earticle

영어

Organizations increasingly use team structures to produce products, to solve the problems, to invent innovative ideas, and so on. So performance drivers have shifted from a reliance on individuals to teams which are performing collaborative tasks. An essential decision relating to the design of teams involves how these teams will be managed. In particular, teams are self managing, they monitor and manage their own performance process. The utilization of work teams requires that organizations develop performance appraisal methods that are useful for evaluating both individual and team performance, especially in a self-managing context. Performance appraisals have been controversial in both research and practices over the past two decades. Many studies have called for expanding inputs for employee performance appraisals from traditional supervisor ratings to multiple ratings such as peer evaluation, 30-degree appraisal, and multi-rater appraisal. All of these methods involve collecting information from a variety of sources knowledgeable about the performance of the individual being rated. The present study investigate peer rating, one of the most commonly used sources of performance rating other than those of supervisors. In particular, because team members work in close proximity,they are able to observe each otherʼs task behaviors and interpersonal behaviors, making them possibly the single best informed source for performance evaluations. On the whole, both field and laboratory studies indicate that peer assessment is a valid and reliable evaluation procedure, but on average,peer rating are not usually accurate. One interpretation of the lack of rating accuracy is that raters may be capable of accurate peer ratings, but they may be unwilling to provide them. Some researchers name this rater errors as adaptive responses in performance appraisal context. We investigate whether an individualʼs relative performance contribution in work team, group satisfaction,and interaction frequency served as a basis for anchoring and adjusting the performance assessment of peers as adaptive responses. Hypotheses are proposed concerning how individualʼs relative performance contribution, group satisfaction,and interaction frequency affects peer rating accuracy defined as discriminating rating tendency. Data from 89 self-managed teams consisting of 445 undergraduate students are analyzed for hypotheses testing. We find strong evidence that the outstanding contributors are more likely to have discriminating rating tendency. Results also show that group satisfaction is related negatively on discriminating rating tendency, but interaction frequency is not significantly related to the rating accuracy. It reveals that outstanding contributors and who are more satisfied with team members were more discriminating evaluators. It could be interpreted that the level of a raterʼ relative performance contribution and the level of satisfaction might affect the peer rating accuracy in a sense. Implications for peer appraisal practices and future research were discussed.

한국어

성과평가는 구성원의 동기 자극 및 조직 공정성 유지에 중요한 역할을 함에도 불구하고 평가오류, 수용성 한계 등 많은 제약을 가지고 있기도 하다. 이러한 한계를 극복하기 위해 최근에는 360도 평가,동료평가, 다차원평가 등 다양한 원천으로부터의 평가정보 획득을 통한 평가방식이 제안되고 있다. 그러나새로운 평가 방식에 대한 연구는 매우 부족한 실정이다. 본 연구는 동료평가에 있어서 평가자의 인지오류및 정보왜곡을 유발할 가능성이 있는 요소로 팀 성과에 대한 기여도, 집단만족과 상호작용빈도가 구체적으로 자율관리팀에서 동료평가 성향에 어떠한 차별적 영향을 미치는지에 대해 파악하였다. 총 445명 89개 팀을 대상으로 자료를 획득하고 가설을 검증한 결과, 평가자의 팀성과 기여도가 높을수록 동료평가의 차별화가 높게 나타났다. 또한, 평가자의 집단만족이 높을수록 동료평가 차별화가 낮게 나타났다. 반면에, 자율작업집단에서의 상호작용 정도와 평가자의 능력수준은 동료평가 차별화에 영향을 미치지 않는 것으로 나타났다. 이는 팀 성과에 열심히 기여한 사람일수록 동료들로부터 그에 합당한 평가를 받고, 기여도가 낮은 사람일수록 차별적으로 평가 받았음을 나타낸다. 집단에 대한 만족이 높으면 일에 대한 이해와 몰입이 높을 뿐만 아니라 동료들과의 관계 및 팀 성과에 관련된 집단 프로세스가 원활하여 공동체적 귀인을 하므로 구성원 모두다 잘한다고 평가하는 귀인효과가 나타났다고 해석 가능하다. 연구결과가 시사하는 바와 향후 연구과제에대해서도 논의하였다.