Introduction
Systematic Literature Review (SLR) is a key part of many research initiatives, especially those in the academic and scientific sectors. In order to synthesize the current state of knowledge, identify knowledge gaps, and reach conclusions about the present state of knowledge, an SLR entails undertaking a thorough and in-depth examination of the existing literature on a particular topic.
SLRs are essential for researchers to advance knowledge and comprehension of a certain subject of study. Scholars can demonstrate their command of the literature in their field of study and their capacity for critical study evaluation by performing an SLR.
In this article, we will discuss the key steps involved in writing an SLR paper, as well as some of the challenges and tips for writing a successful and impactful SLR.
Difference Between an SLR and a Traditional Literature Review?
A systematic, open, and evidence-based literature review is known as a systematic literature review (SLR). It uses a strict and organized procedure to track down, evaluate critically, and synthesize the available data regarding a certain research subject. A conventional literature review, in comparison, may not adhere to a structured or systematic methodology and is more casual and exploratory in nature.
The degree of rigor, structure, and transparency in the research process is the primary distinction between an SLR and a conventional literature review. A typical literature review may place more emphasis on the reviewer’s personal thoughts and interpretations than an SLR, which strives to provide a thorough and objective summary of the status of research on a given issue.
I have written a separate article on traditional literature review with the title, How to write a better Survey Paper in 06 easy steps?. You can visit the post for more details.
How to Conduct a Systematic Literature Review?
Preparation
Preparation is the starting point of any SLR and is critical to the success of the review. It involves several key steps, including selecting the research question, defining the scope of the review, identifying relevant databases and search terms, and defining the inclusion and exclusion criteria for the studies.
i. Selecting the Research Question: The first step in preparation is selecting a clear and focused research question. This question should be relevant to the scholar’s field of study and should guide the scope and direction of the review. The research question should be specific enough to allow for a comprehensive review, but broad enough to generate meaningful results.
| Step | Description | Example |
|---|---|---|
| 1 | Identify the Research Area | Computer Science, Computer Graphics, Artificial Intelligence |
| 2 | Determine the Research Purpose | Answer a specific question, identify trends, compare and contrast different approaches |
| 3 | Identify the Research Gap | Fill gaps in existing research, contribute new information or insights to the field |
| 4 | Formulate the Research Question | “What are the most effective algorithms for image recognition in computer vision?” |
| 5 | Refine the Research Question | “What are the most effective deep learning algorithms for image recognition in computer vision and how do they compare to traditional computer vision algorithms?” |
By following these steps, you can select a focused and relevant research question for your SLR in computer science.
ii. Defining the Scope of the Review: Determining the scope of the review is the next stage after defining the research issue. This entails stating the length of time the review covered, the categories of studies it included, and the locations of those studies. For instance, the review’s scope can cover papers with a specified study design, conducted in a particular nation or region, and published between 2000 and 2020.
| Step | Description | Example |
|---|---|---|
| 1 | Identify Keywords | Image recognition, computer vision, algorithms, deep learning, computer graphics, artificial intelligence |
| 2 | Determine the Time Frame | Latest 5 years, last 10 years, all time |
| 3 | Choose Relevant Databases | IEEE Xplore, ACM Digital Library, Science Direct |
| 4 | Define Inclusion and Exclusion Criteria | Inclusion: Peer-reviewed journal articles, Exclusion: Books, conference papers |
| 5 | Specify the Language | English only |
By following these steps, you can define a clear and specific scope for your SLR, which will help to ensure that your results are relevant and meaningful.
iii. Identifying Relevant Databases and Search Terms: Finding the pertinent databases and search terms that will be utilised to find papers for the review is the next stage. Utilizing academic databases like Google Scholar, Scopus, and Web of Science as well as additional sources like conference transcripts and grey literature may be necessary for this. To achieve a thorough search, the search phrases should be properly picked and pertinent to the research subject.
| Step | Description | Example |
|---|---|---|
| 1 | Choose Relevant Databases | IEEE Xplore, ACM Digital Library, Science Direct |
| 2 | Identify Keywords | Image recognition, computer vision, algorithms, deep learning, computer graphics, artificial intelligence |
| 3 | Use Boolean Operators | AND, OR, NOT |
| 4 | Use Truncation and Wildcard Characters | *, ?, ! |
| 5 | Refine the Search | Add synonyms, eliminate irrelevant results, add constraints (such as date, language) |
By following these steps, you can effectively search for relevant literature in your field, which will serve as the foundation for your SLR.
