The 2001 HRS HUMS contained one section, referred to as Section H: Child Educational Attainment and Expenses. This section contains questions about high school attendance, college attendance (number, duration, type), and the proportion of college expenses (tuition, food, housing) paid for by the parent of the child. Respondents answered the questions for each child (up to and including 10 children). The tuition and room/board data in this file are created by linking college attendance information for each child with two sources:

• the CASPAR college tuition database created and administered by the National Science Foundation.
• the Integrated Postsecondary Education Data System (IPEDS) administered by the National Center for Education Statistics

There are two versions of the HUMS 2001 Tuition Data. The differences between the two are summarized in HRS 2001 HUMS College Tuition Imputations.

1. The restricted version is directly based on tuition imputation procedures described in Section 3. A scheme of data 'fuzzing' is employed: tuition variables (TUITION or ROOMBRD) are rounded to the nearest $1,000 when than $10,000; to the nearest $100 if greater than $1,000 but smaller than $10,000; and to the nearest $10 if greater than $100 but smaller than $1,000. If a school tuition in a particular year is < $10, the variable would be assigned by a random draw from a uniform distribution. The structure of the restricted data file is summarized in Section 4.2.
2. In the public version, the original tuition imputations are transformed into somewhat cruder measures so that the names of colleges attended by HUMS children cannot be identified. Specifically, original tuition (and room and board) imputation values are replaced by quartile means conditional on the type of school (public versus private) and, if public, whether a child attended the school as an in-state or out-of-state student. The structure of the public data file is summarized in Section 4.3.

It is important to note that the two alternative data releases are remarkably comparable, as indicated in the following table. It is up to the researcher to decide whether to use the restricted file or public file.