Researchers have long lamented the elusiveness of developing performance appraisal systems that accurately measure the construct of work performance. Numerous methodologies have been suggested in order to facilitate accuracy in performance ratings. One of the most promising methods to emerge for improving the accuracy of performance ratings is Frame-of-Reference (FOR) training (Bernardin and Buckley, 19811. FOR training establishes a common conceptualization of performance among raters that would serve as a basis for observing and evaluating performance. Establishing a common frame of reference among raters reduces their idiosyncratic rating tendencies. FOR training improves accuracy imparting a shared understanding of performance dimensions and standards for evaluating behavior relevant to these standards--it creates common rater performance expectations. This is accomplished providing the raters with the rating standards as well as examples (critical incidents) of these dimensions. FOR training typically consists of four components: lecture on the multidimensional nature of job performance and the associated problems with rating performance, opportunities to practice, discussion of rating issues, and feedback concerning rating accuracy. FOR training is one of the most effective methods for improving rating accuracy in comparison to other rater training programs (Fletcher and Perry,, 2001). Several studies have reported that FOR training results in more accurate performance ratings than other methods of rater training (e.g., Athey and McIntyre, 1987; Bernardin el al., 2001; Pulakos, 1984, 1986; Smith, 1986; Stamoulis and Hauenstein. 1993; Woehr, 1994). However, Bernardin et al. (2001) summarized that FOR training has yet to deliver these results in field settings, which leads to questions about the generalizability of FOR training in organizational/field settings. In the present article, we discuss the process of FOR training, including the basis for its effectiveness and its applications. Furthermore, we propose a change in the FOR training protocol that increases FOR training's relevance to organizational settings, and we test dais new protocol in a naturalistic setting that allows for random assignment to conditions: a classroom. Fletcher and Perry (2001), in summarizing recent performance appraisal research, suggest that rater training programs often fail to address the influence of the social or environmental context in which performance appraisals take place. As we intend to experimentally test a new FOR training protocol in comparison to a traditional FOR training protocol and a no training control group, it is important to conduct the experiment in a social context. Why FOR Training is More Effective Several studies have attempted to explain why FOR training is more effective than other forms of rater training. One suggestion is that FOR-trained participants are able to remember more of the content presented during the training than those trained other procedures. Athey and McIntyre maintained that FOR participants establish a mental connection between performance dimensions and behavior by essentially rote-memorizing specific dimension-behavioral pairings (1987: 568). The main conclusion of this study is that retention and recall of the training content improves as a result of FOR training. Sulsky and Day (1992) posited that FOR training provides raters with examples of behaviors representing effective and ineffective performance on different dimensions. These examples comprise prototypes that represent the most typical members of a category, (e.g. good employee). This 'allows performance information to be categorized according to its resemblance with the prototypes (Feldman, 1981). Utilization of these prototypes increases accuracy because the raters are able to correctly categorize the performance of each ratee. Woehr (1994) investigated the impact of FOR training on the recall of information about ratees' performance. …
Stored only when both papers are in the corpus — 14 citations exist upstream.