Page 79 - Tallinn University

HARIDUSTEADUSED Tallinna Ülikooli üliõpilaste 2015/2016. õppeaasta PARIMAD TEADUSTÖÖD / Artiklite kogumik
cONclusiON
Literature reviews are an essential part of conducting research at any level and are o en seen as a fundamental aspect of doing research. Reviewing existing literature on a topic can help researchers place the relevance of their own research into a larger context of what has already been done, helping them avoid duplicating research as well as to discover what remains to be done in the eld.
Conducting literature reviews can be done using various approaches, many of which have formally been de ned as methodologies. Each literature reviewing methodology provides researchers with a set of work methods that can be employed to carry out the review. Although most literature reviews vary in their exact work methods, they can similarly be analysed through a four-dimensional framework. is framework, proposed by Grant & Booth, states that all literature reviewing methodologies deal with searching for literature, determining its relevance, analysing it by extracting data and synthesising the collected data.
Grant & Booth [3] proposed fourteen di erent literature reviewing methodologies, based on both scoping literary studies as well as practical experience. ese methodologies were mapped against the previously proposed framework and grouped into a table. is table was extended by complementing it with the work methods each review type employs and the so ware that can be used to assist in carrying out these methods.
An analysis of literature reviewing methodologies showed that most approaches di erentiate on a perfunctory level with di erences appearing only in how data is analysed and presented. is trend is also apparent in the extended table – di erent trends and patterns could be seen in di erent literature reviewing methodologies.
As reviewing literature has played a pivotal role in research, di erent so ware has been proposed to assist conducting literature reviews. While some applications are not initially de ned as tools for conducting literature reviews, they can still be used in the process. Other applications are created implicitly to assist in conducting literature reviews. ese applications vary in terms of functionality and complexity, but are identical in their end-goals. Most of them also share a common shortcoming, since they mostly focus on conducting the analysis phase and assisting in the synthesis phase of the review.
To evaluate the e ect of employing these tools in a literature reviewing process, a case study was carried out. is case study encompassed a literature review in the eld of distributed user interfaces. e two primary tools used in the process were NVivo and Mendeley.
Proposing so ware before conducting the case study was done in „the more the merrier“ approach, proposing di erent tools for each requirement when possible. e e ectiveness of this approach was disproved in the case study, as it turned out that employing a wide variety of tools may hinder and impede the reviewing process.
However, employing literature reviewing so ware was bene cial. Using NVivo to code publications hastened the process of reviewing publications in comparison to using a spreadsheet. It also allowed analysing codes, displaying all references and occurrences of a code in all the publications.
79