Minimum Rank Factor Analysis Research Articles

Latent structure and reliability

There are numerous reliability coefficients and α is the most popular. It is known that different coefficients can be appropriate in specific conditions and that α should not be used indiscriminately. However, some coefficients and conditions, particularly regarding the latent structure, lacked attention in previous research. In four Monte Carlo simulations, this study compared α, λ2, maximized λ4, λ4 based on locally optimal splits, μ2, Gilmer-Feldt, Kaiser-Caffrey α, Heise-Bohrnstedt Ω, Joreskog's ρ, ωtotal, algebraic greatest lower bound, greatest lower bound based on minimum rank factor analysis in every condition and also hierarchical ω and asymptotic ω hierarchical in multidimensional conditions. Findings suggest each of these coefficients can be useful at least in some conditions. Most differences in performance were observed in congeneric conditions and conditions with up to moderate loadings. Some coefficients were found to be more useful than previously considered. Results are discussed in the context of existing theory and previous Monte Carlo studies.

Revisiting internal consistency in hospitality research: toward a more comprehensive assessment of scale quality

PurposeThe purpose of this study is to revisit the measures of internal consistency for multi-item scales in hospitality research and compare the performance of Cronbach’s α, omega total (ωTotal), omega hierarchical (ωH), Revelle’s omega total (ωRT), Minimum Rank Factor Analysis (GLBfa) and GLB algebraic (GLBa).Design/methodology/approachA Monte Carlo simulation was conducted to compare the performance of the six reliability estimators under different conditions common in hospitality research. Second, this study analyzed a data set to complement the simulation study.FindingsOverall, ωTotal was the best-performing estimator across all conditions, whereas ωH performed the poorest. α performed well when factor loadings were high with low variability (high/low) and large sample sizes. Similarly, ωRT, GLBfa and GLBa performed consistently well when loadings were high and less variable as well as the sample size and the number of scale items increased. Of the two GLB estimators, GLBa consistently outperformed GLBfa.Practical implicationsThis study provides hospitality managers with a better understanding of what reliability is and the various reliability estimators. Using reliable instruments ensures that organizations draw accurate conclusions that help them move closer to realizing their visions.Originality/valueThough popular in other fields, reliability discussions have not yet received substantial attention in hospitality. This study raises these discussions in the context of hospitality research to promote better practices for assessing the reliability of scales used within the hospitality domain.

Are Confident Parents Really Aware of Children’s Online Risks? A Conceptual Model and Validation of Parental Self-Efficacy, Mediation, and Awareness Scales

Children’s use of the Internet comes with both risks and opportunities. To minimize risks and maximize opportunities, parents may choose to observe, enable, and/or restrict their children’s Internet use. However, parents’ high confidence in their children’s online safety can itself be a risk factor inhibiting parental awareness of online risks. This research aims to test whether confident parents are accurately aware of how frequently their child has experienced risks online. To this end, construct validity and reliability of scales measuring parental self-efficacy, restrictive-enabling-observant mediation, awareness, and Internet use were established first. Next, a conceptual model of parental awareness was proposed. These results were based on a two-parameter-logistic-model of item response theory, minimum-rank factor analysis, and parallel-mediation analysis of self-reports by a convenience sample of 388 parents in Ireland (Autumn 2019). Confident parents mostly reported their child experienced no online risk in the past couple of months, whereas unconfident parents reported their child experienced an online risk once, twice, or more times. Results of the mediation analysis indicated that confident parents likely underestimated, whereas unconfident parents overestimated, how frequently their child experienced an online risk. The accuracy of parental awareness depended on their mediation strategies, particularly restrictive mediation. Further research is needed to test whether training parents on self-efficacy and mediation of children’s Internet use raises their awareness of the children’s online risks.

