This dataset is designed for teaching the cross validation in machine learning. The dataset is a subset of data derived from the 1996 Adult Census Income dataset, and the example demonstrates how to use the cross validation to search for optimal values for user-specified parameters when using random forest to predict annual income class with individual features such as demographics, working status, marital status, etc. The dataset file is accompanied by a Teaching Guide, a Student Guide, and a How-to Guide for R.