iv. Defining the Inclusion and Exclusion Criteria: The inclusion and exclusion standards for the studies must then be established. Choosing things like the study population, publication date, and study design type may be necessary for this. Only high-quality studies should be included in the review, and the inclusion criteria should be stringent enough to achieve this while still being open-ended enough to allow for a thorough assessment of the literature.
| Step | Description | Example |
|---|---|---|
| 1 | Determine the Purpose of the Review | To evaluate the current state-of-the-art in deep learning for image recognition |
| 2 | Define the Scope of the Review | Peer-reviewed journal articles, published in English, in the last 5 years |
| 3 | Establish Inclusion Criteria | Must relate to deep learning for image recognition |
| 4 | Establish Exclusion Criteria | Conference papers, books, articles not related to image recognition |
| 5 | Evaluate the Relevance of Each Study | Based on inclusion and exclusion criteria, determine if the study should be included in the review |
By doing these actions, you can make sure that the studies you include in your SLR are pertinent to the review’s goal and scope. This will aid in generating reliable and insightful results.
In summary, an SLR’s preparation phase is essential to its success. Researchers can make sure that their SLR is well-defined, thorough, and relevant by following these crucial stages, and they can also make sure that they are able to make useful deductions about the existing state of knowledge in their area of interest.
Searching for Studies
The next phase in performing an SLR is searching for research, which entails using the search terms and databases established during the preparation stage to find pertinent studies. This is commonly done by searching electronic databases, but it is also possible to search other sources like conference proceedings and grey literature.
Electronic database searches are a useful tool for finding many studies on a particular subject. It’s crucial to use numerous databases and the search terms determined during the preparation step when performing a database search. In order to focus their searches, researchers should be careful when utilising Boolean operators like “AND” and “OR.”
| Steps | Description | Computer Science Example |
|---|---|---|
| 1. | Use multiple databases | Use multiple databases such as Google Scholar, IEEE Xplore, and ACM Digital Library to search for relevant studies. |
| 2. | Use search terms identified during the preparation stage | Use search terms such as “machine learning,” “computer vision,” and “deep learning” to locate relevant studies. |
| 3. | Refine searches using Boolean operators | Use Boolean operators such as “AND” and “OR” to refine searches, for example: (“machine learning” AND “computer vision”) OR “deep learning.” |
This table provides an overview of the key steps involved in conducting an electronic database search as part of an SLR, using a computer science example. This process can help researchers to locate a large number of relevant studies on a given topic and is an important step in the overall SLR process.
ii. Screening the Studies: Finding the studies and screening them for eligibility are the following steps. This entails comparing each study to the inclusion and exclusion standards established during the planning phase. Studies that don’t fit the requirements for inclusion are taken out of the review, while those that do are kept in for additional examination.
| Study Title | Inclusion Criteria | Exclusion Criteria | Eligibility | |
|---|---|---|---|---|
| 1 | “Artificial Intelligence in Healthcare” | Studies published in English | Opinion pieces and non-research articles | Included |
| 2 | “Machine Learning Approaches for Predictive Maintenance” | Studies published in the past 10 years | Studies focused on other industries | Included |
| 3 | “Neural Networks in Image Processing” | Studies using neural networks as the main approach | Studies using other techniques | Included |
| 4 | “The Impact of Deep Learning on Speech Recognition” | Studies evaluating the impact of deep learning | Studies that only compare deep learning to other techniques | Included |
| 5 | “Applications of Reinforcement Learning in Robotics” | Studies applying reinforcement learning in robotics | Studies focused on other domains | Included |
Note: The above table is just an example, and the inclusion and exclusion criteria will vary depending on the specific research question and scope of the systematic literature review.
iii. Managing the Studies: The research that is part of the review should be organised to make evaluation and retrieval simple. To do this, you might use a spreadsheet, a database, or reference management software.
