Chapter 11
Situating a Study: The Literature Review
Academic knowledge is not an objective reality that is "out there," waiting to be discovered; rather, it is something that academics actively construct. And as any constructed entity, knowledge, too, can be accepted or rejected by members of the academic community. The task of any academic writer, therefore, is to persuade the reader to accept his or her claims to knowledge as credible and legitimate.
To be accepted as credible and legitimate, new knowledge must build on prior, existing knowledge. Researchers show that their study builds on prior knowledge by posing questions that are warranted by the current state of research, by employing theories and methodologies that are sanctioned by the discipline and its academic community, by using definitions that have been accepted by the researchers working in the target area, and by basing their expectations on previous research—in other words, by situating their study within the body of existing academic literature and demonstrating that it fits that literature.
Textbooks on academic writing often use the term literature review to refer to the idea of situating a study within the existing body of knowledge. Yet, the term literature review is somewhat confusing because it does not clearly convey what the writer needs to do. Students who are new to graduate study often take the term literally and interpret it as an instruction to write a detailed, often chronological, account of everything that has been said on their topic in order to demonstrate familiarity with existing research. Such literature reviews are often recognizable by their organization: Every paragraph is a review of a separate study, starting with "Author so and so investigated..., Author so and so found..., or According to Author so and so,... ." There is no attempt to show how all of those studies are related, and, especially, how they are connected to the student's work.
In fact, rather than reviewing existing literature on the topic for the sake of demonstrating familiarity with it, what authors need to do is make an argument or a series of arguments that would persuade the reader to accept the author's own study as necessary, valid, and perhaps even interesting. Recall from Chapter 5 that an argument is a claim that is supported by evidence—in this case, evidence from previous research. Thus, what you need to do is use
How to Write about Economics and Public Policy. https://doi.org/10.1016/B978-0-12-813010-0.00011-9
© 2018 Elsevier Inc. All rights reserved.
207
208   How to Write about Economics and Public Policy
previous literature to make and support your arguments. These arguments may be about, among other things,
• The importance or timeliness of your topic,
• The state of current knowledge on your topic (both theoretical and empirical),
• The use of specific approaches or methodologies in your area,
• The variables that you have selected to include in your model,
• The specific expectations, and/or
• The definitions you have chosen for your concepts.
A literature review, then, could best be thought of not as a review of existing works on a topic, but as an argument that provides a very selective account of what is known on a particular topic and helps the author build a case for his or her own study and justify his or her own choices and expectations.
The use of literature in an academic study is critical because previous research acts as a way of legitimizing a study and persuading the reader to accept it as needed. There are, however, large differences among studies in how academic literature is organized and presented and how much of it is used in a given study. These differences reflect the requirements of various journals or degree programs, the wishes of individual professors acting as students' advisors, and to some extent, the individual preferences of the authors. Crucially, however, these differences reflect the conventions of different disciplines and research areas.
In this section, I give some general suggestions on situating a study in the context of previous research in public policy and economics. However, because the specific ways in which the literature is used in a study, its organization, and the language that authors use to make arguments about the state of research will depend on the discipline and research area you are working in, make sure to check papers that are most relevant to your research. Study them to learn how authors working in your area use literature to situate their study—what claims they support, how they organize the literature, and where exactly in their paper they talk about the work of others.
SUGGESTIONS FOR USING THE LITERATURE Use (Mostly) Disciplinary Literature
Perhaps the most important criterion for selecting literature for a study is the relevance of the literature you select to the discipline or research area to which your study will be contributing. If you are contributing to a research area in economics, then most of the literature in your study should be related to economics; if you are contributing to a research area in sociology, then most of your literature should come from sociology.
The difficulty, of course, is that many problems in public policy are interdisciplinary and their research often necessitates pooling together concepts,
Situating a Study: The Literature Review Chapter | 11 209
theories, or tools from several disciplines or research areas. For example, the problem of school attainment can be examined from an economic perspective as an educational production function; however, you may also want to employ a psychological theory of motivation to gain further insights into the problem or use research in sociology to identify important school-related variables to include in your model. In this case, you should use literature from all relevant disciplines. However, the extent to which you will use literature from a particular discipline should be proportionate to the importance of that discipline in your research. Ultimately, you should ask yourself who your readers will be: If your research will be read primarily by economists, then you should relate it to research in economics and use mostly economic literature.
