Is exploratory factor analysis always to be preferred? A systematic comparison of factor analytic techniques throughout the confirmatory–exploratory continuum.

Nájera, Pablo; Abad, Francisco; Sorrel, Miguel A.

doi:10.1037/met0000579

Cited by 9 publications

(4 citation statements)

References 95 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In summary, EFA-based methods are not to be seen as methods "competing" with CFA-based ones, for example the methods detailed in Kim et al (2017). Rather, they are a useful addition to the MI-toolbox that broaden the capabilities of investigating MI along an exploratory-confirmatory continuum (Nájera et al, 2023). Especially in the context of scale development, it can be beneficial to apply EFA-based methods to investigate the violations of MI due to cross-loadings, before using CFA-based methods to assess scalar MI.…”

Section: Discussionmentioning

confidence: 99%

“…While the use of CFA allows to incorporate theoretical considerations when investigating MI, it can also be too restrictive in terms of model specification. If the model is slightly misspecified, a CFA-based approach might not accurately recover the true structure of a model (Nájera et al, 2023).…”

Section: Comparison Of Efa-based Approachesmentioning

confidence: 99%

“…As a consequence, MI in this context not only concerns the equivalence of parameters in the measurement model but also the equivalence of its architecture; that is, the number of latent factors and the imposed zero-loadings must hold across groups (De Roover et al, 2022). Needless to say, the strict specification of the measurement model with zero-loadings is often not tenable (Nájera et al, 2023), especially when these restrictions have to be assumed across all groups. If the model is then modified in a data-driven way, its generalizability is diminished because this strategy capitalizes on chance (MacCallum et al, 1992).…”

Section: Cfa Vs Efa In MI Testingmentioning

confidence: 99%

See 2 more Smart Citations

New Developments in Measurement Invariance Testing - An Overview and Comparison of EFA-based Approaches

Sterner,

De Roover,

Goretzko

2024

Preprint

View full text Add to dashboard Cite

When comparing relations and means of latent variables, it is important to establish measurement invariance (MI). Most methods to assess MI are based on confirmatory factor analysis (CFA). Recently, new methods have been developed based on exploratory factor analysis (EFA); most notably, as extensions of multi-group EFA, researchers introduced mixture multi-group EFA, multi-group exploratory factor alignment, EFA trees, and multi-group factor rotation to resolve rotational indeterminacy in EFA. The main advantage of EFA-based (compared to CFA-based) assessment of MI is that no potentially too restrictive measurement model has to be specified. This allows for a more thorough investigation because violations of MI due to cross-loadings can be considered, too. For each method, we address the model specification and recommendations for application, detailing their strengths and weaknesses. We demonstrate each method in combination with multi-group factor rotation in an empirical example. Differences to and possible combinations with CFA-based methods are discussed.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Comparison Of Efa-based Approachesmentioning

confidence: 99%

Section: Cfa Vs Efa In MI Testingmentioning

confidence: 99%

See 1 more Smart Citation

New Developments in Measurement Invariance Testing - An Overview and Comparison of EFA-based Approaches

Sterner,

De Roover,

Goretzko

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…To further assess the structure of the scale, we conducted a non-preregistered Exploratory Factor Analysis (EFA) complemented by Horn's Parallel Analysis (Horn, 1965;Lim & Jahng, 2019). Although the practical differences between PCA and EFA are often negligible (Thompson, 2004), EFA is sometimes preferred (Nájera et al, 2023). Similar to the results from the PCA, analysis of the scree plot and eigenvalues from the EFA indicated a one factor-solution (see Figure S1).…”

Section: Factorial Structure [Aim #1]mentioning

confidence: 95%

The Perceived Economic Scarcity Scale: A valid tool with greater predictive utility than income

Auger,

Sommet,

Normand

2024

British J Social Psychol

View full text Add to dashboard Cite

This article introduces the Perceived Economic Scarcity Scale (PESS), a novel instrument measuring the subjective evaluation and experience of economic scarcity (the feeling of having insufficient financial resources to meet one's needs). We conducted three high‐powered preregistered studies (total N = 1900) to rigorously evaluate the PESS's psychometric properties. In Study 1, we generated a pool of items and used both Principal Component Analysis and Exploratory Factor Analysis to select the most appropriate items. In Study 2, we examined the PESS's construct validity, demonstrating that it measures a distinct construct from related constructs such as subjective social class. In Study 3, we examined the PESS's predictive validity, demonstrating that it is a robust predictor of both affective outcomes (e.g. anxiety‐depressive symptoms) and cognitive outcomes (e.g. economic risk‐taking). Critically, we found that the PESS not only has incremental validity over and above income but also has greater predictive utility than income. We also found that the PESS score varies depending on the distance‐to‐pay and has excellent test–retest reliability. Overall, the PESS appears to be a valid and reliable instrument for assessing perceived economic scarcity, and we encourage researchers to use it to better understand the psychological consequences of ‘not having enough’.

show abstract

Factor structure and measurement invariance of conceptions of happiness in Korea and Canada: an application of penalized structural equation modeling in Mplus

Joshanloo

2024

Qual Quant

View full text Add to dashboard Cite

Is exploratory factor analysis always to be preferred? A systematic comparison of factor analytic techniques throughout the confirmatory–exploratory continuum.

Cited by 9 publications

References 95 publications

New Developments in Measurement Invariance Testing - An Overview and Comparison of EFA-based Approaches

New Developments in Measurement Invariance Testing - An Overview and Comparison of EFA-based Approaches

The Perceived Economic Scarcity Scale: A valid tool with greater predictive utility than income

Factor structure and measurement invariance of conceptions of happiness in Korea and Canada: an application of penalized structural equation modeling in Mplus

Contact Info

Product

Resources

About