Here’s a hypothetical example of a table that could be used to manage the studies in a computer science-related systematic literature review:
| Study ID | Author(s) | Year | Title | Method | Participants/ Samples | Results |
|---|---|---|---|---|---|---|
| 1 | Smith et al. | 2020 | “Evaluation of Machine Learning Algorithms for Image Classification” | Comparative Study | 100 | Algorithm X was found to have the highest accuracy for image classification |
| 2 | Johnson et al. | 2021 | “A Study on Deep Learning Techniques for Speech Recognition” | Experimental Study | 50 | Technique Y showed significant improvement in speech recognition compared to traditional methods |
| 3 | Patel et al. | 2022 | “Comparison of Reinforcement Learning Approaches for Gaming Applications” | Comparative Study | 75 | Approach Z was found to perform the best in terms of speed and accuracy for gaming applications |
This table provides a concise overview of the key information for each study, including the study ID, author(s), year, title, research method, number of participants/samples, and results. By organizing the studies in this manner, it becomes easier to assess the quality and relevance of each study, compare the results of different studies,
In conclusion, searching for studies is an important step in conducting an SLR. By using electronic database searches, screening the studies for eligibility, and managing the studies in an organized manner, researchers can ensure that they have located all relevant studies on their topic of interest and that they are ready to move on to the next stage of the SLR process.
Appraising the Quality of the Studies
An important part of doing an SLR is assessing the calibre of the research, which entails assessing the calibre of each study that was included in the review. This phase helps to reduce the risk of deriving the wrong conclusions from low-quality studies and helps to guarantee that the review’s findings are supported by high-quality studies.
i.Determining Quality Criteria: Choosing the quality standards to be applied is the first stage in evaluating the studies’ level of quality. These guidelines should be based on acknowledged practices in the field and pertinent to the sort of study being conducted. For a randomised controlled trial, the appropriate sample size, the allocation concealment mechanism, and the blinding of participants and outcome assessors might all be considered as quality criteria.
| Step | Description | Example |
|---|---|---|
| 1 | Choose a Quality Assessment Tool | Cochrane Risk of Bias Tool, Newcastle-Ottawa Scale |
| 2 | Define Quality Criteria | Relevance, validity, reliability, generalizability, and objectivity |
| 3 | Assess Each Study | Evaluate the study based on the chosen quality assessment tool and the defined quality criteria |
| 4 | Record Results | Record the results of the quality assessment for each study in a consistent and systematic manner |
| 5 | Evaluate the Overall Quality of the Studies | Based on the quality assessment results, evaluate the overall quality of the studies included in the review |
By following these steps, you can ensure that the studies included in your SLR are of high quality, and that the results of the review are valid and trustworthy. This will help to increase the credibility and impact of your review.
ii. Appraising the Studies: The next stage is to evaluate each reviewed research in light of the quality standards. In order to do this, a conventional quality evaluation tool or a unique tool created especially for the review may be used. Each study’s quality should be systematically recorded, and the final report should contain a summary of the quality findings.
| Step | Description | Example |
|---|---|---|
| 1 | Choose an Appraisal Tool | CASP (Critical Appraisal Skills Programme) or AMSTAR (A Measurement Tool to Assess Systematic Reviews) |
| 2 | Define Appraisal Criteria | Relevance, validity, reliability, generalizability, and objectivity |
| 3 | Appraise Each Study | Evaluate each study based on the chosen appraisal tool and the defined appraisal criteria |
| 4 | Record Results | Record the results of the appraisal for each study in a consistent and systematic manner |
| 5 | Evaluate the Overall Quality of the Studies | Based on the appraisal results, evaluate the overall quality of the studies included in the review |
By following these steps, you can ensure that the studies included in your SLR are critically evaluated and that the results of the review are valid and trustworthy. This will help to increase the credibility and impact of your review.
iii. Assessing Risk of Bias: Another crucial component of evaluating the studies’ quality is determining the risk of bias. This entails analysing the potential sources of bias in every study and determining how they might affect the findings. For instance, in a randomised controlled trial, sources of bias could include the selective presentation of data or the lack of participant and outcome assessor blinding.
| Step | Description | Example |
|---|---|---|
| 1 | Choose a Risk of Bias Tool | Cochrane Risk of Bias Tool or Newcastle-Ottawa Scale |
| 2 | Define Risk of Bias Criteria | Selection, performance, detection, attrition, reporting, and other bias |
| 3 | Assess Risk of Bias for Each Study | Evaluate each study based on the chosen risk of bias tool and the defined risk of bias criteria |
| 4 | Record Results | Record the results of the risk of bias assessment for each study in a consistent and systematic manner |
| 5 | Evaluate the Overall Risk of Bias | Based on the risk of bias results, evaluate the overall risk of bias of the studies included in the review |
These steps will help you recognise and evaluate the potential sources of bias in the studies that make up your SLR. This will improve the validity and dependability of your review’s findings.
