Writing a psychology personal statement

psychology writing sample and steps for writing psychology literature review
Dr.PeterCena Profile Pic
Dr.PeterCena,Swaziland,Researcher
Published Date:02-07-2017
Your Website URL(Optional)
Comment
Writing for Psychology A Guide for Psychology Concentrators Department of Psychology HARVARD Faculty of Arts and Sciences COLLEGE Harvard UniversityWriting for Psychology A Guide for Psychology Concentrators by Shelley H. Carson Jeanne Fama Kate Clancy Jeffrey Ebert Adrienne TierneyCover photo credit: Harvard News Office Copyright 2012, President and Fellows of Harvard CollegeTable of Contents Acknowledgements ........................................................1 Introduction ....................................................................3 Chapter One How to Read Sources Critically .......................................5 Chapter Two Writing a Conceptually Coherent Paper ........................13 Academic Honesty in Writing ........................................22 Chapter Three Do’s and Don’ts of Effective Writing in Psychology ........24 Chapter Four APA Format Guidelines…………………………… ..........30 Appendix: Locating Databases and Sources in the Harvard Library System ..................................................38 Acknowledgements This booklet contains the accumulated wisdom of a number of eminent psychologists and writing experts, including Daryl Bem, Stephen Kosslyn, Brendan Maher, Joseph M. Williams, and the authors responsible for the American Psychological Association style manual. We have also drawn on the keen observations of Baumeister and Leary (1997) and Gordon Harvey (2002). We would like to thank Stephen Kosslyn, C.A. Meyersburg, Kristina Olson, James Herron, and Laura Chivers for their input and assistance in the preparation of this guide. This guide was prepared with the financial support from the Harvard Writing Project. Writing for Psychology page 1 intr introduction oduction Writing for psychology incorporates many of the organizational elements you learned in Expository Writing. In Expos, you were taught general academic guidelines for formulating a thesis, providing a motive for the thesis, supporting this thesis with convincing evidence, and anticipating objections from readers. You were also taught the principles basic to scholarly writing across the curriculum. These critical thinking/writ- ing skills, as well as the ability to form and support an argument, create a foundation on which you will build the more specialized skills required for psychological writing. Writing in the field of psychology (like writing in any specialized field) differs in several respects from the general academic writing style you learned in Expos. Psycho- logical writing is a form of scientific reporting that is based on American Psychologi- cal Association publication style, widely recognized as a standard for scientific writing. The format employed in psychological writing (APA style, discussed below) reflects the principles of clarity, concise wording, and accuracy and facilitates the rapid and logical flow of information from author to reader. Thus, scientific writing values prose that is more straightforward, objective, and less reflective than what you may be used to. In the beginning, you may feel that APA style is dry and colorless, and that it stifles your creativity. However, after a bit of experience, you will find that the guidelines delineated by APA style will help you to write clear, informative, interesting papers. Creativity in psychology tends to come from the ideas behind the writing, not the writing itself. This booklet is designed to acquaint you with the basic principles of psychological writing and to help you avoid pitfalls that beginning writers in the field often encounter. First, in How to Read Sources Critically, we will discuss why it is important to for you to be a discerning reader of the work of other psychologists, and we will present guide- lines to help you read critically. In Writing a Conceptually Coherent Paper, we will go through, step by step, the process of writing an essay or term paper in psychology. The section on Academic Honesty in Writing reinforces information you have previously received about using sources responsibly (and avoiding plagiarism). The Do’s and Don’ts for Effective Writing in Psychology include examples of common mistakes made by beginning writers in the field. Finally, the APA Format Guidelines summarizes some of the basic elements of APA style. For more complete information on science writing, psychological writing, and APA style, we recommend the Publication Manual of the American Psychological Association, 6th Edition (2009) and the APA Style Guide to Electronic References, as well as the other excel- lent references listed at the end of this booklet. Writing for Psychology page 3chapter one How to Read Sources Critically The guidelines that follow are based on the wisdom and advice of numerous research- ers, writers, teachers, and students who have helped us understand what makes a good psychology paper. Much of the information that follows is explained in greater detail by Kosslyn and Rosenberg (2001) and Maher (1978). You are encouraged to read both sources directly. The first step in learning to write well in field of psychology is to learn to read sources critically. There are at least two reasons for this: 1. In order to write well, you must be well informed about the subject matter. 