Estimating confidence intervals for eigenvalues in exploratory factor analysis

Larsen, Ross; Warne, Russell T.

doi:10.3758/brm.42.3.871

Cited by 93 publications

(67 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CIs of eigenvalues were estimated using the mathematical method provided in Larsen and Warne (2010). The means of the eigenvalues and the means of the corresponding 95 % CIs across 2,000 data sets were computed for each condition.…”

Section: Discussionmentioning

confidence: 99%

“…A greater percentage, 78 %, of empirical EFA studies using nonstatistical extraction methods was found in psychological research (Henson & Roberts, 2006). For nonstatistical extraction methods, several criteria are used to help make the decision pertaining to the number of factors to retain, including the eigenvalue-greater-than-one rule (eigenvalue >1; Kaiser, 1960), confidence intervals for eigenvalues (Larsen & Warne, 2010), the scree test (Cattell, 1966), parallel analysis (Horn, 1965), minimum average partial correlations (Velicer, 1976), percentages of the total variance explained, extracted communality, residual matrices, and substantive interpretation of the factor structure; some criteria are used more often than others. A review of empirical EFA studies revealed that the criterion of eigenvalue >1 was overly relied on, with about 19 % of EFA studies in psychological research using an eigenvalue >1 as the sole criterion (Fabrigar, Wegener, MacCallum, & Strahan, 1999), and similar percentages have been reported in reviews of organizational research, with 22 % in 1986 (Ford, MacCallum, & Tait, 1986) and 18 % in 2003 (Conway & Huffcutt, 2003).…”

mentioning

confidence: 99%

See 1 more Smart Citation

On the number of factors to retain in exploratory factor analysis for ordered categorical data

Yang

Xia

2014

Behav Res

View full text Add to dashboard Cite

Conducting exploratory factor analysis (EFA) using statistical extraction methods has been recommended, but little is known about the accuracy of the decisions regarding the number of factors to retain for ordered categorical item data by considering a chi-square test, fit indices, and conventional criteria, such as eigenvalue >1 and parallel analysis. With computer-generated data, the authors examined the accuracy of decisions regarding the number of factors to retain for categorical item data, by combining these pieces of information using weighted least-square with mean and variance adjustment estimation methods based on polychoric correlations. A chi-square difference test was also conducted to compare nested EFA models. The results showed that the eigenvalue >1 criterion resulted in too many factors, in general. The chi-square test, chi-square difference test, fit indices, and parallel analysis performed reasonably well when the number of scale points was four, the number of items was 24, the sample size was at least 200, and the categorical distributions were similar across items. However, parallel analysis had a tendency toward factor underextraction when the correlation among factors was .50, particularly for two-point and 12-item scales.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

On the number of factors to retain in exploratory factor analysis for ordered categorical data

Yang

Xia

2014

Behav Res

View full text Add to dashboard Cite

show abstract

“…An exploratory factor analysis (EFA) (Larsen & Warne ) was carried out using the obtained genetic correlation matrix with a varimax rotation strategy, in order to find the most suitable trait combination to create the genetic indices. The proposed indices were defined according to the traditional criteria of selection in the Lidia cattle breed: aggressiveness (fighting ability or wildness), strength (the ability to attack with strength using the whole body) and mobility (the animal is in continuous movement pacing from one place to another) (Pelayo et al .…”

Section: Methodsmentioning

confidence: 99%

Behavioural linear standardized scoring system of the Lidia cattle breed by testing in herd: estimation of genetic parameters

Pelayo

Solé

Sánchez

et al. 2015

J Animal Breeding Genetics

View full text Add to dashboard Cite

Docility is very important for cattle production, and many behavioural tests to measure this trait have been developed. However, very few objective behavioural tests to measure the opposite approach 'aggressive behaviour' have been described. Therefore, the aim of this work was to validate in the Lidia cattle breed a behavioural linear standardized scoring system that measure the aggressiveness and enable genetic analysis of behavioural traits expressing fearless and fighting ability. Reproducibility and repeatability measures were calculated for the 12 linear traits of this scoring system to assess its accuracy, and ranged from 85.3 and 94.2%, and from 66.7 to 97.9%, respectively. Genetic parameters were estimated using an animal model with a Bayesian approach. A total of 1202 behavioural records were used. The pedigree matrix contained 5001 individuals. Heritability values (with standard deviations) ranged between 0.13 (0.04) (Falls of the bull) and 0.41 (0.08) (Speed of approach to horse). Genetic correlations varied from 0.01 (0.07) to 0.90 (0.13). Finally, an exploratory factor analysis using the genetic correlation matrix was calculated. Three main factors were retained to describe the traditional genetic indexes aggressiveness, strength and mobility.

show abstract

“…The EFA procedure aims at retaining as few factors as possible while explaining most of the variation in the data. The scree plot (31) and the modified eigenvaluegreater-than-one rule (32) were used to properly identify the number of factors in this specific section of the questionnaire. Internal consistency reliability was evaluated using Cronbach alpha coefficients with data from the first completion of the questionnaire.…”

Section: Statistical Analysesmentioning

confidence: 99%

Development and validation of the Perceived Food Environment Questionnaire in a French-Canadian population

Carbonneau¹,

Robitaille²,

Lamarche³

et al. 2017

Public Health Nutr.

View full text Add to dashboard Cite

Objective: The aim of this study was to develop and validate a questionnaire assessing perceived food environment in a French-Canadian population.Design: A questionnaire, the Perceived Food Environment Questionnaire, was developed assessing perceived accessibility to healthy (9 items) and unhealthy foods (three items). A pretest sample of 31 participants were recruited for a pilot testing of the questionnaire. For the validation study, 150 men and women from the Québec City area were recruited and completed the questionnaire twice.! An Exploratory Factor Analysis (EFA) was performed on the items to assess the number of factors (subscales). Cronbach alphas were used to measure internal consistency reliability. Test-retest reliability was assessed with Pearson correlations.Setting: Online survey. Results:The pilot testing did not lead to any change in the questionnaire. The EFA revealed a twosubscale structure. The first subscale is composed of six items assessing accessibility to healthy foods and the second includes three items related to accessibility to unhealthy foods. Three items were removed from the questionnaire due to low loading on the two subscales. The subscales demonstrated adequate internal consistency (Cronbach alpha: 0.77 for healthy foods, and 0.62 for unhealthy foods) and test-retest reliability (r=0.59 and 0.60 respectively, ps<0.0001). Conclusions:The Perceived Food Environment Questionnaire was developed for a French-Canadian population and demonstrated good psychometric properties. Further validation is recommended if the questionnaire is to be used in other populations.

show abstract

Estimating confidence intervals for eigenvalues in exploratory factor analysis

Cited by 93 publications

References 28 publications

On the number of factors to retain in exploratory factor analysis for ordered categorical data

On the number of factors to retain in exploratory factor analysis for ordered categorical data

Behavioural linear standardized scoring system of the Lidia cattle breed by testing in herd: estimation of genetic parameters

Development and validation of the Perceived Food Environment Questionnaire in a French-Canadian population

Contact Info

Product

Resources

About