Replication Data for: Increasing App Engagement Behaviors via Goal-Enabling Technology Features: The Role of Goal Difficulty Dimensions

The underlying data for this study were provided by Raiz Invest under license and cannot be shared publicly. Access to the original data may be granted upon reasonable request to the corresponding author with Raiz Invest’s approval, or by obtaining permission directly from Raiz Invest. To support replication as much as possible within these constraints, we provide a synthetic dataset based on the original data with noise injection, along with accompanying code for applying the analysis procedures. This replication is limited to the key main results reported in Tables 3, 4, and 5 of the paper. All data files are noised datasets generated based on proprietary original datasets. They are intended to allow replication of estimation procedures and approximate results, but they do not reproduce the exact coefficients or standard errors from the proprietary data. • gdata_nc.rds Synthetic data used for the analyses reported in Table 3. • gdata_signins.rds Synthetic data used for the analyses reported in Table 3. • replication_data_retention_after.rds Synthetic data used for the analyses reported in Table 4. • synthetic_data.csv Synthetic data used for the analyses reported in Table 5. Code files • helper.R Helper functions used in the estimation of models for Table 3. • run_model.R Main script to replicate the results reported in Table 3. • Table_4_replication.R Main script to replicate the results reported in Table 4. • fieldexperiment_replication_code.R Main script to replicate the results reported in Table 5. Each script loads the relevant noised dataset, estimates the corresponding models, and produces output suitable for comparison with the reported tables.