Prefer Empirical Literature to Nonempirical
As I explained earlier, there are two kinds of academic literature: empirical studies, which can be quantitative or qualitative, and nonempirical literature, which comes in a wide variety of forms, from theoretical studies to systematic literature reviews to policy-oriented argumentative essays. When researching a problem, you can use both kinds of literature to get an idea about the state of research on your problem. However, when using literature to support claims about the state of current knowledge (e.g., what has and has not been investigated, what is more or less important, what affects what), prefer empirical literature to nonempirical unless you are summarizing theories or theoretical positions.
The use of empirical literature as support for a claim is especially important if you use nonintegral (or parenthetical) citations—citations that are not included in the running text and that usually appear in parentheses at the end of a claim. In fact, readers in many disciplines often assume that the parenthetical citations that they see at the end of authors' claims refer to empirical work. For example, the following statement would be interpreted by many readers as an argument that is being made on the basis of an empirical study:
In post-communist countries, privatized organizations appear to be more efficient than government-controlled organizations (Smith, 1999).
That is, to many readers, this sentence would mean that Smith conducted an empirical study of privatization, in which he found that privatized organizations were more efficient than those controlled by the government, and that the author of this sentence is using Smith's findings to support this claim. If, however, Smith did not conduct an empirical study but merely made a claim somewhere about privatized organizations being more efficient than government-controlled ones, then the use of the parenthetical citation to support this claim may be confusing.
210   How to Write about Economics and Public Policy
If you use nonempirical literature—for example, a researcher's theoretical arguments—to support your claims, make it clear in your narrative that the work you are referring to is nonempirical. Instead of using parenthetical citations at the end of the sentences, consider integrating citations into the text by using signal phrases such as: Smith argued in his review..., Brown theorized..., or
Miller proposed a view.....For more on integral and nonintegral citations
and their use in different disciplines see Chapter 16.
Consider if You Need a Separate Section for the Literature
Students often ask if they should have a separate section dealing with the literature. The answer depends on the type of paper you are writing and the requirements of the journal or educational program to which you are submitting your paper. Many empirical articles in economics and public policy that appear in academic journals do not have a separate literature review section; instead, authors make arguments about the importance of their research, the inclusion of particular variables, or the appropriateness of their approach and methodology in the Introduction. The part that deals with existing literature in such studies can be quite short—as short as a few paragraphs.
Master's theses and especially doctoral dissertations dedicate much more space to situating the study in previous research and usually have separate sections devoted to reviewing relevant literature. In economics papers, it is also common to break a literature review into two parts—theoretical literature and empirical literature—and review theoretical and empirical literature separately. Nonempirical papers almost never have a separate section devoted to the literature because they are usually based entirely on the literature. Such papers are often organized thematically.
Support All Claims that Are Not Based on Your Own Findings
In an academic study, claims that are not based on your own findings should be supported with literature—primarily empirical literature. Recall from Chapter 1 that an academic paper is a dialog in which the author is trying to persuade the reader to accept his or her study. In this dialog, readers can easily reject an author's claims if they find them inappropriate or baseless. The writer's job, therefore, is to anticipate readers' questions or objections and respond to them in an acceptable way.
Novice students sometimes hold mistaken beliefs about the use of literature in a study, thinking that if they cited a lot of studies in their paper, their advisor would think that they were unoriginal. But it is important to realize that a study's originality and, in fact, its usefulness can only be judged by what the author actually did—by the study's design and execution, and not by its claims and citations.
Situating a Study: The Literature Review Chapter | 11 211
In fact, supported claims look stronger than unsupported claims and they make writing appear more professional because they show that the author has researched the issue and has found evidence to support his or her claims. To see that, compare the following two passages. In the first passage, all citations have been removed and the claims are left unsupported; in the second one, all claims are supported with citations. Which one looks more credible to you?
