Overview  Survey Design  Survey Quality Measures  Trend Data  Availability of Data

1. OverviewTop of Page.

a. Purpose

The National Survey of College Graduates (NSCG) is a longitudinal survey conducted since the 1970s, designed to provide data on the characteristics of the nation's college graduates. By surveying college graduates in all academic disciplines, the NSCG provides data useful in understanding the relationship between college education and career opportunities.

The information collected on the NSCG provides a unique source for examining various characteristics of college-educated individuals in the workforce, including occupation, work activities, salary, the relationship of degree field and occupation, and demographic information.

The 1993, 2003, and 2010 cycles of the NSCG provided coverage of the nation's college-educated population as of the survey reference date. In addition to the 1993, 2003, and 2010 survey cycles, the NSCG was conducted biennially or triennially in the periods 1990–99 and 2000–09. For these within-decade iterations of the NSCG, the survey focused on the science and engineering (S&E) workforce component of the college-educated population (health-related and S&E-related fields are included in the S&E definition beginning with 2003).

b. Respondents

NSCG survey respondents are individuals living in the U.S. during the survey reference week, holding a bachelor's or higher degree in any field, and under the age of 76.

c. Key variables

2. Survey Design Top of Page.

a. Target population and sample frame

The target population for the 1993, 2003, and 2010 NSCG survey cycles consisted of all individuals with the following characteristics:

The target population for NSCG survey cycles within the 1990s decade (1995, 1997, and 1999) included the majority of the S&E workforce and consisted of individuals with the following characteristics:

The target population for NSCG survey cycles within the 2000s decade (2006 and 2008) included the majority of the S&E workforce and consisted of individuals with the following characteristics:

The 1993 NSCG selected its sample from the 1990 decennial census long form respondents who indicated they had a bachelor's degree or higher in any field of study. The 1993 NSCG survey respondents served as the sample source for future survey cycles within the 1990 decade (i.e., the 1995, 1997, and 1999 NSCG).

The 2003 NSCG selected its sample from the 2000 decennial census long form respondents who indicated they had a bachelor's degree or higher in any field of study. The 2003 NSCG survey respondents served as the sample source for future survey cycles within the 2000 decade (i.e., the 2006 and 2008 NSCG).

The 2010 NSCG incorporated a dual frame sample design where it selected a portion of its sample from the 2009 American Community Survey respondents who indicated they had a bachelor's degree or higher in any field of study. The remaining portion of the 2010 NSCG sample was selected from respondents to the 2008 NSCG.

b. Sample design

In all survey cycles, the NSCG used a stratification sampling design to select its sample from the eligible sampling frame. Within the sampling strata, the NSCG used probability proportional to size (PPS) or systematic random sampling techniques to select the NSCG sample.

In the 1990s and 2000s, the NSCG sample was stratified by four sampling variables:

Stratification variables were revised in the 2010 NSCG to take advantage of the additional information available on its new sampling frame, the American Community Survey and stratified by three sampling variables.

c. Data collection techniques

The U.S. Census Bureau conducted the NSCG for the National Center for Science and Engineering Statistics (NCSES). Initial data collection used a self-administered mail survey and, beginning in the 2010 survey cycle, a self-administered Web survey. The contact strategy used to request survey participation included a prenotification letter, a first mailing, a reminder letter, and a second mailing.

Nonrespondents to the initial data collection were followed up using computer-assisted telephone interviewing (CATI). In the 1990 decade and the 2003 survey cycle, if the paper questionnaire was not returned and telephone follow-up attempts failed, for certain selected cases, a computer-assisted personal visit interview was conducted.

For each survey cycle, the data collection effort typically lasted 6 to 8 months.

d. Estimation techniques

Every sample case in the NSCG has a sample weight that reflects the portion of the population the case represents. This sample weight reflects weighting adjustments that were conducted to account for the sample selection, nonresponse, trimming procedures to eliminate extreme weights, and raking procedures to ensure the sampling weights agree with sampling frame estimates for certain key sampling variables. The sample weights enable data users to derive survey estimates that reflect the NSCG target population. The variable name on the NSCG data files for the NSCG sample weight is WEIGHT.

