Keywords

CFA, Alignment-within-CFA, Scalar, Metric, Noninvariance, Invariance

Abstract

The need for research instruments adaptable to culturally diverse populations has grown with globalization and digital connectivity. Ensuring measurement invariance (MI) is crucial for generating accurate and comparable scores, especially in comparative studies. Traditional approaches like Multi-Group Confirmatory Factor Analysis (MG-CFA) often involve intricate procedures and can become unwieldy when adjustments for partial invariance are needed. The Alignment-within-CFA (AwC) method emerged as a promising alternative, designed to approximate group-specific factors and produce latent variables with uniform metrics. This study rigorously compares the AwC method and traditional MG-CFA across moderate numbers of groups (3, 4, and 5) under various conditions of noninvariance and sample sizes. By employing Monte Carlo simulations, the study controls study variables and explores a wide range of hypothetical scenarios, enhancing the precision and reliability of MI testing. The findings indicate that the AwC method is similar to or superior to the step-wise partial invariance approach, offering accurate and consistent results in varied scenarios. Specifically, the study examines the conditions under which AwC outperforms traditional MG-CFA and investigates the impact of factors such as different types of invariance, number of groups, and sample size on bias and model fit. This research provides deeper insights into the strengths and limitations of each method, guiding researchers in selecting the most appropriate approach for their specific contexts. The results support the use of the AwC method in scenarios where minimizing bias and error in parameter estimates is critical, paving the way for more streamlined and effective research amidst increasing global diversity.

Completion Date

2024

Semester

Summer

Committee Chair

Sivo, Stephen

Degree

Doctor of Philosophy (Ph.D.)

College

College of Community Innovation and Education

Department

Learning Sciences and Educational Research

Degree Program

Methodology, Measurement & Analysis

Format

application/pdf

Identifier

DP0028628

URL

https://purls.library.ucf.edu/go/DP0028628

Language

English

Rights

In copyright

Release Date

August 2027

Length of Campus-only Access

3 years

Access Status

Doctoral Dissertation (Campus-only Access)

Campus Location

Orlando (Main) Campus

Accessibility Status

Meets minimum standards for ETDs/HUTs

Download

Restricted to the UCF community until August 2027; it will then be open access.

Share

COinS