In conclusion, an important step in doing an SLR is evaluating the calibre of the studies. Researchers may make sure that the findings of their SLR are based on high-quality studies and that they are able to draw meaningful conclusions about their area of interest by utilising quality standards and assessing the possibility of bias. By doing this, they may make sure that their SLR makes a significant contribution to the field and contribute to the legitimacy and reliability of their findings.
Synthesizing the Results of the Studies
Synthesizing the results of the studies is the next step in conducting an SLR and involves summarizing the findings of the studies included in the review. This step helps to identify patterns and trends in the data and to draw meaningful conclusions about the topic of interest.
i. Data Extraction: The pertinent information from each study must be extracted before the findings may be combined. This could entail obtaining information on certain outcomes, like the effect magnitude, or particular variables, like the study population. A systematic record of the extracted data should be kept, and the final report should provide a description of the data extraction procedure.
| Step | Description | Example |
|---|---|---|
| 1 | Choose Data Extraction Template | Choose a standard template or create a custom template to extract relevant data from studies |
| 2 | Identify Data to Extract | Identify the data that is relevant to your SLR, such as study design, sample size, methods, results, and conclusions |
| 3 | Extract Data Consistently | Extract the data consistently across studies, following the chosen template |
| 4 | Record Results | Record the extracted data in a consistent and systematic manner, such as in a spreadsheet or database |
| 5 | Check Data Extraction | Check the extracted data to ensure that it is accurate and complete |
By following these steps, you can extract relevant data from the studies included in your SLR in a consistent and systematic manner. This will help to increase the validity and reliability of the results of your review.
ii. Data Synthesis: The data that was taken out of the studies must then be combined. This could entail performing meta-analyses, creating narrative summaries of the findings, or summarising the data in tables or figures. The type of data and the makeup of the studies included in the review should guide the process of synthesis selection.
| Step | Description | Example |
|---|---|---|
| 1 | Identify Similarities and Differences | Identify the similarities and differences in the results and conclusions of studies investigating the effectiveness of different algorithms for detecting malicious network activity. |
| 2 | Organize Results | Organize the results of the studies into a summary table that includes information about the algorithms used, the dataset tested, the evaluation metric, and the results obtained. |
| 3 | Assess Consistency | Assess the consistency of the results across the studies, looking for patterns and discrepancies. For example, if the majority of studies find that algorithm A is the most effective, but a few studies find that algorithm B is more effective, this may warrant further investigation. |
| 4 | Identify Gaps in the Literature | Identify gaps in the literature, such as studies that use different evaluation metrics or datasets. This may suggest the need for future research to validate the results across different contexts. |
| 5 | Draw Conclusions | Draw conclusions based on the synthesis of the results, taking into account the quality of the studies and the overall consistency of the results. For example, the conclusion may be that algorithm A is the most effective overall, but further research is needed to validate this conclusion for different datasets and evaluation metrics. |
This example demonstrates how the steps for data synthesis can be applied in practice to synthesize the results of multiple studies in a systematic and rigorous manner, in order to draw robust conclusions based on the evidence in the literature.
iii. Drawing Conclusions: Making judgements on the subject of interest is the last step in synthesising the research findings. This could entail highlighting the most important conclusions, pointing out areas of agreement and disagreement, and addressing any data constraints. The findings from the studies cited in the review should be used to support the conclusions, which should be supported by the data.
| Step | Description | Example |
|---|---|---|
| Drawing Conclusions | This step involves summarizing and synthesizing the findings from the previous steps. The conclusions drawn should be based on the data extracted and analyzed and should be supported by the available evidence. | In the example of the SLR on “Performance evaluation of cloud computing systems”, the researcher may draw conclusions about the most effective cloud computing systems based on factors such as scalability, reliability, and security. |
In conclusion, a crucial component of performing an SLR is synthesising the research findings. Researchers can make sure that their SLR offers an extensive and meaningful overview of the literature on their topic of interest by extracting pertinent data from each study, synthesising the data, and deriving meaningful conclusions. They can contribute significantly to the field and help to improve our understanding of the subject by doing this.
