ABSTRACT
32The formulation of detailed objectives is at the heart of any research project. This should include the definition of study participants, interventions, outcome, and results. It is prerequisite to define any basic inclusion criteria to assess the methodological quality of comparable studies and to perform a thorough sensitivity analysis. It is useful to have two observers checking eligibility of candidate studies with disagreement being resolved by discussion or a third reviewer. Selection bias can also occur in making decision about which study to be included and which study to be excluded from meta-analysis. Ideally, all inclusion criteria are set before reviewing the studies identified. The source of search for literature in meta-analysis includes the published literature, unpublished literature, uncompleted research reports, and work in progress. These are also different methods of searching the literature. A backward search involves identifying a publication and then moving to earlier items in the citation. A forward search identifies a publication and then searches all items that later cite the publication. All the studies that fit within the criteria should be located. To provide an accurate estimate of an effect, it is important to find unpublished articles for analysis. Professional meetings are particularly good way to locate unpublished articles. It is good policy to write to the first author of each article that you decide to include in your analysis to see if they have any unpublished research relating to your topic. Research registers are actively maintained lists of studies centered on a common theme. Sometimes the number of studies that fit inside your boundaries is too large for you to analyze them all then choose a random sample of the studies for coding and analysis. The exclusion of gray literature from meta-analysis can lead to exaggerated estimates of intervention effectiveness. A standard strategy would include hand searching of selected journals and checking the references of all the relevant studies identified. Previous reviews, whether they include a meta-analysis or not, are often a fruitful place to look for relevant studies. Foreign studies should be included in the analysis unless you expect that cross-cultural differences would affect the results. The electronic searching, conference proceedings, nonindexed journals would not typically be included. Not all the reports retrieved are appropriate for inclusion in a meta-analysis. You would be well advised to make use database program to assist you in this task. Results of smaller studies can be expected to be more widely scattered around the average. 33The personnel bias exists when some people have economic, social, or political agenda in order to favor their legislations.
