Literature Review

(Source:http://www.sheldoncomics.com/archive/100806.html)

This guide is designed to offer guidance for completing a literature review, and will link you to resources, techniques, and advanced approaches to conducting and writing a literature review.

 

A literature review is a systematic review of the published literature on a specific topic or research question.  The literature review is designed to analyze-- not just summarize-- scholarly writings that are related directly to your research question.  That is, it represents the literature that provides background information on your topic and shows a correspondence between those writings and your research question.

This guide is designed to be a general resource for those completing a literature review in their field.

Different projects involve different kinds of literature reviews with different kinds and amounts of work. And, of course, the "end products" vary.

• Honors paper
• Capstone project
• Research Study
• Senior thesis
• Masters thesis
• Doctoral dissertation
• Research article
• Grant proposal
• Evidence based practice

A literature review is important because it:

• Explains the background of research on a topic.
• Demonstrates why a topic is significant to a subject area.
• Helps focus your own research questions or problems
• Discovers relationships between research studies/ideas.
• Suggests unexplored ideas or populations
• Identifies major themes, concepts, and researchers on a topic.
• Tests assumptions; may help counter preconceived ideas and remove unconscious bias.
• Identifies critical gaps, points of disagreement, or potentially flawed methodology or theoretical approaches.
• Indicates potential directions for future research.

One of the hardest parts of a literature review is to develop a good research question.  You don't want a research question that is so broad it encompasses too many research areas, and can't be reasonably answered. 

Defining your topic may require an initial review of literature on your topic to get a sense of the scope about your topic.   Select a topic of interest, and do a preliminary search to see what kinds of research is being done and what is trending in that topic area.  This will give you a better sense of the topic, and help you focus your research question

In specifying your topic or research question, you should think about setting appropriate limitations on the research you are seeking. Limiting, for example, by time, personnel, gender, age, location, nationality etc. results in a more focused and meaningful topic. 

Writing a literature review will take time to gather and analyze the research relevant to your topic, so it best to start early and give yourself enough time to gather and analyze your sources.  The process of writing a literature review usually covers the following steps:

1. Define your Research question
2. Plan your approach to your research and your review
3. Search the Literature
4. Analyze the material you’ve found
5. Managing the results of your research
6. Writing your Review

• Make a list of the databases which needs to be searched for finding literature
• Remember to include comprehensive databases such as Web of Science, SCOPUS, Google Scholar, WorldCat and Dissertations & Theses, if you need to.

 Where to find databases:

• Find Databases by Subject
• Central Library Databases categorized by discipline
• Find Databases via Research Guides
• Librarians create research guides for all of the disciplines on campus! Take advantage of their expertise and see what discipline-specific search strategies they recommend!

• Review the abstracts of research studies carefully. This will save you time.
• Write down the searches you conduct in each database so that you may duplicate them if you need to later (or avoid dead-end searches that you'd forgotten you'd already tried).
• Use the bibliographies and references of research studies you find to locate others.
• Ask your professor or a scholar in the field if you are missing any key works in the field.
• Use RefWorks to keep track of your research citations. See the RefWorks Tutorial if you need help.

Some questions to help you analyze the research:

• What was the research question of the study you are reviewing? What were the authors trying to discover?
• Was the research funded by a source that could influence the findings?
• What were the research methodologies? Analyze its literature review, the samples and variables used, the results, and the conclusions. Does the research seem to be complete? Could it have been conducted more soundly? What further questions does it raise?
• If there are conflicting studies, why do you think that is?
• How are the authors viewed in the field? Has this study been cited?; if so, how has it been analyzed?

Tips:     

• Again, review the abstracts carefully. 
• Keep careful notes so that you may track your thought processes during the research process.

There are seven basic steps in the search process:

1. Identifying the key words and phrases that reflect the key concepts of your research topic.
2. Formatting these key words and phrases using techniques such as phrase searching and truncation that make the most of them.
3. Turning these key words and phrases into effective searches using a few easy to master techniques.
4. Using the filters and limits in databases to optimise your searches.
5. Reviewing/evaluating your search results.
6. Making the most of your search results – by using the information in database records and article reference lists to find other resources.
7. Search again – Searching is not a linear process. And it is not enough to do just one search. You will need separate searches for each aspect of your topic. You will also need to repeat your searches in multiple databases. As you continue to search and read the literature related to your topic, you will find that you need to modify your searches to include the other keywords you come across, or other aspects of the topic you need to investigate.

Start by writing out your research question. You need to filter out the words that are not going to be useful for your search. Search engines will search for exactly what you type into them, including words like "the". You need to make sure you're only using the words that represent your key concepts.

