Research Data

This dataset originates from a quantitative study investigating the factors influencing purchase intention toward remanufactured laptops among Indonesian undergraduate students. The data were collected in September 2024 from four universities, with a total of 438 valid responses obtained. The study employed a MaxDiff (Best-Worst Scaling) approach supported by Latent Class Analysis (LCA) and TURF (Total Unduplicated Reach and Frequency) analysis to evaluate preference heterogeneity and attribute significance.Data Generation and ProcessingPrior to responding to the MaxDiff questions, all participants were exposed to a short educational module explaining the distinctions between remanufactured, refurbished, and used (BSB) laptops. The survey was delivered online, with consent and guidance provided to ensure clarity and voluntary participation. MaxDiff design involved 21 attributes hypothesized to influence purchase intention. Each respondent was exposed to multiple choice sets where they had to identify the most and least important factor in each set.Raw responses were exported from the survey platform and processed using SPSS and Excel. Root Likelihood (RLH) scores were computed for all respondents to check data consistency. All participants exceeded the minimum RLH threshold (0.047), indicating valid and reliable input. The utility estimation for each factor was computed using hierarchical Bayes modeling, followed by segmentation using LCA.File OverviewThe dataset comprises the following files:Survey pengumpulan data fase 1 – Factor for purchase intention Design & choices.xlsx Contains the design of the MaxDiff survey, including 21 attributes, block configurations, and randomized choice sets used during data collection.Survey pengumpulan data fase 1 – Factor for purchase intention Individual raw scores.xlsx Raw selection data from each respondent indicating which attribute was chosen as best/worst in each task. Includes respondent IDs and task-level selections.Survey pengumpulan data fase 1 – Factor for purchase intention Individual scores.xlsx Contains utility scores for each of the 21 factors on a per-respondent basis. This file is the main input for latent class segmentation.Survey pengumpulan data fase 1 – Factor for purchase intention Scores.xlsx Aggregated utility scores with confidence intervals for each attribute. This is the source for comparative analysis of factor importance.Survey pengumpulan data fase 1 – Factor for purchase intention Counts.xlsx Provides count-based frequency data indicating how often each attribute was selected as best or worst across the sample. Useful for exploratory analysis or validation of utility trends.Data CharacteristicsSample size: 438 undergraduate studentsGeographical scope: Indonesia (Java-based universities)Temporal scope: September 2024Resolution: Individual-level utility scores per attributeMeasurement scale: MaxDiff (Best-Worst) trade-off method; no Likert items usedEach of the 21 columns in the score files corresponds to a decision factor, including quality knowledge, perceived benefits, green knowledge, functionality risk, and social pressure, among others. Scores are standardized utilities representing relative importance and have no units. Missing data is negligible (<1%) and primarily occurs in demographic metadata; they are encoded as NA and were excluded from analysis.Data Access and UsageAll files are stored in .xlsx format and are compatible with Microsoft Excel, Google Sheets, R, Python (pandas), and most statistical software. No proprietary software is required to open or use the files. Researchers interested in replicating the study or applying alternative modeling techniques (e.g., Mixed Logit, Hierarchical Bayes) can use the raw selection data and design matrices provided.This dataset provides empirical evidence for understanding how various psychological, environmental, economic, and social factors drive remanufactured product acceptance in developing countries. It is particularly valuable for scholars of sustainable consumption, circular economy, and market segmentation in emerging markets.