SEARCHING THE LITERATURE
Researchers almost never conduct a study in an intellectual vacuum; their studies are usually undertaken within the context of an existing knowledge base. Researchers undertake a literature review to familiarize themselves with that knowledge base—although, some qualitative researchers deliberately bypass an in-depth literature search before entering the field to avoid having their inquiries constrained or biased by prior work on the topic. This chapter discusses the functions that a literature review can play in a research project and the kinds of material covered in a literature review.
Suggestions are provided on finding references, reading research reports, recording information, and organizing and drafting a written review. Because research reports are not always easy to digest, a section of this chapter offers suggestions on reading them.
Purposes of a Literature Review
Literature reviews can serve a number of important functions in the research process—as well as important functions for nurses seeking to develop an evidence-based practice. For researchers, acquaintance with relevant research literature and the state of current knowledge can help with the following:
• Identification of a research problem and development or refinement of research questions or hypotheses
• Orientation to what is known and not known about an area of inquiry, to ascertain what research can best make a contribution to the existing base of evidence
• Determination of any gaps or inconsistencies in a body of research
• Determination of a need to replicate a prior study in a different setting or with a different study population
• Identification or development of new or refined clinical interventions to test through empirical research
• Identification of relevant theoretical or conceptual frameworks for a research problem
• Identification of suitable designs and data collection methods for a study
• For those developing research proposals for funding, identification of experts in the field who could be used as consultants
• Assistance in interpreting study findings and in developing implications and recommendations
A literature review helps to lay the foundation for a study, and can also inspire new research ideas. A literature review also plays a role at the end of the study, when researchers are trying to make sense of their findings. Most research reports include summaries of relevant literature in the introduction. A literature review early in the report provides readers with a background for understanding current knowledge on a topic and illuminates the significance of the new study. Written research reviews are also included in research proposals that describe what a researcher is planning to study and how the study will be conducted.
Of course, research reviews are not undertaken exclusively by researchers. Both consumers and producers of nursing research need to acquire skills for reviewing research critically. Nursing students, nursing faculty, clinical nurses, nurse administrators, and nurses involved in policy-making organizations also need to review and synthesize evidence-based information.
Scope Of A Literature Search
You undoubtedly have some skills in locating and organizing information. However, a review of research literature differs in many respects from other kinds of term papers that students prepare. In this section, the type of information that should be sought in conducting a research review is examined, and other issues relating to the breadth and depth of the review are considered—including differences among the main qualitative research traditions.
Types of Information to Seek. Written materials vary considerably in their quality, their intended audience, and the kind of information they contain. Researchers performing a review of the literature ordinarily come in contact with a wide range of material and have to decide what to read or what to include in a written review. We offer some suggestions that may help in making such decisions.
The appropriateness of a reference concerns both its content (i.e., its relevance to the topic of the review) and the nature of the information it contains. The most important type of information for a research review are findings from empirical investigations. Cumulatively, research reports sum up what is known on a topic, but the information from such reports is of greatest value when the findings are integrated in a critical synthesis.
For a literature review, you should rely mostly on primary source research reports, which are descriptions of studies written by the researchers who conducted them. Secondary source research documents are descriptions of studies prepared by someone other than the original researcher. Literature review summaries, then, are secondary sources.
Such reviews, if they exist and are recent, are an especially good place to begin a literature search because they provide a quick summary of the literature, and the bibliography is helpful. For many clinical topics, the reviews prepared by the Cochrane Collaboration are a particularly good resource.
However, secondary descriptions of studies should not be considered substitutes for primary sources for a new literature review. Secondary sources typically fail to provide much detail about studies, and they are seldom completely objective.
Our own values and biases are a filter through which information passes (although we should make efforts to control such biases), but we should not accept as a second filter the biases of the person who prepared a secondary source summary of research studies.
Examples of primary and secondary sources:
• Secondary source, a review of the literature on patient experiences in the ICU: Stein-Parbury, J. & McKinley, S. (2000). Patients’ experiences of being in an intensive care unit: A select literature review. American Journal of Critical Care, 9, 20–27.
• Primary source, an original qualitative study on patient experiences in the ICU: Hupcey, J. E. (2000). Feeling safe: The psychosocial needs of ICU patients. Journal of Nursing Scholarship, 32, 361–367.
In addition to locating empirical references, you may find in your search various nonresearch references, including opinion articles, case reports, anecdotes, and clinical descriptions. Some qualitative researchers also review relevant literary or artistic work, to gain insights about human experiences.
Such materials may serve to broaden understanding of a research problem, illustrate a point, demonstrate a need for research, or describe aspects of clinical practice. Such writings may thus may play a very important role in formulating research ideas—or may even suggest ways to broaden or focus the literature search—but they usually have limited utility in written research reviews because they are subjective and do not address the central question of written reviews: What is the current state of knowledge on this research problem?
Depth and Breadth of Literature Coverage. Some students worry about how broad their literature search should be. Of course, there is no convenient formula for the number of references that should be tracked down, or how many pages the written review should be. The extensiveness of the review depends on a number of factors. For written reviews, a major determinant is the nature of the document being prepared. The major types of research reviews include the following:
• A review included in a research report. As we discuss later in this chapter, research reports published in journals usually include brief literature reviews in their introductions. These reviews are succinct and have two major goals: to provide readers with a quick overview of the state of knowledge on the research problem being addressed; and to document the need for the new study and demonstrate how it will contribute to existing evidence. These reviews are usually only two to four double-spaced pages, and therefore only a limited number of references can be cited. This does not mean, of course, that researchers have not conducted a more thorough review, but rather that they are summarizing only what readers need to know to understand the study context. (If the report is published in a book or other format, the literature review section may be longer.)
• A review included in a research proposal. Research proposals designed to persuade funders (or advisors) about the merits of a proposed study usually include a literature review section. As with a review in a research report, a review in a proposal provides a knowledge context and confirms the need for and significance of new research. In a proposal, however, a review also demonstrates the writer’s command of the literature. The length of such reviews may be established in proposal guidelines, but they are often 5 to 10 pages long.
• A review in a thesis or dissertation. Doctoral dissertations often include a thorough review covering materials directly and indirectly related to the problem area. Often, an entire chapter is devoted to a summary of the literature, and such chapters are frequently 15 to 25 pages in length.
• Free-standing literature reviews. Increasingly, nurses are preparing literature reviews that critically appraise and summarize a body of research on a topic, and such reviews play a powerful role in the development of an evidence-based practice. Students are sometimes asked to prepare a written research review for a course, and nurses sometimes do literature reviews as part of utilization projects. Researchers who are experts in a field also may do integrative reviews that are published as journal articles or that contribute to major evidence-based practice projects. Free-standing literature reviews designed to appraise a body of research critically are usually at least 15 to 25 pages long.
The breadth of a literature review also depends on the topic. For some topics, it may be necessary to review research findings in the non-nursing literature, such as sociology, psychology, biology, or medicine. Breadth of the review may also be affected by how extensive the research on the topic has been. If there have been 15 published studies on a specific problem, it would be difficult to draw conclusions about the current state of knowledge on that topic without reading all 15 reports. However, it is not necessarily true that the literature task is easier for little-researched topics. Literature reviews oew areas of inquiry may need to include studies of peripherally related topics to develop a meaningful context. Relevance and quality are the key criteria for including references in a written review of the literature.
With respect to depth in describing studies in a written review, the most important criteria are relevance and type of review. Research that is highly related to the problem usually merits more detailed coverage. Studies that are only indirectly related can often be summarized in a sentence or two, or omitted entirely if there are page restrictions.
Literature Reviews in Qualitative Research Traditions. Qualitative researchers have different views about reviewing the literature as part of a new study. Some of the differences reflect viewpoints associated with various qualitative research traditions.
In grounded theory studies, researchers typically collect data in the field before reviewing the literature.
As the data are analyzed, the grounded theory begins to take shape. Once the theory appears to be sufficiently developed, researchers then turn to the literature, seeking to relate prior findings to the theory. Glaser (1978) warns that “It’s hard enough to generate one’s own ideas without the ‘rich’ detailment provided by literature in the same field” (p. 31). Thus, grounded theory researchers defer the literature review, but then determine how previous research fits with or extends the emerging theory.
Phenomenologists often undertake a search for relevant materials at the outset of a study. In reviewing the literature for a phenomenological study, researchers look for experiential descriptions of the phenomenon being studied (Munhall, 2001).
The purpose is to expand the researcher’s understanding of the phenomenon from multiple perspectives. Van Manen (1990) suggests that, in addition to past research studies, artistic sources of experiential descriptions should be located such as poetry, novels, plays, films, and art. These artistic sources can offer powerful examples and images of the experience under study.
Even though “ethnography starts with a conscious attitude of almost complete ignorance” (Spradley, 1979, p. 4), a review of the literature that led to the choice of the cultural problem to be studied is often done before data collection. Munhall (2001) points out that this literature review may be more conceptual than data-based. A second, more thorough literature review is often done during data analysis and interpretation so that findings can be compared with previous literature.
Locating Relevant Literature For A Research Review
The ability to identify and locate documents on a research topic is an important skill that requires adaptability—rapid technological changes, such as the expanding use of the Internet, are making manual methods of finding information from print resources obsolete, and more sophisticated methods of searching the literature are being introduced continuously. We urge you to consult with librarians at your institution or to search the Internet for updated information.
One caveat should be mentioned. You may be tempted to do a search through an Internet search engine, such as Yahoo, Google, or Alta Vista. Such a search might provide you with interesting information about interest groups, support groups, advocacy organizations, and the like. However, such Internet searches are not likely to give you comprehensive bibliographic information on the research literature on your topic—and you might become frustrated with searching through the vast number of websites now available.
TIP: Locating all relevant information on a research question is a bit like being a detective. The various electronic and print literature retrievaltools are a tremendous aid, but there inevitably needs to be some digging for, and a lot of sifting and sorting of, the clues to knowledge on a topic. Be prepared for sleuthing! And don’t hesitate to ask your reference librarians for help in your detective work.
Electronic Literature Searches. In most college and university libraries, students can perform their own searches of electronic databases—huge bibliographic files that can be accessed by computer. Most of the electronic databases of interest to nurses can be accessed either through an online search (i.e., by directly communicating with a host computer over telephone lines or the Internet) or by CD-ROM (compact disks that store the bibliographic information). Several competing commercial vendors offer information retrieval services for bibliographic databases.
Currently, the most widely used service providers for accessing bibliographic files are the following:
• Aries Knowledge Finder (www.ariessys.com)
• Ebsco Information Services (www.ebsco.com)
• Ovid Technologies (www.ovid.com)
• PaperChase (www.paperchase.com)
• SilverPlatter Information (www.silverplatter.com)
All of these services provide user-friendly retrieval of bibliographic information—they offer menu-driven systems with on-screen support so that retrieval can usually proceed with minimal instruction. However, the services vary with regard to a number of factors, such as number of databases covered, cost, online help, ease of use, special features, methods of access, and mapping capabilities. Mapping is a feature that allows you to search for topics in your own words, rather thaeeding to enter a term that is exactly the same as a subject heading in the database.
The vendor’s software translates (“maps”) the topic you enter into the most plausible subject heading.
TIP: Even when there are mapping capabilities, it may prove useful in your search to learn the subject headings of the database or the key words that researchers themselves identify to classify their studies. Subject headings for databases can be located in the database’s thesaurus. A good place to find key words is in a journal article once you have found a relevant reference.
Several other electronic resources should be mentioned. First, books and other holdings of libraries can almost always be scanned electronically using online catalog systems. Moreover, through the Internet, the catalog holdings of libraries across the country can be searched. Finally, it may be useful to search through Sigma Theta Tau International’s Registry of Nursing Research on the Internet. This registry is an electronic research database with over 12,000 studies that can be searched by key words, variables, and researchers’ names. The registry provides access to studies that have not yet been published, which cuts down the publication lag time; however, caution is needed because these studies have not been subjected to peer review (i.e., critical review by other experts in the field). Electronic publishing in general is expanding at a rapid pace; librarians and faculty should be consulted for the most useful websites.
