Audits & Social Experiments:
A Report Prepared for the U.S. General Accounting Office
PAD-79-1: Published: Oct 1, 1978. Publicly Released: Oct 1, 1978.
Under contract with GAO, the Social Science Research Council conducted a study on methods and techniques for auditing social experiments. Its report focused on the use of reinterviews and other alternatives for monitoring research quality, the general problem of quality assurance, and the roles which GAO might take in meeting its oversight responsibility.
There were not enough instances of the use of the reinterview to predict its impact on research participants, but there seems to be a significant risk of its disrupting research. Possible alternatives to reinterviews are the use of information on research procedures to determine quality, parallel sampling by auditors, surrogate auditors, subsample for reinterviews by auditors, and record-linking and other statistical techniques. GAO possible contributions in social experiments at various stages of the research and evaluation process were identified. The potential costs and benefits of audits were considered as applied to researchers, sponsors, and participants in social experiments. The roles of sponsors were discussed with regard to developing and implementing solicitations, the monitoring of projects, and the dissemination of the project's results. Responsibilities of the researcher were considered in terms of ensuring the quality of research.
- Review Pending
- Closed - implemented
- Closed - not implemented
Recommendation for Executive Action
Recommendation: GAO should recognize the following: (1) unplanned research, such as reinterview, can disrupt research; (2) a reinterview is often unnecessary, and alternatives can generate statistical information on the quality of data; (3) some projects will require that GAO become involved early in the research process; and (4) a small scale effort is more justified than a major in-house testing program to assay the effects of reinterview. GAO should prepare a document which clarifies its likely roles in evaluation of social experiments, and its oversight role should be coordinated with other groups responsible for ensuring quality. GAO efforts to report both deficiencies and proficiencies of social experiments were endorsed. GAO should take an active role in helping to develop guidelines on the management and budgeting of social experiments and of program evaluation.
Comments: Please call 202/512-6100 for additional information.