1. Underline the words that tell you what to do with the information once you've found it, evaluated it and thought about it. They're relevant to the writing you will do later. You don't need them now.
2. Circle any limits in the question. Some examples of these are: geographical locations, periods in history, demographic groups, specific types of clinical tests, or the date range for the literature you are searching. These will help you to refine your search. Some of these limits will be used as keywords. Some will be filters in a database.
3. Now look at the words that are left. Highlight the words and phrases that represent the key concepts you are searching for information on.
4. Draw a table. Write the word or phrase for one concept at the top of each column.
5. If you know of any alternate ways of expressing a concept, write those underneath that concept. This includes alternate spellings and acronyms. Don't worry if you can't think of any alternate words at this stage. You'll find more as you search. This is just a starting point.

Note: If you are studying health, medical or biomedical sciences, you will also need to think about the MeSH terms for your keywords. The Using Health and Medical Sciences guide has information on how to do this.
 

After you identify the key concepts, you need to find information on, you can use these techniques to maximise the potential of each of the words and phrases:

Phrase searching: Sometimes an idea is represented by a group of two or more words. In searching this is referred to as a phrase. To search for a phrase, use double quotation marks around the words, e.g. "ozone depletion". This tells the database to search for occurrences of this specific group of words in exactly this order.

Truncation: When there are various forms of a word, you can cut it back to the root word and add a truncation symbol, e.g. pollut* will search for pollution, pollute, pollutes, polluter, polluters, polluted, polluting, pollutant, pollutants. It's the short version of (pollution OR pollute OR pollutes OR polluter OR polluters OR polluted OR polluting OR pollutant OR pollutants). Truncation symbols vary slightly between databases, so use the 'Help' or 'Search Tips' options to check which one you need.

Wildcards: There may be variations in the spelling of words, e.g. British English and American English have different spellings for some words. If you search with only one spelling, you will miss the relevant results with the alternate spelling. A wildcard character is a symbol that can be used to replace a letter within a word, e.g. organi?ation will search for both organisation and organization. 

Plural terms: Check how the database you are using searches for single/plural versions of keywords - this can have a huge impact on your results, as some databases will automatically search for the plural version of a singular term, but not the reverse 

Keywords can be formatted and combined into searches using specific words and symbols. You can use these techniques to maximise the effectiveness of your searches and get better quality results.

AND – is used to combine words for different concepts, e.g. "ozone layer" AND pollution. It tells the database to find results where all the of the words appear. It narrows your search.

OR – is used to add synonyms or similar concepts to the search. It tells the database to find results where one of the words or phrases appears, e.g. effect OR impact OR consequences OR repercussions. It broadens your search.

NOT – is used to exclude terms you don't want to find. It narrows your search.

brackets () – are used when you are using both AND and OR in a basic search. Because there is only one search box in a basic search, brackets are needed to group the synonyms that are combined using OR. The other words that have been combined with AND go outside the brackets, e.g. (effect OR impact OR consequences OR repercussions) AND "ozone layer" AND pollution. This tells the database to find at least one of the words or phrases from within the brackets as well as all of the words that are outside the brackets. It is a way of doing multiple searches at the same time.
 

Most databases provide a range of options which enable you to refine your search results by manipulating specific elements of your search. Each database has it's own options for refining the search, the following are common to most databases:
Basic limits There are basic limits that usually just require you to tick a checkbox, or select/enter dates. These are usually near the top of the menu and include:

• Publication Date
• Peer-reviewed content
• Type of Publication

Article type Some databases allow you to limit your search to a specific type of document, e.g:

• Academic journal articles
• Case studies
• Conference papers
• Technical papers
• Reports
• Review articles

Specific fields All databases organise information about their documents into specific fields. You can use theses fields to build much more accurate searches. If you go to the Advanced Search you'll find a dropdown list of available fields next to each search box. Enter your words into the box and select the field you want to search for them in. Examples of searchable fields include:

• Author
• Title – this is the article title
• Subject
• Abstract – this is the summary of the article.
• Title, Abstract and Keywords
• All except full text – this is all fields of the record but not the full article. It includes the abstract.
• Publication Name – this is the journal title

You need to review your search results both during, and after, your searches. This helps to ensure that your results are relevant and comprehensive.