TIP: It is rarely possible to identify all relevant studies exclusively through automated literature retrieval mechanisms. An excellent method of identifying additional references is to examine citations in recently published studies or published literature reviews.
Key Electronic Databases for Nurse Researchers. The two electronic databases that are most likely to be useful to nurse researchers are CINAHL (Cumulative Index to Nursing and Allied Health Literature) and MEDLINE® (Medical Literature On-Line). Other potentially useful bibliographic databases for nurse researchers include:
• AIDSLINE (AIDS Information On-Line)
• CancerLit (Cancer Literature)
• CHID (Combined Health Information Database)
• EMBASE (the Excerpta Medica database)
• ETOH (Alcohol and Alcohol Problems Science Database)
• HealthSTAR (Health Services, Technology, Administration, and Research)
• PsycINFO (Psychology Information)
• Rndex (Nursing and managed care database)
The CINAHL Database. The CINAHL database is the most important electronic database for nurses. This database covers references to virtually all English-language nursing and allied health journals, as well as to books, book chapters, nursing dissertations, and selected conference proceedings iursing and allied health fields. References from more than 1200 journals are included in CINAHL. The CINAHL database covers materials dating from 1982 to the present and contains more than 420,000 records. In addition to providing bibliographic information for locating references (i.e., the author, title, journal, year of publication, volume, and page numbers), this database provides abstracts (brief summaries) of articles for more than 300 journals. Supplementary information, such as names of data collection instruments, is available for many records in the database. Documents of interest can typically be ordered electronically.
CINAHL can be accessed online or by CD-ROM, either directly through CINAHL or through one of the commercial vendors cited earlier.
Information about CINAHL’s own online service can be obtained through the CINAHL website (www.cinahl.com).
We will use the CINAHL database to illustrate some of the features of an electronic search. Our example relied on the Ovid Search Software for CD-ROM, but similar features are available through other vendors’ software.
Most searches are likely to begin with a subject search (i.e., a search for references on a specific topic). For such a search, you would type in a word or phrase that captures the essence of the topic (or the subject heading, if you know it), and the computer would then proceed with the search. An important alternative to a subject search is a textword search that looks for the words you enter in text fields of each record, including the title and the abstract. If you know the name of a researcher who has worked on the topic, an author search might be productive.
TIP: If you want to identify all major research reports on a topic, you need to be flexible and to think broadly about the key words that could be related to your topic. For example, if you are interested in anorexia nervosa, you might look under anorexia, eating disorders, and weight loss, and perhaps under appetite, eating behavior, food habits, bulimia, and body weight changes.
In a subject search, after you enter the topic the computer might give you a message through scope notes about the definition of the CINAHL subject heading so you could determine whether the mapping procedure produced the right match. For example, if you typed in the subject “baby blues,” the software would lead you to the subject heading “postpartum depression,” and the scope note would give the following definition: “Any depressive disorder associated with the postpartum period. Severity may range from a mild case of ‘baby blues’ to a psychotic state.” If this is the topic you had in mind, you would then learn the number of “hits” there are in the database for postpartum depression—that is, how many matches there are for that topic.
TIP: If your topic includes independent and dependent variables, you may need to do separate searches for each. For example, if you were interested in learning about the effect of health beliefs on compliance behaviors among patients with AIDS, you might want to read about health beliefs (in general) and about compliance behaviors (in general). Moreover, you might also want to access research on patients with AIDS and their circumstances.
In most cases, the number of hits in a subject search initially is rather large, and you will want to refine the search to ensure that you retrieve the most appropriate references. You can delimit retrieved documents in a number of ways. For example, you can restrict the search to those references for which your topic is the main focus of the document. For most subject headings, you also can select from a number of subheadings specific to the topic you are searching.
You might also want to limit the references to a certain type of document (e.g., only research reports); specific journal subsets (e.g., only ones published iursing journals); certain features of the document (e.g., only ones with abstracts); restricted publication dates (e.g., only those after 1999); certain languages (e.g., only those written in English); or certain study participant characteristics (e.g., only adolescents).
TIP: When searching a database through a commercial vendor, it is usually possible to combine searches to find only those references that include two or more topics. For example, if we were interested in the effect of stress on substance abuse, we could do independent searches for the two topics and then combine the searches to identify studies involving both variables.
To illustrate how searches can be delimited with a concrete example, suppose we were interested in recent research on brain injury, which is the term we enter in a subject search. Here is an example of how many hits there were on successive restrictions to the search, using the CINAHL database through June 2001:
Search Topic/Restriction Hits
Brain injuries 1459
Restrict to main focus 1263
Limit to research reports 481
Limit to nursing journals 28
Limit to 1999 through 2001 publications 12
This narrowing of the search—from 1459 initial references on brain injuries to 12 references for recent nursing research reports on brain injuries—took under 1 minute to perform. Next, we would display the information for these 12 references on the monitor, and we could then print full bibliographic information for the ones that appeared especially promising. An example of one of the CINAHL record entries retrieved through this search on brain injuries is presented in
. Each entry shows an accessioumber that is the unique identifier for each record in the database (the number that can be used to order the full text. Then, the authors and title of the reference are displayed, followed by source information. The source indicates the following:
• Name of the journal (Critical Care Nursing Quarterly)
• Volume (23)
• Issue (4)
• Page numbers (42–51)
• Year and month of publication (2001 Feb.)
• Number of cited references (23)
The printout also shows all the CINAHL subject headings that were coded for this particular entry; any of these headings could have been used in the subject search process to retrieve this particular reference. Note that the subject headings include both substantive/topical headings (e.g., brain injuries, quality of life) and methodologic headings (e.g., interviews). Next, when formal, named instruments are used in the study, these are printed under Instrumentation. Finally, the abstract for the study is presented.
Based on the abstract, we would then decide whether this reference was pertinent to our inquiry.
Once relevant references are identified, the full research reports can be obtained and reviewed. All the documents referenced in the database can be ordered by mail or facsimile (fax), so it is not necessary for your library to subscribe to the referenced journal.
Many of the retrieval service providers (such as Ovid) offer full text online services, so that, for certain journals, documents can be browsed directly, linked to other documents, and downloaded.
The MEDLINE®Database. The MEDLINE® database was developed by the U.S. National Library of Medicine (NLM), and is widely recognized as the premier source for bibliographic coverage of the biomedical literature. MEDLINE® incorporates information from Index Medicus, International Nursing Index, and other sources. MEDLINE® covers more than 4300 journals and contains more than 11 million records. In 1999, abstracts of reviews from the Cochrane Collaboration became available through MEDLINE® (they are also available directly on the internet at www.hcn.net.au/cochrane).
Because the MEDLINE® database is so large, it is often useful to access a subset of the database rather than the unabridged version that has references dating from 1966 to the present. For example, some subsets of the database cover only references within the previous 5 years. Other subsets include core medical journals, specialty journals, and nursing journals.
The MEDLINE® database can be accessed online or by CD-ROM through a commercial vendor (e.g., Ovid, Aries Knowledge Finder) for a fee. This database can also be accessed for free through the Internet at various websites, including the following:
• PubMed (http://www.ncbi.nlm.nih.gov/PubMed)
• Infotrieve (http://www4.infotrieve.com/newmedline/search.asp)
The advantage of accessing the database through commercial vendors is that they offer superior search capabilities and special features.
Print Resources. Print-based resources that must be searched manually are rapidly being overshadowed by electronic databases, but their availability should not be ignored. It is sometimes necessary to refer to printed resources to perform a search to include early literature on a topic. For example, the CINAHL database does not include references to research reports published before 1982.
Print indexes are books that are used to locate articles in journals and periodicals, books, dissertations, publications of professional organizations, and government documents. Indexes that are particularly useful to nurses are the International Nursing Index, Cumulative Index to Nursing and Allied Health Literature (the “red books”), Nursing Studies Index, Index Medicus, and Hospital Literature Index.
Indexes are published periodically throughout the year (e.g., quarterly), with an annual cumulative index. When using a print index, you usually first need to identify the appropriate subject heading.
Subject headings can be located in the index’s thesaurus. Once the proper subject heading is determined, you can proceed to the subject section of the index, which lists the actual references.
Abstract journals summarize articles that have appeared in other journals. Abstracting services are in general more useful than indexes because they provide a summary of a study rather than just a title. Two important abstract sources for the nursing literature are Nursing Abstracts and Psychological Abstracts.
TIP: If you are doing a completely manual search, it is a wise practice to begin the search with the most recent issue of the index or abstract journal and then to proceed backward. (Most electronic databases are organized chronologically, with the most recent references appearing at the beginning of a listing.)
Reading Research Reports
Once you have identified potential references, you can proceed to locate the documents. For research literature reviews, relevant information will be found mainly in research reports in professional journals, such as Nursing Research. Before discussing how to prepare a written review, we briefly present some suggestions on how to read research reports in journals.
What Are Research Journal Articles? Research journal articles are reports that summarize a study or one aspect of a complex study. Because journal space is limited, the typical research article is relatively brief—usually only 15 to 25 manuscript pages, double-spaced. This means that the researcher must condense a lot of information into a short space.
Research reports are accepted by journals on a competitive basis and are critically reviewed before acceptance for publication. Readers of research journal articles thus have some assurance that the studies have already been scrutinized for their scientific merit.
Nevertheless, the publication of an article does not mean that the findings can be uncritically accepted as true, because most studies have some limitations that have implications for the validity of the findings. This is why consumers as well as producers of research can profit from understanding research methods.
Research reports in journals tend to be organized in certain format and written in a particular style. The next two sections discuss the content and style of research journal articles.
Content of Research Reports. Research reports typically consist of four major sections (introduction, method section, results section, discussion section), plus an abstract and references.
The Abstract. The abstract is a brief description of the study. placed at the beginning of the journal article. The abstract answers, in about 100 to 200 words, the following questions:
What were the research questions?
What methods did the researcher use to address those questions?
What did the researcherfind? and
What are the implications for nursing practice? Readers can review an abstract to assess whether the entire report is of interest.
Some journals have moved from having traditional abstracts—which are single paragraphs summarizing the main features of the study—to slightly longer, more structured, and more informative abstracts with specific headings. For example, abstracts in Nursing Research after 1997 present information about the study organized under the following headings: Background, Objectives, Method, Results, Conclusions, and Key Words.
presents abstracts from two actual studies. The first is a “new style” abstract for a quantitative study entitled “Family reports of barriers to optimal care of the dying” (Tolle, Tilden, Rosenfeld, & Hickman, 2000). The second is a more traditional abstract for a qualitative study entitled “Families of origin of homeless and never-homeless women” (Anderson & Imle, 2001). These two studies are used as illustrations throughout this section.
The Introduction. The introduction acquaints readers with the research problem and its context. The introduction, which may or may not be specifically labeled “Introduction,” follows immediately after the abstract. This section usually describes the following:
• The central phenomena, concepts, or variables under study. The problem area under investigation is identified.
• The statement of purpose, and research questions or hypotheses to be tested. The reader is told what the researcher set out to accomplish in the study.
• A review of the related literature. Current knowledge relating to the study problem is briefly described so readers can understand how the study fits in with previous findings and can assess the contribution of the new study.
• The theoretical framework. In theoretically driven studies, the framework is usually presented in the introduction.
• The significance of and need for the study. The introduction to most research reports includes an explanation of why the study is important to nursing.
Thus, the introduction sets the stage for a description of what the researcher did and what was learned.