3. Survey Quality Measures Top of Page.

a. Sampling variability

NSCG variance estimates were calculated using the successive difference replication method. Due to the large amount of data collected in the NSCG, it is not practical to make direct calculations of a variance estimate for every NSCG survey estimate. As a result, generalized variance functions were developed to model the variance estimates for certain characteristics. Generalized variance parameters were then calculated through the use of the generalized variance function. Generalized variance parameters can be used to calculate standard errors for various types of characteristics.

Please contact the NSCG Project Officer to obtain the NSCG generalize variance parameters and for instructions for using the generalized variance parameters to calculate standard errors.

b. Coverage

The concept of coverage in the survey sampling process is the extent to which the total population that could be selected for sample "covers" the survey's target population. Any missed housing units or missed individuals within sample households in the ACS or the decennial census long form would create undercoverage in the NSCG. Additional undercoverage errors may exist because of errors in the NSCG sampling frame that led to incorrectly classifying individuals as not having bachelor's degrees or above when in fact they held such a degree.

c. Nonresponse

(1) Unit nonresponse. The weighted response rate for the 1993, 2003, and 2010 survey cycles ranged from 73% to 80%. For the subsequent rounds of the NSCG, the weighted response rates ranged from 87% to 95%. Results from the research and analysis of NSCG nonresponse trends have been used in the development of the nonresponse weighting adjustments. Careful selection of factors for constructing the nonresponse weighting adjustment cells minimizes the potential for nonresponse bias in the NSCG estimates.

(2) Item nonresponse. The NSCG item nonresponse rate for key items (employment status, type of employment, occupation, and primary work activity) ranged from 0.0% to 0.6%. Other variables, especially those involving sensitive information, had higher nonresponse rates. For example, salary and earned income had item nonresponse rates of approximately 10% to 12%. A Hot Deck imputation method was used to compensate for the item nonresponse.

d. Measurement

The NSCG is a survey of individuals and thus subject to reporting errors from differences in interpretation of questions. It is also true for any multimodal survey (mail, CATI, personal visit) that some measurement errors will differ by modality. To reduce measurement errors the NSCG questionnaires were pretested and focus groups held, and a set of survey concurrent evaluation research was done throughout the history of the survey. Please contact the NSCG Project Officer to obtain information on the NSCG survey evaluation research.

4. Trend Data Top of Page.

In 2010 the NSCG covered all those holding a college degree as of January 2009. In 2003 the NSCG covered all those holding a college degree as of April 2000. In 1993 the NSCG covered all those holding a college degree as of April 1990. Decade comparisons can be made among the 1993, 2003, and 2010 NSCG data as many of the core questions remained the same. However, small but notable differences exist, such as the collection of occupation and education data being based on the different versions of the Standard Occupational Classification (SOC) and Classification of Instructional Programs (CIP). Also due to the reference month differences in some survey cycles, seasonal differences may occur when making comparisons across decades. As a result, use caution when interpreting across-decade comparisons.

5. Availability of Data Top of Page.

a. Publications

The data from this survey are included in numerous NCSES InfoBriefs and Special Reports. All NCSES reports are available at http://www.nsf.gov/statistics/.

Information from this survey is also included in Science and Engineering Indicators and Women, Minorities, and Persons With Disabilities in Science and Engineering. Information from this survey is also published in detailed statistical tables from the Scientists and Engineers Statistical Data System (SESTAT).

b. Electronic access

The NSCG data are available in the SESTAT data tool and in public use files available through the SESTAT data download page.

c. Contact for more information

Additional information about this survey can be obtained by contacting:

John Finamore
Project Officer
Human Resources Statistics Program
National Center for Science and Engineering Statistics
National Science Foundation
4201 Wilson Boulevard, Suite 965
Arlington, VA 22230

Phone:(703) 292-2258
E-mail: jfinamor@nsf.gov