Reporting the Results of the Review
The last step in the procedure entails summarising the review’s findings and presenting them in a clear, succinct manner. This is known as reporting the results of the SLR. Researchers, practitioners, and policymakers should all be able to access and comprehend the results when they are presented.
i. Organizing the Results: The first step in reporting the results is to organize the information in a logical and easy-to-follow manner. This may involve using headings and subheadings, tables, and figures to summarize the key findings. The results should be presented in a way that is clear and concise, and that accurately reflects the findings of the review.
Here’s the information presented in a tabular format:
| System | Strengths | Weaknesses |
|---|---|---|
| A | High availability, low latency | Low data storage capacity |
| B | High data storage capacity, efficient data transfer | High cost, low scalability |
| C | High scalability, low cost | Low availability, high latency |
ii. Including Key Information: It’s crucial to include vital details in the report, such as the review’s history, its methodologies, the findings of the studies’ quality assessment, and a summary of the conclusions. A discussion of the results’ implications and any data constraints should also be included in the report.
| Steps | Example |
|---|---|
| 1. Identify the information to include in the table | Relevant information such as the study’s authors, publication year, methodology, results, and conclusions |
| 2. Organize the information in a logical and clear format | Use a tabular format, where each study is represented in a separate row and each column represents a specific piece of information (e.g., authors, year, methodology, results, conclusions) |
| 3. Ensure that the table is easy to read and comprehend | Use clear and concise language, use headings and subheadings, ensure that the information is visually separated, and consider using shading or color to emphasize important information |
| 4. Highlight key findings | Use bold or italic text, use symbols or icons, or highlight findings that are particularly relevant to the research question |
Example:
| Study | Authors | Publication Year | Methodology | Results | Conclusions |
|---|---|---|---|---|---|
| Study 1 | Smith et al. | 2020 | Qualitative | Results showed that users preferred a more personalized experience with technology | Concluded that personalization is crucial for user satisfaction with technology |
| Study 2 | Jones et al. | 2021 | Quantitative | Results indicated that users have a strong desire for technology that is easy to use and understand | Concluded that usability is a key factor in user satisfaction with technology |
| Study 3 | Davis et al. | 2022 | Mixed Methods | Results showed that users value technology that is both personalized and easy to use | Concluded that both personalization and usability are important factors in user satisfaction with technology |
iii. Ensuring Transparency:When presenting the results of an SLR, transparency is essential, and it’s important to make sure that the steps taken and the outcomes of the study’s quality assessment are clearly described. As a result, the conclusions will be more reliable and valid, and other academics will be able to gauge the quality of the review.
| Step | Description | Example (Computer Science) |
|---|---|---|
| 1 | Define the purpose of transparency | The purpose of transparency in the SLR is to ensure that the review process is open, accountable and can be easily followed by others. |
| 2 | Identify potential sources of bias | Potential sources of bias in the SLR process include researcher experience, study design, and funding source. |
| 3 | Document the review process | The review process should be documented in detail including the research question, inclusion and exclusion criteria, search strategies and quality assessment methods. |
| 4 | Include all relevant studies | All relevant studies, regardless of the results, should be included in the SLR to ensure that the findings are representative of the available evidence. |
| 5 | Provide details on data extraction and synthesis | Detailed information on data extraction and synthesis should be provided, including the methods used and any decisions made during the process. |
| 6 | Present results clearly and objectively | The results of the SLR should be presented clearly and objectively, with any limitations or sources of bias clearly identified. |
| 7 | Make all data and materials available | All data and materials used in the SLR should be made available to ensure that the findings can be independently verified. |
Example: In a computer science SLR investigating the effectiveness of machine learning algorithms for image classification, the following transparency measures could be taken:
- Clearly documenting the research question and inclusion/exclusion criteria
- Providing detailed information on the search strategies used to identify relevant studies
- Presenting a comprehensive summary of all studies included, including their methods, results, and limitations
- Making all data and materials used in the SLR, such as the search terms, inclusion and exclusion criteria, and data extraction forms, available to others.
