Lawrence B. Mohr

Determinants of Innovation in Organizations

The present study is an attempt to identify the determinants of innovation in public agencies, i.e., the degree to which they adopt and emphasize programs that depart from traditional concerns. Innovation is suggested to be the function of an interaction among the motivation to innovate, the strength of obstacles against innovation, and the availability of resources for overcoming such obstacles.The significance of the research can be viewed in terms of Hynemans observation nearly twenty years ago that bureaucratic agencies “… may fail to take the initiative and supply the leadership that is required of them in view of their relation to particular sectors of public affairs. …†His concern was the responsiveness of the public sector not only to expressed wants but to public wants that may go unexpressed, or be only weakly expressed, and whose utility is much more easily recognized by the informed bureaucratic official than by the ordinary citizen.While the results and conclusions to be reported appear to be largely valid for organizations in general, the empirical focus will be local departments of public health which, as a class, have had a rather dramatic succession of opportunities to respond to new public problems over the past twenty-five years. A brief introductory paragraph will orient the reader to the applied setting.

Organizational Technology and Organizational Structure.

Contrary to impressions that might be obtained from the literature on technology and structure, there is not a great deal of reliable evidence that the social structure of organizations is strongly affected by technology. The present research finds a very weak relationship between technological manageability and subordinate participation in decision making. When the conceptualization of technology is expanded to include more than manageability, a moderate relationship with participativeness emerges. No support is found for the hypothesis that the effectiveness of an organization is determined by the consonance between its technology and its social structure. Technology and structure are both multidimensional concepts that cannot be expected to be related in a simple manner. Future research on the determinants of most dimensions of social structure in organizations should probably emphasize independent variables other than technological characteristics.

The Concept of Organizational Goal

The organizational goal concept is important for significant types of organizational research but its utility has been downgraded in recent scholarship. This paper reviews critically key contributions to conceptualizing the organizational goal and synthesizes many of their elements into a more concrete and comprehensive conceptualization. The efforts of Etzioni, Seashore and Yuchtman, Simon, and Thompson to bypass the need for a goal concept in evaluative and other behavioral research are unconvincing in important respects. However, they are persuasive in underscoring the importance of viewing organizational goals as multiple and as empirically determined. Perrow, Gross, and others convincingly suggest a dual conceptualization, so that goals are dichotomized into those with external referents (transitive goals) and those with internal referents (reflexive goals). Deniston et al. contribute the desirability of subsetting the goals of organizations into “program goals” and of differentiating goals from both subgoals and activities. The existence and relative importance of organizational goals and an allied concept, “operative goals,” may be operationally determined by current social science methods. The goal concept as presented here has implications for the evaluation of organizational effectiveness, for research on organizational behavior, for organization theory, and for views of the role of organizations in society.

Developing Public Policy Theory: Perspectives from Empirical Research

There has been considerable interest in the development of theories of public policy formation, but theoretical efforts to date have not demonstrated adequate recognition of the distinctive qualities of the dependent variable as a focus of research. Facets of public policy are far more difficult to study systematically than most other phenomena investigated empirically by political scientists. Our attempt to test hypotheses with some rigor demonstrated that public policy becomes troublesome as a research focus because of inherent complexity-specifically because of the temporal nature of the process, the multiplicity of participants and of policy provisions, and the contingent nature of theoretical effects. We use examples of policy making taken from the case study literature to show concretely how such complexity makes it essentially impossible to test apparently significant hypotheses as they are presented by Lowi, Dahl, Banfield, and others. Our effort here is to enhance theoretical development by carefully specifying and clarifying the major shortcomings and pointing out the apparent directions of remedy.

The Qualitative Method of Impact Analysis

Consider the qualitative approach to evaluation design (as opposed to measurement) to be typified by a case study with a sample of just one. Although there have certainly been elaborate and emphatic defenses of the qualitative approach to program evaluation, such defenses rarely attempt to qualify the approach explicitly and rigorously as a method of impact analysis. The present paper makes that attempt. The problem with seeking to advance a qualitative method of impact analysis is that impact is a matter of causation and a non-quantitative approach to design is apparently not well suited to the task of establishing causal relations. The root of the difficulty is located in the counterfactual definition of causality, which is our only broadly accepted formal definition of causality for social science. It is not, however, the only definition we use informally. Another definition, labeled “physical causality,” is widely used in practice and has recently been formalized. Physical causality can be applied to the present problem. For example, it explains the persuasiveness of Striven’s “Modus Operandi” approach tailored case study design with a sample size of one in principle as strong a basis for making inferences program impact as a randomized experiment. Crucial program evaluation finding that people’s “operative reasons” for doing what they do are the physical actions. it is shown that external validity using this qualitative approach would have exceptional strengths.