Passage 1
Although women are achieving greater equity in certain labor sectors, they are not making much progress in the higher echelons of the business world. Women account for about 5% of senior management in corporate America. While the earnings differential between male and female managers has decreased in recent years, this gap continues to exceed that reflected in the labor force as a whole. In one 1991 study, the greatest salary gap between men and women occurs with managerial positions, where women in full-time management jobs receive only 61 % of the salaries received by men in similar positions. (Tsui, 1998, p. 364)
Passage 2
Although women are achieving greater equity in certain labor sectors (Blau, 1994), they are not making much progress in the higher echelons of the business world. Women account for about 5% of senior management in corporate America (Bickley, 1996). While the earnings differential between male and female managers has decreased in recent years, this gap continues to exceed that reflected in the labor force as a whole (Jacobs, 1992). In her 1991 analysis, Crispell reported that the greatest salary gap between men and women occurs with managerial positions, where women in full-time management jobs receive only 61% of the salaries received by men in similar positions. (Tsui, 1998, p. 364)
In Passage 1, there are no citations, and as readers, we may wonder, quite naturally, where all this information comes from. Why should we accept the author's claims, for example, that "women are achieving greater equality" or that "the earnings differential between male and female managers has decreased"? In contrast, in Passage 2, we can see plenty of supporting evidence, which is shown in the form of citations. We can even check this evidence to see if the author's interpretation is accurate or if we agree with that interpretation. These citations increase the author's credibility in the eyes of the readers and the readers' confidence in the author's knowledge of the subject matter. Situating a study, therefore, means framing it as a response to the ultimate readers' questions: What is the basis for your claims? Where is your evidence?
212   How to Write about Economics and Public Policy
Evaluate Rather than Merely Describe
Situating a study in previous research means not only describing relevant research but also, crucially, evaluating it—or, as Feak and Swales (2009) put it, "taking a stance toward the literature" (p. 71). In fact, a review of the literature without some sort of evaluation is hard to imagine as such a review would be difficult for readers to interpret. How, for example, can readers interpret the fact that Smith found one thing, Brown found another thing, and Miller found something else?
Evaluation helps the author organize the material and present disparate studies as coherent arguments showing what is more or less important, significant, or well-established; how common a particular finding or approach is; or if there is another view on the problem. Evaluation also helps the author show his or her own stance toward the literature as a whole or toward a particular study. For example, authors may describe a question as "big" or "important," an analysis as "accurate" or "myopic," or a consensus as "lacking" or "overwhelming," thereby indicating their own position toward what they are describing, which may encompass a wide range of feelings, from criticism to admiration.
To see the importance of evaluative statements, compare these two sentence openings:
1. There is some literature that shows that...
2. A large and growing literature has shown that...
In the first sentence, the author merely states the fact that there are studies showing something. Such a statement would be difficult to interpret beyond its literal meaning and it might leave at least some readers wondering, So, what? In the second sentence, however, the author makes an evaluative statement about the body of existing literature: the words "large and growing" show the author's own assessment of the state of the field and his or her belief that there may be a potential consensus. Such a statement also helps the reader quickly grasp the state of the field and see how the author's study fits into it.
Note that evaluative statements are typically followed by nonintegral (parenthetical) citations to studies that have found support for the author's claims or by explanations and examples from relevant studies, which in turn are followed by citations to those studies. This is important because many readers would find it difficult to accept unsupported evaluations.
Below are some examples of evaluative statements. Read them and notice how they help the author organize the literature. What stance toward the literature does the author show in each statement?
A large and growing literature has shown that maternal health determines offspring's health and productivity.
Situating a Study: The Literature Review Chapter | 11 213
Aid effectiveness has been a subject of much debate in the past decade.
There is a considerable lack of consensus on the importance of nonfarm activities to the incomes of rural households.
The empirical work on the relationship of land reform and economic growth is mired in controversy.
The early literature on economic development was strongly influenced
by the work of Brown (1965).
In a pioneering contribution, Brown (1998) demonstrated a strong association between economic development and life expectancy.
Tsang (2002) argues that in China, unified inflation targeting may not be the optimal solution because differences in inflation rates among the provinces may undermine the effectiveness of a unified inflation-targeting monetary policy. His arguments, however, are based on an empirical analysis of a rather limited set of data. Furthermore, in his analysis, Tsang did not compare social welfare loss under different inflation targeting policies, which significantly weakens the overall conclusion.
ORGANIZING THE LITERATURE TO MAKE A POINT
Situating a study in previous research typically requires making three kinds of statement. They are briefly described below.
Statements about the Overall State of the Field
Statements about the overall state of the field include statements about the direction in which the field is going, its major findings, its consensuses and disagreements, its main theoretical positions, and so on. The purpose of these statements is to summarize current research in a succinct manner in order to establish a niche (Feak & Swales, 2009) for the author's study. These statements are usually made on the basis of empirical work and they take the form of claims followed by nonintegral, parenthetical citations to studies that support those claims. For example:
Do elections increase public spending? Some say yes (citations) and others say no (citations).
