Abstract: In order to explore the present situation and problems of using hierarchical linear modeling in psychological researches in china, this article reviewed and analyzed 50 literatures of China psychological journals from 2002 to 2011 included in CNKI database. A checklist, derived from the methodological literature on hierarchical linear modeling and focusing on the issues of model development and specification, data preparation, estimation method and hypothesis testing, was used to analyze those 50 articles using hierarchical linear modeling in psychological researches in china. The results of sample description are as follows: (1) HLM method has been widely used in our psychological researches, for example, Management, development and education psychology. (2) The most common applications were two-level models where individuals were nested within contexts. In addition, it deserve to be mentioned that the sample size of level 2 is significant larger than the sample size of level 1. Although HLM method in our psychological researches has been widely used, there are a number of problems in its application: (1) In model development and specification, one could not determine how many models were estimated in five articles, what type of centering if any was used in most of the articles, Whether the null model was tested in half articles. (2) In data preparation, no article reports the test whether the data were consistent with assumptions, for example, distributional assumptions; no article implements the prior power analysis; In many articles, one could not determine whether the data were complete, if they were not complete, the treatment method of missing value has not been described. (3) In estimation method and hypothesis testing, except six articles, one could not determine how the models were estimated; many articles have not report enough information for a reader to critique the reported analyses, for example, standard errors for the parameters of interest and proportion of explained variance. Four guidelines for researchers reporting hierarchical linear analyses are provided.

Key words: hierarchical linear modeling, bibliometric analysis, content analysis

摘要： 以2002-2011年中国期刊网收录的50例应用多层线性模型(HLM)的心理学期刊论文为研究对象，从样本描述、模型发展与规范、数据准备、估计方法与假设检验4个角度进行文献计量和内容分析，对我国心理学研究中HLM方法的使用现状进行评估。结果表明，HLM方法主要用于管理、发展和教育心理学，绝大多数应用都是两层模型且层2样本量较大。HLM方法在广泛应用的同时仍存在忽略前提假设检验、分析过程中的重要信息和结果报告不完整等问题，随后提供了4条建议。

关键词: 多层线性模型, 文献计量, 内容分析