Examples from an introductory paragraph:
The homeless in the
In this paragraph, the researchers described the background of the problem, the population of primary interest (homeless women), and the study purpose.
The Method Section. The method section describes the method the researcher used to answer the research questions. The method section tells readers about major methodologic decisions, and may offer rationales for those decisions. For example, a report for a qualitative study often explains why a qualitative approach was considered to be especially appropriate and fruitful.
In a quantitative study, the method section usually describes the following, which may be presented as labeled subsections:
• The research design. A description of the research design focuses on the overall plan for the collection of data, often including the steps the researcher took to minimize biases and enhance the interpretability of the results by instituting various controls.
• The subjects. Quantitative research reports usually describe the population under study, specifying the criteria by which the researcher decided whether a person would be eligible for the study. The method section also describes the actual sample, indicating how people were selected and the number of subjects in the sample.
• Measures and data collection. In the method section, researchers describe the methods and procedures used to collect the data, including how the critical research variables were operationalized; they may also present information concerning the quality of the measuring tools.
• Study procedures. The method section contains a description of the procedures used to conduct the study, including a description of any intervention. The researcher’s efforts to protect the rights of human subjects may also be documented in the method section.
Table 5-1 presents excerpts from the method section of the quantitative study by Tolle and her colleagues (2000).
Qualitative researchers discuss many of the same issues, but with different emphases. For example, a qualitative study often provides more information about the research setting and the context of the study, and less information on sampling. Also, because formal instruments are not used to collect qualitative data, there is little discussion about data collection methods, but there may be more information on data collection procedures. Increasingly, reports of qualitative studies are including descriptions of the researchers’ efforts to ensure the trustworthiness of the data. Some qualitative reports also have a subsection on data analysis. There are fairly standard ways of analyzing quantitative data, but such standardization does not exist for qualitative data, so qualitative researchers may describe their analytic approach.
Table 5-2 presents excerpts from the method section of the study by Anderson and Imle (2001).
The Results Section. The results section presents the research findings (i.e., the results obtained in the analyses of the data). The text summarizes the findings, often accompanied by tables or figures that highlight the most noteworthy results.
Virtually all results sections contain basic descriptive information, including a description of the study participants (e.g., their average age). In quantitative studies, the researcher provides basic descriptive information for the key variables, using simple statistics. For example, in a study of the effect of prenatal drug exposure on the birth outcomes of infants, the results section might begin by describing the average birth weights and Apgar scores of the infants, or the percentage who were of low birth weight (under
In quantitative studies, the results section also reports the following information relating to the statistical analyses performed:
• The names of statistical tests used. A statistical test is a procedure for testing hypotheses and evaluating the believability of the findings. For example, if the percentage of low-birth-weight infants in the sample of drug-exposed infants is computed, how probable is it that the percentage is accurate? If the researcher finds that the average birth weight of drug-exposed infants in the sample is lower than that of nonexposed infants, how probable is it that the same would be true for other infants not in the sample? That is, is the relationship between prenatal drug exposure and infant birth weight real and likely to be replicated with a new sample of infants—or does the result reflect a peculiarity of the sample? Statistical tests answer such questions. Statistical tests are based on common principles; you do not have to know the names of all statistical tests (there are dozens) to comprehend the findings.
• The value of the calculated statistic. Computers are used to compute a numeric value for the particular statistical test used. The value allows the researchers to draw conclusions about the meaning of the results. The actual numeric value of the statistic, however, is not inherently meaningful and need not concern you.
• The significance. The most important information is whether the results of the statistical tests were significant (not to be confused with important or clinically relevant). If a researcher reports that the results are statistically significant, it means the findings are probably validand replicable with a new sample of subjects. Research reports also indicate the level of significance, which is an index of how probable it is that the findings are reliable. For example, if a report indicates that a finding was significant at the .05 level, this means that only 5 times out of 100 (5 100 [1] .05) would the obtained result be spurious or haphazard. In other words, 95 times out of 100, similar results would be obtained with a new sample. Readers can therefore have a high degree of confidence—but not total assurance—that the findings are reliable.
The Discussion Section. In the discussion section, the researcher draws conclusions about the meaning and implications of the findings. This section tries to unravel what the results mean, why things turned out the way they did, and how the results can be used in practice. The discussion in both qualitative and quantitative reports may incorporate the following elements:
• An interpretation of the results. The interpretation involves the translation of findings into practical, conceptual, or theoretical meaning.
• Implications. Researchers often offer suggestions for how their findings could be used to improve nursing, and they may also make recommendations on how best to advance knowledge in the area through additional research.
• Study limitations. The researcher is in the best position possible to discuss study limitations, such as sample deficiencies, design problems, weaknesses in data collection, and so forth. A discussion section that presents these limitations demonstrates to readers that the author was aware of these limitations and probably took them into account in interpreting the findings.
Example from a discussion section of a quantitative report:
Overall, one third of the sample of family respondents indicated moderate to severe decedent pain in the final week of life. Although this rate is somewhat better than rates reported elsewhere, it still raises concern that control of pain for dying patients is simply not good enough….
Interestingly, families had more complaints about the management of pain for decedents who died at home, even though they did not report higher levels of pain. Perhaps this is because in the home setting, family members are more aware of pain management problems and bear more responsibility for direct care of such needs (Tolle et al.,2000, p. 315).
References. Research journal articles conclude with a list of the books, reports, and journal articles that were referenced in the text of the report. For those interested in pursuing additional reading on a substantive topic, the reference list of a current research study is an excellent place to begin.
The Style of Research Journal Articles. Research reports tell a story. However, the style in which many research journal articles are written—especially reports of quantitative studies—makes it difficult for beginning research consumers to become interested in the story. To unaccustomed audiences, research reports may seem stuffy, pedantic, and bewildering. Four factors contribute to this impression:
1. Compactness. Journal space is limited, so authors try to compress many ideas and concepts into a short space. Interesting, personalized aspects of the investigation often cannot be reported. And, in qualitative studies, only a handful of supporting quotes can be included.
2. Jargon. The authors of both qualitative and quantitative reports use research terms that are assumed to be part of readers’ vocabulary, but that may seem esoteric.
3. Objectivity. Quantitative researchers normally avoid any impression of subjectivity and thus research stories are told in a way that makes them sound impersonal. For example, most quantitative research reports are written in the passive voice (i.e., personal pronouns are avoided). Use of the passive voice tends to make a report less inviting and lively than the use of the active voice, and it tends to give the impression that the researcher did not play an active role in conducting the study. (Qualitative reports, by contrast, are more subjective and personal, and written in a more conversational style.)
4. Statistical information. In quantitative reports, numbers and statistical symbols may intimidate readers who do not have strong mathematic interest or training. Most nursing studies are quantitative, and thus most research reports summarize the results of statistical analyses. Indeed, nurse researchers have become increasingly sophisticated during the past decade and have begun to use more powerful and complex statistical tools.
A major goal of this textbook is to assist nurses in dealing with these issues.
Tips on Reading Research Reports. As you progress through this textbook, you will acquire skills for evaluating various aspects of research reports critically. Some preliminary hints on digesting research reports and dealing with the issues previously described follow.
• Grow accustomed to the style of research reports by reading them frequently, even though you may not yet understand all the technical points. Try to keep the underlying rationale for the style of research reports in mind as you are reading.
• Read from a report that has been photocopied. Then you will be able to use a highlighter, underline portions of the article, write questions or notes in the margins, and so on.
• Read journal articles slowly. It may be useful to skim the article first to get the major points and then read the article more carefully a second time.
• On the second or later reading of a journal article, train yourself to become an active reader.
• Keep this textbook with you as a reference while you are reading articles initially. This will enable you to look up unfamiliar terms in the glossary at the end of the book, or in the index.
• Try not to get bogged down in (or scared away by) statistical information. Try to grasp the gist of the story without letting formulas and numbers frustrate you.
• Until you become accustomed to the style and jargon of research journal articles, you may want to “translate” them mentally or in writing. You can do this by expanding compact paragraphs into looser constructions, by translating jargon into more familiar terms, by recasting the report into an active voice, and by summarizing the findings with words rather thaumbers. As an example,
presents a summary of a fictitious study, written in the style typically found in research reports. Terms that can be looked up in the glossary of this book are underlined, and the notes in the margins indicate the type of information the author is communicating.
presents a “translation” of this summary, recasting the information into language that is more digestible. Note that it is not just the jargon specific to research methods that makes the original version complicated (e.g., “sequelae” is more obscure than “consequences”). Thus, a dictionary might also be needed when reading research reports.
• Although it is certainly important to read research reports with understanding, it is also important to read them critically, especially when you are preparing a written literature review. A critical reading involves an evaluation of the researcher’s major conceptual and methodologic decisions. Unfortunately, it is difficult for students to criticize these decisions before they have gained some conceptual and methodologic skills themselves. These skills will be strengthened as you progress through this book, but sometimes common sense and thoughtful analysis may suggest flaws in a study, even to beginning students.
Some of the key questions to ask include the following: Does the way the researcher conceptualized the problem make sense—for example, do the hypotheses seem sensible? Did the researcher conduct a quantitative study when a qualitative one would have been more appropriate? In a quantitative study, were the research variables measured in a reasonable way, or would an alternative method have been better?
Preparing A Written Literature Review
A number of steps are involved in preparing a written review, as summarized in Figure 5-1. As the figure shows, after identifying potential sources, you need to locate the references and screen them for their relevancy.
Screening References. References that have been identified through the literature search need to be screened. One screen is totally practical—is the reference readily accessible? For example, although abstracts of dissertations may be easy to retrieve, full dissertations are not; some references may be written in a language you do not read. A second screen is the relevance of the reference, which you can usually (but not always) surmise by reading the abstract. When abstracts are not available, you will need to take a guess about relevance based on the title. For critical integrated reviews, a third criterion is the study’s methodologic quality—that is, the quality of evidence the study yields.
Abstracting and Recording Notes. Once a document has been determined to be relevant, you should read the entire report carefully and critically, identifying material that is sufficiently important to warrant note taking and observing flaws in the study or gaps in the report. As noted earlier, it is useful to work with photocopied articles so that you can highlight or underline critical information. Even with a copied article, we recommend taking notes or writing a summary of the report’s strengths and limitations. A formal protocol is sometimes helpful for recording information in a systematic fashion. An example of such a protocol is presented in Figure 5-2. Although many of the terms on this protocol are probably not familiar to you at this point, you will learn their meaning as you progress through this book.
Organizing the Review. Organization of information is a critical task in preparing a written review. When the literature on a topic is extensive, we recommend preparing a summary table. The table could include columns with headings such as Author, Type of Study (Qualitative versus Quantitative), Sample, Design, Data Collection Approach, and Key Findings. Such a table provides a quick overview that allows you to make sense of a mass of information.
Example of a summary table:
Abercrombie (2001) reviewed research related to strategies that have been found to improve follow-up after an abnormal Papanicolaou (Pap) smear test. Her review included a table that summarized nine studies. The headings in her columns were: Author and date; sample size; objectives; design/intervention; and results.
Most writers find it helpful to work from an outline when preparing a written review. If the review is lengthy and complex, it is useful to write out the outline; a mental outline may be sufficient for shorter reviews. The important point is to work out a structure before starting to write so that the presentation has a meaningful and understandable flow. Lack of organization is a common weakness in students’ first attempts at writing a research literature review. Although the specifics of the organization differ from topic to topic, the overall goal is to structure the review in such a way that the presentation is logical, demonstrates meaningful integration, and leads to a conclusion about what is known and not known about the topic.
TIP: An important principle in organizing a review is to figure out a way to cluster and compare studies. For example, you could contrast studies that have similar findings with studies that have conflicting or inconclusive findings, making sure to analyze why the discrepancies may have occurred. Or you might want to cluster studies that have operationalized key variables in similar ways. Other reviews might have as an organizing theme the nature of the setting or the sample if research findings vary according to key characteristics (e.g., comparing research with female subjects and research with male subjects, if the results differ.) Doing a research review is a little bit like doing a qualitative study—you must search for important themes.