Several studies suggest that the new policy has several benefits including a reduction in child mortality (citations), improved maternal health (citations), and a greater access to sanitation facilities (citations).
The root causes of corruption in politically centralized systems reflect not only economic realities but also perceived cultural norms (citations).
214   How to Write about Economics and Public Policy
If you look closely at these statements, you will notice that they summarize relevant research by imposing some sort of order on the literature—by grouping and presenting studies according to some criterion. This is how the overall state of research in a field is usually described.
Common criteria for grouping studies include
• Approaches or methods that have been used to study a problem or phenomenon (e.g., experimental vs. observational methods or the use of one particular model vs. another);
• Major findings, especially if they have been mixed (e.g., some studies have found a positive effect and others, a negative effect);
• Main controversies (e.g., different opinions about the threshold value of inflation needed for inflation to exert positive effect on economic growth);
• Main theoretical positions (e.g., neoclassical view vs. Keynes's view of the relationship between government expenditure and economic growth);
• Different kinds of determinants of some outcome (e.g., policy-related vs. other determinants of economic development);
• Different views of a phenomenon (e.g., linear vs. nonlinear view of economic development); or
• Different settings (e.g., predictors of economic growth in developed vs. developing countries).
The way you organize and present relevant literature should help you frame your study in relation to that literature. For example, if you wish to frame your study as an attempt to resolve a controversy, focus on mixed results and present relevant literature according to whether it shows a positive or a negative effect of some variable on another variable or variables that you are interested in. If you wish to present your study as an attempt to extend existing knowledge about the determinants of some outcome from developed to developing countries, you may want to group studies by economic status. You may further want to show that there are important differences between studies conducted in developed and in developing countries and that these differences may account for the different results that have been obtained. Or if you are interested in the effects of a particular policy, you may want to organize your literature according to the type of effect that the policy has had.
Below are two examples from published studies showing how authors impose order on the material they present. In the first example, the authors review two types of argument on gender wage differences. In the second example, the authors present studies of privatization that have been grouped by study design. Notice the use of citations to support claims in both examples.
Situating a Study: The Literature Review Chapter | 11 215
Example 1
The economic literature on gender wage differentials has identified multiple reasons why women and men have generally different levels of earnings. Basically, there are two types of arguments. The first is concerned with differences in human capital. Due to the higher incidence of expected career breaks, women make different human capital choices (both education and on-the-job training choices), and this in turn leads to job segregation. Many studies have shown that segregated labor markets are the main reason for gender wage differences (e.g., Meyersson-Milgrom et al., 2001; Korkeamaki and Kyyra, 2006; Wolf and Heinz, 2007). The other main argument is based on labor market discrimination.
PRP [performance-related pay] increases wages due to its impact on selection and effort (Lazear, 2000; Pekkarinen and Riddel I, 2008). Tying pay to performance attracts high-ability employees and provides incentives to increase their effort. Thus, PRP may affect the gender wage gap through several mechanisms: (1) discrimination, (2) segregation, (3) differences in selection effects between sexes and (4) differences in effort effects. We now consider each of these mechanisms in turn. (Kangasniemi & Kauhanen, 2013, p. 5133)
Example 2
The assumption behind privatization in many parts of the world is that private ownership improves corporate performance. The empirical evidence for this assumption comes from two kinds of studies. The first, exemplified by Megginson, Nash, and Van Randenborgh (1994) and La Porta and Lopez-de-Silanes (1997), compares pre- and postprivatization performance of selected privatized firms. The second focuses on comparing the performance of state firms with either private (Boardman and Vining 1989) or privatized (Pohl et al., 1997) firms operating under reasonably similar conditions. (Frydman et al., 1999, p. 11 53)
Statements about Most Relevant Studies
Statements about the overall state of the field and major findings are usually followed by more detailed reviews of empirical studies that are particularly relevant to your research. Rudestam and Newton (2001) call such studies "very relevant literature" (p. 64) and explain that these are empirical studies that focus largely on the same relationship as the one you are interested in, incorporating all or most of the variables that you will focus on in your own study. These studies form the foundation on which your study will build and they are used to justify expectations, definitions, or the choice of particular variables, measures, methods, or models. Authors usually devote considerable space to reviewing very relevant studies, describing how a study was done (i.e., its methodology), what results were obtained, and what those results mean.