Psychometric evidence of a perception scale about covid-19 vaccination process in Peruvian dentists: a preliminary validation

BackgroundIn the current pandemic context, dental professionals have greater occupational risks due to their healthcare activity, placing their expectations on the vaccine as a means of protection and at the same time hoping that the immunization process will be safe, reliable and comfortable, giving them greater peace of mind when they return to work. Therefore, the aim of the present study was to develop and provide a preliminary validation of a scale to measure perception of the COVID-19 vaccination process in Peruvian dental professionals.MethodsCross-sectional study with instrumental design. The scale was self-administered virtually. It was distributed through social networks to 220 dental professionals from two universities in the Peruvian capital between June and August 2021. The Aiken V was used for content analysis, while descriptive statistics such as mean, variance, kurtosis and skewness were used for construct validation, in addition to Pearson’s correlation matrix for analysis of the 18 items. Subsequently, a Parallel Analysis based on minimum rank factor analysis was performed. Finally, the reliability of the total scale and its dimensions was evaluated with Cronbach’s alpha.ResultsThe Aiken V coefficient values were favorable for all items. Parallel analysis indicated the existence of three dimensions. Principal component analysis with rotation suggested grouping eight items for the first dimension, six items for the second dimension and four items for the third dimension. These dimensions showed good reliability, as Cronbach’s alpha was 0.87, (95% confidence interval [CI]: 0.84–0.90), 0.80 (95% CI: 0.75–0.84) and 0.82 (95% CI: 0.78–0.86), respectively. In addition, the overall reliability of the scale was 0.89 (95% CI: 0.86–0.91), being acceptable.ConclusionsThe perception scale of the COVID-19 vaccination process in dental professionals proved preliminarily to be a valid and reliable scale that can be used for research purposes. However, it is recommended to extend its application and evaluate its metric properties in other health professionals.

Cross-cultural adaptation and validation of the Athlete Fear Avoidance Questionnaire in Italian university athletes with musculoskeletal injuries.

The aim of this study was to translate, culturally adapt and validate an Italian version of the Athlete Fear Avoidance Questionnaire (AFAQ-I). We conducted a cross-sectional evaluation of the psychometric properties of the AFAQ-I in university athletes with musculoskeletal injuries, culturally adapting it in accordance with international standards. Psychometric evaluation included the assessment of structural validity (exploratory factor analysis), internal consistency (Cronbach's alpha and inter-item correlation), test-retest reliability [intraclass correlation coefficient, (ICC) (2,1)], measurement error and minimum detectable change (MDC). To examine construct validity, we compared (Spearman ρ) the AFAQ-I with a numerical pain rating scale (NPRS), the Pain Catastrophizing Scale (PCS) and the Fear Avoidance Beliefs Questionnaire (FABQ) subscales [FABQ-Physical Activity (FABQ-PA) and FABQ-Work (FABQ-W)]. The AFAQ-I was administered to 133 university athletes with musculoskeletal injuries (95 males and 38 females; mean age 25 years, SD 5; mean average pain duration 5.6 months, SD 8.7). Factor analysis revealed an acceptable 1-factor 10-item solution (explained common variance at minimum rank factor analysis: 0.74) although a couple of items (#6 and 9) presented low factor loadings, suggesting the presence of a small secondary dimension. Cronbach's alpha was 0.78 and the average inter-item correlation was 0.27. ICC (2,1) was 0.95 and the MDC was 4.4 points. As hypothesized a priori, the AFAQ-I moderately correlated with NPRS (ρ = 0.42), PCS (ρ = 0.59), FABQ-PA (ρ = 0.40) and FABQ-W (ρ = 0.34). In conclusion, the AFAQ-I is a valid Italian translation of AFAQ that demonstrates acceptable psychometric properties. However, we recommend further analysis of the construct definition of the AFAQ and additional examination of its structural validity.

