L'évaluation de la formation en entreprise selon le modèle, de Donald L. Kirkpatrick: Un regard critique

In-house training has become an aspect of human resource management that today's companies can no longer neglect. Over the past 2 years, business and industry have invested considerable sums of time and money in developing the skills of their personnel. However, training evaluation efforts have evolved little and have for the most part lacked rigor. A literature review reveals that the majority of companies have favored an approach to evaluating training results that is based on the four step model first introduced by D.L. Kirkpatrick. This article points out some of the limitations of each step in the model and proposes solutions.

Effectiveness of UI Non-Compliance Detection

This article reports on the different methods of investigation employed to detect noncompliance with UI regulations, and their relative effectiveness. It considers a previous evaluation of the Unemployment Insurance Program for which a national random sample of UI claimants was selected from the active claimload and referred to Investigation and Control Officers for in-depth enhanced investigations. Results of these investigations, along with details of the specific control and investigation activities, were reported for analysis. A variety of investigation methods were generally combined, which meant that these were inter-correlated, making it difficult to apply standard regression analysis. Instead, principal components method is used for grouping of variables, and these are then entered into a logit regression model.

La mesure des résultats dans le domaine de la réadaptation

There are many obstacles to introducing outcome measures into the physical impairment rehabilitation sector. The developmental research report in this article discusses and proposes solutions for three of the problems identified. The first concerns a lack of consensus on the definitions of some of the concepts used as criteria for treatment success. The second pertains to the choice of performance indications for the chosen success criteria, and the third is methodological. Two theoretical models, measurement instruments, and methodological considerations are proposed as solutions.

Effects of a neutral answer choice on the reliability and validity of attitude and opinion items

The objective of this study is to determine whether a middle alternative in the response choices to a questionnaire influences the reliability and validity of survey responses. A 32-item questionnaire regarding maternity care was administered to a sample of 1,390 persons consisting of 597 physicians, 723 nurses, and 70 midwives. The sample was randomly divided into two groups, one receiving a questionnaire with five answer choices, including a middle option, and the other receiving a similar questionnaire but without the middle option. Results showed that including a neutral option had little effect on overall reliability and validity; however, it did allow better psychometric coherence when the items were considered globally. The effect of the middle option also depends on the opinions the surveyed persons hold about the study issue.

Ethical and methodologucal improvements to mail survey research: an alternative follow-up method

Follow-up contacts increase response rates to mail surveys substantially. However, for individuals who do not wish to participate such repeated contacts are intrusive, creating an ethical concern. An alternative follow-up procedure was developed that allows contacted individuals to choose not to participate by returning a “Reasons for Not Responding” (RNR) form. This study was designed to compare the RNR method with a traditional follow-up procedure, the Total Design method (TDM). Specifically, a two (follow-up method) by two (early versus late return) between-subjects design was used to examine satisfaction with and quality of responses to a survey mailed to 300 randomly selected households. The results showed that respondents in the RNR conditions were more satisfied with the research than respondents in the TDM conditions, particularly after repeated contacts. Further, although response quality was equivalent for the two methods in the early returns, it was lower in the TDM group, but not in the RNR group, after repeated contacts. These results suggest that mail survey researchers should seriously consider using the RNR method, especially because (a) the overall response rate to the two methods was very similar, (b) the nonrespondents frequently indicated on their RNR forms that they had made an informed decision not to participate, and (c) the RNR form allows researchers to collect information on the demographic characteristics of nonrespondents.

Au-delà de mdèles «rationnel» et «pluraliste» dans l'analyse des politiques publiques: le cas de la CTMSP

Most criticisms of the summative/experimental model of program evaluation, like those of the "technocratic" model of public policy making, assume that the deficiencies of these models can be corrected by adopting a more "pluralistic" view of the policy process (Elmore, 1976, 1978; Lindblom, 1965; Monnier, 1992; Pressman & Wildavsky, 1979). Though partially accepting the validity of this point of view, the author considers the pluralist perspective not in itself sufficient to correct the deficiencies of the summative and technocratic models. Instead, he proposes a "public policy communities" view of the policy process. This approach is illustrated by examining the emergence of a new health care policy in Quebec, the extended care needs classification system [Classification par types en milieu de soins prolongés (CTMSP)].

Emancipatory Politics, Critical Evaluation And Government Policy

The aspirations of emancipatory or empowerment-based social interventions, such as those inscribed in Health Canada's and the World Health Organization's Ottawa Charter, require a significant reconstruction of traditional evaluation practices. The prevailing response to this need has been the inclusion of "empowering" research techniques in the evaluation activity. The author argues that empowering research techniques cannot, in and of themselves, effectively inform and support the political aspirations of emancipatory intervention, such aspirations instead demand a dramatic shift in the institutional role of evaluation research and the focus of the evaluative gaze. Drawing from elements of Habermas's theory of communicative action, the author outlines the dimensions of a critical approach to evaluation that is more consistent with the principles under-scoring emancipatory intervention. The Treasury Board's guide to program evaluation is used as illustration, to demonstrate the implications of such an approach for conventional evaluation questions.

A General Theory of Delegation, Accountability and Empowerment

Delegation, accountability, and empowerment are parts of' a system aimed at high performance. This article presents definitions and purposes of these terms, describes the nature of their interrelationships and impacts on performance, and comments on how advanced program evaluation methods can render delegation and accountability more effective and empowerment less risky.