當前中國大陸公眾參與政府績效管理的有效形式之一，就是「民主評議政風行風」。本文旨在通過對「民主評議政風行風」活動現狀的研究，提出完善行風評議中公眾參與的建議。從西方新公共管理運動的成功經驗看，政府績效管理必須以「顧客」為導向，以公眾滿意為標準，請公眾充分參與政府工作績效的評估。在中國大陸持續開展的「民主評議政風行風」活動中，主持活動的政府主管部門聘請「行評代表」，同時向「第三方」發放調查問卷，是這個活動中的兩種「公眾參與」形式。通過分別研究「民主評議政風行風」活動，及這個活動中的兩種公眾參與形式，可以看到，現在公眾參與的主體構成趨於多元化、科學化；評議對象和內容更加廣泛；評議形式更為多樣。現在行風評議工作，已經極大地促進了政府部門政風行風的轉變，但也在一定程度上存在一些問題。例如評議主體與被評對象資訊不夠對稱；「第三方」一定程度地被動參與、形式參與；「問卷調查表」設計不夠科學；「第三方」的構成職業比例不夠科學等問題。針對這些問題，建議進一步實施政務公開，加大政府工作的透明度；分類管理，增強「問卷調查表」的科學性和專業性；科學組織，提高「第三方」參與評議的能力；以「第三方」評議形式，促進政府專項工作的開展。

“Democratic Review of Administrative and Business Style” (DRABS) isone of methods of public participative government performance management inMainland China. This paper is aimed to propose how to improve the practice ofthe DRABS by reviewing its implementation. From the perspective of new publicmanagement, government performance management must be “customeroriented”and assessed based on the public satisfaction, so it needs to absorbadequate public participation. In the exercise of the DRABS, the office-inchargewould employ “assessment representatives” and conduct questionnairesurveys among the public, or so-called “the third party.” These are two kinds ofpublic participation. Through the analysis of the DRABS and the two kinds ofpublic participation, it can be found that the sources of public participation becomeincreasingly diverse and the assessment also becomes more scientific; the targets and contents of assessment become broader; methods of assessment becomemore diverse. Under the DRABS, the service of government agencies hasbeen greatly improved but some problems can be found in the process of theDRABS. There is information asymmetry between the assessment targets andassessors; the participation of the “third party” is too passive and too superficial;the design of the questionnaire is not scientific enough; the sources of the“third party” assessors do not match the assessment targets. For these problems,the author suggests that the openness and transparency of the governmentworks should be strengthened; different assessment methods should be appliedto different kinds of agencies; the questionnaire survey should be more scientificand professional; the capacity of the “third party” participation should beenhanced; government affairs should be improved by the “third party” assessment.