Once the main topics and their order of presentation have been determined, a review of the notes is in order. This not only will help you recall materials read earlier but also will lay the groundwork for decisions about where (if at all) a particular reference fits in terms of your outline. If certain references do not seem to fit anywhere, the outline may need to be revised or the reference discarded.
Writing a Literature Review. At this point, you will have completed the most difficult tasks of the literature review process, but that process is not complete until you have drafted and edited a written product. Although it is beyond the scope of this textbook to offer detailed guidance on writing research reviews, we offer a few comments on their content and style. Additional assistance is provided in books such as those by Fink (1998) and Galvan (1999).
Content of the Written Literature Review. A written research review should provide readers with an objective, well-organized summary of the current state of knowledge on a topic. A literature review should be neither a series of quotes nor a series of abstracts. The central tasks are to summarize and critically evaluate the evidence so as to reveal the current state of knowledge on a topic—not simply to describe what researchers have done. The review should point out both consistencies and contradictions in the literature, and offer possible explanations for inconsistencies (e.g., different conceptualizations or data collection methods).
Although important studies should be described in some detail, it is not necessary to provide extensive coverage for every reference (especially if there are page constraints). Reports of lesser significance that result in comparable findings can be summarized together.
Example of grouped studies:
McCullagh, Lusk, and Ronis (2002, p. 33) summarized several studies as follows: “Although noise-induced hearing loss is preventable through appropriate use of hearing protection devices, studies among farmers consistently show a low level of use (Broste et al., 1989; Engstrand, 1995; Hallet, 1987;Karlovich et al.,1988; Langsford et al., 1995).”
The literature should be summarized in your own words. The review should demonstrate that consideration has been given to the cumulative significance of the body of research. Stringing together quotes from various documents fails to show that previous research has been assimilated and understood.
The review should be objective, to the extent possible. Studies that conflict with personal values or hunches should not be omitted. The review also should not deliberately ignore a study because its findings contradict other studies. Inconsistent results should be analyzed and the supporting evidence evaluated objectively.
The literature review should conclude with a summary of the state of the art of knowledge on the topic. The summary should recap study findings and indicate how credible they are; it should also make note of gaps or areas of research inactivity.
The summary thus requires critical judgment about the extensiveness and dependability of the evidence on a topic. If the literature review is conducted as part of a new study, this critical summary should demonstrate the need for the research and should clarify the context within which any hypotheses were developed.
TIP: The literature review section of a research report (or research proposal) usually includes informatioot only about what is known about the problem and relevant interventions (if any), but about how prevalent the problem is. In research reports and proposals, the authors are trying to “build a case” for their new study.
You will become increasingly proficient in critically evaluating the research literature. We hope you will understand the mechanics of writing a research review once you have completed this chapter, but we do not expect that you will be in a position to write a state-of-the art review until you have acquired more skills in research methods.
Style of a Research Review. Students preparing their first written research review often have trouble adjusting to the standard style of such reviews. For example, some students accept research results without criticism or reservation, reflecting a common misunderstanding about the conclusiveness of empirical research. You should keep in mind that no hypothesis or theory can be proved or disproved by empirical testing, and no research question can be definitely answered in a single study. Every study has some limitations, the severity of which is affected by the researcher’s methodologic decisions. The fact that theories and hypotheses cannot be ultimately proved or disproved does not, of course, mean that we must disregard evidence or challenge every idea we encounter—especially if results have been replicated. The problem is partly a semantic one: hypotheses are not proved, they are supported by research findings; theories are not verified, but they may be tentatively accepted if there a substantial body of evidence demonstrates their legitimacy.
TIP: When describing study findings, you should generally use phrases indicating tentativeness of the results, such as the following:
• Several studies have found
• Findings thus far suggest
• Results from a landmark study indicated
• The data supported the hypothesis . . .
• There appears to be strong evidence that
A related stylistic problem is an inclination of novice reviewers to interject opinions (their own or someone else’s) into the review. The review should include opinions sparingly and should be explicit about their source. Reviewers’ own opinions do not belong in a review, with the exception of assessments of study quality.
The left-hand column of Table 5-3 presents several examples of stylistic flaws. The right-hand column offers recommendations for rewording the sentences to conform to a more acceptable form for a research literature review. Many alternative wordings are possible.
Research Examples Of Research Literature Reviews
The best way to learn about the style, content, and organization of a research literature review is to read several reviews that appear in the nursing literature. We present two excerpts from reviews here and urge you to read other reviews on a topic of interest to you.
Research Example From a Quantitative Research Report. Teel, Duncan, and Lai (2001) conducted a study about the experiences of 83 caregivers of patients who had had a stroke. A segment of their literature review that was included in their introduction follows (Teel et al., 2001, pp. 53–54). Over half a million Americans suffer strokes each year. Approximately 75% survive, yet most have residual neurologic impairment that requires supportive care (Gresham et al., 1995)*. Long-term assistance for many stroke patients is provided in home settings, by family caregivers who must acquire a number of new skills to successfully manage the outcomes of stroke… (Biegel et al., 1991; Evans et al., 1992; Jacob, 1991; Matson, 1994). Family home care management after stroke is essential, yet it is often stressful and demanding. The physical care requirements, vigilance, and altered roles that are often part of the stroke sequelae contribute to caregiving stress (Davis & Grant, 1994).
Caregiving demands can have negative emotional and physical consequences for the family caregivers, which can, in turn, have negative implications for continuation of the caregiving role. Because of the potential effects, many caregiver outcomes have been studied. Mental health outcomes, including depression, perceived burden and strain, anxiety, and alternations in mood, have been examined relative to caregiving (Matson, 1994; Periard & Ames, 1993). Stroke caregivers were found to have higher depression scores than noncaregivers, and the elevated levels persisted at 1 year post-stroke (Schultz et al., 1988).
For caregiving wives, increased social support was correlated with less depression (Robinson & Kaye, 1994).
Physical health outcomes for caregivers have included assessments of general health and chronic illness (George & Gwyther, 1986), number of physician visits or days hospitalized (Cattanach & Tebes, 1991), assessment of immune function (Kiecolt-Glaser et al., 1991), and fatigue (Jensen & Given, 1991; Nygaard, 1988; Rabins et al., 1982; Teel & Press, 1999). Stroke caregivers have reported impaired physical health (Deimling & Bass, 1986), yet Rees and colleagues (1994) found no immunologic alterations in stroke caregivers who had been caring for at least 6 months compared with caregivers in a cross-sectional analysis of immune function.
Evans and colleagues (1992) have suggested that the influence of the family also may affect stroke outcome.
For example, the family can have a buffering effect on patient coping, with family emotional, informational, and practical support enhancing post-stroke coping ability. Patients categorized as having suboptimal home environments at 1 year after stroke had caregivers who were more likely to be depressed, less likely to be a spouse caregiver, had below average knowledge about stroke care, and reported more family dysfunction than caregivers in the optimal group (Evans et al., 1991).
Because each of these variables has been associated with stroke outcome, post-stroke evaluation and treatment should include attention to minimizing caregiver depression and family dysfunction, while promoting knowledge about stroke care (Evans et al., 1991, 1992).
Overall, the research literature is extremely limited in reports of outcomes for caregivers of stroke patients. In particular, studies about caregiving in the first several months after a stroke, a time in which there can be significant change in patient condition and caregiving routine, is virtually absent from the literature.
Research Example From a Qualitative Research Report. Boydell, Goering, and Morrell-Bellai (2000) conducted a study of the experiences of 29 homeless individuals. A portion of the literature review for their research report follows (Boydell et al., 2000, pp. 26–27). Studies show that homelessness involves much more thaot having a place to live. Individuals often lose their sense of identity, self-worth, and self-efficacy (Buckner, Bassuk, & Zima, 1993). Hallebone (1997) studied 38 homeless men ethnographically and found that psychosocial identities tend to be fragmented….
Snow and
Matousek (1991) describes how the profound loss of self, which is associated with homelessness, presents a spiritual challenge to define one’s very existence.
SUMMARY POINTS
• A research literature review is a written summary of the state of existing knowledge on a research problem. The task of reviewing researchliterature involves the identification, selection, critical analysis, and written description of existing information on a topic.
• Researchers review the research literature to develop research ideas, to determine knowledge on a topic of interest, to provide a context for a study, and to justify the need for a study; consumers review and synthesize evidence-based information to gain knowledge and improve nursing practice.
• The most important type of information for a research review are findings from empirical studies. Various nonresearch references—including opinion articles, case reports, anecdotes, and clinical descriptions—may serve to broaden understanding of a research problem or demonstrate a need for research, but in general they have limited utility in written research reviews.
•A primary source with respect to the research literature is the original description of a study prepared by the researcher who conducted it; a secondary source is a description of the study by a person unconnected with it. Primary sources should be consulted whenever possible in performing a literature review.
• An important bibliographic development for locating references for a research review is the widespread availability of various electronic databases, many of which can be accessed through an online search or by way of CD-ROM. For nurses, the CINAHL and MEDLINE ® databases are especially useful.
• In searching a bibliographic database, users usually perform a subject search for a topic of interest, but other types of searches (e.g., textword search, author search) are available.
• Although electronic information retrieval is widespread, print resources such as print indexes and abstract journals are also available.
• References that have been identified must be screened for relevance and then read critically. For research reviews, most references are likely to be found in professional journals.
• Research journal articles provide brief descriptions of research studies and are designed to communicate the contribution the study has made to knowledge.
• Journal articles often consist of an abstract (a brief synopsis of the study) and four major sections: an introduction (explanation of the study problem and its context); method section (the strategies used to address the research problem); results section (the actual study findings); and discussion (the interpretation of the findings).
• Research reports are often difficult to read because they are dense, concise, and contain a lot of jargon. Qualitative research reports are written in a more inviting and conversational style than quantitative ones, which are more impersonal and include information on statistical tests.
• Statistical tests are procedures for testing research hypotheses and evaluating the believability of the findings. Findings that are statistically significant are ones that have a high probability of being reliable.
• In preparing a written review, it is important to organize materials in a logical, coherent fashion. The preparation of an outline is recommended, and the development of summary charts often helps in integrating diverse studies.
• The written review should not be a succession of quotes or abstracts. The reviewers’ role is to point out what has been studied, how adequate and dependable the studies are, what gaps exist in the body of research, and (in the context of a new study), what contribution the study would make.
Searching the literature – locating your resources
An effective literature search will:
- Reduce the time spent looking for information
- Maximise the quality and appropriateness of results
- Teach you how to use the “tools of the trade” – i.e. scholarly subject databases and indexes
- Help clarify the scope of your research topic
- Help find a niche in the literature
- Assist in identifying the experts and the important and influential published works in your field
- Identify publishing options – relevant journals, publishers, conferences, etc.
You need to make sure you are getting a good range of the available literature in your area of study. Keep in mind as you are searching that the age of the material is relevant. The best method is to start from the most recent sources and work backwards. This also puts you in a better position to let good studies guide your reading. You can always go back and check that seminal reference everyone else quotes.
- Use a variety of resources to cover a range of media- a literature review should include a range of literature, such as books, journal articles, or Internet sites. Theses, conference papers, eprints and government or industry reports can also be included. Do not rely solely on electronic full-text material which is more easily available
- Be aware of the importance of evaluating information. Is a journal refereed/peer reviewed, is a source authoritative?
- When beginning a literature search, reference sources such as dictionaries can assist in defining terminology, and encyclopedias may be useful in introducing topics and listing key references
- Most searching is done by subject. You can also do author searches and search using citation indexes (i.e. Web of Science or Scopus databases.)
