Yet it should be stressed that explicit theory, quantitative data, and careful, rigorous analysis have not been as frequently utilized in studying policy as would be possible or desirable. Thus, political scientist Marver H. Bernstein's hoary contention that regulatory agencies pass through a four-stage life cycle (gestation and birth, youth, maturity, and old age), frequently culminating in their “capture" (which is not well-specified) by the regulated groups, is often cited as though it were a clearly supported phenomenon.45 Bernstein provides impressionistic support but by no means strong proof for his life-cycle theory. (He does not follow a single commission through all of the stages of the cycle.) It still lacks systematic empirical support. Conventional wisdom of this sort frequently rests on a rather frail intellectual foundation. Another example, also in the regulatory area, is economist George Stigler's theory of economic regulation. It holds that, as a rule, regulation is sought by the affected industry and operated for its benefit. 46 This theory will not do much to explain a raft of consumer protection, industrial health and safety, and environmental programs, or the deregulation legislation of the late 1970s and early 1980s
Many perceptive and informative studies of policy formation employ little or no statistical analysis. Examples are Charles O. Jones's Clean Air, Barbara J. Nelson's Making an Issue of Child Abuse, I. M. Destler's American trade politics, and Adam Sheingate's The Rise of the Agricultural Welfare State. 47 The quality of intellectual analysis and careful use of sound data (or information) are more important than whether and to what extent quantitative analysis is employed when it comes to determining the worth of a study.
To be rigorous, analysis does not have to be quantitative, and not all quantitative analysis is rigorous. Those who use quantitative techniques have been known to quarrel with enthusiasm and even some rancor over the reliability or appropriateness of their techniques, the quality of their data, and the validity of their findings. (In the chapter titled "The Policy-Makers and Their Environment," there is a discussion on whether socioeconomic or political variables better explain policy.) Also, to be fair-minded, one should avoid developing a phobia for quantitative or statistical analysis, as some did in reaction to the behavioral movement in political science. Much can be learned through quantitative analysis.