216   How to Write about Economics and Public Policy
Often, you will need not only to review very relevant studies but also to evaluate them and explain why they are insufficient—and why, therefore, your study is needed. One way to do that is to point to the differences between those studies and your own study, which may be differences in purpose, focus, or methodology. Another common approach is to point out flaws in the design of the most relevant studies and show how your study will fix those flaws. Below are examples from research papers showing how authors review most relevant studies. The first two come from student papers and the last one, from a published paper.
The first example comes from a paper by Efita Fitri Irianti, a student in Economics, Planning, and Public Policy, who examined the relationship between fiscal decentralization and human development. She begins her review with a statement that implies the importance of her topic: Several studies have examined.... She then describes in detail the purpose, design, and results of a most relevant study before offering her interpretation of the study's results in the last sentence. Read this example and notice the phrases she uses to review the study and draw her own conclusion.
Several studies have examined the relationship between fiscal decentralization and human development. For example, Habibi et al. (2003) studied the impact of fiscal decentralization on human development in Argentina using health (infant mortality rates) and education (secondary school enrollment) as dependent variables. The authors used panel data analysis at the provincial level over a period of 25 years from 1970 to 1994. The variables included as independent variables were per-capita income of public employees, per-capita total expenditure, ratio of provincial taxes to own revenue, ratio of own revenue to total revenue, ratio of royalties to own revenue, and ratio of conditional transfers to total transfers. They found that infant mortality rates had a significant negative association with the ratio of provincial taxes to own revenue and the ratio of own revenue to total revenue. For educational output, the ratio of provincial taxes to own revenue and the ratio of own revenue to total revenue were positively and significantly associated with secondary school enrollment. These results imply that fiscal decentralization may have a positive impact on human development. (Irianti, 2014, pp. 4-5)
The second example is from a paper by Itai Maparara, a student in Public Finance, who examined the direction of causality in the relationship between government expenditure and economic growth. In his theoretical framework, Itai outlined several theoretical views on the relationship he was interested in including those of Keynes and Wagner (see his extract in Chapter 10). Here, he reviews two empirical studies to show how they are related to those theories and to justify his own approach. Notice the use of the word "interesting" to describe the second study. Why do you think he uses this word? Also notice that after reviewing this study, he interprets its findings in a way that helps him justify his own research.
Situating a Study: The Literature Review Chapter | 11 217
Examining the case of a developing country, Nasiru (2012) employed a bound test to cointegration and Granger causality tests to determine causality in Nigeria for the period from 1961 to 2010. The study categorized government expenditure into capital and recurrent expenditure. The results show that no causality exists for these expenditure categories in the long run. However, in the short run, causality flows from government capital expenditure to economic growth, thus supporting Keynes' views. The study concluded that government can influence economic growth by shifting expenditure from recurrent to capital formation.
One interesting study was carried out by Loizides and Vamvoukas (2004). The study examined causality in a bivariate and trivariate framework using an error correction model and a Granger causality test, enabling the comparison of results from different frameworks. In the bivariate framework, data analysis supported Keynes' views for the United Kingdom and Ireland in both the short and long run whereas the data for Greece validated Wagner's inferences. However, when causality was examined in the trivariate framework by adding inflation as a third variable, the UK data supported Wagner's views. These results imply that omitting certain explanatory variables may influence results on the direction of causality.
Therefore, the present study directly proceeds into the trivariate framework of analysis in order to confirm the findings of Loizides and Vamvoukas (2004). This study will also employ error correction model and Granger causality tests as a strategy. The uniqueness of this study is the choice of a third explanatory variable, trade, included in the model. By including trade, the study can examine whether the regional group has influenced government expenditure and economic growth in its member countries through agreed-upon trade policies. (Maparara, 201 6, pp. 3-4)
Statements about Other Authors' Arguments or Theoretical Positions
Although empirical studies often constitute the bulk of the literature that authors review in order to situate their study, nonempirical literature is also often used. For example, authors may need to describe theories or theoretical positions, theoretical or policy-related arguments advanced by others, or conclusions from a comprehensive literature review on their topic.