During Your Searches:

• Think about the number of results. New areas of research may only have a few dozen articles published whilst a well-established area will have hundreds (or more). Generally, aim for results lists with no more than 100-150 articles per search - larger numbers will be difficult to work through, and may indicate that your search has not focused closely enough on specific aspects of your topic.
• Briefly review each article. Don't rely on the article title. Read the abstract to get a clearer picture of what is covered in the article. This will help you short-list the articles that are most likely to be relevant. You will still need to read the full article for everything in your short-list, to determine whether you can use it or not, but you will have a smaller number to read in full.
• As you work through the articles or thesis, you might like to check whether you have a balance between research and review articles, and recent and older research. The preferred balance will vary according to the nature of the topic.

Databases are designed to be mined for information, so make the most of them and get as much information as you can from your search results:

• Once you have located a relevant citation, look at the complete record to see if there are other terms listed which might be useful for searching. Depending on the database you are using, these may be called subject headings, descriptors, concepts, codes etc.
• Check the reference lists of relevant articles for other relevant citations.
• If you have identified significant researcher/s, try an Author search on their name/s.
• If you are using the SCOPUS database, use the cited by links to see how other authors have utilised a particular article. For Web of Science use the Times cited links.
• If a particular search has yielded good results, consider creating an alert for it. This instructs the database to run the search automatically on a regular basis, and notify you when new articles are published in your area of interest.

Searching is not a linear process, and it is not enough to do just one search. You will need separate searches for each aspect of your topic. You will also need to repeat your searches in multiple databases. As you continue to search and read the literature related to your topic, you will find that you need to modify your searches to include the other keywords you come across, or other aspects of the topic you need to investigate. After following the above steps, if the results are not satisfatory, review the process and conduct search again.

Literature Databases

• Web of Science
• SCOPUS
• PubMed

Available Database:

• SCIFinder Scholar

Available Business Databases

• EBSCHost Business Source Premier
• EMIE Economic Outlook
• CNKI
• IndiaStat
• Sage Business Cases

Available Legal Databases

• Manupatra

Available Databases:

• MathSciNet

After you have located your literature, read it, analyzed and evaluated it, it’s time to embark on the essential next stage of actually writing it up. It’s always worthwhile to lavish care on the actual writing of your literature review.
 
Some tips:
 

• Keep your audience in mind as you write your literature review. Your writing should be pitched at the level of expected readers. Use the terminology appropriate to them, i.e. physics terms for physicists; sociology terms for sociologists.
• If you are writing for the ordinary reader, avoid all jargon. Generally, "plain English" is the best strategy.
• It’s usually a good idea to keep your paragraphs short.
• Subheadings should be used to clarify the structure. They break up the material into more readable units as well as give the reader a place to "dive in" if she doesn't want to read all of the material.
• It’s often a good idea to write the first draft straight through and quickly – this can help preserve continuity and give coherence. Once you have text down on paper (or on a computer) it’s often far easier to make needed revisions.

Some common errors include:

• a failure to focus by going off on tangents;
• failure to cite essential pertinent studies;
• failure to maintain a coherent, logical flow;
• weak organization;
• poor language, grammar etc.
• Use direct quotation sparingly and judiciously. Paraphrasing writers' works is often preferable to quoting direct passages.
• Be prudent in the number of studies you discuss and cite. Referring to almost everything on the subject is useless.
• Don't cite references that you haven't read.
• A review is NOT a group of linked abstracts, one per paragraph.

At the end of the review the reader, captivated by both the style and content, should be able to declare: “This is precisely the study that must be carried out now to advance the needed research in this field.”

Traditional or Narrative literature Review

• Critiques and summarizes a body of literature
• Draws conclusions about the topic
• Identifies gaps or inconsistencies in a body of knowledge
• Requires a sufficiently focused research question

        Weaknesses:

• A large number of studies may make it difficult to draw conclusions
• The process is subject to bias that supports the researcher's own work.

Systematic Literature Review

• More rigorous and well-defined approach
• Published and unpublished studies relating to a particular subject area
• Details the time frame within which the literature was selected
• Details the methods used to evaluate and synthesize findings of the studies in question
• Check Right Panel for PRISMA Framework of Systamatic Review

Meta-analysis

• A form of systematic review (reductive)
• Takes findings from several studies on the same subject and analyzes them using standardized statistical procedures
• Integrates findings from a large body of quantitative findings to enhance under-standing (study=unit of analysis)
• Draws conclusions and detect patterns and relationships

Meta-synthesis

• Non-statistical technique
• Integrates, evaluates and interprets findings of multiple qualitative research studies
• Identifies common core elements and themes
• May use findings from phenomenological, grounded theory or ethnographic studies
• Involves analyzing and synthesizing key elements

The PRISMA framework or “Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)”, is a set of guidelines or steps developed by Moher et al. (2009). These steps are useful for systematic literature reviews, critical literature analyses and meta-analyses. The PRISMA tool or framework uses a set of methods to systematically search papers and literature for review-based studies. In addition, PRISMA is also based on the formulated inclusion and exclusion criteria in any type of study that systematically assesses the quality of chosen papers and either includes for the study or excludes for the study.