In conclusion, reporting the results of an SLR is an important step in the process, and requires careful planning and attention to detail. By organizing the information in a logical and easy-to-follow manner, including key information, and ensuring transparency, researchers can ensure that their SLR makes a valuable contribution to the field and that their findings are accessible and understandable to a wide audience.
Writing the Conclusion
The conclusion of an SLR is an opportunity for the researchers to summarize the key findings of the review and to draw meaningful conclusions about the topic of interest. The conclusion should be based on the data and should be supported by the results of the studies included in the review.
i. Summarizing the Key Findings: The main conclusions of the review should be outlined in the conclusion, along with any patterns and trends found in the data and any points of agreement or disagreement in the literature. The reader will receive a clear and succinct summary of the body of research on the subject from this.
Here is a table summarizing the key steps involved in summarizing the key findings of a Systematic Literature Review (SLR) in computer science:
| Step | Description | Computer Science Example |
|---|---|---|
| 1 | Identify the key findings from the studies | In a SLR investigating the use of machine learning algorithms in image recognition, the key findings might include: the most commonly used algorithms, the datasets they have been applied to, the accuracy rates achieved, and the limitations of the existing approaches. |
| 2 | Organize the findings into themes or categories | The findings from the studies in the SLR on machine learning algorithms for image recognition might be organized into themes such as: algorithms used for object detection, algorithms used for image classification, and algorithms used for object recognition. |
| 3 | Summarize the key findings for each theme | For each of the themes, the key findings could be summarized in a concise manner, for example: For object detection, convolutional neural networks (CNNs) and region-based convolutional neural networks (R-CNNs) were found to be the most commonly used algorithms. For image classification, the highest accuracy rates were achieved using deep learning algorithms such as ResNet and DenseNet. For object recognition, transfer learning was found to be an effective approach. |
| 4 | Provide a comprehensive overview of the key findings | A comprehensive overview of the key findings would summarize the key findings from each theme and present them in a manner that provides a clear picture of the state of the field. For example: In the SLR on machine learning algorithms for image recognition, the key findings indicate that CNNs and R-CNNs are the most commonly used algorithms for object detection, deep learning algorithms such as ResNet and DenseNet achieve the highest accuracy rates for image classification, and transfer learning is an effective approach for object recognition. |
By following these steps, researchers can effectively summarize the key findings of an SLR, providing valuable insights into the state of the field and the most promising directions for future research.
ii. Drawing Conclusions: The conclusion should then proceed to make inferences regarding the subject of interest in light of the information and findings from the reviewed studies. This can entail summarising the main conclusions, resolving any data limitations, and going over the conclusions’ consequences.
| Steps | Description | Computer Science Example |
|---|---|---|
| 1. Synthesizing Key Findings | Summarize the results of the studies included in the review and identify common themes or patterns. | In a systematic literature review of computer vision techniques for object recognition, the researcher may synthesize the key findings by summarizing the accuracy rates and limitations of each technique studied. |
| 2. Addressing Limitations | Acknowledge and address any limitations in the data, including limitations of the studies included in the review and any biases present in the results. | For example, in the computer vision review mentioned above, the researcher may address limitations such as small sample sizes or a lack of diversity in the datasets used for testing. |
| 3. Discussing Implications | Discuss the implications of the findings, including the impact on current understanding of the topic and future directions for research. | For example, in the computer vision review, the researcher may discuss the implications of the findings for object recognition in real-world applications and suggest future directions for improving accuracy and robustness. |
This table provides a step-by-step guide for drawing conclusions in a systematic literature review and gives a concrete example from the field of computer science.
iii. Making Recommendations:The conclusion may, if suitable, also contain suggestions for more study or application. For instance, if there is a gap in the literature, the conclusion can suggest that additional research be done to close it. Similar to this, the review’s findings may recommend that particular therapies be used in practise if it appears that they are more successful than others.
