Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling [Oral Presentation]

Tian-fang Frank Ye; Emma E. Buchtel

Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling [Oral Presentation]

Department of Applied Social Sciences

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

Abstract

[Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling] In this talk, I will introduce how we utilized Rasch modeling and Bayesian structural equation modeling (BSEM) to develop and validate the Particularism vs. Corruption Tolerance Scale (PCTS; Buchtel et al., 2021). The PCTS is a scenario-based, self-reported instrument that measures participants’ acceptability of two types of stories regarding breaking the rules: particularism and corruption. First, in two pilot studies, 19 pairs of scenarios (a total of 38 items) were generated that represent the two types of behaviors. Next, in Study 1, these pairs were rated by a total of 499 participants (Mainland =283, Hong Kong =216). We utilized Rasch analysis to examine the scale functioning and to conduct item selection based on item fit indices. In addition, Differential item functioning (DIF) was used to exclude items that were perceived differently across cultures. As a result, Study 1 retained 8 pairs of scenarios (16 items) for the PCTS final version, which demonstrated good construct validity, internal consistency, and discriminant validity. Finally, in Study 2, the 16-item PCTS were rated by 991 adult online participants (Hong Kong = 455, Mainland China = 536). Because the assumptions of frequentist CFA are considered too restrictive, less practical, and often lead to model rejection (Asparouhov et al., 2015; Sellbom Tellegen, 2019), we utilized BSEM to examine the measurement model, approximate measurement invariance, and external validity of the PCTS. In BSEM, trivial model misspecifications and random noise (i.e., cross-loadings, correlated residuals) were estimated with small informative priors. The results supported the construct validity, partial scalar invariance, and discriminant validity of the scale.

Original language	English
Title of host publication	25th International Congress of the International Association for Cross-Cultural Psychology
Place of Publication	Prague, Czech Republic
Publication status	Published - 27 Jul 2021

Cite this

@inproceedings{1a5b289394834ce6a628e27dfd1884d4,

title = "Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling [Oral Presentation]",

abstract = "[Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling] In this talk, I will introduce how we utilized Rasch modeling and Bayesian structural equation modeling (BSEM) to develop and validate the Particularism vs. Corruption Tolerance Scale (PCTS; Buchtel et al., 2021). The PCTS is a scenario-based, self-reported instrument that measures participants{\textquoteright} acceptability of two types of stories regarding breaking the rules: particularism and corruption. First, in two pilot studies, 19 pairs of scenarios (a total of 38 items) were generated that represent the two types of behaviors. Next, in Study 1, these pairs were rated by a total of 499 participants (Mainland =283, Hong Kong =216). We utilized Rasch analysis to examine the scale functioning and to conduct item selection based on item fit indices. In addition, Differential item functioning (DIF) was used to exclude items that were perceived differently across cultures. As a result, Study 1 retained 8 pairs of scenarios (16 items) for the PCTS final version, which demonstrated good construct validity, internal consistency, and discriminant validity. Finally, in Study 2, the 16-item PCTS were rated by 991 adult online participants (Hong Kong = 455, Mainland China = 536). Because the assumptions of frequentist CFA are considered too restrictive, less practical, and often lead to model rejection (Asparouhov et al., 2015; Sellbom Tellegen, 2019), we utilized BSEM to examine the measurement model, approximate measurement invariance, and external validity of the PCTS. In BSEM, trivial model misspecifications and random noise (i.e., cross-loadings, correlated residuals) were estimated with small informative priors. The results supported the construct validity, partial scalar invariance, and discriminant validity of the scale.",

author = "Ye, {Tian-fang Frank} and Buchtel, {Emma E.}",

year = "2021",

month = jul,

day = "27",

language = "English",

booktitle = "25th International Congress of the International Association for Cross-Cultural Psychology",

}

Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling [Oral Presentation]. / Ye, Tian-fang Frank; Buchtel, Emma E.
25th International Congress of the International Association for Cross-Cultural Psychology. Prague, Czech Republic, 2021.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling [Oral Presentation]

AU - Ye, Tian-fang Frank

AU - Buchtel, Emma E.

PY - 2021/7/27

Y1 - 2021/7/27

N2 - [Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling] In this talk, I will introduce how we utilized Rasch modeling and Bayesian structural equation modeling (BSEM) to develop and validate the Particularism vs. Corruption Tolerance Scale (PCTS; Buchtel et al., 2021). The PCTS is a scenario-based, self-reported instrument that measures participants’ acceptability of two types of stories regarding breaking the rules: particularism and corruption. First, in two pilot studies, 19 pairs of scenarios (a total of 38 items) were generated that represent the two types of behaviors. Next, in Study 1, these pairs were rated by a total of 499 participants (Mainland =283, Hong Kong =216). We utilized Rasch analysis to examine the scale functioning and to conduct item selection based on item fit indices. In addition, Differential item functioning (DIF) was used to exclude items that were perceived differently across cultures. As a result, Study 1 retained 8 pairs of scenarios (16 items) for the PCTS final version, which demonstrated good construct validity, internal consistency, and discriminant validity. Finally, in Study 2, the 16-item PCTS were rated by 991 adult online participants (Hong Kong = 455, Mainland China = 536). Because the assumptions of frequentist CFA are considered too restrictive, less practical, and often lead to model rejection (Asparouhov et al., 2015; Sellbom Tellegen, 2019), we utilized BSEM to examine the measurement model, approximate measurement invariance, and external validity of the PCTS. In BSEM, trivial model misspecifications and random noise (i.e., cross-loadings, correlated residuals) were estimated with small informative priors. The results supported the construct validity, partial scalar invariance, and discriminant validity of the scale.

AB - [Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling] In this talk, I will introduce how we utilized Rasch modeling and Bayesian structural equation modeling (BSEM) to develop and validate the Particularism vs. Corruption Tolerance Scale (PCTS; Buchtel et al., 2021). The PCTS is a scenario-based, self-reported instrument that measures participants’ acceptability of two types of stories regarding breaking the rules: particularism and corruption. First, in two pilot studies, 19 pairs of scenarios (a total of 38 items) were generated that represent the two types of behaviors. Next, in Study 1, these pairs were rated by a total of 499 participants (Mainland =283, Hong Kong =216). We utilized Rasch analysis to examine the scale functioning and to conduct item selection based on item fit indices. In addition, Differential item functioning (DIF) was used to exclude items that were perceived differently across cultures. As a result, Study 1 retained 8 pairs of scenarios (16 items) for the PCTS final version, which demonstrated good construct validity, internal consistency, and discriminant validity. Finally, in Study 2, the 16-item PCTS were rated by 991 adult online participants (Hong Kong = 455, Mainland China = 536). Because the assumptions of frequentist CFA are considered too restrictive, less practical, and often lead to model rejection (Asparouhov et al., 2015; Sellbom Tellegen, 2019), we utilized BSEM to examine the measurement model, approximate measurement invariance, and external validity of the PCTS. In BSEM, trivial model misspecifications and random noise (i.e., cross-loadings, correlated residuals) were estimated with small informative priors. The results supported the construct validity, partial scalar invariance, and discriminant validity of the scale.

M3 - Conference article published in proceeding or book

BT - 25th International Congress of the International Association for Cross-Cultural Psychology

CY - Prague, Czech Republic

ER -

Development and Validation of the Particularism vs. Corruption Tolerance Scale: An Application of Rasch Analysis and Bayesian Structural Equation Modeling [Oral Presentation]

Abstract

Fingerprint

Cite this