PRISMA is used mostly in medical and pharmaceutical researches, however, in recent years the use of PRISMA is also done in management studies and other non-medical studies.
There are a few important things to note while formulating PRISMA. The PRISMA is based on the 27-item quality checklist broadly divided into:

• Title,
• Abstract,
• Introduction,
• Methods,
• Results,
• Discussion and,
• Funds (Moher, 2013).

It is not mandatory to use all the checklist items; however, the inclusion and exclusion criteria are mostly based on the checklist items. The main role of the PRISMA checklist is to identify the quality of the papers identified using search strategies and the databases. In addition, the flow chart of PRISMA is based on 4 stages:

• identification,
• screening,
• eligibility and,
• included.

  

The PRISMA framework or “Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)”, is a set of guidelines or steps developed by Moher et al. (2009). These steps are useful for systematic literature reviews, critical literature analyses and meta-analyses. The PRISMA tool or framework uses a set of methods to systematically search papers and literature for review-based studies. In addition, PRISMA is also based on the formulated inclusion and exclusion criteria in any type of study that systematically assesses the quality of chosen papers and either includes for the study or excludes for the study.

PRISMA is used mostly in medical and pharmaceutical researches, however, in recent years the use of PRISMA is also done in management studies and other non-medical studies.
There are a few important things to note while formulating PRISMA. The PRISMA is based on the 27-item quality checklist broadly divided into:

• Title,
• Abstract,
• Introduction,
• Methods,
• Results,
• Discussion and,
• Funds (Moher, 2013).

It is not mandatory to use all the checklist items; however, the inclusion and exclusion criteria are mostly based on the checklist items. The main role of the PRISMA checklist is to identify the quality of the papers identified using search strategies and the databases. In addition, the flow chart of PRISMA is based on 4 stages:

• identification,
• screening,
• eligibility and,
• included.

Section/topic	Sl. No.	Checklist item	Level of importance
Title	1	Identify the report as a systematic review, meta-analysis, or both, or the type of paper and the title scope.	High
Abstract	2	Clarity of the study with; background, aim, methods used, findings, conclusion and other relevant information.	High
Introduction	3	Must have a rationale.	High
	4	Must mention the objectives of the study	High
Methods	5	Protocol registration.	Low
	6	Eligibility criteria mentioned.	Moderate
	7	Information sources or the data collection sources.	High
	8	Search strategies or the research paradigm/philosophy.	Moderate
	9	Study selection or the research approach.	Moderate
	10	Data collection process or the sample size.	High
	11	Data items.	Moderate
	12	Risk of bias in individual studies.	Low
	13	Summary measures or the tools of analysis	High
	14	Synthesis of results or the methods of analysis	High
	15	Risk of bias across studies	Low
	16	Additional analyses	Low
Results	17	Study selection or the key findings	High
	18	Study characteristics or statistical presentations	High
	19	Risk of bias within studies	Low
	20	Results of individual studies or graphs and tabular presentations	High
	21	Synthesis of results	Low
	22	Risk of bias across studies	Low
	23	Additional analysis	Low
Discussion	24	Summary of evidence	Moderate
	25	Limitations	High
	26	Conclusions	High
Funding	27	Information about funding or author conflict	Low

Step 1: Identify the papers on the basis of search strategies and the relevant databases used for searching and collecting research papers. Most common databases of literature include:

• Google Scholar,
• Embase,
• IEEE Xplore,
• EBSCO,
• Cochrane and,
• PubMed.

The first step does not involve inclusion or exclusion of papers.

Step 2: Screening the papers will involve using the quality checklist in stages. The first screening uses both the inclusion or exclusion criteria and the quality checklist to exclude the irrelevant papers.
 
Step 3: Check if the title of the papers found are corroborating the current study. In addition, include or exclude papers on the basis of availability of full text papers.
 
Step 4: Then check if the papers are in their specified language. Most scholars use English; however other may use their local languages like French, Japanese, Spanish and others. Inclusion or exclusion of papers are in this steps also seeks identify duplicate papers. This step only checks the 1st and 2nd checklist item.
 
Step 5: The main and the most important step is Eligibility of the papers. Prioritize to use the quality checklist in finding the eligibility of the papers. Use all or some of the checklist items on the basis of the criticalness of the papers.
 
Step 6: The last step involves critically reviewing the papers to address the aim of the current study.