- Home
- Data Catalog
- National Survey of College Graduates
National Survey of College Graduates
Add to My BasketDescription
The National Survey of College Graduates (NSCG) began in 1993 and has been conducted biennially to provide individuals, educational institutions, businesses and the Federal Government with the information they need to make important decisions. The survey provides data on the number and characteristics of individuals with a bachelor's or higher degree, with a special focus on individuals with education and/or employment in science or engineering. The survey includes age, educational history, citizenship, disability status, occupational/employment information, race and ethnicity, salary, sex, student loan debt, job satisfaction and work-related training. It samples individuals who are living in the United States or Puerto Rico during the survey reference week, have at least a bachelor's degree, and are younger than 76.
PLEASE NOTE: There are two versions of the NSCG. This metadata file refers to the NSCG data available through the NCSES Secure Data Access Facility (SDAF), a virtual data enclave that can only be accessed within the United States. This version of the NSCG data includes the variables found in the public use data files but allows for the ability to conduct longitudinal analyses of linked cases starting with the 2010 survey cycle. Additional linkages using this version are very limited. This version of the NSCG data does not include the restricted use data variables, such as state and institution level data. Researchers can apply for the NSCG restricted use data by selecting the NSCG version available through the U.S. Census Bureau's Federal Statistical Research Data Centers.
NCSES is updating their data linkage policy to better meet the data linkage needs of NCSES and external researchers. Currently, NCSES does not support researcher access to direct Personally Identifiable Information (PII). NCSES encourages researchers interested in developing an SAP application that includes linking NCSES restricted data to non-NCSES data sources to contact NCSES at NCSES_Licensing@nsf.gov to assess feasibility and appropriateness.
Detailed Methodology
The NSCG uses a four-panel rotating panel design that began with the 2010 NSCG. As part of this design, every new panel receives a baseline survey interview and three biennial follow-up interviews before rotating out of the survey.
Since the 2013 NSCG, the sample size has ranged from approximately 124,000 to 164,000 cases. For each survey cycle, the sample includes returning sample members from the prior NSCG and new sample members selected from a recent American Community Survey (ACS) frame.
Year-to-year comparisons can be made among the NSCG survey cycles because many of the core questions remained the same. Small but notable differences exist across some survey years, such as the collection of occupation and education data based on more recent taxonomies. Also, because of the use of different reference months in some survey cycles, seasonal differences may occur when making comparisons across years.
There is overlap in the cases included in every four sequential survey cycles (e.g., 2017, 2019, 2021, and 2023). This overlap among cases allows for the ability to conduct longitudinal analysis of this subset of the NSCG sample. To reduce the risk of disclosure, longitudinal analyses can be conducted only within a restricted environment.
The NSCG uses a trimodal data collection approach: Web survey, mail survey, and computer-assisted telephone interview (CATI). Data collection lasts approximately 6 months. The data collected in the NSCG are subject to both editing and imputation procedures. The NSCG uses both logical imputation and statistical (hot deck) imputation as part of the data processing effort. Every sample case in the NSCG has a final sample weight that reflects the portion of the overall population the case represents. This final sample weight reflects weighting adjustments that were conducted to account for the following:
Sample selection Nonresponse Trimming procedures to eliminate extreme weights Raking procedures to ensure the sampling weights agree with sampling frame estimates Overlap procedures to convert weights that reflect the population of each individual ACS frame into a final sample weight that reflects the NSCG target population
The final sample weights enable data users to derive survey-based estimates of the NSCG target population.
The NSCG data available on the NCSES SDAF includes the same variables as the public use data files but cases can be linked over time using the REFID variable. This allows for longitudinal analyses starting with the 2010 wave of the survey.