You are viewing a Web site, archived on 07:54:53 Nov 18, 2004. It is now a Federal record managed by the National Archives and Records Administration.
External links, forms, and search boxes may not function within this collection.

Survey Methodology:
National Survey of College Graduates

1. Overview top

a. Purpose

The National Survey of College Graduates is a longitudinal survey, designed to provide data on the number and characteristics of experienced individuals with education and/or employment in science or engineering (S&E) in the United States. The results of this survey are vital for educational planners within the Federal Government and in academia. The results are also used by employers in all sectors (education, industry, and the government) to understand trends in employment opportunities and salaries in S&E fields and to evaluate the effectiveness of equal opportunity efforts. This survey is designed to complement the other surveys of scientists and engineers conducted by SRS in order to provide a comprehensive picture of the number and characteristics of individuals with education and/or employment in S&E in the United States. This combined system is known as SESTAT (Scientists and Engineers Statistical Data System).

b. Respondents

In the 1993 survey respondents were individuals living in the US in the reference week of April 15th and holding a bachelor's or higher degree. In later years, the coverage of the 1995, 1997 and 1999 panel surveys are limited to those either holding one or more degrees in an S&E field or those working in an S&E occupation.

c. Key variables

2. Survey Design top

a. Target population and sample frame

The 1993 survey target population consisted of all individuals under the age 76 with at least a bachelor's degree as of the day of the 1990 decennial census, April 1, 1990. This survey was a baseline survey for the decade of the 1990s. The 1995, 1997, and 1999 panel surveys follow those individuals identified in the 1993 survey as having an S&E degree and/or an S&E occupation. However, those holding a U.S. earned PhD in an S&E field were not followed as these individuals are covered in a companion SESTAT survey—the Survey of Doctorate Recipients (SDR).

b. Sample design

The sample frame for the 1993 survey consisted of those individuals who indicated on the long form of the 1990 Census that they had at least a bachelor's degree and had a birth date consistent with their being under the age of 76 as of the 1993 survey reference date of April 1993. A highly stratified design was used along with oversampling for somewhat small groups of interest including women, minorities, the disabled, and foreign born individuals. The 1993 survey sample size was 215,000. Later survey sample sizes ranged from 60,000 to 40,000.

c. Data collection techniques

The survey is conducted for NSF by the Bureau of the Census. Initial data collection was done through the use of a self-administered mail survey using a prenotification letter, a first mailing, a reminder letter, and a second mailing.

Nonrespondents to the mail questionnaire were followed up using computer-assisted telephone interviewing (CATI). A personal visit follow-up of nonrespondents to the CATI phase was also conducted.

Information in the 1993 survey was collected for the week of April 15, 1993. Data collection took place between May 1993 and March 1994. Follow up surveys used a May through December collection schedule.

d. Estimation techniques

The base weight for each case was the inverse of the probability of sample selection. This weight was adjusted for nonresponse by multiplying by the inverse of the response rate cases. Nonresponse adjustment strata were primarily those used in sampling. In the 1993 survey, a final ratio adjustment by stratum was made using the 1990 decennial census data as the control.

3. Survey Quality Measures top

a. Sampling variability

The sample size is sufficiently large that estimates based on the total sample are subject to minimal sampling error. However, sampling error is larger with estimates of small subgroups of the population. Estimates of the sampling errors for the SESTAT system as a whole are accessible through the SESTAT web page, http://sestat.nsf.gov.

b. Coverage

Coverage errors exist because some individuals were not included in the Decennial Census. Additional coverage errors may exist because of errors in the decennial data file that led to our incorrectly classifying individuals as not having bachelor's degrees or above when in fact they held such a degree. In later research it was found that about 11 percent of those recorded as holding a bachelor's degree or higher in the decennial census did not report such a degree even after follow-up on the NSCG. This over coverage was eliminated from the NSCG through weighting.

c. Nonresponse

(1) Unit nonresponse - The unweighted response rate for the 1993 survey was 78 percent; the weighted response rate was 80 percent. Response rates for subsequent rounds of the NSCG range from 91 to 95 percent. While these are relatively high response rates, nonresponse error is a possible source of bias in this survey. In order to minimize the impact of this source of error, results are adjusted for nonresponse through the use of statistical weighting techniques.

A nonresponse study was conducted on the 1993 NSCG to determine if demographic information from the 1990 Decennial Census (the sample frame for the NSCG) could be used to reliably distinguish respondents from nonrespondents[1]. Through this study, it was determined that no strong characterizations could be made of the conditions that would allow the determination of whether a sample person is a respondent or nonrespondent. However, the study did yield other useful results. For example, it was determined that further reduction of nonresponse may be possible in the pool of individuals where the reason for nonreponse was inability to locate. An analysis of the costs by response rate indicated that there might be certain groups in the population for whom more expensive modes of data collection (CATI, personal visit) do not yield higher response rates than mail alone. Finally, it was also determined that follow-up of nonrespondents in subsequent rounds of the NSCG would not be as beneficial as only following respondents.

In 1999, a further analysis was conducted of nonresponse in the 1993 NSCG to explore the nonresponse patterns of highly educated people as compared to the overall population. Two separate interactions with the population were studied: whether or not contact was made, and whether or not cooperation was achieved. The demographic variables studied included age, gender, marital status, parental status, and race/ethnicity. Additionally, occupation and income were also studied. The results of this research may assist in the development of targeted strategies for improving contact and cooperation rates within the NSCG population, which shows different trends than the general population.

(2) Item nonresponse - In 1997, the item nonresponse rate for key items (employment status, type of employment, occupation, and primary work activity) ranged from 0.0 percent to 0.6 percent. Other variables, expecially those that ask sensitive information, had higher nonresponse rates. For example, salary and earned income had item nonresponse rates of approximately 9.0 percent. Information on other variables with high item nonresponse is available in the SESTAT web system in the technical notes section for the variable in question.

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 in the 1993-97 period. Results from this research can be found in http://sestat.nsf.gov in the Research Compendium section, and in the technical notes section for the individual variables.

4. Trend Data top

In 1993 the NSCG covered all those holding a college degree as of April 1990. Subsequent NSCGs covered only persons with S&E degrees or working in an S&E field—and excluded those holding a US-earned S&E doctorate. Because of the 1993 to 1995 large decrease in coverage, analyzing trend data for the NSCG can be problematic. A more useful source of trend data can be obtained from the SESTAT combined file.

5. Availability of Data top

a. Publications

The data from this survey are combined with other data on scientists and engineers and are published in publication series such as Science and Engineering Indicators and Women, Minorities, and Persons with Disabilities, all available on the SRS Web site. Information from this survey is also included in reports such as those on the engineering workforce.

b. Electronic access

Data from this survey are available on the SRS Web site and on SESTAT. Selected aggregate data are available in public use data files upon request.

c. Contact for more information

Additional information about this survey can be obtained by contacting:

Julia D. Oliver
Human Resources Statistics Program
Division of Science Resources Statistics
National Science Foundation
4201 Wilson Boulevard, Suite 965
Arlington, VA 22230

Phone: (703) 292-7809,
E-mail: joliver@nsf.gov


[1]See Tremblay, Antoinnette, and Moore, Thomas F. III. (1995), "Nonresponse Issues of the National Survey of College Graduates", Washington, DC: U.S. Department of Commerce, Bureau of the Census. This paper may be viewed at http://sestat.nsf.gov/research/compend.htm.

Last Modified: Apr 13, 2004 Comments to srsweb@nsf.gov