reported on the progress towards this goal (European Commission 2000; 2001b; 2002;2003). Recently, it became clear that the target will be missed (European Commission
2004). After this mid-term review, the 2010 time horizon was replaced by a three year cycle of planning and evaluation (European Commission 2005).
Goal displacement thus is either functional or dysfunctional. Essentially, the original formulation of the concept by Merton was about rules in bureaucratic
organizations (1968). Conformance to regulation is not seen as a measure for a specific purpose, but for becomes an immediate value in life-organization of the
bureaucrat (p. 253). We would expect that goal displacement, and the different strategies to get there, is more likely with performance systems that have the qualities
of regulation. The hypothesis claims the more a measurement system infringes on the discretion of an organization, the more goal displacement processes will show.
The reduction of discretion however is assumed a result of the way information is used and not of the mere provision of performance information. Thus a more precise
hypothesis is that diverse uses reduce discretion of organization to a dissimilar extent, and therefore bring about goal displacement strategies.
9.4. Central hypothesis: use determines effects.
Before we elaborate on the central hypothesis, we briefly repeat the three categories of use we defined above (p. 160). Three functions are identified: research,
internal management and accountability (The Working Party on Performance Monitoring in the Public Services of the Royal Statistical Society 2005).
1. First, performance information may be collected in order to find out what works? This is the research and learning function. The main rationale is indeed learning.
The rise of evidence based policy initiatives may be seen in this light (Davies, Nutley, and Smith 2003; Boaz and Nutley 2003). People are intrinsically motivated to
perform well and to seek responsibility. This also motivates them learn how to do the job. How can policy or management be improved? Policy planning and evaluation,
strategic planning and evaluation and business process re-engineering are examples of policy and management tools with a primarily research orientation. Performance
information can be used for process evaluation and outcome evaluation, which envisages respectively service improvement as well as policy improvement (Scheirer 1994).
The use of performance information for research and learning has the lowest impact on the degrees of freedom of the organization.
2. Secondly, the internal management function of performance information is about identifying and sanctioning well or underperforming institutions or public servants
(to control, to motivate) and about allocating resources (to budget). The purpose is not to learn from mistakes, but to agree upon targets. The use of performance
information for management tools such as internal performance contracts, performance based pay, performance budgeting and performance mandates fall in this category
(Bovaird and Loffler 2003; Kamensky and Morales 2005). The use for internal management may have a significant impact on the degrees of freedom within the organization.
3. The third purpose is accountability. The main propoNeed a Professional Writer to Work on this Paper and Give you Original Paper? CLICK HERE TO GET THIS PAPER WRITTENsition is that the public sector should be accountable to the citizens/taxpayers and politicians (Bolton 2003).
Therefore, performance of public bureaucracies should be disclosed to the public. Two categories of external pressure can be identified. First, pressure is assumed by
showing results to the general public. In case of a monopoly, the potential criticism of the public is expected to exert enough pressure on the organization. In case
of (quasi-) markets, rankings are supposed to provoke market pressures. Typically, league tables, citizen charters and upgraded annual reporting are examples of this
accountability role (Bowerman 1995; Gormley and Weimer 1999). Secondly, pressure is instituted by the political system. Performance contracts with agencies are an
example. These contracts give autonomy to agencies provided the agency commits itself to output or outcome targets. The use for accountability usually has a high
impact on the degrees of freedom of the organization.
Figure 37 represents the central hypothesis. Utilization of performance information determines effects. The more performance information is used for accountability,
management and research, the more effects will show. The function starts at the origin. No utilization will not cause changes in organizational outputs and behavior.
