood (ML) estimation of confirmatory factor analysis (CFA) models can show serious estimation problems such as non-convergence or parameter estimates outside the admissible parameter space. In this article, we distinguish different Bayesian estimators that can be used to stabilize the parameter estimates of a CFA: the mode of the joint posterior distribution that is obtained from penalized maximum likelihood (PML) estimation, and the mean (EAP), median (Med), or mode (MAP) of the marginal posterior distribution that are calculated by using Markov Chain Monte Carlo (MCMC) methods. In two simulation studies, we evaluated the performance of the Bayesian estimators from a frequentist point of view. The results show that the EAP produced more accurate estimates of the latent correlation in many conditions and outperformed the other Bayesian estimators in terms of root mean squared error (RMSE). We also argue that it is often advantageous to choose a parameterization in which the main parameters of interest are bounded, and we suggest the four-parameter beta distribution as a prior distribution for loadings and correlations. Using simulated data, we show that selecting weakly informative four-parameter beta priors can further stabilize parameter estimates, even in cases when the priors were mildly misspecified. Finally, we derive recommendations and propose directions for further research.

scholarly journals Analisis Faktor Eksploratori Konstruk Risiko Online

Warta ISKI ◽  
2018 ◽  
Vol 1 (01) ◽  
pp. 63
Author(s):  
Amia Luthfia ◽  
Pinckey Triputra ◽  
Hendriyani Hendriyani
Keyword(s):  
Factor Analysis ◽  
Maximum Likelihood ◽  
Exploratory Factor Analysis

Semua risiko dan sisi merugikan sebagai konsekuensi dari penggunaan Internet dan eksplorasi anak dan remaja di Internet dinyatakan sebagai risiko online. Risiko online merupakan sebuah konstruk yang sifatnya baru dibangun dan definisi dari konsep ini masih diperdebatkan. Selain itu, pengelompokkan risiko (dimensi) dari konsep ini didasarkan pada hasil koding dari respon anak-anak dari penelitian kualitatif. Maka, penelitian ini bertujuan untuk menemukan dimensi-dimensi untuk variabel risiko online dan membangun skala risiko online untuk penelitian sejenis di Indonesia. Penelitian ini menguji validitas konstruk risiko online dan menggali berapa faktor yang terbentuk dari konstruk risiko online dengan dengan menggunakan Exploratory Factor Analysis. Metode ekstraksi faktor yang digunakan pada penelitian ini adalah Maximum Likelihood (ML) dengan rotasi faktor Varimax. Penelitian ini menggunakan survei online www.surveymonkey.com untuk memperoleh respon dari responden. Responden pada penelitian ini sebanyak 161 remaja DKI Jakarta berusia 13-18 tahun. Analisis faktor eksploratori menunjukkan hasil bahwa keseluruhan 33 item layak untuk dijadikan indikator karena memiliki koefisien korelasi anti-image yang baik.  Dari ekstraksi dan rotasi faktor dihasilkan 6  faktor dengan nilai kumulatif varian sebesar 69.557%, nilai ini cukup baik karena total varian yang dapat dijelaskan melebihi 50% varian. Enam faktor yang dihasilkan ini dapat dijadikan dimensi dari variabel risiko online. Maka, dimensi dari variabel risiko online adalah (1) Tindakan berisiko; (2) Nilai & informasi menyesatkan; (3) Pornografi; (4) Target tindakan berisiko; (5) Perundungan dan pelecehan; (6) Orang tidak dikenal.

scholarly journals A Comparison of Penalized Maximum Likelihood Estimation and Markov Chain Monte Carlo Techniques for Estimating Confirmatory Factor Analysis Models with Small Sample Sizes

2020 ◽  
Author(s):  
Oliver Lüdtke ◽  
Esther Ulitzsch ◽  
Alexander Robitzsch
Keyword(s):  
Monte Carlo ◽  
Factor Analysis ◽  
Markov Chain ◽  
Markov Chain Monte Carlo ◽  
Maximum Likelihood ◽  
Posterior Distribution ◽  
Parameter Estimates ◽  
Penalized Maximum Likelihood ◽  
Confirmatory Factor ◽  
Joint Posterior Distribution

With small to modest sample sizes and complex models, maximum likelihood (ML) estimation of confirmatory factor analysis (CFA) models can show serious estimation problems such as nonconvergence or parameter estimates that are outside the admissible parameter space. In the present article, we discuss two Bayesian estimation methods for stabilizing parameter estimates of a CFA: Penalized maximum likelihood (PML) estimation and Markov Chain Monte Carlo (MCMC) methods. We clarify that these use different Bayesian point estimates from the joint posterior distribution—the mode (PML) of the joint posterior distribution, and the mean (EAP) or mode (MAP) of the marginal posterior distribution—and discuss under which conditions the two methods produce different results. In a simulation study, we show that the MCMC method clearly outperforms PML and that these performance gains can be explained by the fact that MCMC uses the EAP as a point estimate. We also argue that it is often advantageous to choose a parameterization in which the main parameters of interest are bounded and suggest the four-parameter beta distribution as a prior distribution for loadings and correlations. Using simulated data, we show that selecting weakly informative four-parameter beta priors can further stabilize parameter estimates, even in cases when the priors were mildly misspecified. Finally, we derive recommendations and propose directions for further research.

