Many managers devote substantial time to the performance appraisal process and, in most instances, arrange formal meetings with employees to give them feedback (Eichel and Bender, 1984). Research has consistently shown that the receipt of feedback is essential for employees to improve their performance (Ashford and Cummings, 1983; Greller, 1980; Ilgen et al., 1979). And more broadly, the overall process of setting goals, measuring performance, and providing feedback is vital as a form of management control (Snell, 1992). Several writers have argued that, given the sometimes delicate nature of the performance appraisal interview (PAI), different approaches may be more or less suitable under different circumstances (see Ilgen et al., 1993, for an overview). Maier (1958), for example, suggested that managers should design and administer the PAI based on the purpose of the interview as well as the characteristics of the participants. Similarly, Cummings and Schwab (1973) argued that the PAI process should be tailored to a particular employee with full consideration of the function, frequency and type of feedback to be given. Although contingency approaches such as these good intuitive sense, they rest on the following three assumptions: (1) there is no one best way to conduct a PAI, (2) all ways of conducting the interview are not equally effective, and (3) determining the most appropriate (i.e., best) PAI processes depends upon factors which define the interview context. Despite the prevalence of the PAI, coupled with its documented importance, it is somewhat surprising to note that empirical studies that actually test these propositions are conspicuously absent (Cederblom, 1982). Without more solid evidence, it is difficult to place much confidence in prescriptive models. The purpose of this research was to explore some of the variables (and their combinations) that comprise a contingency approach to the PAI. Much of this work is based on an earlier paper by Klein, Snell and Wexley (1987) that laid out an open systems model of the PAI that included: (1) inputs--forces that define the context in which the PAI is conducted, (2) throughputs--processes that actually transpire during the PAI, and (3) outputs--employees attitudes and performance subsequent to the PAI. From our point of view, it is not simply one or two of these variables that make or break a PAI, but the interactive composition of factors that ultimately leads to a successful or unsuccessful experience. DEVELOPMENT OF A THEORETICAL MODEL Past research on the PAI has almost exclusively examined direct univariate effects--for example, the value of criticism (Kay et al., 1965) or the value of goal setting (Burke et al., 1978). A contingency perspective would imply that the value of any PAI processes, such as criticism or goal setting, depends on judicious attention to when, why and how such a practice would be used. Performance Appraisal Interview Processes Several studies have shown that the way in which the PAI is conducted can significantly influence subsequent employee reactions (See Cederblom, 1982; Klein et al., 1987; Goodson and McGee, 1991, for complete reviews). There are, however, numerous inconsistencies across studies. For example, goal setting has been found to result in greater motivation to improve by some researchers (Burke, et al., 1978; Burke and Wilcox, 1969; Nemeroff and Cosentino, 1979) but not by others (Dipboye and Pontbriand, 1981; Ivancevich, 1982). Similarly, while Kay, Meyer, and French (1965) found the use of criticism to be related to dissatisfaction with the PAI, Nemerroff and Wexley (1979) did not find criticism to be related to any PAI outcomes. Finally, regarding participation in the PAI, Burke and Wilcox (1969) and others (Burke et al., 1978; Dipboye and Pontbriand, 1981; Greller, 1975; Nemeroff and Wexley, 1979) concluded that participation was positively related to employee satisfaction with the PAI. …
Stored only when both papers are in the corpus — 24 citations exist upstream.