Systematic reviews are the main type of research scholars at ARPL are encouraged to do. They capture the core of good research and are a great way for a new scholar to learn the process while doing work that can have a big impact.
Imagine you are trying to answer an important question about health in your community. Let’s say that you are wondering: "What are the most effective ways to prevent malaria in African communities?" To answer such question properly, you cannot rely on just one or two studies. You need to look at all the available research on the topic and bring it together in a organized way. This is exactly what a systematic review does.
A systematic review is a special type of research paper that collects, evaluates, and summarizes all existing studies on a specific research question. Unlike a regular literature review where you might select a few interesting papers to discuss, a systematic review follows strict rules to find every relevant study, judge their quality, and combine their findings. Think of it as being like a detective who must follow a careful procedure to gather all the evidence about a case, rather than just picking a few pieces of evidence that seem interesting.
<aside> 🐾
The word "systematic," in this context, refers to following a clear and step-by-step method that other researchers could repeat and get the same results. This makes systematic reviews very reliable and trustworthy, unlike literature reviews, whereas mentioned, we never do at ARPL. They are considered one of the highest forms of evidence in research because they do not depend on just one study's findings, which might be incorrect or biased. Instead, they look at the big picture by examining many studies together. For example, while one study might find that vitamin supplements prevent colds and another study might find they do not help at all, a systematic review would examine both studies plus dozens of others to determine what the overall evidence really shows.
</aside>
Systematic reviews are different from other types of reviews you might encounter. A narrative review is more like an essay where the author discusses various studies on a topic but does not follow strict rules about which studies to include or how to analyze them. The author might focus on studies that support their own views. A scoping review maps out what research exists on a broad topic without trying to answer a specific question or judge study quality. For instance, a scoping review might explore "What research exists about youth mental health in Africa?" without trying to answer whether a specific treatment works. A systematic review, however, starts with a precise question and follows rigorous methods to answer it. (Please note that you don’t need to pay much attention for those other types mentioned above, yet if you are curious, reach out to your mentor or refer to Types of Research modules in the curriculum).
<aside> 2️⃣
There are two main types of “systematic” reviews. Some systematic reviews include a meta-analysis, which means the authors use statistical methods to mathematically combine the results from different studies into one overall result. For example, if ten studies tested whether a new teaching method improved history scores, a meta-analysis would calculate the average effect across all ten studies. Other systematic reviews do not include meta-analysis, either because the studies are too different to combine mathematically or because the research question does not involve numbers. These reviews synthesize findings through careful description and comparison instead. Both types are valuable and follow the same careful systematic process.
</aside>
Systematic reviews serve several important purposes in research and decision-making. First, they help solve the problem of information overload. In today's world, thousands of research studies are published every year on almost any topic you can imagine. A busy doctor, teacher, or policymaker cannot possibly read all these studies. A systematic review does this work for them by finding all relevant studies and presenting a clear summary of what the research shows overall.
Second, systematic reviews help identify where studies agree and disagree. Sometimes different studies reach different conclusions, which can be confusing. A systematic review examines why these differences exist. Perhaps the studies used different methods, tested different groups of people, or measured outcomes differently. By carefully comparing studies, a systematic review can explain these differences and determine what conclusions are most supported by evidence.
For example, if you wanted to know whether exercise reduces depression in teenagers, you might find some studies say yes while others show no effect. A systematic review would examine factors like how much exercise was done, what type of exercise, how depression was measured, and the characteristics of the teenagers in each study to understand these different results.
Third, systematic reviews reveal gaps in research. By collecting all existing studies on a topic, researchers can clearly see what questions have been studied extensively and what questions need more research. Perhaps many studies have examined a health intervention in adults but very few have tested it in children. Or perhaps all the studies on an educational program were conducted in wealthy countries, and we do not know if the program works in low-resource settings. These gaps guide future researchers toward the questions that still need answers.
You can remember that in the Opening Ceremony and in many areas of the curriculum, it was mentioned that reviews are important for young scientists like yourself (and even myself writing this module right now) as they allow us to have a deep understanding of a niche topic of the field and thus identify gaps for future empirical research (potentially during undergrad or even grad studies).
One final merit of systematic reviews is that they really save resources by preventing unnecessary duplication of research. If a systematic review shows that a question has been thoroughly answered with strong evidence, researchers can move on to new questions instead of conducting yet another study on the same topic. However, if a systematic review reveals that existing studies have important limitations, it clarifies what kind of new research is needed.
Just like any other type of research paper, every systematic review begins with a clear, focused research question. You can refer to the module on research question for guidance. To refresh your mind, this question must be specific enough to guide your search for studies but broad enough to be meaningful and useful. A vague question like "Does education help people?" is too broad for a systematic review. A better question would be "Does providing adult literacy programs in rural African communities improve women's health-seeking behaviors?" This question is specific about the intervention (adult literacy programs), the population (women in rural African communities), and the outcome (health-seeking behaviors).
First, you need to understand the field and your topic before you land on a research question. Most of us are just starting out and haven't read many journal articles yet. So it’s fine to begin with something as simple as a Google search. That means you already need a basic idea in mind.
When you’re building your initial knowledge, it’s okay to use easy-to-understand sources with pictures and videos. You can look at less formal resources, like reading a Wikipedia page, watching a YouTube video, or checking out infographics. At this early stage, you’re just trying to get familiar with the main words and ideas related to your topic. Make sure you understand these "keywords" clearly and maybe write them down.
Once you have that basic understanding, you can start looking into the academic literature. Try to find a systematic review or meta-analysis about your topic: pick the latest and most relevant one and read through it. You may try to ask questions during this time. You may then see if other journal articles have already answered those questions. If not, or if it hasn’t been done well enough, that could be a good research question.
Once you have developed your research question, you should write a detailed outline (also called a protocol) before you begin your systematic review. A protocol is like a blueprint or recipe that explains exactly what you will do and how you will do it. Writing a protocol before starting prevents you from making biased decisions later. For example, without a protocol, you might be tempted to include or exclude certain studies based on whether their results support what you hoped to find. The protocol commits you to a method in advance, making your review more objective and trustworthy.
Your protocol should include several key sections. The background explains why your research question is important and what is already known about the topic. The objectives state clearly what your systematic review aims to achieve. The methods section is the most detailed part and describes your eligibility criteria (which studies will be included and excluded), your search strategy (where and how you will look for studies), your selection process (how you will screen studies), your data extraction plan (what information you will collect from each study), and your synthesis methods (how you will bring the findings together). We will be mainly using the term outline, instead of protocol.
If you are doing a systematic review in public health, biology, neuroscience, or similar fields, take a look at the following page.