Type I and Type II Error Rates and Overall Accuracy of the Revised Parallel Analysis Method for Determining the Number of Factors

Green, Samuel B.; Thompson, Marilyn S.; Levy, Roy; Lo, Wen‐Juo

doi:10.1177/0013164414546566

Cited by 34 publications

(48 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The success rates in the four levels of critical eigenvalues were respectively 32.9%, 49.9%, 70.9%, and 87.3%. The overall success rate of 61% makes this version of MLFA more successful that PA, thus changing, with the current set of simulation conditions, the order observed by Green et al (2015). The improvements observed for the four levels of critical eigenvalues are, however, not large enough to bring MLFA to levels of success comparable to those of the NEST variants, which all respectively exceeded 50%, 70%, 85%, and 90%.…”

Section: CDmentioning

confidence: 87%

“…Their principal axis factoring model simply used the squared multiple correlation of each variable as communalities, without iterative improvement. The same authors (Green, Thompson, Levy, & Lo, 2015) later compared their own preferred version of R-PA, that based on PAFA, with the MLFA approach, reporting advantage for their own method.…”

Section: Candidate Methodsmentioning

confidence: 99%

“…CD applies a criterion widely different from R-PA, based on the The Quantitative Methods for Psychology stabilization of the full pattern of surrogate data eigenvalues around the actual data eigenvalues, once enough factors are included in the model. To our knowledge, no direct comparison of CD and R-PA has yet been published, although Green et al (2015), after becoming aware of the CD method, expressed that "it would be interesting in the future to compare the revised PA and the Ruscio-Roche methods to understand how and why they differ".…”

Section: Candidate Methodsmentioning

confidence: 99%

“…Green et al (2015) compared MLFA and PA to their R-PA version that corresponds to the present NESTrf, and observed than although there was no uniformly superior method, R-PA generally fared better than PA and the latter better than MLFA. Many of their simulations included singlet variables or less than two factors, but the two or three factor conditions yielded this same general pattern.…”

Section: CDmentioning

confidence: 97%

See 3 more Smart Citations

Testing the number of required dimensions in exploratory factor analysis

Achim¹

2017

TQMP

View full text Add to dashboard Cite

While maximum likelihood exploratory factor analysis (EFA) provides a statistical test that k dimensions are sufficient to account for the observed correlations among a set of variables, determining the required number of factors in least-squares based EFA has essentially relied on heuristic procedures. Two methods, Revised Parallel Analysis (R-PA) and Comparison Data (CD), were recently proposed that generate surrogate data based on an increasing number of principal axis factors in order to compare their sequence of eigenvalues with that from the data. The latter should be unremarkable among the former if enough dimensions are included. While CD looks for a balance between efficiency and parsimony, R-PA strictly test that k dimensions are sufficient by ranking the next eigenvalue, i.e. at rank k + 1, of the actual data among those from the surrogate data. Importing two features of CD into R-PA defines four variants that are here collectively termed Next Eigenvalue Sufficiency Tests (NESTs). Simulations implementing 144 sets of parameters, including correlated factors and presence of a doublet factor, show that all four NESTs largely outperform CD, the standard Parallel Analysis, the Mean Average Partial method and even the maximum likelihood approach, in identifying the correct number of common factors. The recommended, most successful NEST variant is also the only one that never overestimates the correct number of dimensions beyond its nominal α level. This variant is made available as R and MATLAB code as well as a complement incorporated in a Microsoft Excel file.

show abstract

Section: CDmentioning

confidence: 87%

Section: Candidate Methodsmentioning

confidence: 99%

Section: Candidate Methodsmentioning

confidence: 99%

Section: CDmentioning

confidence: 97%

See 2 more Smart Citations

Testing the number of required dimensions in exploratory factor analysis

Achim¹

2017

TQMP

View full text Add to dashboard Cite

show abstract

“…Thus, there are conflicting results from the eigenvalue approach and the scree plot approach. That is why a parallel analysis (Monte Carlo simulation) was developed to determine how many components this method suggests to retain (O’Connor, 2000; Green et al, 2015). Parallel analysis suggested nine factors above chance (see Supplementary Table S5 ).…”

Section: Methodsmentioning

confidence: 99%

Music Preferences and Personality in Brazilians

2018

View full text Add to dashboard Cite

This article analyzes the relationship between musical preference and type of personality in a large group of Brazilian young and adult participants (N = 1050). The study included 25 of 27 states of Brazil and individuals aged between 16 and 71 years (M = 30.87; SD = 10.50). Of these, 500 were male (47.6%) and 550 were female (52.4%). A correlational study was carried out applying two online questionnaires with quality parameters (content-construct validity and reliability), one on musical preference and the other on personality. The results indicate four main findings: (1) the musical listening of the participants is limited to a reduced number of styles, mainly Pop music and others, typical of Brazilian culture; (2) the Brazilian context supposes a determining aspect in the low preference of non-Brazilian music; (3) there is a positive correlation between most personality types analyzed and the Latin, Brazilian, Classical and Ethnic musical styles. A negative correlation between these types of personality and the consumption of Rock music was also observed; (4) musical preferences are driven not only by personality but in some cases they are also driven by socio-demographic variables (i.e., age and gender). Likewise, this work shows how participants make use of music in personality aspects that may be of interest for the analysis of socio-affective behavior (personality) as well as according to different socio-demographic variables (e.g., age and gender). More cross-cultural research on musical preference and personality would need to be carried out from a global perspective, framed in the context of social psychology and studies of mass communication.

show abstract

Beyond “Just Saying No”: A Preliminary Evaluation of Strategies College Students Use to Refuse Sexual Activity

Marcantonio

2018

Arch Sex Behav

View full text Add to dashboard Cite

Preventing sexual assault is a core goal for universities as prevalence rates of sexual assault remain high, particularly among college students. A key mechanism thought to decrease rates of sexual assault is teaching college students how to give clear, explicit, verbal refusals. However, there is a paucity of research regarding how college students refuse sex. Thus, the purpose of this study was to understand different behavioral strategies college students would use to refuse sex. A sample of 773 heterosexual college students (523 women, 250 men) were recruited from two large southern universities in the USA to complete a survey on sexual communication. Thirty-eight items assessing verbal and behavioral cues that college students would use to refuse vaginal-penile sex were written based on previous, formative research. Items were assessed by the research team through an exploratory factor analyses, followed by a confirmatory factor analysis (CFA). The results yielded a three-factor structure: direct nonverbal refusals, direct verbal refusals, and indirect nonverbal refusals; CFA results suggested a good fit index for the model. Two independent sample t tests were conducted to examine differences in refusal cues across gender and relationship status; significant differences in refusals emerged for both. The three-factor structure depicting refusal cues was similar to previous work depicting cues college students use to communicate sexual consent; such information could inform sexual assault prevention programming.

show abstract

Type I and Type II Error Rates and Overall Accuracy of the Revised Parallel Analysis Method for Determining the Number of Factors

Cited by 34 publications

References 26 publications

Testing the number of required dimensions in exploratory factor analysis

Testing the number of required dimensions in exploratory factor analysis

Music Preferences and Personality in Brazilians

Beyond “Just Saying No”: A Preliminary Evaluation of Strategies College Students Use to Refuse Sexual Activity

Contact Info

Product

Resources

About