The Causes of Human Behavior: Implications for Theory and Method in the Social Sciences

Acknowledging that though the disciplines are supposed to be cumulative, there is little in the way of accumulated, general theory, this work opens a dialogue about the appropriate means and ends of social research based in analysis of fundamental issues.This book examines two root issues in the methodology of explanatory social research--the meaning of the idea of causation in social science and the question of the physiological mechanism that generates intentional behavior. Conclusions on these as well as on several derived problems emerge through the analysis. Among the latter, the analysis shows that neither universal nor probabilistic laws governing human behavior are possible, even within the positivist or empiricist traditions in which laws are a central feature. Instead, the analysis reveals a more modest view of what an explanatory social theory can be and do. In this view, the kind of theory that can be produced is basically the same in form and content across quantitative and qualitative research approaches, and similarly across different disciplines. The two streams of analysis are combined with resulting implications for large-sample, small-sample, and case study research design as well as for laws and theory.Written for the practicing empirical researcher in political science and organization theory, whether quantitative or qualitative, the major issuesand findings are meant to hold identically, however, for history, sociology, and other social science disciplines.Lawrence B. Mohr is Professor of Political Science and Public Policy, University of Michigan.

Analysis of No-Difference Findings in Evaluation Research

Conclusions of no difference are becoming increasingly important in evaluation research. We delineate three major uses of no-difference findings and analyze their meanings. (1) No-differ ence findings in randomized experiments can be interpreted as support for conclusions of the absence of a meaningful treatment effect, but only if the proper analytic methods are used. (2) Statistically based conclusions in quasi-experiments do not allow causal statements about the treatment impact but do provide a metric to judge the size of the resulting difference. (3) Using no-difference findings to conclude equivalence on control variables is inefficient and potentially misleading. The final section of the article presents alternative methods by which conclusions of no difference may be supported when applicable. These methods include the use of arbitrarily high alpha levels, interval estimation, and power analysis.

Authority and Democracy in Organizations

Participativeness as a style of supervision is considered as a dependent variable. The data do not directly support the initial hypotheses in any important way, but severalfindings of interest emerge indirectly from a more detailed, multivariate analysis. First, supervisory behavior in these terms is probably influenced significantly by affect between supervisor and subordinates. Second, when affect is not a factor, supervisory style is more rational, with participativeness depending on the training of the subordinates and their perceived capacity to contribute constructively to decision-making. Third, the measurement of participativeness in research affects theoretical development in an important way, since the technical and professional level of subordinates probably has a significant effect upon their objective level of participation but not upon the degree of participation as reported by them. Fourth, the latter finding is significant for the much more common body of research in which participation is treated as an independent variable; it may help to clear up many puzzling and inconsistent findings in that tradition. Lastly, the data analysis suggests a new normative and descriptive conceptualization of democracy in the workplace, which is offered in conclusion as a competitor of the power-equalization approach.

Regression artifacts and other customs of dubious desert

Abstract This paper was written as a belated response to an invitation by Campbell to debate the subject of regression artifacts and related topics on which we apparently differed. The thesis is that there appears to be a certain tendency in social science and program evaluation to adhere to some methodological practices by force of custom or habit rather than because of their reasoned applicability. These include the ideas of regression artifacts, random measurement error, and change or gain scores. A defense of this position in regard to all three ideas is presented.

The Impact Profile Approach to Policy Merit The Case of Research Grants and the University

In writings on the theory of valuing, many take the position that impacts on the relevant outcome dimensions should be aggregated to arrive at one summary assessment of program merit. A contrary position is taken here, specifying that the impacts should be kept separate and unweighted and expressed only in their own original measurement scales. All impacts, however, should be portrayed, including those for which no rigorous data analysis has been carried out. It is argued that aggregating, even by the individual stakeholder, is both futile and misleading. An extensive evaluation of the effects of research grants on the university is included as a full-scale illustration of the method. Financial impacts are considered, as well as impacts on faculty and student quality, on university prestige, and on the quality of instruction.