Dataset on turnover intention determinants among Generation Z micro-hospitality workers in a developing economy: Evidence from Indonesia
Description
This dataset provides empirical insights into the determinants of turnover intention among Generation Z employees within the informal micro-hospitality sector in a developing economy (Indonesia). The data were collected through an on-site, self-administered survey using a multistage stratified sampling strategy across five administrative districts in Malang City, East Java. The final validated dataset comprises 185 respondents who met strict inclusion criteria (Generation Z cohort, minimum six months of tenure). The instrument measures six theoretically derived constructs: Co-worker’s Warmth, Co-worker’s Competence, Job Satisfaction, Self-Resilience, Job Insecurity, and Turnover Intention, operationalized through 69 items on a 7-point Likert scale. The dataset has undergone rigorous preprocessing and psychometric evaluation, including missing data assessment, straight-lining detection, outlier screening, and normality checks. Exploratory Factor Analysis (EFA) confirmed a robust six-factor structure, supported by exceptional sampling adequacy (overall KMO = 0.903), strong internal consistency (Cronbach’s α = 0.919–0.967; McDonald’s ω = 0.920–0.968), and established discriminant validity (all HTMT ratios < 0.85). This dataset is a valuable resource for researchers and practitioners investigating workforce dynamics, employee retention strategies, and psychometric instrument validation, particularly within the context of informal, small-scale service economies in the Global South.
Files
Steps to reproduce
1. Download the Files Download all provided files, which include: Cleaned_dataset.xlsx (The final validated dataset with 185 respondents, ready for analysis) Raw_dataset.xlsx (The original unprocessed data with 203 responses, for transparency) Codebook_variables_and_scaling.pdf (Detailed dictionary of all variables, scales, and reverse-coded items) Survey_questionnaire_English.pdf & Survey_questionnaire_Indonesian.pdf (The original data collection instruments) 2. Review the Codebook Open the Codebook_variables_and_scaling.pdf to understand the variable names, the 7-point Likert scale coding (1 = Strongly Disagree to 7 = Strongly Agree), and demographic categories. Note: Items marked with "(R)" in the codebook have already been reverse-coded in the Cleaned_dataset.xlsx so that higher scores consistently reflect higher levels of the respective construct. 3. Import the Data Open the Cleaned_dataset.xlsx file in Microsoft Excel, or import it directly into your preferred statistical software (e.g., R, SPSS, Stata, Python, or SmartPLS). The dataset contains no missing values, as cases with >10% missing data or straight-lining patterns were removed during preprocessing. 4. Compute Construct Scores (Optional) To use the data for structural equation modeling (SEM) or regression, you can compute composite scores for each of the six constructs by calculating the mean of their respective items. For example, the "Co-worker's Warmth" score is the mean of items CoW1 through CoW12. 5. Replicate Psychometric Validation By running standard statistical procedures on the Cleaned_dataset.xlsx, secondary users can replicate the key validation metrics reported in the associated Data in Brief article, including: Descriptive statistics and normality assessment (Skewness and Kurtosis) Sampling adequacy (Kaiser-Meyer-Olkin measure and Bartlett’s test of sphericity) Exploratory Factor Analysis (EFA) using Principal Axis Factoring with Promax rotation Internal consistency reliability (Cronbach’s Alpha and McDonald’s Omega) Discriminant validity assessment using the Heterotrait-Monotrait (HTMT) ratio.
Institutions
- State University of MalangEast Java, Malang