Nonempirical literature is often used to support theoretical definitions of concepts, various classifications, or theoretical views and expectations. For example, in a study of political instability and economic growth, you may use theoretical literature to support a definition of conflict or political instability; in a study of farmers' motivations to convert to organic farming, you may use a comprehensive literature review to derive a classification for those motivations, dividing them, for example, into economic and noneconomic ones. However, nonempirical literature should generally not be used to support claims about empirical relationships unless you are deriving predictions from a theoretical study.
218   How to Write about Economics and Public Policy
As I explained earlier, you should make it clear for the reader when you use nonempirical literature. Avoid the use of nonintegral, parenthetical citations at the end of sentences when reviewing nonempirical literature; instead, introduce nonempirical works with signal phrases (e.g., Brown argued..., Smith contends...) and briefly indicate the type of study you are referring to: Brown's (1999) theoretical analysis has shown that... ; Using a systematic review of relevant literature, Smith (1999) derived....
When reviewing nonempirical literature, it is important to explain how it is related to your purpose. This often requires that you evaluate and critique authors' arguments in relation to your own study. For example, if you are summarizing a theoretical view, explain how it relates to your expectations or why it may or may not be applicable to your particular context; or if you are describing several possible definitions, evaluate them in relation to your methodology and explain which one is preferable in your context and why.
Below are some examples from research papers showing how authors use nonempirical literature to situate and justify their study. The first example comes from Itai Maparara's paper on the relationship between government expenditure and economic growth. Here, Itai describes a theory relating government expenditure and economic growth. Notice that after describing Wagner's views, he also explains what they mean for his own study. Notice also how he justifies the need for his study in the last sentence. What argument does he make?
The law of increasing state activity (Wagner's law) developed by Wagner (1 893) resulted from his empirical analysis of government expenditure and economic growth for five Western European countries. It states that government spending increases faster than economic growth in progressive economies. Wagner (1893) contended that such trends are evident because governments suffer pressure from social progress, which demands changes in relative spheres of private and public economy. As governments respond to such demand, their expenditure increases. He further asserted that since governments are financially handicapped, growth of public spending cannot precede economic growth. In other words, the basis for financing additional expenditure is growth of the economy (Peacock & Wiseman, 1961). His views support the perception that the direction of causality should flow from economic growth to government expenditure. However, Wagner (1893) analyzed data from a century ago and for countries with governments that were different from those of today. The role of government has gone through transitions so that a reexamination of the empirical evidence for his claim has become imperative. (Maparara, 2016, p. 2)
In the second example, Nana Mensah Otoo, a student in Public Finance, summarizes theoretical arguments that connect intellectual property rights (IPR) protection and foreign direct investment (FDI). Notice that after reviewing these arguments, he tries to explain what they mean and gives an overall
Situating a Study: The Literature Review Chapter | 11 219
evaluation. Notice also that he uses another author's argument to support his assertion that theory may not provide a clear insight into the relationship between IPR and FDI. Why do you think he does that?
It is useful to briefly review why IPR protection might matter for FDI theoretically. IPR protection is often discussed in the context of innovation and creative activity. According to Maskus (2004), intellectual outputs have the characteristics of a public good: once produced, they are available on a nonexclusive basis. The author further indicates that intellectual outputs are also nonrivalrous in use, that is, additional par- ties can benefit from them at zero additional cost. Hence, these factors make it difficult for producers of intellectual outputs to appropriate the returns to their investments and recoup costs. Maskus stresses that in the absence of property rights, the market for intellectual outputs would fail oryield an inefficient supply of output. Regarding the connection to FDI and trade, Braga and Fink (1998) have argued that because many producers of intellectual output are engaged in both domestic and foreign markets, risks of unauthorized copying and imitation exist both at home and abroad. The authors argue that in regions where IPR protection is weak, incentives to market (via trade or FDI) might also be weak. Moreover, weak IPR protection (and smaller markets as a consequence) may adversely affect incentives to innovate and produce, thereby affecting the potential to export and invest abroad. Similarly, the theoretical study by Taylor (1994) indicates that stronger patent protection increases technology transfer when competition exists between foreign and domestic investors. However, theory does not necessarily provide clear insights into the IPR/FDI relationship, as Maskus (2000) noted in the case of patents and trade:
Theoretical models do not clearly predict the impacts of variable patent rights on trade volumes. Much depends on local market demand, the efficiency of imitative production, and the structure of trade barriers. Also important are the reactions of imperfectly competitive firms. Thus, a clear picture can emerge only from empirical studies, (p. 113).
