Concise Two-Domain Teacher Professional Identity Scale

Description

This repository contains the data and supporting materials for the study “Who I Am and How I Act: Development and Initial Psychometric Evaluation of a Concise Two-Domain Teacher Professional Identity Scale.” The study developed and evaluated a brief measure of teacher professional identity using survey data from 623 in-service high school teachers in China.​ The repository is organized into five folders. The data folder contains the de-identified data files in CSV format. The figures folder contains the path diagrams generated from the statistical models. The report folder contains readable text reports summarizing the results of all analyses. The result folder contains the intermediate statistical results produced throughout the analyses. The script folder contains all R scripts used to conduct the analyses.​ Together, these materials cover multivariate normality testing, exploratory and confirmatory factor analyses, repeated random split-half cross-validation, bifactor analysis, reliability and construct validity assessment, measurement invariance testing, and latent mean comparisons across gender and teaching-subject categories. They are provided to support transparency, verification, and reproduction of the findings reported in the article and its supplementary material.

Steps to reproduce

# Steps to Reproduce 1. Download and extract the complete repository while retaining the original folder structure. The repository root should contain the `data`, `figures`, `report`, `result`, and `script` folders. 2. Install R and, if preferred, RStudio. The scripts require several R packages, including `MVN`, `moments`, `psych`, `GPArotation`, `nFactors`, `corrplot`, `lavaan`, `semTools`, `semPlot`, `dplyr`, `readr`, `tibble`, and `car`. Missing packages are installed automatically by the scripts when an internet connection is available. 3. Open each R script and replace the original root path, `X:/DataTeacherIdentity`, with the local path to the extracted repository. Retain the existing subfolder names and the filename `data/raw_data.csv`. 4. Run the scripts in the following order: 1. `script/mvn/mvn_initial.R` 2. `script/efa/efa_initial.R` 3. `script/efa/efa_round2.R` 4. `script/efa/efa_round3.R` 5. `script/efa/efa_round4.R` 6. `script/mvn/mvn_round2.R` 7. `script/cfa/cfa.R` 8. `script/split_half/split_half.R` 9. `script/bifactor/bifactor.R` 10. `script/validity_reliability/validity_reliability.R` 11. `script/invar_gender/invar_gender.R` 12. `script/invar_course_category/invar_course_category.R` 5. The exploratory factor analysis scripts must be run sequentially because each later round uses the retained dataset produced by the preceding round. The other scripts use `data/raw_data.csv` and can be run after completion of the item-selection procedure. 6. Newly generated intermediate and final statistical outputs will be saved in the corresponding subfolders under `result`. Readable summaries of the analyses can be compared with the archived text files in `report`, while the model path diagrams can be compared with those in `figures`. 7. The split-half cross-validation script uses a fixed random seed, allowing the 10 random sample divisions to be reproduced. Small numerical or graphical differences may nevertheless occur when different versions of R or the required packages are used.

Institutions

• Yunnan University

  Yunnan, Kunming

Categories

Licence