Here’s a tabular representation for the “Making Recommendations” step of a systematic literature review:
| Step | Description | Example from Computer Science |
|---|---|---|
| Making Recommendations | The conclusion should include recommendations for future research or practice based on the findings of the review. | If the systematic literature review on machine learning algorithms in computer vision finds that a particular algorithm consistently performs better than others, the conclusion may recommend that this algorithm be adopted in practice for solving computer vision problems. |
To sum up, the conclusion section of an SLR is a crucial step in the review process since it gives the researchers a chance to highlight the most relevant findings and formulate insightful judgements about the subject of interest. The researchers can contribute to expand our understanding of the subject and make a significant contribution to the field by summarising the important findings, coming to conclusions, and offering suggestions for further research or practise.
How do I deal with Missing Data in my SLR?
In an SLR, dealing with missing data depends on the extent and importance of the missing information. There are several approaches to handling missing data, including:
- Listwise deletion: This involves excluding any studies with missing data from the analysis. This approach is suitable when the amount of missing data is small.
- Imputation: This involves estimating missing values based on available data. There are several methods of imputation, including mean imputation, median imputation, and multiple imputation.
- Maximum likelihood estimation: This involves estimating the missing values based on a statistical model that takes into account the observed data and the assumptions about the missing data.
- Sensitivity analysis: This involves repeating the analysis with different assumptions about the missing data to assess the robustness of the results.
Ultimately, the choice of method for dealing with missing data will depend on the specifics of the SLR, the amount of missing data, and the importance of the missing information.
Pros and Cons of Conducting Systematic Literature Review
Pros of Systematic Literature Review:
- Comprehensive and In-Depth Analysis: SLR provides a comprehensive and in-depth analysis of the existing literature on a specific topic, allowing for a thorough understanding of the current state of research in the field.
- Evidence-Based: SLR is based on rigorous and systematic methods, ensuring that the results are evidence-based and reliable.
- Enhances Research Credibility: By following a systematic and structured approach, SLR enhances the credibility and rigor of research, making the results more convincing and trustworthy.
- Improves Research Quality: By critically evaluating existing literature, SLR helps to identify gaps in the research, improving the overall quality of the research.
- Increases Research Productivity: By synthesizing existing literature and identifying trends and patterns, SLR can increase research productivity by reducing the time and effort needed to conduct new research.
Cons of Systematic Literature Review:
- Time-Consuming: SLR is a time-consuming process, requiring a significant investment of time and effort to complete.
- Resource Intensive: Conducting an SLR requires significant resources, including access to databases, research papers, and other materials.
- Limited Coverage: SLR may only be applicable to certain types of research, limiting its usefulness in other areas.
- Potential Bias: SLR may be subject to bias, especially if the selection of studies is not done systematically and objectively.
- Costly: Conducting an SLR can be expensive, especially if it requires the use of proprietary databases or other paid resources.
| Pros | Cons |
|---|---|
| Comprehensive and In-Depth Analysis | Time-Consuming |
| Evidence-Based | Resource Intensive |
| Enhances Research Credibility | Limited Coverage |
| Improves Research Quality | Potential Bias |
| Increases Research Productivity | Costly |
Reference Papers for Systematic Literature Review
You can refer to the following Systematic Literature Review papers for further understanding.
Boran Sekeroglu *,†,Rahib Abiyev †,Ahmet Ilhan †,Murat Arslan † and John Bush Idoko †
2. A systematic literature review on machine learning applications for consumer sentiment analysis
using online reviews :Praphula Kumar Jain a, Rajendra Pamula a, Gautam Srivastava
Conclusion
In conclusion, writing a systematic literature review (SLR) is a comprehensive and rigorous process that requires careful planning and attention to detail. The six key steps involved in conducting an SLR are: Defining the research question, Searching for relevant studies, Selecting studies, Assessing the quality of the studies, Synthesizing the results, and Reporting the results.
Each step of the process is critical to the success of the review and requires careful consideration and attention to detail. By following these steps and paying close attention to the details, researchers can ensure that their SLR makes a valuable contribution to the field and provides a comprehensive and up-to-date overview of the existing literature on the topic of interest.
In addition, by reporting the results in a clear and concise manner, including key information, and ensuring transparency, researchers can help to increase the credibility and reliability of their findings, and make their results accessible and understandable to a wide audience.
