This file is identical in format to a Case Data file. The contents, however, are reorganized by the transpose (XPOSE) program to switch from one record per case to one record per task in the inventory. In a statistical sense, "observations" are changed from cases to tasks, and hence the new file name. This file is currently used only to accomplish the task clustering within a study.
Although task clustering results are typically displayed using CLUSMAP and PRTFAC, the Task Data file does have rudimentary data capable of being displayed by PRTVAR or PRTJOB. Instead of case data background information, a Task Data file has only two background items. The variable V0001 is the Task ID (e.g., "B 27"), and V0002 is the first line of text from the original TASK statement (e.g., "Remove and Replace spark plugs from").
If one computes and prints a job description for a group of tasks, instead of task statements, one sees a roster of the case/booklet control numbers (for each case who performs tasks in this group) along with text "Case Number nnnn". This information, taken from V0001 and C0001 on the original Case Data file from which this file was created, is encoded to create pseudo task statements.