- Some databases allow you to save searches and set up email alerting. These can help you keep your literature search up to date
- Ensure you take care with recording your references. Keep systematic and accurate records. Software such as EndNote can assist. Be careful to record page numbers where you found quotes.
- Good time-management skills can help you to feel in control! Aim to find the most important relevant material early. Read as you go and make critical and evaluative notes as you read.
by the Library
Finding things to read is the first step, but once you have your lists of references, it may be difficult to get hold of them. You will probably need to do some work to collect all the material you need. Travelling to other libraries, using the Interlibrary Loan service and ordering articles from overseas are all options.
Keeping tracking of reading materials
Because the nature of a literature review is to be dependent on what other authors have said, it is vital that you develop good referencing skills. These skills should be put into use the moment you first pick up a piece of written material. As soon as you pick up an article, put the publication details into your selected referencing system. The University Library has site licenses for three referencing products: Endnote, Endnote Web and RefWorks. Additionally there are free bibliographic management programs available on the web, such as Zotero.
EndNote – EndNote is a full-featured bibliographic management package and is widely used by researchers at the University. It can be used to:
- store bibliographic records (e.g. details of books, chapters, journal articles, websites, conference papers, theses, reports, etc.), PDF files, other attachments, links and research notes added manually or imported from web-based search engines/databases
- automatically generate citations and bibliographies in specific styles (e.g. APA 5th, Chicago 15a, Vancouver) in word-processed documents
- search and retrieve bibliographic records from library catalogues and journal indexing databases.
- retrieve articles by querying the University Source It system for web locations.
How to undertake a literature search and review
Literature search is a search designed to identify existing research and information about your chosen topic.
From the materials you find you will produce a literature review. This is a written piece
summarizing and analyzing the literature you have found through your search.
Why is the literature search and review important?
Using the published literature is a core part of the academic communication process. It connects your work to wider scholarly knowledge, demonstrates your understanding, and puts any research you have done in a wider context.
You need to use the published literature in order to:
• provide an academic basis to your research
• clarify your ideas and findings
• find data and research methods
• identify potential issues with the work you plan to do
Some projects involve conducting your own studies, in others you may be analyzing the literature itself, or other primary sources. In all these cases, the information you find in your literature search should inform and underpin everything you do, including the methods you use and your discussion of your findings.
How long does a literature search take?
This depends on your topic. Bear in mind you will need to:
• Build in time to read and digest what you find.
• Allow time to get hold of materials that aren’t locally available. Interlibrary loans (discussed later in this guide) take an average of 2 weeks to arrive.
• Don’t leave it all until the last minute – it will only make life stressful and mean that you won’t get the marks you deserve.
1
A literature search can be a daunting process, but there are some simple steps which can help you plan and manage the process:
1. Picking your research question
2. Planning your search
3. Evaluating and recording your results
4. Reviewing your search plan
5. Synthesizing your results
The diagram below shows how these steps interact as you move through the literature search and review.
As you go through this process you may find you need additional help. You can get help from your colleagues, your tutor, by going to an information desk or by arranging to see your subject librarian.
Suggestions of where to find help on particular issues are made throughout the guide.
1. Picking your research question
First you will need to decide what you want to find out. You will probably start with a fairly broad idea of the area you want to look at, e.g.
‘The impact of television advertising’
Asking yourself some questions will help you focus down on the exact topic you want to study.
2
Being as specific as possible in the question you ask will make your literature search and your project more manageable.
For example, when looking at television advertising you would need to specify whether you were interested in: children in terms of behavior, eating habits, lifestyles, consumerism, school meals, eating habits, policy in terms of health, advertising regulation.
You would also need to think about the limits to your investigation. You might want to restrict by some of the following:
Time current issues, rather than historic trends
Country
Discipline a Media Studies, Health, Psychological, Marketing, or Political perspective
Gender studying children, just boys, or only girls
Age studying particular age groups like the under 5s or 8-12s
Type of material using only research material or including popular and practitioner/trade publications
Focusing at an early stage is a good idea, as it helps you get an idea of how much information is out there. You can always re-focus your question later if you find too little or too much information about your topic.
2. Planning your search
You can take a number of approaches to your search:
• Systematic – you try to find all relevant material
• Retrospective – you find the most recent material and work backwards
• Citation – you follow up references from useful articles, books and reading lists
• Targeted – you restrict your topic and focus on a narrow area of the literature
In practice, most people use a mixture of approaches. You might:
• be systematic in looking at everything relevant in the library
• adopt a retrospective approach when looking at journal articles
• use citation searching to get useful leads if your topic crosses several disciplines
• be more targeted when you have a clear picture of what you need to find out.
Choosing your Sources
There are many different sources of information. You will almost certainly need to use both books and journal articles, but you may need to use other information sources related to your subject, such as government reports, or sources of statistical information.
TIP: Reading a general text or doing some browsing on the Internet can be a helpful way of clarifying your thoughts at this stage, and picking exactly what you want to research.
3
Finding books
Books are often a good starting point. Textbooks summarize key theories and more specialised texts often present research findings in a clear and comprehensive way. There are three key places to look:
• The library catalogue
• Other library catalogues list what is available in libraries elsewhere.
• Internet booksellers (e.g. www.amazon.co.uk), may have the details of newly published books.
Finding journal articles
In many subjects journals are the key resources for a literature search. They are the principal place where research and practice are discussed and new work presented. This means a specialist or new topic will often be better covered by journal articles than by books.
You will need to use databases to find journal articles on your topic. Some databases will give you references so you can trace an article, others allow you to access the full text straight away. Your Subject Guide will list the most useful databases for searching the journal and report literature, and the ‘How to use Journals’ guide gives further guidance.
Keywords
When using either library catalogues or databases you need to pick your search terms carefully. Search engines and library databases are not intelligent, and will match up words without considering their meaning. This means a search for ‘apple’ will find information about both the fruit and company, regardless of your intention.
Selecting keywords – words or phrases that describe your topic as simply and distinctively as possible – can make searching much easier. Selecting keywords can be a straightforward process, if the words describing your topic have a single meaning, but more often you need to think carefully about the keywords you use to express your ideas.
Here are some approaches to try when selecting keywords:
Specific terms Start your search by using words that are specific to your research topic and, ideally, not common elsewhere.
Similar and Are there other words with similar meanings? Using these alternative related terms terms will find a different set of results.
Spellings and Can your search term be spelt in different ways?
e.g. behavior vs. behavior. Some databases and search engines don’t automatically call up the
TIP: Your Subject Guide, available from the Library web pages, highlights the most useful starting points in your subject area. You can also consult related Subject Guides if your topic covers multiple subject areas.
4
Singulars and Some databases don’t automatically look for single and plural
plurals versions of a word. Try both. Usually people and things are plural, ideas are expressed as singular.
Combining terms You can usually search for phrases using quotation marks e.g. “television advertising”, and can combine terms using AND, OR and NOT, e.g. television AND advertising (will find documents containing both words), television OR advertising
(will find documents containing either word), television NOT advertising (will find documents which do not mention advertising).
Truncating terms Most databases will allow you to search for terms that begin with the same set of letters, using a symbol such as * $ or ? For example, politi* can search for politic, politics and political. The symbol used will vary between databases so check the help screens to find out which one to use.
3. Evaluating and recording your results
Once you have an initial set of search results, have a look at the material you have found. This will give you an overview of what you’ve found so far, helping you identify what kind of information is out there, and whether you’re finding the right kind of material.
As you find materials, it is also important to record the outcomes of your search plan. This will save you time and effort when you need to re-run a search or locate references at the end of your project. You will definitely want to record the full reference of everything you find.
It is essential that you reference all items as it:
• ensures good academic practice by acknowledging other people’s ideas
• enables your tutor to see what sources you have used
• gives more authority to your arguments
• shows the scope and breadth of your research
• avoids plagiarism
Referencing is important regardless of whether you are directly quoting or paraphrasing the original source. All sources you use, regardless of format, need to be referenced so ensure you reference images and diagrams as well as printed or online material.
5
Plagiarism (presenting someone else’s material as your own) is a serious academic offence and can result in a reduction of the mark awarded or a module failure. In extreme cases you may be expelled from the University. However, plagiarism is easily avoided by keeping track of where you get your ideas and referencing appropriately.
4. Reviewing your search plan
Once you have evaluated and recorded your initial results, you will need to review and revise your search plan in order to fill in gaps in the material you have found, and address any other issues you have noted.
Literature searching is a cycle and for a typical project you will need to go through the process of evaluating and revising several times before you have found the right material. Here are some commonly encountered problems and potential ways in which you might revise your search plan to cope with them.
Finding too much
Sometimes you will find that there is just too much information. This might be because:
• Lots has been written on your main topic
• Your topic has links with many other subject areas
Returning to your research question and re-focusing can solve this problem by giving you a clearer idea of what you really want to find out (see the section on Picking your research question for further details).
If your research question is already specific, you may need to revise your search plan. Things to try include:
• Using more precise terms (advertising rather than marketing)
• Adding in limits (television and advertising and
• Concentrating on key authors and books
• Using database features like the help menu, or subject headings (see Tip).
Finding too little
Sometimes you will struggle to find much relevant material. You will need to think of ways of broadening the scope of your project. In particular you can think about:
• Making the project (or just your keywords) more general.
• Searching for comparative or related information, e.g. looking at all children, rather than a particular age group.
TIP: Using subject headings within library databases is a great way to narrow down your search.
These are labels which summarise the main themes of an article or paper in a few words. Every article listed in the database covering the same topic will have the same subject heading, even if different words are used to describe the topic within the article.
Subject headings may be suggested by the database when you search, or listed in the information about the article alongside the article title, author and abstract.
6
Your tutor is a good source of advice: they can often tell you if a topic is very new, or little-studied, and they may be able to suggest related areas of research to investigate.
Finding materials which are not ‘academic’ enough
You may worry that the materials you’ve found are not appropriate to your level of study, or your tutor may have told you that you need to make your work more academic. This just means you need to be more selective in your use of sources.
Two helpful ways of accessing scholarly material are:
• Limiting your search to the academic (or scholarly or peer-reviewed) journal search options available on many of the databases, and making sure you are using academic rather than trade journals.
• Limiting your Internet search to sites which end in .ac or .edu.
Continue reviewing and revising your search plan and recording and evaluating your results until you are happy with the materials you have found. It is a good idea to start this process early in your project, and then re-run your search as the project progresses to update yourself oewly published material.
4. Synthesising your results
You will also need to present an analysis of the literature you have found: synthesizing the results of your literature search into a literature review.
In many projects and dissertations a literature review forms a chapter of the finished piece of work, and may be assessed as a separate assignment, handed in at an earlier stage.
The literature review is a way of demonstrating two things:
• Literature search – the materials that you have found
• Understanding and analysis – how you have put what you found into the context of your project.
The key elements of a literature review are providing an overview and argument, reading critically, writing analytically, and identifying areas for further research. These are discussed further below.
i. Providing an overview and an argument
Start by discussing your research question and your initial thoughts. It is a good idea to provide an overall summary of the literature you have found, in particular highlighting any gaps in research and conflicts in theory.
Make sure that you also state your own research perspective and the scope of your investigation, in particular what limits you established, and why you have chosen to approach the topic in a particular way.
TIP: Make an appointment to attend a drop-in to see your subject librarian who is an expert at finding information in your topic area and will help you with selecting the right sources and the best search strategies.
TIP: Remember that the literature search underpins the whole of the work. You should also include references to the literature when discussing your methods and findings.
7
ii. Reading critically
The literature you find will not be unbiased. Think and read critically. For each item you read, you might want to consider:
• Has the author clearly defined the topic and question? Is it an effective analysis and account of the subject?
• Is there any bias evident (political, ideological, disciplinary?)
