Skip Navigation
small NCES header image

Data Information


Have Questions Regarding ... Please contact...
Key statistics or general question(s) about NCES studies? NCES.Info@ed.gov
(866) 677-6987
Obtaining or modifying a restricted-use data license or decrypting a restricted-use data file? https://nces.ed.gov/statprog/rudman/
IESData.Security@ed.gov
(202) 502-7307
Analyzing restricted-use data or using NCES data tools? NCES.Info@rti.org
(866) 677-6987

Dec 2013

Final set of NPSAS:12 variables released
For a codebook of variables and to build tables and run regressions off the data, visit the DataLab: http://nces.ed.gov/datalab.

Aug 2013

Note on enrollment totals for NPSAS:12 First Look Report

The NPSAS:12 First Look Report indicates that the sample represents 26 million undergraduates and 4 million graduate students enrolled in postsecondary education between July 1, 2011 and June 30, 2012. These counts include students who were enrolled in multiple institutions in the 2011-12 academic year. However, the estimates in PowerStats and the restricted use file are based on an unduplicated enrollment count of 23 million undergraduates and 4 million graduate students. NPSAS:12 uses a multiplicity adjustment to unduplicate the count of undergraduate and graduate students. This unduplicated count is the weighted sum users will find on the NPSAS:12 dataset.

Aug 2013

Revision of NPSAS:08 Weights

NCES has reweighted NPSAS:08 data to match weighting procedures used in NPSAS:12. At the time NPSAS:08 was originally released, only 2006-07 12-month enrollment counts were available from the Integrated Postsecondary Education Data System (IPEDS) for poststratification (weighting estimates to known population totals). The revised weights, which use 2007-08 12-month enrollment counts, provide better estimates in sectors where significant enrollment shifts occurred between 2006-07 and 2007-08.

Revised weights for NPSAS:08 were released simultaneously with the release of NPSAS:12 data. The release includes revisions to the public NCES analysis tools, PowerStats and QuickStats, and to restricted-use data files. (The Institute of Education Sciences Data Security Office will provide instructions to current NPSAS:08 restricted-use data licensees on how to obtain the revised weights.) Prior NPSAS iterations did not use IPEDS 12-month enrollment counts for poststratification and, as such, are unaffected.

Why is it important to use the revised NPSAS:08 weights?

Since the release of NPSAS:08, NCES has observed substantial year-to-year volatility in IPEDS 12-month unduplicated enrollment totals, particularly among for-profit institutions. These fluctuations are large enough to influence NPSAS poststratification adjustments and, consequently, the resulting estimates. This volatility, combined with the ability to rely upon preliminary IPEDS data available earlier in NPSAS data processing, has led NCES to change its poststratification procedures for NPSAS:12 and use preliminary current year 12-month enrollment counts rather than final, adjudicated data from the prior academic year.

To maximize the benefit of these improved procedures, NCES has revised NPSAS:08 weights to reflect 12-month enrollment in the NPSAS year. As a result, analysts can expect greater accuracy in estimates for institutions with enrollment volatility during that time. The most notable improvement will be observed in those sectors where enrollment volatility between 200607 and 200708 was greatest, in particular private for-profit institutions. Because private for-profit institutions experienced an enrollment surge between 200607 and 200708, NPSAS:08 underestimated enrollment in that sector thereby overestimating ratio-based measures.

Differences in estimates before and after revision can be appreciable. For example, at private for-profit 2-year or more institutions, in NPSAS:08 the percentage of undergraduates who received federal aid decreased from 96 percent to 87 percent after reweighting; for those with federal aid, the average amount decreased from $7,700 to $7,300. Across sectors, the overall percentage of undergraduates who received federal aid remained the same at 47 percent; however, the average amount of federal aid decreased from $6,600 to $6,500.

Aug 2013

Institutions in Puerto Rico not sampled in NPSAS:12

Puerto Rican institutions were not included in the 2012 administration of NPSAS. Because Puerto Rican institutions have appeared in prior NPSAS studies with the exception of NPSAS:1987, analysts wishing to compare NPSAS:12 to other NPSAS administrations may filter those data sets on the COMPTO87 variable to exclude Puerto Rico. NCES plans to include Puerto Rican institutions in future NPSAS cycles.

Feb 2010

NPSAS:08 Student employment and remedial course variables were revised and new variables added

In the NPSAS:08 undergraduate PowerStats, the variables pertaining to student employment have been revised to correct the underestimates of students working full time and considering themselves to be primarily employees rather than students. For example, the estimated percentage of undergraduates considering themselves to be primarily employees (JOBROLE2) has been increased from 15 percent to 24 percent, the percentage working 35 or more hours per week (JOBENR2) has been increased from 26 percent to 34 percent, the average hours per week worked (JOBHOUR2) has been increased from 27 to 30 hours, and the average earnings per year (JOBEARN2) have been increased from $10,700 to $14,400. The employment variables will also be revised for the NPSAS:08 graduate DAS in the near future.

The variables for whether the student took remedial courses in 2007-08 (REMETOOK) and the number of remedial courses taken in 2007-08 (REMENUM) have been revised to include all undergraduates in the denominator. REMETOOK is now consistent with how the variable was coded in NPSAS:04. To restrict the denominator to include only first- and second-year students and those who were not in a degree program, use the variable REMETK2, Remedial courses: took in 2007-08 (1st and 2nd year students).

There has also been a minor revision to Undergraduate degree program (UGDEG). A few cases were recoded, causing no impact on the estimates. New variables have been added for net price, aid ratio, high school curriculum, college majors, and institution types.

Dec 2009

Revision of the weights in NPSAS:1996, NPSAS:2000, and NPSAS:2004 Data Analysis System files have been revised

The NPSAS:1996, NPSAS:2000, and NPSAS:2004 weights were revised in order to make the federal Stafford loan estimates comparable across the various cycles of NPSAS. The original weights were adjusted to sum to the total of Stafford loan gross commitments. In NPSAS:2008, however, the weights were adjusted to sum to the total Stafford loan net disbursements, which is the amount that students actually received after cancellations. The revised weights for NPSAS:1996, NPSAS:2000, and NPSAS:2004 (revised in June 2009) have now all been adjusted to sum to the actual net Stafford loan amounts disbursed and are now all comparable to the NPSAS:2008 Stafford loan data.

There has been no change in the NPSAS:1990 and NPSAS:1993 weights, because those cycles of NPSAS did not adjust to Stafford loan totals.

The revised NPSAS:1996 and NPSAS:2000 weights will produce estimates that differ from the estimates in any tables or publications produced before December 2009. The NPSAS:04 weight will produce estimates that differ from any tables or publications produced before June 2009. Using the revised weights has the largest effect on the estimates of the average Stafford loan amounts, and therefore on the averages of all composite financial aid variables that include Stafford loans, such as total loans, total aid, and cumulative loans.

For undergraduates, applying the revised weights reduced the average Stafford loan amount by about $600 in 2003-04, $300 in 1995-96, and by $500 in 1999-00. The average amount of total aid in those years was reduced by about one-half of these amounts. There were only minor changes in average grant amounts. However, the revised weights will result in some changes in the estimates for nearly all variables in NPSAS:1996, NPSAS:2000, and NPSAS:2005, although the differences are usually very small.


Would you like to help us improve our products and website by taking a short survey?

YES, I would like to take the survey

or

No Thanks

The survey consists of a few short questions and takes less than one minute to complete.