Help – Census Enrollment – General


(copied from the Overview tab on the Census Autumn Enrollment dashboard)

The census is particularly useful for comparisons over time, as it represents a stable, consistent, official view of Stanford’s student population.

WHAT:  An annual autumn snapshot of the student population. 

AVAILABLE YEARS:  From the academic year 2002-2003 to the most recently available census year.

WHEN:  The census is taken about a week after students have confirmed their study plans (mid to late October).

WHO:  This dashboard includes active, degree-seeking students who are registered as of the census date and not on leave.  Non-degree-seeking students are also captured in the census; for now, use ad hoc to report on them.

Students are counted in 3 different ways in the census; the numbers will vary depending on which count is used:

Please refer to the Headcount Info tab, for more information on the different headcounts and what they mean.


Census data is used for required reporting to the Federal Government; it is also used for a wide range of external reporting requests and surveys (e.g. Common Data Set, US News & World Report), and for internal records as well (Stanford Facts, Internal Statistics Book, etc.).

THE TIMING OF THE CENSUS (for those who care)

The Federal Government asks institutions of higher learning to take an autumn census as close to Oct. 15th as possible.  For Stanford, with its quarter system, that date is often before the Final Study List date, and a census would not be very accurate.  Stanford has always waited until after the Final Study List date, and after the subsequent “Term Cancel” process is run by the Registrar’s Office, which cleans out people who didn’t end up being here this quarter after all (in some cases because they graduated, yay!).  The exact date of the Census is recorded (in an Ad Hoc field); in the years included in SIRIS it has varied between 10/14 to 10/23, and in recent history is typically the 9th day after the Final Study List day – the Final Study List deadline is generally on a Friday, the census is as of the Sunday one week and two days later.  In the past, the availability of the census data could be several days past the actual snapshot; with the new system, we hope the data can be available within a couple days of the actual snapshot, or about 11 days after the Final Study List.


There are some aspects of student records where we can go back in time and check information or even add information that was true as of the census date.  For instance, we have added in what subplans students were in on the census date, which was not captured in the original census data.  We could also potentially go back and add the missing “half” of coterm students, though we have not done that yet.  (Only one side of a coterm student is included in the “PS1” era below; both sides are captured in the PS2 era.)

However some aspects of student records get overwritten in PeopleSoft, so for those aspects, we have to rely on what was captured by the census process:

  • We can’t recapture the basic population – who was term enrolled on the census date, other than what was captured at the time, because that information gets overwritten in PeopleSoft.  For instance, if a student withdraws after the census date, you would not be able to tell they were “not withdrawn” on the census date.  We can’t go back and start capturing census data on every quarter, for instance.
  • We can’t recapture biodemo or citizenship data as of the census date, other than what was captured at the time, because PeopleSoft only maintains current info – past info is overwritten.


This can be relevant because history reflects what kind of information is available.  The most recent years have more information, but you have to know how to pull it out properly (e.g. only take one quarter at a time or you will get inflated results).

NSI no 1987 through 2000-2001 a complex manual process involving SPIRES, probably SAS text files, converted and merged into a single table in DSS We have not yet made these years available, because significant bridging work would need to be done to convert NSI departments and majors and make sure they match up properly to the PeopleSoft ones. In theory, it would be possible though, if there’s enough demand for it.
PS0 – transition year no 2001-2002 a complex manual process involving

Brio Query tool, Excel, SAS

text files, converted and merged into a single table in DSS The transition to PeopleSoft took more than one year. The very first year of being in PeopleSoft, the data is not that complete/consistent, so we opted not to include it in this release of census data.
PS1 – first 10 years yes 2002-2003 through 2011-2012 a complex manual process involving

Brio Query tool, Excel, SAS

text files, converted and merged into a single table in DSS Only rows that would be counted in the census were included in the file
PS2 – second era yes 2012-2013 through now a mostly automated process involving views & Informatica PowerCenter single table in DSS Census file includes all student/plans for students not withdrawn and active on the census date, but only certain ones are tagged for inclusion in the various headcounts. Also, census started being taken every quarter, and the census started having advance runs for error-checking.