• How scholarly is the piece of work? Trade and professional journal articles and websites are often current but not scholarly.
• Is the argument coherent, or does the piece contradict itself?
• Are there references to sources the author consulted? Have any sources or theories been ignored or omitted?
• Is the item as relevant to the topic which you are investigating as it initially appeared?
iii. Writing analytically
Try to summarize the arguments of different authors in relation to your own research question or topic.
Can you compare or contrast different authors or theories, or identify any emerging themes? You will need to present an argument, or a series of points: do not just describe what different authors have written.
Often you will find that your topic overlaps different subject disciplines, bringing in multiple perspectives and different sets of literature. Pointing this out and discussing it emphasisesthe thoroughness of your work.
iv. Identifying areas for further research
Finally, relate the literature review to the bigger questions within the literature or subject. You should try and identify areas where not a lot of literature existed, as opportunities for gaps to be filled.
v. Afterwards
Once you have completed your literature review, it is a good idea to use it as an opportunity to reflect on your own progress, both in terms of finding information and in critical reading. These are very valuable graduate skills, and often you do not see how well you have worked until after the task is completed.
Remember that an understanding of the literature is a core part of the academic process.
The process of undertaking a literature review
The process of undertaking a literature review is an integral part of doing research. While this may be considered to be its primary function, the literature review is also an important tool that serves to inform and develop practice and invite discussion in academic work. Whatever its purpose, the task of doing a literature review is often viewed as an onerous and confusing one by students. The aim of this chapter is to provide you, the student, with a comprehensive understanding of what a literature review is and, equally, what it is not. It explores its purpose and relevance and the differences between the literature review and other forms of academic writing. The fundamental steps involved in undertaking a literature review will also be considered. Whether or not you have previously embarked on the literature review journey, this chapter is designed to help you understand the process and skills involved iavigating the literature and reaching your ultimate destination.
Whether it is for clinical or academic purposes (or your own innate curiosity!) it is important to understand what a literature review is before you start sourcing and immersing yourself in copious amounts of research and theoretical concepts. A literature review is a synopsis of other research. Moreover, it is a critical appraisal of other research on a given topic that helps to put that topic in context (Machi and McEvoy, 2009). A comprehensive review should provide the reader with a succinct, objective and logical summary of the current knowledge on a particular topic. Therefore, it is not an essay of one’s own personal views and opinions. Similarly, it is not a series of quotes or lengthy descriptions of other people’s work. Quite simply, the literature review provides a critical discussion on the topic of interest while pointing out similarities and inconsistencies in existing relevant literature. It is important to note that while a literature search is the means of helping you to unearth literature that is appropriate to your task in hand, a literature review is the process of critically evaluating and summarizing that literature.
The Purpose of the Literature Review: The Question and Context
Conceptualising the Literature Review
Think of a topic that interests you in clinical practice. Imagine this as a wide-rimmed, intricate crystal vase that tapers to a very narrow stem. There are some imperfections in the crystal. The rim represents the body of knowledge surrounding your chosen topic, and the stem represents your ultimate research question. Imagine the vase as your literature review. What does it mean? What do you need to do?
In the above activity you have already embarked on the process of starting your literature review. By undertaking an initial, broad literature search and then review, you will eventually be at the stage of fine-tuning and narrowing down your research idea or question in the context of other literature. A thorough and critical review of the literature will enable you to do just that. It is important to mention here that some literature reviews are preceded by a pre-determined research question and, therefore, how and when they are conducted varies according to the type of approach used. For example, quantitative research studies are usually driven by the context of previous knowledge, with specific research questions in mind based on conducting an extensive literature review before data collection commences.
Qualitative studies typically adopt a less structured approach to doing a literature review at the outset. They may start with a broad research question or topic of interest that is refined and honed as data are collected. Therefore, while the literature may be consulted briefly at the outset, a more thorough and in-depth review is done based on emerging data rather than pre-existing knowledge. Whatever the approach, the purpose of the review is to provide an analysis and synthesis of all the available literature on a given subject in a critical fashion. This then allows for further understanding of the subject in the context of what is already known. Furthermore, it can lead to the development of new research questions. Using the analogy of the crystal vase, each individual piece of crystal (including the imperfections) fits with the others to make the vase whole and cannot be viewed in isolation. Similarly, for the literature review to be complete and comprehensive, it needs to be carried out and evaluated in light of all other relevant literature, in order to get the fullest picture possible.
The Importance of Reviewing the Literature
The importance of the literature review is directly related to its aims and purpose. Nursing and allied health disciplines contain a vast amount of ever increasing literature and research that is important to the ongoing development of practice. The literature review is an aid to gathering and synthesising that information. The purpose of the literature review is to draw on and critique previous studies in an orderly, precise and analytical manner. The fundamental aim of a literature review is to provide a comprehensive picture of the knowledge relating to a specific topic. For example, if one is proposing to undertake a research project, then the purpose of the literature review is to situate that project in its relevant context or background. It does this by drawing on previous work, ideas and information. In addition, a good review will extract and critically evaluate the pertinent findings and issues that have emerged from previous work (Hart, 2010). By doing so, it provides justification for the proposed research and demonstrates a thorough grasp of background knowledge. Going back to our analogy of the crystal vase, it is evident that some imperfections exist. These we can take to represent evidence that is not strong and cannot be viewed in isolation. To do so would give an incomplete picture. Therefore, the literature needs to be reviewed in the context of all other information relating to the topic. One single viewpoint or article will not give the full story and may serve to bias the review.
1.1 Defining Your Research Question
In health and social sciences, research questions usually stem from practice and serve to inform and develop practice. Defining your question can be a difficult task. Think of a particular area of practice that interests you. Ask yourself what you know about it. Now think, what is it you want to find out? For example, you may have an interest in the impact of type II diabetes. Ask yourself, Why? What? Who? Where? How? Decide what your question is and keep close to it. Your literature review should proceed from the known to the unknown, guided by your research question.
The importance of the literature review cannot be overstated. It is the tool to advancing practice. Furthermore, it can help to inspire and generate new ideas by highlighting inconsistencies in current knowledge (Aveyard, 2010). Literature reviews are not undertaken solely for the purpose of doing research. They have an important function in evaluating current practice and making recommendations for policy development and change. They are also useful for exploring existing theoretical or conceptual frameworks concerning a given subject. Similarly, they facilitate the development of theoretical or conceptual frameworks through exploration and critical evaluation of existing knowledge. The manner in which the review is written is a crucial component of understanding what a literature review is and often poses the greatest challenge to students. Before we discuss the steps in undertaking a literature review we need to examine what differentiates it from other types of academic writing.
Essay Writing versus Writing a Literature Review
Essay writing is a process that communicates ideas to an intended reader, and is usually written according to pre-determined academic conventions. An essay may have many purposes but its basic format is structured as an introduction, main body and conclusion that convey information relating to the essay question. The question serves to focus and direct the student as to what is required in the essay. For example, the question may require the student to produce a general overview of a topic. On the other hand, the purpose of the essay may be to produce a specific analysis of a particular subject; therefore the question will be phrased differently. Words such as discuss, explain, describe, analyse and assess may be an integral part of the question and will determine what approach to writing should be adopted. In some instances, a particular topic may not have been assigned and, therefore, it is up to the student to decide the purpose of the essay and what they are trying to achieve – that is, is the purpose to inform or educate, or perhaps create a persuasive argument?
Common essay types include judgement essays, exploratory essays and reflective essays (Shields, 2010). Judgement essays typically defend a particular argument or viewpoint and attempt to persuade or convince the reader to adopt the writer’s stance. Judgement essays are based on an evaluation of relevant evidence and theories surrounding a topic and are written with the intention of constructing a sound argument that defends the writer’s viewpoint. The judgement essay requires the writer to produce a subjective account of an issue, based on a discussion and interpretation of existing evidence. In contrast, exploratory essays do not require the writer to adopt a particular stance. This type of essay is concerned with producing a reasoned explanation of a given subject or phenomenon. It explores an issue in a logical and thorough fashion and presents and explains factual information in a balanced manner, without defending a particular viewpoint. The exploratory essay is concerned with a review, comparison and discussion of theories relating to a specific subject area. Both judgement and exploratory essays require a level of critical analysis of the essay question (Shields, 2010). The reflective essay is based on the premise that learning occurs and is enhanced through reflection. While reflective essays require a format similar to other essays, this is typically less structured as reflection is personal to the writer and so the style and language used will differ from more formal essay writing. Reflective essays are not based on a question but tend to focus on description and analysis of a personal incident or experience with a view to learning from that event. While reflective essays are descriptive iature, it is necessary to discuss relevant theories and concepts relating to the event in order to analyse it.
So what then of the literature review, and does it differ from other types of essay writing? There are some similarities. As when writing an essay, one needs to have a structure and focus when undertaking a literature review. There is a sequence of events that should be adhered to in order to create a comprehensive review. The style and language must adhere to academic convention. Sections need to be cohesive and flow logically. Concepts and theories should be compared and contrasted, grouped for similarities and inconsistencies. Evidence needs to be analysed and discussed in relation to its context and significance.
The difference between essay writing and writing a literature review is that while the purpose of essay writing is generally to discuss ideas with respect to the essay question, the aim of the literature review is to summarise and synthesise all that is published on a given topic. The literature review is undertaken to present results of research and key information in an objective and discursive manner. In contrast to the essay, the literature review should summarise the key concepts, theories and empirical studies while discussing their strengths and limitations. If this is done for the purposes of research, then it needs to be focused, in-depth and relevant to the research question. Critical discussion is a crucial component of writing a good review. A literature review is not a descriptive, exploratory essay. Similarly, unlike the judgement essay, it does not seek to defend a particular viewpoint, nor does it offer personal opinion or speculation, as does the reflective essay. It is not a criticism, but rather is a critical review that goes beyond description to the level of critique, analysis and synthesis. Writing a literature review can be considered difficult when one has been accustomed to essay-style writing. The basic structure of introduction, main body and conclusion still applies, with some specific additional steps that serve to further differentiate literature review writing versus essay writing.
Steps in Undertaking a Literature Review
A literature review is an essay of sorts; however, it tends to be more formulaic iature. The key steps to undertaking any literature review firstly involve developing a structure for the review. This entails selecting the review topic, carrying out a literature search, reading, critiquing and analysing the literature, and finally writing the review. The approach to writing a literature review will vary slightly according to the type of review undertaken. For example, a systematic review will have a specific format that must be adhered to throughout. Structuring the review so that it is presented in a clear, coherent and consistent manner is vital, and it is necessary to develop a framework for this before starting to write. A well-organised literature review will consist of an introduction, a main body that critiques the findings of previous work, addressing both theoretical and empirical literature, a discussion and a conclusion.
Selecting a topic to review is the first step in the process. Selecting a topic is guided by your overarching research objective or the problem you wish to explore. Carrying out a literature search involves using both primary and secondary sources, as well as theoretical or anecdotal papers relevant to the topic. The search strategy and terms used should be stated in order to provide evidence that the review was thorough and comprehensive. It is important to narrow down the review topic or research question to avoid generating overwhelming amounts of information. A general idea of the area of interest is a good starting point, but it is necessary thereafter to hone this to a specific aspect of interest in order to make the review manageable. Searching the literature effectively will enable you to do this. The importance of a thorough literature search cannot be overstated, as this will enable you to broaden your knowledge of your chosen topic. It also helps situate the research question in the context of existing knowledge.
Searching and reading the literature is a continuous process in the early stages of doing a literature review.
Analysis and synthesis are the next stages in the process of doing a literature review, and at this point all of the relevant information should have been gathered. Once an initial overview has been done, it is necessary to critique and critically analyse the literature in order to obtain a critical review of the content. When writing the final review it is important that key information is critically evaluated, rather than described. Evidence needs to be summarised and presented logically, comparing and contrasting findings, and offering new insights where possible. Chapters 6 to 8 discuss how to analyse, synthesise and write the review.