Yet, this may not always be the case. Organizations that are not measuring may experience the effects of measurement. This may occur when peer organizations are
measuring and they are subjected to some pressure to align to their peers. In that case, the function would not start in the origin
intensity of effects
ACC + MGMT + RES
Figure 37: Central hypothesis of measurement effects: utilization determines effects
The function is hypothesized to be exponential. This is because it adds up the three functions for the different categories of use. The hypothesis is that with
increased utilization, the effects will also increase. Yet, the alteration (delta) differs for categories for use. Use for research and learning will cause fewer
effects than use for management or accountability.
intensity of effects
Figure 38: Central hypothesis of measurement effects: breakdown for categories of use
This then sets the following hypotheses. First, high use will lead to high effects. Given our methodology (interview-setting), this hypothesis states that
organizations that are measuring more will perceive more effects of measurement. Second, high utilization for accountability will lead to high effects, no matter what
the utilization for research or management is. The different alteration (delta) of the three functions implies that effects of accountability overtake effects of other
uses in case of high use for accountability. Secondly, high utilization on management, but low on accountability will lead to moderate effects, no matter what the
utilization for research is. Thirdly, high use for research will lead to low effects, under the condition that use for management and accountability is low.
H1. high use will lead to high effects: organizations that are measuring more will attach a higher sense of reality to the effects.
H2 high use for accountability will lead to the highest effects, no matter what the utilization for research or management is.
H3 high use on management, but low on accountability will lead to moderate effects, no matter what the utilization for research is.
H4 high use for research will lead to low effects, under the condition that use for management and accountability is low.
The underlying assumption about the relation between the categories of use is that high impact use drives out low impact use. Impact in this case is the impact on the
degrees of freedom of the organization. This is a common observation in social psychology. Intrinsic motivation is likely to be undermined by extrinsic motivation
(Deci 1975; Lepper and Greene 1978). Research and learning are
built upon intrinsic motivation. Internal management control and accountability are based on extrinsic motivation. Therefore, the latter uses may override the former.
Figure 39 gives an oversight of the research process. The emphasis lies on twelve semi structured interviews. We chose face-to-face interviews for two reasons. First,
it complements the survey data. Secondly, and more importantly, the research issue – the effects of performance information – is not suited for survey or document
analysis. Perverse and negative effects in particular are prone to social desirable answers. Therefore, a survey would yield deceptive results. Equally, these perverse
effects are not found in documents, since organizations will precisely attempt to avoid publication of these mechanisms. Exception might be documents of external
actors such as audit institutions or the media. However, audit is in the Ministry of the Flemish community in its infancy. The problem with the media is its selective
attention. They usually focus on some policy sectors more than others and they overweigh the negative over the positive. Therefore, we opted for face-to-face
interviews with the managers of the sections.
sectoral and departmental spread and concentration
“user profiles” based on survey data
Semi structured interviews
triangulation of “user profile”
likelihood of effects “effect profile”
Linking user profiles with effects profiles
Figure 39: Overview of the methodology and data
The twelve sections of the Ministry of the Flemish Community were selected based on two criteria. First, we required combination of concentration and spread over
policy sectors and departments. This should allow for comparison within and between sectors and departments. This also allows for us to
keep constant the characteristics of the department of which the sections are a part. Yet, we do not lose the possibility to compare across departments82.
Secondly, we selected sections that have moderate to high experience with performance measurement. Obviously, this is a necessity for studying the effects of
performance measurement. We determined the experience with performance measurement by using the survey of 138 sections of the Ministry of the Flemish Community (see p.
121). In order to differentiate between the uses of performance measurement, we selected sections with different user profiles. The user profiles are based on the
categories of use.
In order to establish the initial user profiles, we used survey data. Based on indications of use, scores on three categories of use were calculated. The indications
are represented in Table 37. The calculation is straight. Every indication gets a standardized score, which is 1 or 0 (present not present), or the score on a scale
divided by the maximum value of the scale. Next, we add up the scores and divide the sum by the total indications for the category (4). In case of ordinal scales,
missing values were replaced by the series mean. In case of nominal (1/0) missing values are set to be 0. We assume that in general if organizations do have for
instance self-assessment or indicators in an annual report, they will report on that. If they do not have it, they may not answer the question for reasons of social
Latest completed orders:
|#||Title||Academic Level||Subject Area||# of Pages||Paper Urgency|