The systematic protocol of ‘Preferred Items for Systematic Review Recommendations’ (PRISMA) outlined in Moher et al., Pickering and Byrne and Pickering et al., were used to conduct a systematic literature review. This approach aimed to comprehensively synthesize, evaluate and track the scientific literature on a certain topic of interest. The method involved selecting and categorizing papers according to the specific eligibility criteria in order to minimize the potential biases, which occur in a narrative review. A systematic review has higher accuracy, consistency and transparency if exclusion and inclusion criteria are strictly defined and followed. After fixing the topic, the inclusion and exclusion criteria were implemented according to the aims and objectives of the research.
Scopus, Web of Science, and Science Direct were selected as the appropriate databases for this study because most of the articles pertaining to the relevant discipline, were found in these databases. Only full-text peer-reviewed original research articles were selected, because these articles maintain a standard of work with detailed methods, results, etc., and have credibility as a result of the peer-review process. The time chosen for the search was between the years 2000 and 2018 because significant research into adaptation began in 2000 and has continued. Only articles written in English were considered for this review because English was the main language used in research and academic publishing, with over 90% of academic articles published in English.
2.1. Search Protocol and Selection Methods A total of 809 articles were initially obtained for possible inclusion from the 3 databases, and these papers were assessed using the four major steps for selecting articles for a systematic review, i.e., identification, screening, eligibility and inclusion. Articles were initially screened to ensure their relevance to the study’s aims and objectives. After removing duplicated papers, we obtained 380 articles in the identification step. Then, in the screening step, 228 papers were excluded due to having an unrelated focus. Therefore, 152 papers were assessed for eligibility by studying their full text. In this eligibility step, 37 articles were excluded because the study area did not include coastal areas, and another 59 articles did not meet other eligibility criteria. In the final step, 44 papers were included, and after that, a snowball search was undertaken to include more articles and to find out if any relevant articles had been missed. A snowball search uses the reference list or citations in a paper to identify additional papers. Snowball sampling is an effective method for identifying sources published in obscure journals that may be missed from common databases, although potentially onerous because the sample grows at an exponential rate. In this study, articles cited within the set of articles were assessed for inclusion, with 10 found to meet the criteria. Therefore, finally, 54 full-text relevant articles were obtained and reviewed.
2.2. Review Limitation This study was focused on peer-reviewed articles in English. However, this means the study excluded work on adaptation published in other sources, such as non-scientific literature or in non-English journals. Another limitation was that the study considered only 3 databases to identify articles concerned with adaptation and resilience in coastal agriculture in Bangladesh. The inclusion of a different combination of databases may add further outcomes. The most obvious limitation of this study was that no inter-sectoral studies were found. This may be due to a lack of inter-sectoral studies, but the focus on agriculture may also have excluded studies where agriculture was not the primary focus. Analysis of inter-sectoral papers may inform unanticipated results and reveal more severe consequences of climate change. There was also a large gap in the knowledge of how sector-based policies and actions make it difficult to respond to the inter-sectoral impacts of climate change.