1.2 Putting it Together
A checklist is useful to have when undertaking a literature review. In order to stay focused throughout the process, think about the following issues:
Have you identified a specific research question?
What type of literature review are you conducting?
How will you conduct your literature search?
What is the scope of your review? What type of publications will you use and from what discipline?
Have you identified the type of literature that will help you address your research question?
Has your literature search been extensive enough to include all relevant material? Has it appropriate breadth and depth?
• Have you critically analysed the literature and discussed its strengths and weaknesses?
• Have you identified conflicting findings and offered possible reasons?
• Is your literature review balanced and objective?
• Have you discussed the significance of the findings?
• Is your literature review written in a logical and succinct manner?
• Have you identified and discussed implications for practice and further research?
Summary
A literature review is a critical evaluation of extensive research and theory relating to a specific topic. It is the process of analysis and synthesis of previous work in order to produce a summary of the knowledge on that topic. It gives insights into the background and context of a proposed study and is a logical, coherent argument that arises from a critical analysis of the state of knowledge in a specific topic area. This chapter has addressed the concept of what the literature review is and its importance and relevance in practice and research. We have also explored how it differs from other forms of academic writing. The steps involved in undertaking a literature review have been outlined and discussed briefly.
Key Points
A literature review is a critical evaluation of knowledge of a topic.
The literature review may be used as a basis for research, practice and policy development or academic purposes.
The steps to undertaking a literature review involve selecting a topic and literature searching, reading and organising the literature, analysis and synthesis, and writing up the final review.
Why are literature reviews important?
Literature reviews are important because they seek to summarise the literature that is available on any one topic. They make sense of a body of research and present an analysis of the available literature so that the reader does not have to access each individual research report included in the review. This is important because there is an increasing amount of literature available to health and social care professionals, who cannot be expected to read and assimilate all the information on any one topic. Everyone who works within health and social care has a professional duty to be up-to-date with recent developments and ideas that inform their practice. Yet, it is virtually impossible for any one practitioner to assimilate, process and decide how to implement all this information in their professional lives.
Why is there so much available information?
The amount of information available to all health and social care professionals is vast and expands on a daily basis. Every day there are media headlines, reports from conferences, reports of research from scientific journals, expert opinion followed by an opposing expert opinion. There are many reasons for this increase in information available to professionals. It is partly due to the increase in information technology which has led to the increasing availability of information from on-line journals and other websites offering information about health and social care. However, the main reason for the increase in information available within this field stems from the recent emphasis on evidence-based practice, which has led to the increasing demand for research evidence upon which practice decisions should be based. Evidence-based practice has been described as a new paradigm within health and social care which has gradually emerged since the 1970s. Practitioners began to question their practice and to search for a scientific rationale for the care they delivered, which previously might have been given according to tradition and experience.
As more and more research was carried out and the body of evidence within health and social care expanded, concern about getting this research into practice also increased. The term evidence-based practice is used to refer to the appropriate application of this research knowledge to practice. Evidence-based practice has been described by David Sackett, founder of the NHS Research and Development Centre for Evidence Based Medicine in
Why does one piece of research often contradict another?
It often seems to be the case that a piece of research is published one month which contradicts the findings of a piece of research published the month before. For example, one week working mothers are told that preschool care benefits their child and the next they are told that it is better for the child to stay at home. There is often then an outcry – people are confused by the differing messages conveyed and wonder why the results can vary so much. This can be due to the media portrayal of the research in which a complex set of results is reduced to a simplified message. However, it is also due to the fact that any one individual piece of research, or indeed any single piece of health care information, is like just one part of a large jigsaw. It does not represent the whole picture – it represents merely a section of that picture and needs to be set in the context of other information. An individual piece of health care information, taken in isolation, does not necessarily help the reader to achieve a better understanding of the bigger picture towards which the information contributes. There are many reasons for this. For example, the research might have been undertaken in a specific area of practice or with a specific group of people, or sample, and is not generalisable to other areas. Alternatively, there might be flaws in the research design which affect its overall validity. Therefore, when you read a report that seems to conflict with a report you read the previous week and are uncertain as to which report you should consider the most reliable, it is important to consider the merits of each individual report and to remember that each single piece of research contributes just part of the bigger picture and should not be viewed in isolation. This is why literature reviews are so important in health and social care because they enable the reader to view one piece of research within the context of others. For example, consider the MMR (measles, mumps and rubella) vaccination media stories in 1998–1999. In 1998, Professor Wakefield published an article in The Lancet suggesting that there was a possibility of a link between the vaccination, autism and bowel disorders. This article was based on a small case study of 12 children who had attended
Literature reviews help you to see the full picture
The strength of the evidence presented by
Uncovering new evidence
The MMR controversy provides one clear example as to why it is important to review all the evidence together and how one piece of information can give a misleading picture. Without the comprehensive review of the literature which followed
Encouraging objective thinking
There are other similar examples that illustrate the importance of the evidence provided by literature reviews that are carried out systematically compared to reviews that are not. Take, for example, Linus Pauling, the world accredited scientist, who wrote a book entitled How to live longer and feel better. In this book he quoted from a selection of articles that supported his opinion that vitamin C contains properties that are effective against the common cold. This book makes an interesting and convincing read. However, the arguments presented in the book were challenged some years later by Professor Knipschild (1994), who undertook a systematic review of all of the evidence surrounding the effectiveness of vitamin C and came to very different conclusions. He argued that Pauling had not looked systematically at all the research and had only selected articles that supported his view, while apparently ignoring those that did not. This is why when you read a report by an expert in a particular area, you should remember that his or her report represents just an expert view which might not be substantiated by evidence. This is why expert opinion is generally not considered to be a strong form of evidence. In summary, literature reviews are important in health and social care because they enable information and research about health and social care to be viewed within its particular context and set amid other similar information and research, so that its impact can be evaluated systematically. Reviewing the literature provides a complete picture, which remains partially hidden when a single piece of research or other information is viewed in isolation.
The importance of a systematic approach to the literature review
The literature review is a vital tool because it provides a synthesis of research and information on a particular topic. It is important that the review is approached in a systematic manner so that all the available information is incorporated into the review. When you read literature reviews, you will discover that some are undertaken in more detail than others. The most detailed type of literature review is often referred to as a systematic review.
The systematic review
A systematic review strives to identify comprehensively and track down all the available literature on a topic, whilst describing a clear, comprehensive methodology. Systematic reviews have been defined as ‘concise summaries of the best available evidence that address sharply defined clinical questions’ (Mulrow et al. 1997). The most well known method for conducting a systematic review is produced by the Cochrane Collaboration. The Cochrane Collaboration was established in 1993 and is a large international organisation whose purpose is to provide independent systematically-produced reviews about the effectiveness of health care interventions. The Cochrane Collaboration provides detailed guidance about how to undertake the review. One of the main features of a systematic review is that reviewers follow a strict protocol to ensure that the review process undertaken is systematic by using explicit and rigorous methods to identify, critically appraise, and synthesise relevant studies in order to answer a predefined question. The reviewers then develop a comprehensive searching strategy, and leave no stone unturned in the search for relevant literature, and do not regard the process complete until the search is exhausted. For example, reviewers search for unpublished research and might talk to researchers about unpublished data or articles not accepted for publication, in addition to published data on the topic in question. The reason for this is that there is evidence that a publication bias exists; that results which show clear benefit of an intervention are more likely to be published than those which do not. Thus using only published data could bias the result of the review. Reviewers then develop inclusion and exclusion criteria in order to assess which information they retrieve should be incorporated into the review, and to ensure that only those papers that are relevant to the question(s) addressed by the literature review are included. The reviewers then critique the selected papers according to predetermined criteria in order to assess the quality or validity of the research identified. Studies that do not meet the inclusion criteria are excluded from the review. This is to ensure that only high quality papers which are relevant to the literature review question are included. The results of research that has been poorly carried out are likely to be less reliable and may bias the findings of the review. Finally, the findings of all the papers that are identified and incorporated for the review are then pulled together and combined using a systematic approach. For example, a meta-analysis might be undertaken if the results of the research included in the review are reported using statistics, or a metaethnography can be undertaken if the results of the research included are mainly qualitative. This enables new insights to be drawn from the summary of the papers that was not available before. The methods of undertaking a systematic review are rigorous and timeconsuming. The production of a systematic review usually requires the dedication and effort of a team of experienced researchers over a period of time. Because of the comprehensive nature of the searching strategy, critique and synthesis of the literature, a systematic review undertaken in the detail required by the Cochrane Collaboration is usually considered to be the most detailed and robust form of review that exists. For example, in the
Less detailed approaches to reviewing the literature
Even if the stringent requirements of a Cochrane Collaboration style systematic review may not be within the capacity of a novice researcher, it is still possible to undertake a ‘systematic approach’ to reviewing the literature. The term systematic review is used by the Cochrane Collaboration to describe the reviews they produce which are carried out according to strict protocol. However, a literature review can be approached in a systematic manner even if the detail required by the Cochrane Collaboration is not attained. While the term systematic review is often used to refer to reviews undertaken according to the Cochrane Collaboration method of reviewing, there is no reason why this term cannot refer to a review of the literature that has been undertaken using a systematic approach, but which is less rigorous and detailed than the methods described above. This means there can be some confusion concerning the meaning of a systematic review. One reader might interpret the term systematic review to mean nothing less than a review conducted using the methods advocated by the Cochrane Collaboration approach, while another reader might accept that a systematic review incorporates a systematic approach but may not reach the same exacting standards. Undergraduate and postgraduate students who are undertaking a literature review for their dissertation would not normally be expected to achieve a systematic review of the standard required by the Cochrane Collaboration. They would, however, be expected to apply the general principles and guidelines of this approach to produce a literature review that used a systematic approach in the search for, critique and synthesis of the literature. For those new to literature reviewing, it is possible – indeed essential – to achieve a systematic approach to reviewing the literature, otherwise there can be no assurance that the review has been undertaken in a rigorous manner. If a literature review is to be submitted for an academic degree, the method undertaken to review the literature should always be systematic.
