A similar study was carried out by Melvin Kohn in the 1960s and published in 1971. The data in this research came from answers to interview questions rather than from the interpretation of TAT stories, and the subjects were drawn from a somewhat different sample--about three thousand men employed in a variety of civilian occupations. The degree to which their employing organization was bureaucratic was scored by counting the number of hierarchical levels in it. The results suggest that men employed in the more bureaucratic organizations were more likely to display intellectual flexibility and value self-direction and new experiences than was the case of men employed in less bureaucratic organizations. One reason for this finding may be that the employees of bureaucratic organizations tended to have more schooling than those of other kinds of organizations, and education (or the individual traits that correlate with education, such as intelligence) may produce those qualities detected in the survey.54
Charles Goodsell has reviewed a number of studies of the same sort as Kohn's, and most produce the same results.55 There are exceptions. An experiment in which business managers and public-school administrators played a game involving choices between big payoffs with high risk and low payoffs with small risk revealed that the businessmen were more likely to accept risk than the school officials.56 Given the greater rewards in a successful business career and the greater security in many (but not all) government careers, it would be surprising if there were no differences in the kinds of people found in each. But there is as yet no strong, consistent evidence that the two careers attract different personalities.
Prior experiences and professional norms (and their attendant career opportunities) certainly influence the behavior of rank-and-file bureaucrats; political ideology may have an effect, but we do not have enough evidence to speak confidently about this. Experience, professionalism, and ideology are likely to have their greatest influence when laws, rules, and circumstances do not precisely define operator tasks.
For decades, the number of agencies conferring broad discretion on operators steadily increased. Driven by an optimistic belief in the ability of nonpartisan experts effectively to manage complex economic affairs, Congress told many regulatory agencies with respect to some problem to do whatever would serve "the public interest." In these cases, the powers of the state were turned over, at least initially, to appointed officials who could make choices based on their beliefs and professional norms. Moreover, more professionals were hired to staff the government partly because