Self-handicapping strategies in educational context: construction and validation of the Brazilian Self-Handicapping Strategies Scale (EEAPREJ)

Self-handicapping strategies refer to the set of choices and attitudes adopted to minimize blame for failure and increase the value of success in achievement situations. This paper aims to describe the stages of construction and the psychometric analysis of a scale to measure the self-handicapping strategies of university students. In study 1, the major steps for the construction of the scales and initial results are reported. The internal consistency indices were acceptable and the principal component analysis revealed factors with little explanation of data variance. In study 2, data from a sample of 834 students from several undergraduate courses of different Brazilian universities were subjected to exploratory factor analysis using the minimum rank factor analysis (MRFA) method and the matrix of polychoric correlations. The parallel analysis criterion for factor retention indicated the one-factor solution as the best fit to data. The importance of having a valid and reliable measure to assess self-handicapping strategies in educational contexts and the promising use of the scale in actions to improve learning in higher education are discussed.

Assessment of Participation within the International Classification of Functioning, Disability, and Health (ICF): The Turkish Validity and Reliability of the Participation Scale.

Objective The aim of the study was to investigate the reliability and validity of the Turkish version of the Participation Scale (P-Scale). Methods A total of 152 students, with a mean age of 20.27 ± 2.19, participated in the study. Sociodemographic information (age, gender, and family income) was recorded; all participants completed the P-Scale twice with a 15-day interval. Translation and cross-cultural adaptation were performed to analyze the validity and reliability of the P-Scale. Cronbach's alpha and McDonald's alpha were used for scale reliability statistics and explanatory (EFA) and confirmatory (CFA) factor analysis; Mardia's multivariate normality and assumption tests were used for the validity of the scale. The factor extraction methods were minimum rank factor analysis in EFA and weighted least squares mean and variance adjusted estimator polychoric correlation matrix in CFA. Results Internal consistency of the scale was found good with Cronbach's alpha (0.852) and excellent with McDonald's alpha (0.924). The EFA and CFA resulted in two-factored structure, with the explained variance found to be higher than 30%. Conclusions Analysis demonstrated that the P-Scale had a satisfactory level of reliability and validity in Turkish university students.

Reliability and Validity of the Turkish Version of the Morningness – Eveningness Questionnaire

Background: People are divided into different chronotypes in relation to their circadian rhythm features. The Morningness–Eveningness Questionnaire (MEQ), developed by Horne and Östberg, was the first and is the most used questionnaire for determining the chronotype of people. The reliability and validity of MEQ were tested in this study due to some deficiencies and inconsistencies in previous reliability and validity studies for the Turkish versions of MEQ and deficiencies regarding the validity and reliability of other versions of MEQ. Subjects and Methods: MEQ was translated into Turkish based on the opinions of five academics who were experts in their respective fields and the suitability and content validity of the questionnaire were ensured. The Turkish MEQ was administered to 419 undergraduate students aged 18–34 years. Robust minimum rank factor analysis (MRFA) was used to reveal the construct validity of the questionnaire. Results: As a result of the analyses, the questionnaire was examined in three factors. None of the items were excluded from the study and all items were collected under three factors. The explained total variance value was 42.7%, and the factors explained 64% of common (shared) variance in MRFA. According to the reliability analysis, Cronbach's alpha value for the MEQ Turkish version was satisfactory for the overall questionnaire (0.765). Conclusions: The MEQ version developed in this study is more valid and reliable compared with other versions. However, these results support that a more detailed instrument for morningness and eveningness can be developed.

