Abstract: R&D personnel are the core force of knowledge creation and technological innovation in an organization. In order to promote the organization’s innovation and maintain its sustainable development, how to motivate R&D personnel to share their knowledge effectively has become an important issue of common concern in the field of industrial practice and academic research. Performance appraisal is an important tool for manager to assess, motivate and control the knowledge sharing behavior of employees. Traditional performance appraisal mainly relies on the administrative management under the leadership of superior authorities or teams. With the wide application of cross-departmental cooperation, autonomous team and virtual team in the process of technical study and knowledge sharing, there is an urgent need to rely on peer evaluation of internal group to improve management efficiency. Peer evaluation is the thinking or reflection of employees on their colleagues’ work. Peer evaluation exists in the real and virtual communication environment. R&D team or group will inevitably be affected by their colleagues’ evaluation in the process of face-to-face evaluation, work feedback, information sharing, comments on virtual community and the other types of organizational daily activities. However, few studies directly explore the relationship between peer evaluation and knowledge sharing, especially among R&D groups within organization. In addition, Chinese culture has the characteristics of high group orientation and strong social concern. Does high-intensity group concern result in overload of cognitive resources of R&D personnel, which leads to the change of their preference or exclusion on specific types of peer evaluation, and then have a special impact on knowledge sharing? Starting from the phenomenon of Chinese culture, takingR&D personnel as the research samples, this paper explored the relationship between peer evaluation and employees knowledge sharing. The research process was as follows. Firstly, theoretical framework and research hypothesis were constructed through integrating performance evaluation theory, motivation theory, social exchange theory and social anxiety theory together. Secondly, visited and learned about R&D Department within different companies, 12 R&D personnel were selected for interviews, and the types of peer evaluation were excavated and refined to form the initial questionnaire by content analysis. Then, carried out exploratory factor analysis and reliability analysis through 85 small sample tests, revised and formed the final measurement terms. Thirdly, a large sample of empirical research was conducted and 535 valid questionnaires were collected, through reliability and validity analysis, common method deviation analysis and regression analysis, tested the influence mechanism of different types of peer evaluation on employees knowledge sharing.The results showed that: (1) The conceptual structure of four dimensions with two variables had been proposed and tested, which can be divided into positive evaluation and negative evaluation from the perspective of nature, and be divided into process-oriented evaluation and outcome oriented evaluation from the perspective of orientation. (2) The positive impact of peer positive evaluation on R&D employee’s knowledge sharing and the negative impact of peer negative evaluation on it had been verified. (3) Both outcome-oriented evaluation and process-oriented evaluation can promote employees knowledge sharing, process-oriented evaluation can promote explicit knowledge sharing more significantly, while outcome-oriented evaluation can promote tacit knowledge sharing more significantly. That was to say, process-oriented evaluation which was relied on the attitude, method and procedure of technological research process as evaluation criteria, can encourage explicit knowledge sharing more effectively, since it was easy to be expressed and measured. However, the unique experience knowledge of R&D employee was difficult to monitor and measure in the process of knowledge sharing, so tacit knowledge will be more effective to be motivated through outcome-oriented evaluation. (4) Outcome-oriented evaluation weakened the relationship between the relationship of positive evaluation and explicit knowledge sharing. Process-oriented evaluation strengthened the relationship between the relationship of negative evaluation and tacit knowledge sharing, while outcome-oriented evaluation play a weakening role between the relationship of negative evaluation and tacit knowledge sharing. (5) Group attention had produced important boundary effects. When group attention was higher, the promoting effect of outcome-oriented evaluation on explicit knowledge sharing was stronger, and the promoting effect of outcome-oriented evaluation on tacit knowledge sharing became insignificant.The conclusions of this study can provide important management inspiration for organizations to make better use of peer evaluation tools and promote the efficiency of knowledge sharing. Firstly, the organizations should abandon the single or one-size-fits-all evaluation mode of thinking, take differentiated measures to monitor the advantageous side and flip side of emotional components on peer evaluation. Timely controlled the downsides of negative evaluation, and avoid them turning into bamboo telegraph or gossip behind employees’ back, so as to create a disharmonious atmosphere of "literati are light on each other" between knowledge workers andR&D personnel. Secondly, according to the characteristics of knowledge, organizations should make use of the rational components of peer evaluation, and seriously consider the practical effect of evaluation orientation, so as to promote the healthy development of organizational knowledge management mechanism. For example, when R&D personnel are sharing explicit knowledge, the organization can guide R&D personnel to adopt the final work performance or goal completion as evaluation criteria. When R&D personnel were sharing their researching experience or inspiration, the organization should advocate R&D personnel to adopt process elements, such as task methods and work procedures, as evaluation criteria. Thirdly, the organization should create a good atmosphere of public opinion in order to enhance the effectiveness of peer evaluation. On the one hand, for explicit knowledge sharing, organizations can advocate open outcome-oriented evaluation among employees. On the other hand, the organization should monitor the evaluation environment of colleagues, especially when R&D personnel share their researching experience in the face-to-face situations, their mangers should help dilute outcome-oriented evaluation thinking habits in teams or departments. It is undeniable that China is a country with group-oriented cultural characteristics, and people tended to pay attention to each other. Therefore, organizations should guide R&D personnel to reduce comparisons on their research performance in public. By creating a good atmosphere of organizational public opinion, it can provide a beneficial organizational environment for R&D employee to share their personal valuable experience and knowledge.

Key words: peer evaluation, explicit knowledge sharing, tacit knowledge sharing, group attention