Factor Analysis and Latent Structure: IRT and Rasch Models

Abstract Background The Centre for Epidemiologic Studies Depression Scale (CES-D) is a commonly used psychometric scale of depression. A four-factor structure (depressed affect, positive affect, somatic symptoms, and interpersonal difficulties) was initially identified in an American sample aged 18 to 65. Despite emerging evidence, a latent structure has not been established in adolescents. This review aimed to investigate the factor structure of the CES-D in adolescents. Methods We searched Web of Science, PsychINFO and Scopus and included peer-reviewed, original studies assessing the factor structure of the 20-item CES-D in adolescents aged ≤18. Two independent researchers screened results and extracted data. Results Thirteen studies met the inclusion criteria and were primarily from school-based samples in the USA or Asia. Studies that conducted confirmatory factor analysis (CFA; n = 9) reported a four-factor structure consistent with the original factor structure; these studies were primarily USA-based. Conversely, studies that conducted exploratory factor analysis (EFA) reported distinct two or three factor structures (n = 4) and were primarily based in Asia. Limitations Studies in a non-English language and those that included individuals aged > 18 years were excluded. Ethnic or cultural differences as well as different analytical methods impacted generalisability of results. The use of CFA as the primary analysis may have biased towards a four-factor structure. Conclusions A four-factor CES-D structure was an appropriate fit for adolescents in Western countries; further research is required to determine the fit in in Asian countries. This has important implications for clinical use of the scale. Future research should consider how cultural differences shape the experience of depression in adolescents.

Download Full-text

Comparing item parameter estimates and fit statistics of the Rasch model from three different traditions

Jurnal Penelitian dan Evaluasi Pendidikan ◽

10.21831/pep.v24i1.29871 ◽

2020 ◽

Vol 24 (1) ◽

Author(s):

Bahrul Hayat ◽

Muhammad Dwirifqi Kharisma Putra ◽

Bambang Suryadi

Keyword(s):

Factor Analysis ◽

Rasch Model ◽

Measurement Model ◽

Rasch Measurement ◽

Item Parameter ◽

Rasch Models ◽

Rasch Measurement Model ◽

Item Factor Analysis ◽

The Rasch Model ◽

Special Case

Rasch model is a method that has a long history in its application in the fields of social and behavioral sciences including educational measurement. Under certain circumstances, Rasch models are known as a special case of Item response theory (IRT), while IRT is equivalent to the Item Factor Analysis (IFA) models as a special case of Structural Equation Models (SEM), although there are other ‘tradition’ that consider Rasch measurement models not part of both. In this study, a simulation study was conducted to using simulated data to explain how the interrelationships between the Rasch model as a constraint version of 2-parameter logistic (2-PL) IRT, Rasch model as an item factor analysis were compared with the Rasch measurement model using Mplus, IRTPRO and WINSTEPS program, each of which came from its own 'tradition'. The results of this study indicate that Rasch models and IFA as a special case of SEM are mathematically equal, as well as the Rasch measurement model, but due to different philosophical perspectives people might vary in their understanding about this concept. Given the findings of this study, it is expected that confusion and misunderstanding between the three can be overcome.

Download Full-text

Clinical target syndromes in psychiatry: latent structure analysis versus factor analysis

European Psychiatry ◽

10.1017/s0924933800000420 ◽

1991 ◽

Vol 6 (6) ◽

pp. 301-306 ◽

Cited By ~ 1

Author(s):

P Bech

Keyword(s):

Factor Analysis ◽

Structure Analysis ◽

Statistical Models ◽

Rating Scales ◽

Rating Scale ◽

Latent Structure ◽

Depression And Anxiety ◽

Sufficient Statistics ◽

Latent Structure Analysis ◽

Symptom Rating

SummaryThe algorithms for the demonstration of shared phenomenology of psychiatric syndromes in DSM-III are resistant to quantification. In contrast, the rating scale approach quantifies clinical target syndromes in psychiatry. The two most useful statistical models for quantifying shared phenomenology by symptom rating scales have been reviewed; namely factor analysis and latent structure analysis. Results have shown that factor analysis has demonstrated dimensions of dementia, delirium, schizophrenia, mania, outward aggression, depression and anxiety. Latent structure analysis has confirmed that the items of brief rating scales (such as the Melancholia Scale) are additively related implying that their total scores are sufficient statistics for the measurement of these factors or dimensions. Latent structure analysis should be considered as a psychometric “glasnost” compared to algorithm-resistant logic of quantification in DSM-III.

Download Full-text