T226. PSYCHOMETRIC INVESTIGATION OF SELF-REPORTED THOUGHT DISORDER IN NON-CLINICAL SAMPLES

BackgroundSome dimensional models of psychosis predict that clinical phenomena are continuous with non-clinical phenomena. As such, Barrera et al. (2015) developed a self-report questionnaire to measure experiences akin to thought disorder (TD) in non-clinical populations. However, this measure requires further psychometric investigation. The current work seeks to replicate the three-factor structure of responses to the questionnaire that have been reported previously in a non-clinical sample, and to determine whether these responses predict other measures of schizotypal experiences, as well as measures of other personality characteristics (neuroticism and extraversion), and social adversity.MethodsThe 29-item questionnaire was used within an anonymous online survey to collect self-reported TD ratings from samples of 300 university students and 166 people from the general population. The survey also included the Schizotypal Personality Questionnaire (SPQ), neuroticism and extraversion scales from the International Personality Item Pool (IPIP), and the Social Defeat Scale.ResultsOf the self-report TD ratings collected from the student sample, an exploratory minimum rank factor analysis indicated the presence of three factors (Odd Speech, Poor Memory and Monitoring of Speech, and Difficulties with Initiating and Sustaining Conversation), which explained 56.5% of the variance and showed internal consistency (ordinal α ≥ 0.88). Odd Speech and Poor Memory and Monitoring of Speech factor scores showed stronger correlations with scores from the Odd Speech Scale of the SPQ (r ≥ 0.55, p < 0.001) than with other SPQ scale scores and IPIP personality scores. However, Difficulties with Initiating and Sustaining Conversation factor scores correlated strongly with other negative-like symptom scales from the SPQ, as well as with IPIP extraversion scores. A multiple regression analysis showed that, out of the three factors, Poor Memory and Monitoring of Speech scores predicted greater Social Defeat Scale scores independently of SPQ total scores and the IPIP extraversion and neuroticism scores, though the latter was the most important predictor. The results were very similar when the correlation and regression analyses were repeated in the non-student sample from the general population.DiscussionThe findings replicate Barrera et al.’s (2015) work, suggesting that TD-like experiences are multidimensional when measured in non-clinical samples using the self-report questionnaire. Although items intended to capture negative TD-like experiences might reflect introversion, items assessing oddities of speech and speech monitoring may not be entirely accounted for by existing measures of personality and schizotypy. Furthermore, since social defeat is linked to the risk factors of psychosis, some aspects of self-reported TD might prove useful in exploring the mechanisms of developing psychosis. Additional data is currently being collected to further investigate the replicability and generalizability of these findings.

Psychometric Properties of the Externalizing Spectrum Inventory: Replication and Extension across Clinical and Non-Clinical Samples

The Externalizing Spectrum Inventory aims at assessing personality features that underlie externalizing disorders such as substance abuse and antisocial behaviors. The objective was to replicate the psychometric properties of the 160-item Externalizing Spectrum Inventory in Dutch clinical and non-clinical samples. First, Cronbach’s alpha, test-retest reliability and the factor structure were analyzed on a mixed sample of inpatients (n = 149), undergraduates (n = 227), and community participants (n = 178). The factor structure was evaluated through confirmatory and exploratory factor analyses; for the latter Parallel Analysis was used, based on Minimum Rank Factor Analysis. Next, the criterion validity was analyzed using the Aggression Questionnaire and the NEO-Five Factor Inventory as external measures. The Dutch Externalizing Spectrum Inventory subscales showed sufficient reliability (α=.68-.94; ICC=.68-.91), except in the undergraduate sample (α=.49-.96; ICC=.43-.97). The factor structure of the Externalizing Spectrum Inventory was not confirmed and the exploratory analysis yielded different factor solutions across samples. The criterion validity was supported with regard to trait aggression and partly supported with regard to the Five Factor Model. The results suggest that the ESI-160 and its original factor model can be used for prediction purposes. However, further research of the factor structure is strongly recommended.

Validation of a new tool for evaluating subjects' satisfaction with medicine package leaflets: a cross-sectional descriptive study.

