This paper described and illustrated the meta-analytic approach to synthesis of research results. Meta-analysis is a term coined by Glass, who defined it as the “analysis of analyses” or “the statistical analysis of a large collection of analysis results from individual studies for the purpose of integrating the findings.” The meta-analytic approach to research integration has the following characteristics. First, it is quantitative in that it uses a variety of statistical methods for organizing and extracting information from a large mass of data which are practically incomprehensible by other means. Second, meta-analysis does not prejudge research findings in terms of research quality. It trys to represent all the research literature on a specific topic. Third, it seeks general conclusions. Thus the meta-analyst must aggregate findings from almost all the studies on a given topic in disregard of their superficial differences. Fourth, it is a systematic and replicable approach to integrating findings from a set of studies. Generally meta-analysis proceeds as follows: A meta-analyst selects and defines a research problem for which a large mass of studies exists. The next step is to define the population of research studies and all identified studies that fit this definition must be acquired. If the population is too large, probability sampling techniques may be employed. Collecting the research results to be synthesized, it is necessary to describe, classify, and quantify their features. The findings of the studies are transformed into a common metric so that they can be aggregated statistically. Once the features of the studies have been quantified and their findings standardized, the meta-analyst proceeds with data analysis. Data analysis in meta-analysis consists of inferential statistics as well as simple descriptive statistics such as means and standard deviations of effect sizes. Significant testings of the differences in mean effect sizes among important subgroups and combining independent tests of significance can be conducted. Meta-analysis usually presents such statistics as the average effect size, combining significant level, and fail-safe number. There are some criticism and difficulties about the application of meta-analysis. For example, the critics argue that the meta-analysis is like mixing apples and oranges, and thus it makes no sense to integrate the findings of different studies. It is also criticized that it advocates low standards of judgment of the research quality by integrating ‘good’ and ‘bad’ studies together. One of the problems about the application of meta-analysis is that it is sometimes difficult to get standardized measures of the findings from some studies because of insufficient data. Though it has some serious difficulties and problems, the meta-analysis is considered to be an effective approach to research integration. It will prove to be quite valuable when applied and interpreted with care. Under the pressure of accumulating numbers of studies on a research topic, the need for quantitative techniques of research integration like meta-analysis will certainly grow.