At present, there are several problems in health science popularization in China, such as unbalanced development among science popularization institutions, insufficient investment in science popularization, difficulties in refined evaluation of science popularization effectiveness. Besides, with the advent of "science popularization fever", science popularization work presents new characteristics and a new evaluation system is urgently needed. Taking health science popularization as an example, the research on the work evaluation of institutions is relatively weak, and the existing evaluation system has some practical problems such as single dimension, difficulties to quantify, lack of systematization and operability.

To construct an evaluation index system of health science popularization, so as to provide reference for the establishment of evaluation criteria for health science popularization work for social institutions.

Based on the theory of "structure-process-result" and literature reports, a correspondence questionnaire was formulated, 14 senior practitioners in the field of health science popularization were selected to form an expert panel, and Delphi method was used to conduct three rounds of correspondence by email from April to May 2022. SPSS 25.0 software was used for statistics and analysis of data. The evaluation index system of science popularization work for social institutions was constructed according to expert opinions.

The recovery rate of the first round of correspondence questionnaire was 100%. A total of 9 experts gave 34 revised opinions on 19 indicators with the rate of expert opinions was 64.29% (9/14), indicating a high degree of experts' enthusiasm. The authority coefficient of experts of the correspondence experts was 0.814, with the judgment basis of 0.886 and the familiarity degree of 0.742. Kendall's W coefficient of the first round and the second round of expert correspondence was 0.238 and 0.246, respectively (P<0.001). The CVs of the two rounds of expert correspondence were both less than 0.250, indicating a good consistency of expert opinions. The evaluation index system of science popularization work for social institutions was determined and constructed, including 3 primary indicators (input, implementation and effectiveness of science popularization), 16 secondary indicators and 77 tertiary indicators.

The evaluation index system of health science popularization work for social institutions is constructed based on Delphi method, including 3 primary indicators (input, implementation and effectiveness of science popularization), 16 secondary indicators and 77 tertiary indicators, its application is helpful to improve the efficiency of health science popularization work in social institutions.

目前我国的科普工作存在着科普机构之间发展不平衡、科普投入不足、科普成效难以精细化评价等问题，且随着"科普热"的来临，科普事业呈现出新的特点，亟需新的评价体系。以健康科普为例，针对机构的工作评价研究相对薄弱，现有评价体系存在维度单一、难以量化、缺少系统性和可操作性等现实问题。

构建健康科普工作评价指标体系，为建立面向社会机构的健康科普工作评价标准提供参考依据。

基于"结构-过程-结果"理论和文献报道，制订函询问卷，遴选14名健康科普工作领域的资深工作者组成专家小组，采用德尔菲法通过电子邮件的方式于2022年4—5月进行3轮函询，采用SPSS 25.0软件进行数据的统计与分析，根据专家意见构建机构科普工作评价指标体系。

第1轮函询问卷回收率为100%，共有9名专家对其中19项指标给出了34条修订意见，专家意见提出率为64.29%（9/14），专家积极程度较高。函询专家的权威系数为0.814，判断依据为0.886，熟悉程度为0.742。第1、2轮专家函询的Kendall's W系数分别0.238、0.246（P<0.001）。两轮专家函询的变异系数均<0.250，专家意见具有较好的一致性。确定并构建出机构科普工作评价指标体系，其中包含3项一级指标（科普投入、科普实施、科普成效）、16项二级指标和77项三级指标。

基于德尔菲法构建出机构科普工作评价指标体系，其中包含3项一级指标（科普投入、科普实施、科普成效）、16项二级指标和77项三级指标，其应用有助于提升社会机构的健康科普工作效率。