Policy Data as a Social Process: A Qualitative Approach to Quantitative Data

Abstract

E VALUATION AND SOCIAL POLICY research have existed for a long time but have proliferated in the last two decades.' Increasingly, social scientists are involved in assessing the impact of social welfare programs such as income maintenance, manpower development, deinstitutionalization, preschool intervention, and mainstreaming handicapped youngsters, as well as studying the effects of particular treatment modes (behavior modification, halfway houses, methadone maintenance, parole). In addition, they are involved in assessing the extent and nature of such public concerns as child abuse, violence in schools, handicapped people, and substance abuse, as well as the services available to combat them. The major thrust of such research is based on positivist assumptions and quantitative a n a l y s i ~ . ~ For program evaluation, this means developing operational definitions or program goals (increased I.Q., abstention from drugs, employment), and measuring success against the yardstick of these desired outcomes. In policy research where investigators are assessing the extent and nature of a particular social problem, there is reliance on compilations from official statistics, but data are also generated from random samples of the population or the choice of particular geographic areas to survey, and/or consultation of professionals for estimates of the extent of the problem and the number of people being served. From time to time, social scientists are reminded of the meager reliability of data produced by official agencies (Merton 1957; Seidman and Couzens 1974; Kitsuse and Circourel 1963). For the most part, these critiques have been understood by those who use and create the data as a call to be more methodical, as if the problem were inaccurate counting. In the

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 51 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.