II The Soviet Union
III Student evaluation of teaching in the U.S.
IV Job training in the U.S.
V Aircraft baggage handling
Peter Smith, in his 1993 article “Outcome-related Performance Indicators and Organizational Control in the Public Sector” identifies seven dysfunctional consequences induced by an ill- conceived or poorly implemented performance measurement system.
The first category proposed by Smith (1993) named ‘tunnel vision’ emphasises the narrow focus on areas for which the performance measurement is designed, leaving other important areas without further attention. Second, ‘suboptimization’ whereby managers tend to implement their own objectives rather than engaging in necessary strategic coordination to achieve the organisation’s objectives. ‘Myopia’ is the term for concentration on short-term issues instead of a long-term strategy. Fourth, the term ‘convergence’ is used synonymously for behaviour characterised by the desire to be outstanding as opposed to being exposed to the performance measurement process. The term ‘ossification’ expresses the mental reluctance to adopt or implement new and innovative ideas possibly yielding advantages. Sixth, ‘gaming’ which stands for a modification of behaviour to gain strategic advantage. Finally, a ‘misrepresentation’ of facts and results including fraud and ‘creative’ accounting is recognised.
The previously defined categories of dysfunctional behaviour were introduced in the context of the National Health Sector in the UK, but can nevertheless be easily extended to other sectors. The following paragraphs provide examples of dysfunctional behaviour in another context but NHS which will be subordinated under the seven headings proposed by Smith (1993).
II. The Soviet Union
The relevance of this historical case is given by the ideology formerly driving the Soviet Union. Evidence for three types of dysfunctional behaviour - ‘tunnel vision’, ‘misrepresentation’ and ‘myopia’ - can be provided.
The Soviet Union featured a system in which production was based on an incentive system that allowed 30 to 100% increase in salary for fulfilling the factory’s output plan and additional 2 to 10% for ‘over-fulfilment’, (Berliner, 1988). Output targets were set unreasonably high and constantly raised if successfully achieved. Therefore, a strong incentive for underperformance, in order avoid a subsequent rise in output level imposed on production, was embedded in the system, (Estok, 1999; Lukka, 1988; Briers and Hirst, 1990).
Performance measures concentrated on output levels, leaving other criteria such as quality aside, and can therefore be seen as a too narrow in focus (‘tunnel vision’) as only one key area is taken into consideration. Managers operating under such conditions took on a shortterm perspective (‘myopia’) as the system provided little or no incentives for long-run management. The achievement of monthly targets was all that mattered, (Berliner, 1988).
In addition, stretching accounting rules (‘misrepresentation’) by exaggerating output levels was quite commonly made use of. This was possible as the USSR did not comply with International Accounting Standards, (Horngren, 1999). If the prespecified output was not reached, it was simply ‘borrowed’ from the next month, (Berliner, 1988). Managers showed a tendency towards underreporting their firm’s capacity, in hope for lower production targets i.e. deliberately false information about resources was communicated, (Berliner, 1988).
Criticisms on the system employed in the Soviet Union abound. Only one single measurement criteria was administered and emphasised on; while the process and behaviour were of no interest. This resulted in output being overstressed. Additional criteria would have been necessary to provide direction in this highly incentive-based system. Multiple dimensions of performance should be rewarded in form of a balanced set of performance dimensions balancing task and social, internal and external, and short-term and long-term activities, (Landy et al. 1983). However, there is also a danger in over-designed reward systems as they often work too well and therefore generate undesired outcomes, (Landy et al. 1983).
III. Student evaluation of teaching in the U.S.
An explicit case of ‘gaming’ occurred in the business department of the Texas A&M University as a female instructor known for being a relatively strict marker gave only D’s and F’s in a basic business class and therefore received a SET-score in the one range (of a five- point scale, with five being the best and one the worst result). As a consequence she was removed from that class and placed in a non-required, graduate course. She then decided upon giving an equal split of As and Bs, announcing this new ‘assessment policy’ to students at the beginning of the semester; her SET-score level off at 4.9. Crumbley (1995) assumes that an instructor, provided that he or she can choose teaching styles, grade difficulty, and course content, will maximise SET-score through them i.e. a that persons is more likely if assessed with this measure to try to maximise it rather than working towards the overarching goal of ‘educated’ and knowledgeable students. This is in line with Medley (1979) who argues that teachers try to meet criteria they are assessed with. Worthington and Wong (1979) additionally identified dysfunctional consequences in this context such as ‘antilearning’ (resulting in grade inflation and coursework deflation) and ‘pander-pollution behaviour' i.e. actions taken by the instructor to intervene inside and outside the classroom with regards to the maximisation of the SET-score.
Since single numerical measures do not have the capacity to accurately mirror all aspects of a construct (Newton, 1988) SET should be used in combination with other measurement criteria like classroom visitations, self-appraisals, video machines, recorded material, and evaluation of teaching material for example, (Crumbley 1995). SET could be used as valuable feedback about teaching but should not be the main criterion in deciding upon wages and employment, (Landy and Farr, 1983). Cahn (1986) gives cause for thought whether those having the consequences of professional malfeasance (the students) have the required knowledge and farsightedness to make judgement about the quality of teaching they undergo. In line with Crumbley (1995) I believe that in the long run little or even uneducated students will have serious consequences for universities as the society will find new ways to recruit knowledgeable employees. A dilution of the grading system is likely if marks do not accurately reflect knowledge.
IV. Job training in the U.S.
A large-scale job training program for economically disadvantaged in the U.S was administered with the mission to raise the earnings ability and to lower the welfare dependency of the poor. Managers were measured on the time required to ‘produce’ a job placement which was found to be a short run perspective (‘myopia’), in the long-run training was considered to be more important. In 1982 performance measurement was changed as Job Training Partnership Act came in force. Managers were then evaluated according to client’s labour market success at the end of their training.
The recognition of a dysfunctional outcome (‘myopia’) and the endeavour of eliminating its cause induced another dysfunction (‘gaming’). The system offers room for ‘gaming’ as managers could be tempted to only enrol participants likely to perform well i.e. the most ‘job- ready’ ones regardless of how much they might gain from the program, (Courty and Marschke, 2003). The attempt to assess managers seems inappropriate as the success of clients might be linked to other extraneous factors which go beyond the control of the manager. A more complex performance measure (including mediation attempts, successful job placements, average time of client working for the proposed company, company’s satisfaction with new employee) has to be employed in order to capture the diverse nature of this process.
- Quote paper
- MMag. Stefanie Gapp (Author), 2004, Dysfunctional consequences of an ill-conceived or poorly implemented performance measurement system, Munich, GRIN Verlag, https://www.grin.com/document/149434