Narrative reviews
It is generally accepted that a Cochrane Collaboration systematic review offers the most robust form of evidence for health and social care professionals. However, not all reviews are conducted to this level. Literature reviews vary in the extent to which they are conducted in a systematic manner. For example, a literature review can incorporate a systematic approach but not in the amount of detail as described in the previous section. This approach would be expected of all those submitting a literature review as a component for an academic degree. At the other end of the spectrum there are literature reviews which are undertaken with no defined method or systematic approach. These are often referred to as narrative reviews. There is general concern that narrative reviews do not produce reliable evidence. The lack of a systematic approach to a narrative review is described by Greenhalgh (1997), who makes reference to essays written by medical students who may ‘browse through the indexes of books and journals until [they] came across a paragraph that looked relevant and copied it out. If anything did not fit in with the theory [they] were proposing [they] left it out’ (p. 672). The narrative literature review is one that does not use specific identified methods for searching for, critiquing and synthesising the literature. Instead the methods used are undefined and only a small selection of available literature is incorporated in the review, which may or may not have been appraised (Hek et al. 2000). There is not a clear indication as to how the study was conducted and therefore the study is not repeatable. Consequently, the conclusions drawn are likely to be inaccurate. These ‘traditional’ or narrative reviews have a number of biases. There is normally the personal bias of the author(s), a bias in the selection of included material, and with no clear methodology they cannot be reproduced independently, so conclusions cannot be verified easily and may be misleading. The example given earlier about the evidence for the use of vitamin C illustrates this point. Professor Knipschild challenged the findings presented in a narrative style review when he undertook a more systematic approach to a review on the same topic. The danger of a narrative review: it can lead to misleading conclusions because a comprehensive search for and critique of literature is not undertaken. In a narrative review, the searching strategy is not clearly defined or organised. There is no specific structure to the searching strategy and it is not clear how the authors search for the literature they identify, how much of the identified literature is incorporated in the review or whether any strategies for critical appraisal of the literature were used. As a result, a narrative review might be no more than a biased collection of research papers and other information about a given topic. This may lead to a biased and one-sided review of the literature which is not comprehensive. Individual research papers that are relevant to the review question may be identified but because the search is not systematic, other similar papers may not be identified. The research papers that are identified are theot set in their context but remain like single pieces of a jigsaw. Furthermore, in a narrative review, there is ofteo clear statement about which studies to include in a review and how these should be critiqued. In a systematic review, predefined inclusion and exclusion criteria are set which determine the relevance of each study identified. In a narrative review, these standards are not defined and any literature might be included without justification or rationale. Therefore, the reader of the review is unclear how much relative importance should be attached to each individual research report included as its merits are not discussed. While it is acknowledged that a fully systematic approach is beyond the scope of most novice researchers, the narrative review is not a strategy that should be resorted to. One of the first researchers to raise concern about the quality of the narrative review was Mulrow in 1997, who criticised the lack of rigour with which many reviews were carried out. Mulrow (1997) examined 50 literature reviews published in four major medical journals and identified that 49 had no statement of the methods used and 47 had inappropriate summaries of the information included. She concluded that, at that time, medical reviews did not routinely use scientific methods to identify, assess and synthesise information. The main differences between a narrative and a systematic review are summarised below: • Narrative review • no focussed research question • no focussed searching strategy • no clear method of appraisal or synthesis of literature • not easily repeatable. • Systematic review • well focussed research question • well focussed searching strategy with comprehensive and explicit methods • rigorous methods of appraisal and synthesis of the literature • method of undertaking review is explicit and repeatable • the most detailed reviews require a rigorous and demanding process – not for the faint hearted! If you are undertaking a review of the literature, you are strongly advised to adopt a systematic approach to the review and to avoid a narrative approach where possible. Those new to reviewing the literature are not normally expected to undertake a systematic review in the detail as required by the Cochrane Collaboration. However, you are required to undertake a systematic approach to the literature review; the possible methods for achieving a systematic approach to a literature review are outlined in the subsequent chapters of this book.
The literature review as a research methodology
It is important to remember that a literature review that is carried out systematically is a research methodology in its own right. Your review will have a defined research question and you will follow a systematic approach to answering that question. Even if you are not undertaking a Cochrane-style systematic review, you need to follow a systematic process when you are undertaking your review and you will need to document this process very clearly when you come to write up your review. It is important that you document clearly how you undertook the steps you have taken. The reader needs to know that you undertook a comprehensive and systematic approach to your literature review and the only way to determine this is to give a full account of your literature review process. If you do not document a process that was undertaken, the reader will be given the impression that this process was not undertaken. There should be a clearly defined section detailing the methods used to address the question. The methods section will usually commence with how you identified your research question. Discuss the rationale for your research question and explore its origins. You can draw on related literature at this point. Remember also to justify your use of a literature review as your chosen research methodology. Why did you not choose another research methodology, such as one involving primary data collection? You should then document how you searched for appropriate literature. You are advised to include a report of the search terms you used and your search strategy. You should then document how this literature was critiqued and justify your choice of critical appraisal tools. Finally, you need to document how you brought this information together. Present information in a graph or chart if this is appropriate. Overall, your methods section will contribute a large portion of the overall review and is likely to amount to approximately one-fifth to one-quarter of the overall word count. Finally, your literature review is likely to contain the following components: • a clearly defined research question • a clearly documented methods section • a clear presentation and analysis of the results of your literature search. Relevant literature might include primary research reports, books, discussion articles and other published information. The literature is analysed in order to shed new light on the topic question. • a final discussion section, in which you make conclusions and give recommendations based on the findings.
In summary
You should be starting to see how and why literature reviews are such an essential tool for health and social care professionals. First and foremost, they enable us to gain a comprehensive overview and summary of the available information on a particular topic. Literature reviews are generally more useful to the health and social care practitioner than any one individual piece of research because they allow one piece of research to be viewed within the wider context of others. The process of undertaking a literature review has also been introduced in this chapter. Emphasis has been placed on the importance of the literature review as a research method in its own right and its relevance as a research methodology for an undergraduate or postgraduate dissertation. We have also discussed the need to review the literature using a systematic approach in order to achieve an understanding of the body of literature as a whole in relation to a particular research. As a general rule, when you set out to review the literature, you should aim to undertake a systematic approach as outlined in this chapter, irrespective of whether it is feasible to achieve the detail in the review as required by the Cochrane Collaboration, for example.
Key points
• Literature reviews are an essential tool for those who work in health and social care in order to make sense of the range of information that may be published on any given topic.
• The literature review process is a research methodology in its own right and should commence with a research question, followed by a research design, presentation of results and finally, a discussion of the results.
• The literature review process can and should be approached systematically when undertaken by a novice researcher.
When do we write literature reviews?
· A literature review could be set as an assignment at the beginning of a course in order to get students ‘up to speed’ with the existing research in a given field of study
· A literature review often forms part of a dissertation or research project proposal, making an argument or producing justification for the new work that the researcher proposes to carry out.
· A literature review is thus a report on the current state of research in a given area of study that comments on the strengths and weaknesses of the research (landmark studies or writings) within a given field and perhaps gives recommendations for possible future study
· One last use for a literature review is that it can form the introductory section for a completed dissertation or project which gives the context into which the new research fits
Being critical
A literature review should gradually develop and refine a research problem or question that requires further academic investigation. As such, it should be a critical review of the landmark studies demonstrating their strengths and shortcomings in relation to a specific area of study (we shall discuss the process of getting started and defining an area of investigation below).
Theatre critics, despite being called critics, will often say what is good about a play or stage musical as well as what is bad. Likewise, a good review will consider the pros and cons of its sources. Being critical does not mean that you have to be wholly negative. You should however, only be discussing aspects that are relevant to the developing research issue. Your research will have two aspects to it. Firstly, your reading will begin to illuminate some research problems (i.e. areas that need further research). Secondly, when you write up your review you will use these sources to build up an incremental argument for exactly why this research is to be undertaken.
Above all else, you should avoid merely describing the research that you are reviewing. A literature review is not an annotated list, something akin to a ‘furniture catalogue’. The literature review has a purpose – to define the nature of the research problem and to produce a justification or recommendation of fresh research which needs to be carried out to solve that problem.
Narrowing the topic
A literature review may well form part of a larger task such as producing a dissertation. More than likely, a lot of time will be spent on it and on the new research that stems from it. Dissertations or final year projects may take an entire year to be completed. Therefore the topic area needs to be an interesting one or your enthusiasm may eventually wane. Try to choose something that you are passionate about or want to add your ‘two penny worth’ to. If you think your work is important and you have something to say, it will sustain you.
First choose a broad(ish) area and then do some preliminary reading to narrow the topic down. ‘Diabetes’, for example, would be too broad and unfocused for a final research topic. A little reading might help you to find a more sharply defined research area such as ‘treatment compliance in diabetes sufferers’.
Sources
Before we think about where to look, let’s take a moment to think about what sorts of sources we want to review. Obviously, our sources should relate to our investigation. They need to be of a good academic standard. Therefore it is important to think about the type of source – academic journals, books and websites are all valid. Again, be critical. Who is the author? What is the date? Is it a landmark study or peripheral to your investigation?
Use your initial keywords to search the university book and journal databases. A useful tip is to introduce yourself to your subject librarian. He/she will be able to show you how to get the most out of your searches and may even be able to introduce you to some material too! Keep a note of your searches so that you do not go over old ground and note any new useful keywords for further searching.
Other places to look:
· Your source’s sources! When you find a good source then look at its reference list for its sources and investigate those!
· The sources referred to in text books (avoid using the text book itself as these are secondary – use primary sources where possible)
· Any relevant course reading lists
· A little trick… type in your topic and the words ‘literature review’ to an internet search engine – you may get some useful resources but don’t be tempted to plagiarise! It is easily discovered and will not help you in the long term
With journals it is important to use the abstracts which are normally included in most cases. These are summaries of what the article contains: its evidence, argument and conclusions. Sometimes this may be all the information you need. It may also be an idea to practise skimming the article text for keywords (and also scanning the headings and bold type etc.). Read what you need to read and no more. Academic books are often more useful in that they provide content and especially indexes to look up keywords.
Taking notes and keeping records
When doing your reading it is important to take notes. It is important to extract the information that you need rather than to try to write up your review from a pile of photocopies. It is best to write your notes in your own words. This will help you to understand the information in a deeper way; you have to understand it to translate it into your own words and it will stop you relying too heavily on questions when you write up the review. It is also vitally important to keep good records of the source of your information for referencing purposes.
One method of note making and record keeping is to use note cards. The following ais examples of this two-card system. It consists of large note cards and smaller reference list cards.
Note Card
N
8×5” (203x127mm) record cards are ideal for this. Write only one piece of information from a source per card. Number your cards as you take your notes (this one is number eight) as this will allow you to refer to another note on the same subject (this one is reminding me that I took a note on a similar theme on card number five). Keep an in-text Harvard reference (Jones (1997:26). As you type up this note in the review, the in-text reference is therefore conveniently to hand. You’ll note that a page number (26) has been recorded. This is not normally necessary for an in-text Harvard reference where a quotation is not used, but is kept just in case there is a need to find the page again to check the details.
Reference List card
|
This note card will be used to write up your reference list. 5×3” (127x76mm) record cards are ideal. Smaller cards are used so that they can be put into a pocket and taken back to the library to enable you to find the source again, if necessary. This is also the reason for keeping the class mark (shelf number) and library name in the bottom left hand corner. The author and title are at the top of the card in order to ease finding again in a card index system. When writing up the reference list, all you have to do is arrange your cards into a pile in alphabetical order by the authors’ last names and type!
Writing up your review
Before writing up, be sure to plan your main body first. Statistics show that students who plan their work consistently gain the higher marks. The easiest way to plan is to put your note cards into piles by theme. These piles will form the core of your paragraphs (see the example paragraph below). Structuring your review by theme is much more effective than going over your sources chronologically as it will help you to avoid repetition and will enable you to develop your argument in relation to the research question and/or research recommendations.
On the following page is an example of a thematic main body paragraph which employs the three elements of Point (theme), Evidence (from your note cards) and Evaluation (the element which makes the paragraph ‘critical’).
POINT – EVIDENCE – EVALUATION
Studies in the field of diabetes treatment have touched on the issue of patient compliance. Investigations by Jones (1997) and Bloggs (2002) both uncovered the problem of the non-compliance of patients with their treatment regimes. However, this issue was not the main focus of either study and neither investigated the reasons for non-compliance.
You can see that the core of this paragraph is the evidence from notes 5 and 8 from the note cards. So plan by making piles and when you write up your rough draft of your main body you need to make the point (theme) of the paragraph clear and also state the value (evaluate) of the evidence in relation to the argument. You can see that Bloggs’ and Jones’ research touched on the right issue but did not go into more detail. We can therefore see the contribution this paragraph is making towards defining a research problem (i.e. that more research into the reasons for non-compliance is needed – this may be further refined as the review goes on).
The final draft
You may need to add an introduction to the review (if this review is an exercise in itself – it may not be necessary if the review is part of a proposal or bigger dissertation introduction). The review should however have a concluding paragraph which brings the main points together and states the research problem and recommendations.
The completed review should:
· Show a clear understanding of the overall topic and its development
· Discuss pros and cons of landmark research
· Show a range of definitions and approaches to the topic area
· Develop to show a clear research problem
· Reach sound conclusions/recommendations for further research, using a well-based argument.
