ABSTRACT
The focus of this chapter is on the identification of relevant primary studies. This process forms the first step of the conduct phase of the systematic review process, as highlighted in Figure 5.1.
An important element of any systematic review or mapping study is to devise a search strategy that will find as many primary studies as possible that are relevant to the research questions. The likelihood is that the strategy will involve a combination of search methods. One widely used method is automated searching of the literature using resources such as digital libraries and indexing systems. Other methods include manual searching of selected journals and conference proceedings, checking papers that are cited in the papers included in a review (backwards snowballing) and checking papers that cite the papers included in a review (forwards snowballing). The search strategy will aim to achieve an acceptable level of completeness (see Section 5.1) within the review’s constraints of time and human resources. The level of completeness that might be targeted will depend on the type of review being undertaken. Generally, for a quantitative systematic review, that is, one which compares software engineering technologies, a high level of completeness
FIGURE 5.1: Searching stage of the systematic review process.
