Survey Designs

The National Science Foundation (NSF) collects data about college graduates and doctorate degree recipients living in the US through three surveys: the National Survey of College Graduates (NSCG), the Survey of Doctorate Recipients (SDR), and the National Survey of Recent College Graduates (NSRCG). Respondents from each of these three surveys with science or engineering degrees or occupations are compiled into one dataset: the SESTAT (the Scientists and Engineers Statistical Data System). Health field degree holders or practictioners are included in SESTAT for all respondents in the SDR survey and 2003-forward versions of the NSCG and NSRCG.

IPUMS-HigherEd harmonizes the SESTAT samples from survey years 1993-onward, and the original full SDR survey samples. Due to sampling methods, all respondents in the SDR have science, engineering, and health doctorate degrees, and are all also found in the SESTAT for that year. In contrast, not all respondents to NSCG are included in the SESTAT sample because the the non-scientists and engineers are not included. However, in most years, a number of questions that are asked only of SDR respondents are not included in SESTAT files. There are also circumstances where full SDR variables have more detail than the same variable in the SESTAT version of the sample.

More detailed information about each survey component of the SESTAT is available through the following links:

National Survey of College Graduates (NSCG)

The National Survey of College Graduates is a survey of individuals residing in the United States, under the age of 76, who hold a bachelor's degree or higher.

Survey of Doctorate Recipients (SDR)

The Survey of Doctorate Recipients collects information from individuals under the age of 76 who received a doctorate degree from a US institution.

National Survey of Recent College Graduates (NSRCG)

The National Survey of Recent College Graduates collects data about individuals under the age of 76 who received a science or engineering degree from a US academic institution within the previous 2 years.

External Resources

Diagram from the National Science Foundation that illustrates the target population of the SESTAT that is drawn from each of the NSF surveys.
Described in the caption

Source: National Science Foundation, Division of Science Resources Studies

Back to Top

National Survey of College Graduates

The National Survey of College Graduates (NSCG) is a longitudinal survey that collects information from college graduates living in the United States. Until the 2010 Census, eligible individuals for the survey were identified by the education attainment responses to the US Census long form. The target population included non-institutionalized individuals living in the US under the age of 76 who had completed at least a bachelor's degree by the date of the Census. The first precursor to the NSCG, the Postcensal Manpower Survey, was conducted in 1962. The most recent redesign of the survey was fielded in 1993 which drew the sampling cohort from the 1990 US Decennial Census.

The sample design is essentially a two-stage sampling scheme. Census long-form households in 1990 and 2000, and ACS households in 2010 were selected using a stratified systematic sampling method. Then, respondents for the NSCG surveys were chosen from individuals in the target population from long-form or ACS households using a stratification sampling scheme using age, race, highest degree type, occupation, and sex. In 2010, sex was dropped as a stratification variable. The survey data were collected by mail survey, with follow-up computer-assisted telephone interviews for initial non-respondents. A self-administered Web survey was introduced in 2010.

In the current design, a new cohort is selected for each decennial census, and follow-up surveys are sent to the cohort every 2 to 3 years for the rest of the decade. Starting with the 2010 US Decennial Census, college degree-holders were identified using the American Community Survey (ACS), which effectively replaced the Census long-form. With each iteration of the survey, except for the first NSCG survey of the decade, respondents for the NSRCG are surveyed again with the NSCG survey. Please see our User Note on the longitudinal aspects of SESTAT surveys. An exception to the NSCG follow-up pattern is that the 2010 NSCG survey also surveyed respondents from the 2008 NSCG.

Each year, the NSCG includes between 77,000 (in 2010) and 150,000 respondents (in 1993). According to the National Research Council (2008)*, the NSCG represents the stock of scientists, engineers, and health scientists in the US. Unlike the other two components of the SESTAT database, the NSCG sampling frame includes scientists and engineers that did not receive higher education in the United States, as well as non-scientists and engineers with a bachelor's degree or higher. The full NSCG files include all respondents, however the SESTAT versions of the NSCG files only include individuals who have either a science and engineering degree or a science and engineering occupation, or both. In other words, individuals without a science or engineering post-secondary degree and not currently working in a science or engineering occupation were dropped for the creation of a SESTAT file.

Back to Top

Survey of Doctorate Recipients

The Survey of Doctorate Recipients (SDR), first conducted in 1973, gathers data from individuals who have earned a doctorate degree in science, engineering, or health in the United States. The pool of possible respondents comes from the Survey of Earned Doctorates, which is given to every new recipient of a research doctorate from a US academic institution. Respondents for the SDR were chosen as a stratified sample from the eligible individuals in the Doctorate Records File, which is the resulting database of the Survey of Earned Doctorates. The SDR is fielded every 2 to 3 years, and on average includes 47,000 records. Response rates range from above 80 percent in the 1990's to approximately 76 percent in the 2013 SDR survey. All individuals in the SDR survey are included in the SESTAT of that year. After being selected for the SDR, respondents receive follow-up SDR surveys until they turn 76. It is possible to link these individuals across samples - for more information, see our Longitudinal Data User Note.

Starting in 2010, respondents to the SDR who were currently living abroad were moved to a separate file that became the International Survey of Doctorate Recipients (ISDR). Over time, the sample size of the ISDR has increased.

According to the National Research Council (2008)*, the SDR represents the both the stock and flow of scientists, engineers, and health scientists with doctorate degrees. Respondents were contacted by mail, and non-respondents were followed-up with computer-assisted telephone interviews. Later surveys also used web surveys for data collection.

Back to Top

National Survey of Recent College Graduates

The National Survey of Recent College Graduates (NSRCG) interviews individuals who earned a bachelor's or master's degree from select US academic institutions within two years prior to the survey. The NSRCG has been conducted every 2 to 3 years since 1993. The NSRCG respondents are identified in a two-stage sampling procedure. First, a stratified sample of US colleges and universities is chosen to provide a list of graduates from the previous two years. The stratification scheme involved the creation of a composite size measure using the number of graduates, degree level, field of major, racial diversity, and gender ratio. The probability of an institution being chosen increased proportionally with the number of degrees awarded by the institution. Information about these institutions was collected using the Integrated Postsecondary Education Data System (IPEDS). Of the 302 institutions selected in the first sampling stage in 2010, 290 responded with the requested lists of graduates.

The second stage of sampling selects individuals from the lists of graduates who received a science, engineering, or health degree. Graduate response rates range from 72 to 84 percent, with some attrition within cohorts over time.

According to the National Research Council (2008)*, the NSRCG represents the flow of new scientists, engineers, and health scientists from US academic institutions. Each cohort from the NSRCG is then surveyed in the NSCG 2 to 3 years later. NSRCG respondents were surveyed using computer-assisted telephone interviews, with mail follow-up for non-respondents. The NSRCG was discontinued in 2010.

* National Research Council. Using the American Community Survey for the National Science Foundation's Science and Engineering Workforce Statistics Programs. Washington, DC: The National Academies Press, 2008.

Back to Top