Package leaflets of medicines need to be intelligible, but tools for their evaluation are scarce. To validate a new tool for assessing subjects' satisfaction with medicine package leaflets (LiS-RPL). Cross-sectional descriptive study conducted in two regions of Portugal (Lisbon and Centre). 503 participants (53.1% male) were selected according to convenience and homogenously distributed into three groups: 1 to 6; 7 to 12; and > 12 years of schooling. LiS-RPL was developed based on international regulation guidelines and was initially composed of 14 items. Twelve package leaflets were tested. Dimensionality calculations included: exploratory factor analysis and minimum rank factor analysis; Kaiser-Meyer-Olkin index and Bartlett's sphericity test to assess matrix adequacy for exploratory factor analysis; exploratory bifactor analysis with Schmid-Leiman solution to detect possible existence of a broad second-order factor; and Bentler's Simplicity Index and Loading Simplicity Index to assess factor simplicity. Diverse coefficients were calculated to assess reliability. Minimum rank factor analysis detected a two-factor or single-factor structure. Exploratory factor analysis with 12 items showed a two-factor structure, explaining 69.11% of the variance. These items were strongly correlated with each other (r = 0.80). Schmid-Leiman: all items seemed to represent the general factor (loadings above 0.50), which was 76.4% of the extracted variance. Simplicity indices were good (percentile 99): Bentler's Simplicity Index of 0.99 and Loading Simplicity Index of 0.48. Internal consistency indexes indicated good reliability. LiS-RPL was shown to be homogenous. LiS-RPL is a validated tool for evaluating subjects' satisfaction with medicine package leaflets.

Determining the number of factors using parallel analysis and its recent variants.

Parallel analysis (PA) is recommended as one of the best procedures to determine the number of factors but its theoretical justification has long been questioned. The current study discussed theoretical issues on the use of eigenvalues for dimensionality assessment and reviewed the development of PA and its recent variants proposed to address the issues. The performances of 13 different PAs including PA with minimum rank factor analysis, revised PA, and comparison data method were investigated through a Monte Carlo simulation under a wide range of factor structures that produce small factor-representing and nonrepresenting eigenvalues for different types of measurement scales. Results showed that the traditional PA using full correlation matrices performed best in most of the conditions, especially when population error was involved. However, the overall accuracy of PA was not satisfactory when factor-representing eigenvalues were small, that is, when factor loadings were low and factor correlations were high. From these results, we suggest that the original PA be used to determine the number of factors but the estimated number should not be taken as a fixed estimate. The number of factors within ±1 range of the estimate can be considered as viable candidates and interpretational validity of the competing models should be consulted for the decision. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Calculus Self-Efficacy Inventory: Its Development and Relationship with Approaches to learning

This study was framed within a quantitative research methodology to develop a concise measure of calculus self-efficacy with high psychometric properties. A survey research design was adopted in which 234 engineering and economics students rated their confidence in solving year-one calculus tasks on a 15-item inventory. The results of a series of exploratory factor analyses using minimum rank factor analysis for factor extraction, oblique promin rotation, and parallel analysis for retaining extracted factors revealed a one-factor solution of the model. The final 13-item inventory was unidimensional with all eigenvalues greater than 0.42, an average communality of 0.74, and a 62.55% variance of the items being accounted for by the latent factor, i.e., calculus self-efficacy. The inventory was found to be reliable with an ordinal coefficient alpha of 0.90. Using Spearman’ rank coefficient, a significant positive correlation ρ ( 95 ) = 0.27 , p &lt; 0.05 (2-tailed) was found between the deep approach to learning and calculus self-efficacy, and a negative correlation ρ ( 95 ) = − 0.26 , p &lt; 0.05 (2-tailed) was found between the surface approach to learning and calculus self-efficacy. These suggest that students who adopt the deep approach to learning are confident in dealing with calculus exam problems while those who adopt the surface approach to learning are less confident in solving calculus exam problems.