Measuring Impact of Stress in Sandwich Generation Caring for Demented Parents

GeroPsych ◽  
2014 ◽  
Vol 27 (4) ◽  
pp. 171-179 ◽  
Author(s):  
Laurence M. Solberg ◽  
Lauren B. Solberg ◽  
Emily N. Peterson
Keyword(s):  
Factor Analysis ◽  
Exploratory Factor Analysis ◽  
Social Worker ◽  
Resource Availability ◽  
Well Being ◽  
Sandwich Generation ◽  
Item Questionnaire ◽  
Significant Difference ◽  
Adult Caregivers ◽  
The Impact

Stress in caregivers may affect the healthcare recipients receive. We examined the impact of stress experienced by 45 adult caregivers of their elderly demented parents. The participants completed a 32-item questionnaire about the impact of experienced stress. The questionnaire also asked about interventions that might help to reduce the impact of stress. After exploratory factor analysis, we reduced the 32-item questionnaire to 13 items. Results indicated that caregivers experienced stress, anxiety, and sadness. Also, emotional, but not financial or professional, well-being was significantly impacted. There was no significant difference between the impact of caregiver stress on members from the sandwich generation and those from the nonsandwich generation. Meeting with a social worker for resource availability was identified most frequently as a potentially helpful intervention for coping with the impact of stress.

The Hierarchical Model of Exercise Dependence

2015 ◽  
Vol 36 (4) ◽  
pp. 247-257 ◽  
Author(s):  
Gayatri Kotbagi ◽  
Laurence Kern ◽  
Lucia Romo ◽  
Ramesh Pathare
Keyword(s):  
Factor Analysis ◽  
Confirmatory Factor Analysis ◽  
Physical Exercise ◽  
Exploratory Factor Analysis ◽  
Convergent Validity ◽  
Large Population ◽  
Assessment Tools ◽  
Negative Consequences ◽  
Confirmatory Factor ◽  
Preliminary Version

Abstract. Physical exercise when done excessively may have negative consequences on physical and psychological wellbeing. There exist many scales to measure this phenomenon. The purpose of this article is to create a scale measuring the problematic practice of physical exercise (PPPE Scale) by combining two assessment tools already existing in the field of exercise dependency but anchored in different approaches (EDS-R and EDQ). This research consists of three studies carried out on three independent sample populations. The first study (N = 341) tested the construct validity (exploratory factor analysis); the second study (N = 195) tested the structural validity (confirmatory factor analysis) and the third study (N = 104) tested the convergent validity (correlations) of the preliminary version of the PPPE scale. Exploratory factor analysis identified six distinct dimensions associated with exercise dependency. Furthermore, confirmatory factor analysis validated a second order model consisting of 25 items with six dimensions and four sub-dimensions. The convergent validity of this scale with other constructs (GLTEQ, EAT26, and The Big Five Inventory [BFI]) is satisfactory. The preliminary version of the PPPE must be administered to a large population to refine its psychometric properties and develop scoring norms.

Subclinical Sadism and the Dark Triad

2019 ◽  
Vol 40 (3) ◽  
pp. 127-133 ◽  
Author(s):  
Laura K. Johnson ◽  
Rachel A. Plouffe ◽  
Donald H. Saklofske
Keyword(s):  
Factor Analysis ◽  
University Students ◽  
Personality Traits ◽  
Exploratory Factor Analysis ◽  
Convergent Validity ◽  
Dark Triad ◽  
Antisocial Personality ◽  
Factor Solution ◽  
Online Study ◽  
Antisocial Personality Traits

Abstract. The Dark Triad is a constellation of three antisocial personality traits: Machiavellianism, narcissism, and psychopathy. Recently, researchers have introduced a “Dark Tetrad” that includes subclinical sadism, although others suggest considerable overlap between psychopathy and sadism. To clarify the position of sadism within the Dark Triad, an online study was conducted with 615 university students. Exploratory factor analysis revealed that a six-factor solution fit the data best, representing Machiavellianism, psychopathy, physical sadism, verbal sadism, narcissism, and vicarious sadism. Furthermore, convergent validity was supported through sadism’s correlations with the HEXACO personality traits. The results support sadism’s inclusion within the Dark Tetrad as a unique construct but with some conceptual overlap with psychopathy.

Equivalent Factorial Structure of the Brief Symptom Inventory (BSI) in Clinical and Nonclinical Jordanian Populations

2010 ◽  
Vol 26 (2) ◽  
pp. 116-121 ◽  
Author(s):  
Fawzi S. Daoud ◽  
Amjed A. Abojedi
Keyword(s):  
Factor Analysis ◽  
Confirmatory Factor Analysis ◽  
Exploratory Factor Analysis ◽  
Clinical Sample ◽  
Brief Symptom Inventory ◽  
Factorial Structure ◽  
Symptom Inventory ◽  
Confirmatory Factor ◽  
Item Checklist ◽  
General Psychological Distress

This study investigates the equivalent factorial structure of the Brief Symptom Inventory (BSI) in clinical and nonclinical Jordanian populations, using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). The 53-item checklist was administered to 647 nonclinical participants and 315 clinical participants. Eight factors emerged from the exploratory factor analysis (EFA) for the nonclinical sample, and six factors emerged for the clinical sample. When tested by parallel analysis (PA) and confirmatory factor analysis (CFA), the results reflected a unidimensional factorial structure in both samples. Furthermore, multigroup CFA showed invariance between clinical and nonclinical unidimensional models, which lends further support to the evidence of the unidimensionality of the BSI. The study suggests that the BSI is a potentially useful measure of general psychological distress in clinical and nonclinical population. Ideas for further research are recommended.

Sign in / Sign up

Export Citation Format

Share Document