How to Write a Systematic Literature Review for Psychology Research

By Martin Dempster, Donncha Hanna

Literature reviews are an important part of psychology research proposals and research reports. However, sometimes literature reviews are produced as research reports in their own right. These reviews usually follow a particular format and are known as systematic reviews.

The systematic review is a powerful research methodology that answers questions on the basis of good evidence and provides researchers with a valuable, comprehensive and up-to-date summary of work conducted in a specific area. Systematic reviews are not a solo effort; a team of several people is required for this type of review. The following steps outline how to tackle a systematic literature review:

  1. Specify the question to be answered by the review.

    A systematic review aims to answer a specific question, which must be clarified at the outset. For example, a systematic review may aim to answer the question: ‘What is the effect of hypnosis (compared to no treatment) on the anxiety levels of people with heart disease?’

  2. Write a protocol.

    The protocol begins with a rationale for the review and includes details about the methodology of the review. Key information for your protocol includes:

    • Eligibility criteria. Outline your list of criteria that articles must meet before you include them in your review. Often, this includes information about the research design used in the studies that you want to review and the type of participants included in these studies.

    • Search strategy. Identify and retrieve articles for your review by searching electronic databases. Specify the search terms that you use in these searches.

    • Validity criteria. Decide how you’ll assess the validity (or quality) of each study. This depends on the types of studies that you include in your review.

    • Data extraction, analysis and dissemination. Describe how you’ll extract data from the research articles during your review, and how you’ll use/combine this information.

  3. Retrieve eligible literature.

    Conduct a search for the articles and retrieve them using the search strategy outlined in your protocol. Assess these research articles to ensure that they meet your eligibility criteria.

    At least two reviewers should conduct the eligibility assessment of your selected articles and agree that the articles meet the eligibility criteria. This alerts the review team to any potential bias.

  4. Collect data.

    After you exclude ineligible articles from your review, assess the remaining articles for quality. Conduct this assessment independently, using at least two reviewers and checking that they’re both in agreement. Extract data from all the articles that you include in your review; by doing this, you reduce the information that you present from each article to a manageable and analysable amount.

  5. Analyse data, draw conclusions and report findings.

    When your review includes quantitative information and the studies in your review are sufficiently similar, you can conduct statistical meta-analysis (where you combine the data from the various studies and provide an overall effect size for the phenomenon under investigation).

    In other situations, you can instead provide a narrative synthesis of the data (where you summarise the findings from the various studies and present the reader with an answer to the original review question).