The German version of the high-performance work systems questionnaire (HPWS-G) in the context of patient safety: a validation study in a Swiss university hospital

BackgroundHigh performance work systems (HPWSs) are successful work systems in the context of safety climate and patient safety. The 10-item HPWS questionnaire is a validated instrument developed to assess existing HPWS structures in hospitals. The objectives of this cross-sectional study were to translate the English HPWS questionnaire into German (HPWS-G), to rate its content validity, and to examine its psychometric properties.MethodsContent validity was examined by a panel of 12 physicians and nurses, and I-CVI and S-CVI calculated. For internal consistency, Cronbach’s α and item-scale correlations were determined. Construct validity was measured via confirmatory factor analysis.A convenience sample of 782 nurses and physicians in a University hospital setting in Switzerland’s German-speaking region was surveyed. Four inclusion criteria were applied: working in intensive care, emergency department or operating room; having daily patient contact; having worked in the current clinical area for more than three months; and more than 40% employment.ResultsA total of 281 questionnaires were completed (response rate: 35.9%). Overall, the 10-item HPWS-G questionnaire showed good content validity (I-CVI = .83–1; S-CVI = .86) and internal consistency (Cronbach’s α = .853). HPWS-G scores correlated significantly with safety climate (rs = .657, p < .01) and teamwork climate (rs = .615, p < .01). The proposed 1-factor model was accepted considering results of applied minimum rank factor analysis; a confirmatory factor analysis indicated an acceptable to good model fit (GFI = .968; CFI = .902; RMSEA = .043).ConclusionsThe HPWS-G showed good psychometric properties. In clinical practice it can be used to assess HPWS practices and for intra- and inter-hospital benchmarking. Some minor adaptions to the wording could be made as well as reassessing the psychometric properties at other clinical sites.

Cross-cultural adaptation and validity of the Italian version of the Central Sensitization Inventory

BackgroundCentral sensitization (CS) is an important feature in patients with chronic pain. The Central Sensitization Inventory (CSI) was developed with the goal of detecting the patients' symptoms related to CS. ObjectivesThis study aimed at cross-culturally adapting the CSI into Italian, and at assessing its structural and construct validity in patients with different chronic pain disorders. DesignClinimetric study. MethodsThe Italian version of the CSI (CSI-I) was generated following forward and backward translations, expert committee review, and pilot-testing. Patients with pain for ≥3 months were eligible if diagnosed with: low back pain (LBP), temporomandibular disorder (TMD), hand osteoarthritis (HOA), fibromyalgia (FM), or rheumatoid arthritis (RA). Structural validity was assessed with exploratory factor analysis and parallel analysis based on minimum rank factor analysis; construct validity was evaluated by testing ten hypotheses on: 1) expected differences between relevant subgroups, 2) expected correlations with other instruments measuring pain intensity, physical functioning, psychological functioning, headache symptoms, and pain self-efficacy. Results220 patients were included: 35% with LBP, 17% with TMD, 19% with HOA, 9% with FM, and 20% with RA. Factor analyses revealed that the CSI-I is a unidimensional instrument. Construct validity was satisfactory since 80% of the hypotheses were met. ConclusionsThe CSI-I was successfully developed and exhibited satisfactory validity in patients with chronic pain. Its reliability, responsiveness and content validity should be investigated in future studies. Until robust evidence indicates a strong relationship between CS and the CSI-I, caution should be adopted in claiming that the CSI-I measures CS.

Proportion of Indicator Common Variance Due to a Factor as an Effect Size Statistic in Revised Parallel Analysis.

Past research suggests revised parallel analysis (R-PA) tends to yield relatively accurate results in determining the number of factors in exploratory factor analysis. R-PA can be interpreted as a series of hypothesis tests. At each step in the series, a null hypothesis is tested that an additional factor accounts for zero common variance among measures in the population. Integration of an effect size statistic-the proportion of common variance (PCV)-into this testing process should allow for a more nuanced interpretation of R-PA results. In this article, we initially assessed the psychometric qualities of three PCV statistics that can be used in conjunction with principal axis factor analysis: the standard PCV statistic and two modifications of it. Based on analyses of generated data, the modification that considered only positive eigenvalues ( ) overall yielded the best results. Next, we examined PCV using minimum rank factor analysis, a method that avoids the extraction of negative eigenvalues. PCV with minimum rank factor analysis generally did not perform as well as , even with a relatively large sample size of 5,000. Finally, we investigated the use of in combination with R-PA and concluded that practitioners can gain additional information from and make more nuanced decision about the number of factors when R-PA fails to retain the correct number of factors.

