2019
DOI: 10.1177/0146621619893785
|View full text |Cite
|
Sign up to set email alerts
|

Fit Indices for Measurement Invariance Tests in the Thurstonian IRT Model

Abstract: This study examined whether cutoffs in fit indices suggested for traditional formats with maximum likelihood estimators can be utilized to assess model fit and to test measurement invariance when a multiple group confirmatory factor analysis was employed for the Thurstonian item response theory (IRT) model. Regarding the performance of the evaluation criteria, detection of measurement non-invariance and Type I error rates were examined. The impact of measurement non-invariance on estimated scores in the Thurst… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
9
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6
2

Relationship

1
7

Authors

Journals

citations
Cited by 14 publications
(9 citation statements)
references
References 41 publications
(69 reference statements)
0
9
0
Order By: Relevance
“…To assume measurement invariance, the value assigned to (ΔCFI) must not be greater than 0.01. In this sense, it is possible to observe, through the ΔCFI values, that the scale structure is stable, with no response bias in the samples for the different groups (Lee and Smith, 2020).…”
Section: Measurement Invariancementioning
confidence: 90%
“…To assume measurement invariance, the value assigned to (ΔCFI) must not be greater than 0.01. In this sense, it is possible to observe, through the ΔCFI values, that the scale structure is stable, with no response bias in the samples for the different groups (Lee and Smith, 2020).…”
Section: Measurement Invariancementioning
confidence: 90%
“…To assume measurement invariance, the value assigned to (ΔCFI) must not be greater than 0.01. It was observed, through the ΔCFI values, that the scale structure is stable, with no response bias in the samples for the different groups (Lee & Smith, 2020). The invariance data by multigroup con rmatory factor analysis (AFCMG), through the values of con gurable, metric, and scalar invariance, shows that the structures measured between the groups and the scalar units are the same for the groups studied since the values of difference from the Comparative Fit Index (ΔCFI) are below 0.01.…”
Section: Measurement Invariancementioning
confidence: 94%
“…For the Likert‐scale data, χ 2 difference tests between two nested measurement models, such as between configural invariance and metric invariance, and between metric invariance and scalar invariance, were used (Millsap, 2011; Vandenberg, 2002). For the forced‐choice data, cutoffs values suggested in Lee and Smith (2020b) 5 were employed to determine measurement invariance; if a change in CFI between configural and metric invariance models is larger than 0.007, metric invariance is not met; if a change in CFI between metric and scalar invariance models is larger than 0.001 or larger than 0.004 in NCI, scalar invariance is not met. Changes in CFI and NCI were computed with the adjusted degrees of freedom due to the redundancy in each item block composed of three or more statements.…”
Section: Methodsmentioning
confidence: 99%
“… χ 2 difference tests (deviance tests) are not readily available through the DIFTEST function in Mplus due to the adjustment in degrees of freedom for forced‐choice data (see Lee and Smith, 2020b, for technical details about measurement invariance tests for forced‐choice data).…”
mentioning
confidence: 99%