Veteran Recreational Therapy Engagement Survey for Machine Learning Prediction

57 survey records from veterans aged 18 and above at the New York State Veterans Home at Oxford. The data was collected for a cross-sectional observational study to develop machine learning models predicting recreational therapy participation. The dataset was created by Mohammad Najeh Samara and last updated on 2026-04-30.

Use Cases

• Predicting high recreational therapy participation based on activity preference diversity and total preferences.
• Identifying veterans at risk of non-participation based on facility tenure and limited preferences.
• Analyzing the influence of group and spiritual activities on engagement patterns.
• Validating machine learning algorithms like Random Forest for clinical prediction tasks in small-sample settings.

Strengths

• Data includes 57 veteran records with demographic characteristics, participation patterns, and activity preferences.
• Machine learning models were validated using Leave-One-Out and 5-fold Stratified Cross-Validation, with reported performance metrics like an F1-score of 0.860 for high participation prediction.
• Feature importance analysis identified specific predictors, such as activity preference diversity (Gini importance: 0.293) and facility tenure (0.268).

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is small (416.7 KB), limiting the scope and generalizability of findings.

Provenance

Source

New York State Veterans Home at Oxford

Collection Method

Comprehensive self-administered survey

Freshness

Last updated 2026-04-30 05:26:31; freshness should be verified.

Geography

New York State, United States