A Comparison of Unidimensionality and Measurement Precision of the Narcissistic Personality Inventory and the Narcissistic Admiration and Rivalry Questionnaire.

The current study compares the closeness to unidimensionality (CU) and measurement precision (MP) of the Narcissistic Personality Inventory (NPI)-with either a pairwise forced-choice or 5-point Likert-type scale response format-to the Narcissistic Admiration and Rivalry Questionnaire (NARQ). Minimum rank factor analysis and item information curves from item response models were utilized. The results mainly confirmed our expectations that NPI subscales are lower in CU and MP compared with NARQ subscales when the NPI was administered with its traditional forced-choice response format. When the NPI was administered with a 5-point Likert-type scale response format, the NPI subscale Leadership/Authority and NPI Grandiose Exhibitionism showed similarly high levels of CU and MP as the two NARQ subscales. While the NPI subscale Entitlement/Exploitativeness had a higher CU than the NARQ subscales it showed considerably lower levels of MP.

A new method for simultaneous estimation of the factor model parameters, factor scores, and unique parts

In the common factor model the observed data is conceptually split into a common covariance producing part and an uncorrelated unique part. The common factor model is fitted to the data itself and a new method is introduced for the simultaneous estimation of loadings, unique variances, factor scores, and unique parts. The method is based on Minimum Rank Factor Analysis and allows for the percentage of explained common variance to be computed. Taking into account factor indeterminacy, an explicit description of the complete class of solutions for the factor scores and unique parts is given. The method is evaluated in a simulation study and fitted to a dataset in the literature.

Multi-set factor analysis by means of Parafac2.

We consider multi-set data consisting of Nk observations, k=1,…, K (e.g., subject scores), on J variables in K different samples. We introduce a factor model for the J×J covariance matrices Σk, k=1,…, K, where the common part is modelled by Parafac2 and the unique variances Uk, k=1,…, K, are diagonal. The Parafac2 model implies a common loadings matrix that is rescaled for each k, and a common factor correlation matrix. We estimate the unique variances Uk by minimum rank factor analysis on Σk for each k. The factors can be chosen orthogonal or oblique. We present a novel algorithm to estimate the Parafac2 part and demonstrate its performance in a simulation study. Also, we fit our model to a data set in the literature. Our model is easy to estimate and interpret. The unique variances, the factor correlation matrix and the communalities are guaranteed to be proper, and a percentage of explained common variance can be computed for each k. Also, the Parafac2 part is rotationally unique under mild conditions.

¿Matrices Policóricas/Tetracóricas o Matrices Pearson? Un estudio metodológico

The use of product-moment correlation of Pearson is common in most studies in factor analysis in psychology, but it is known that this statistic is only applicable when the variables related are in interval scale and normally distributed, and when are used in ordinal data may to produce a distorted correlation matrix . Thus is a suitable option using polychoric/tetrachoric matrices in item-level factor analysis when the items are in level measurement nominal or ordinal. The aim of this study was to show the differences in the KMO, Bartlett`s Test and Determinant of the Matrix, percentage of variance explained and factor loadings in depression trait scale of Depression Inventory Trait - State and the Neuroticism dimension of the short form of the Eysenck Personality Questionnaire -Revised, regarding the use of matrices polychoric/tetrachoric matrices and Pearson. These instruments was analyzed with different extraction methods (Maximum Likelihood, Minimum Rank Factor Analysis, Unweighted Least Squares and Principal Components), keeping constant the rotation method Promin were analyzed. Were observed differences regarding sample adequacy measures, as well as with respect to the explained variance and the factor loadings, for solutions having as polychoric/tetrachoric matrix. So it can be concluded that the polychoric / tetrachoric matrix give better results than Pearson matrices when it comes to item-level factor analysis using different methods.