(Otoo, 201 3, p. 2)
The above suggestions for organizing the literature apply in equal measure to short reviews presented in Introductions and to longer reviews presented in separate sections. In the Introduction, one or more statements about the overall state of research in the field are often followed by a more detailed focus on just a few studies, which are reviewed in detail and critiqued. In papers where there is a separate section for reviewing the literature, detailed reviews and critiques of relevant studies are usually presented in the Literature Review section whereas statements about the overall state of research are presented in the Introduction.
The specific organization of the literature will also depend on whether the paper is quantitative or qualitative. If you are writing a qualitative paper, your literature review can be organized thematically. A common strategy is to break down the main research question into subquestions and use those subquestions as section headings in the literature review.
220   How to Write about Economics and Public Policy
If you are writing a quantitative paper examining a relationship between variables, you may want to have two separate sections, a theoretical framework section and an empirical evidence section. In the theoretical framework section, review the theory on which your study is based. If there is currently a theoretical debate on your topic, review both sides of the debate and explain the expectations from both sides. In the empirical evidence section, group studies describing the background of your research according to the criteria that are relevant to your purpose, for example, the context in which the studies were done (e.g., developed vs. developing countries) or the time period in which they were done (e.g., before or after the introduction of a policy).
COMMON PROBLEMS
Below are some common problems that students who are new to graduate study may have when working with academic literature and reviewing studies in order to situate their own research. Some of these problems have been described earlier in this chapter; here, I summarize them in one place.
Not imposing order on the material. The student reviews disjointed studies without imposing any order on the material—without showing how the studies are related to one another and/or to the student's own study.
Flipping dependent and independent variables. The student writes a paper about the role of X in Y (e.g., urban poverty in childhood obesity). Here, the hypothesized direction of causality is from X to Y (from poverty to obesity). However, the literature review focuses on studies that have examined the causes of urban poverty—in other words, on studies that treated urban poverty as the dependent variable and that would be largely irrelevant to the relationship between urban poverty and obesity. Thus, when reviewing literature, always keep in mind the direction of the hypothesized relationship—what are you hypothesizing to affect what?—and make sure to review studies that bear directly on that relationship.
Using the wrong literature. Sometimes students use literature that is almost exclusively policy-related or that is not disciplinary. Other times, a student may include studies that, while being on the same topic as the student's paper, do not actually support the student's arguments. If you are writing an academic paper, make sure that most of your literature is scholarly rather than policy-related, that most of it comes from the discipline you are contributing to, and that it actually supports your claims.
Not developing an argument. A common mistake is simply to review a study, describing what was done and/or what was found and leave it at that. Such a review is rather difficult to interpret, making readers wonder what the author is trying to say. Use literature to develop an argument, not just show that you have read it. This means drawing conclusions from studies; interpreting
Situating a Study: The Literature Review Chapter | 11 221
them; explaining what they mean, individually or collectively; and evaluating and critiquing them.
Presenting a claim without any evidence. It is often difficult for novice researchers to know which claims do, and which ones do not, require literature support. As a general rule, assume that all claims that are not a result of your own investigation require support from the literature. For example, consider the following statements:
Women are less financially literate than men.
Globalization leads to inequality.
Firms with employee-friendly work environments achieve greater innovative success.
All of these claims require some sort of support. If you are using them to justify the importance of your topic or your expectations, support them with previous research. If these arguments are based on your own findings, indicate this in the narrative by using such phrases as
We/I find that...
Our/my study indicates... Our/my results suggest...
Supporting claims with other claims. Students sometimes believe that as long as there were a parenthetical citation at the end of their claim, their claim would be valid. This is not necessarily true. As I explained earlier, evidence in academic research usually means empirical evidence, so claims should be supported with empirical evidence rather than with other people's claims or opinions.
Presenting evidence and making illogical claims. Sometimes students misunderstand what a study really shows; other times they may misinterpret a study's findings or make an illogical claim from a perfectly legitimate study. To avoid such mistakes, check carefully what the study you are reviewing really shows, what its author(s) claim, and whether the study does support your own arguments.
Criticizing rather than critiquing. Students often know that when reviewing the literature, they need to critique it. However, some confuse critiquing with criticizing. The difference is that the former focuses on analyzing and evaluating, whereas the latter focuses on finding faults. In your review of the literature, focus on critiquing and avoid criticizing.