PLDF3 Variables

The tables below list the variables that are in PLDF3 and provide short definitions.

The intent in the organization of these tables was to group all like variables in the various years in one variable name. Rarely was there any ambiguity in reading the documentation from year to year. In one case, HRS_OPEN, in 1987-88, the variable was defined as a count from one week and from 1989, the number is annual. For 1987-88, as is noted below, I multiplied the number by 52 in order to convert the numbers for those first two years to something (roughly) comparable to the data from later years.

The documentation proved helpful and only one obvious error was discovered. According to the documentation, in 1989 the variables PSUNDUP and SVCINUSE were not reported but they are in the dataset. It is not clear whether they were reported for all institutions or only those who reported in spite of the data's not being collected or whether the documentation is in error and the data are collected appropriately.

Four variables have been added to each year's data in order to ease processing and subsequent analysis: newkey and span, state and year were added. Year was hard coded in one of the programs and is the nominal year of collection, that is, the fiscal year and the same as the year that appears on publications. State is the two-digit postal code. This variable was made from various variables in the dataset as noted below.

Five variables were discarded completely but could be readily included in a separate dataset if someone were to want any of them. They are HRSCD, POPCD, PUBSEQ, SEQ, and STPOP. They are discussed below. Of course, when variables such as ZIP replaced FLDE and LIBZIP, those replaced variables were discarded also but the data they reported are in PLDF3 in the variable ZIP. The data from these discarded five are not in PLDF3.

A spreadsheet (.csv, .ods, xlsx) presents these variables in another format. It shows that four variables (ELACCEXP, ELMATS, EMSVCACC, and INETUSE) were last collected in FY 2002. Six variables (LSABOUND, PRMATEXP, OTHMATEX, CAP_REV, EBOOK, and DATABASE) were added in FY 2003. And so on as variables are added (mostly) and subtracted from the data collection as data sought changed or were refined. The spreadsheets summarize this experience with adjustments being made here and there over the years. With the exceptions noted below the red line in the spreadsheet, any data reported to NCES or IMLS are in PLDF3.

The following table has the variable in the topical order used in the documentation. The convention followed is to use the latest variable name except where the data are no longer collected and the short definitions and notes come, largely, from the XXXdocumentation for 2012 although the definitions largely do not change over time. STATSTRU, however, does change slightly in the last few years and given its importance, the years with differences are listed separately. The wording differs on occasion for variables documented in the earlier years but the definitions overall are pretty consistent.

I am indebted to my colleague, Don McMorris, who spotted a problem in the FY 1987 employee variables that I had noted in previous documentation while I had missed the reason. He read the documentation for the 1987 data and found the problem: these variables that year were reported without an explicit decimal point. I have amended the input program to correct the error.

In the process of rereading the earlier documentation to make sure that there were no other similar problems, I found a practice in the version of the ICPSR data that is consistent but appears benign. One example of many can be found in the Indentification Variables below with the PLDF3 variable, ADDRESS. Note that from 1988 to 1989 (the years the data come to us from ICPSR) the variable was named "LIBADDR" while in the documentation it is "LIB_ADDR." The 1990 data were preserved—apparently the last copy—by a model librarian in the FSCS community in the original 5.25" floppy disk. (Take a bow, Libby). I adapted the ICPSR 1989 program to read these 1990 data and used the same convention of deleting the underscores.

Examination of the early NCES documentation and comparing it with the documentation from ICPSR shows a consistent practice of dropping the underscores. I am not a fan of such changes but I have seen no evidence that it has caused harm and the Hippocratic Oath of data compilation is: first do no harm. The work of ICPSR that I see reflected through their work on these data is consistently excellent and "fixing" the variable names to agree with the documentation seems to me more likely to do harm than leaving well enough alone.

November 22, 2022