2. By training yourself to spot strengths and weaknesses in others’ theses, ar- guments, methodologies and conclusions, you will become more aware of strengths and weaknesses in your own work. This awareness should help you to become a critical reader of your own writing. Researchers write review papers, theoretical papers, and reports of empirical studies in order to advance a point of view. You, as a reader, are more likely to adopt a writer's point of view if the writer backs up this point of view with solid evidence. As a critical reader of psychological literature, you must consider both the quality and the quantity of the evidence that a writer uses to support his or her argument. You will also need to consider the following: • implicit or explicit author bias (the author’s agenda) • soundness of the methodology • appropriateness of the statistical analyses • whether the strength of the conclusions matches the strength of the evidence This section of the guide will teach you to make these determinations about the quality of evidence, which will help you to read sources critically and to write your own papers persuasively. I. Consider Whether the Information in the Source is Evidence What is viewed as evidence in one academic discipline is not always viewed as evi- dence in others. In some fields in the humanities or social sciences, logic and rhetoric are forms of evidence. In the field of psychology, however, evidence is in the form of empirical research results: data becomes evidence once it is evaluated in the context of a hypothesis. Empirical data arise from observation or experimentation under controlled conditions; in contrast, opinions are personally-held convictions that may or may not be based on controlled observations, and may even contradict well-gathered data. Writing for Psychology page 5 Example: In many scientific fields, the opinion held by the majority has been that men are more attracted to science than women, and that this explains not only the larger numbers of men in science, but also their greater success. This opinion was so widely held that for many years little was done in any scientific institution to improve access to science for women. However, in the last decade several empirical studies have demonstrated that bias against women explains a significant portion of the difficulty women have in succeeding in science and that blinding fellowship committees and article reviewers leads to better rep- resentation of women in science. Wenneras and Wold (1997) found that men were regarded as more competent than women in a funding competition, even when their objectively-measured productivity was identical. More recently, Bud- In the field of psychology, den et al. (2008) showed that double-blinding the review of articles for publica- evidence consists of tion in the journal Behavioral Ecology (neither reviewer nor reviewee know the other’s identity) led to a significant increase in the publication of articles where empirical research results the first author was female. These results do not completely condemn the opin- rather than quotations and ion that men are better equipped or are more attracted to science as a profes- opinions of scholars. sion. But they provide positive evidence in support of the argument that bias against women has played some role in their underrepresentation. The differ- ence between opinion and empirical data demonstrated by this example is that while many people may have had this opinion about women in science, only the empirical studies provide evidence on the issue. Empirical studies go beyond the conjecture of opinion—as informed as that opinion may be—and provides data that serve as evidence. It is perfectly okay to consider the opinion of an author as well as the opinions of other writers cited by that author (secondary sources). However, remember that opinions do not constitute evidence. II. Consider the Quantity of Evidence: In addition to considering the type of evidence a writer uses, you should consider the amount of evidence that supports the writer’s argument. In psychology (as in all science), it is common to find empirical studies that report contradictory findings. This is why researchers place so much emphasis on the replication of results. Multiple data points (i.e., converging evidence from multiple studies or multiple measures) with the same results are more convincing than a single observation. An author may present contradic- tory findings, but his or her conclusions should be supported by the preponderance of In psychological writing, data presented. the opinions of authors Example: In 1994, psychologists D. Bem and C. Honorton conducted and ana- lyzed eleven studies on extrasensory perception (ESP). They used a method called do not constitute evidence. the ganzfeld procedure, in which two participants sit in separate rooms and try to convey information to each other. Bem and Honorton (1994) concluded that participants were able to do this at greater rates than chance, thus supporting the concept of ESP. In a later analysis of 30 ganzfeld experiments conducted in seven different labs, other researchers (Milton & Wiseman, 1999) concluded that participants were not able to communicate telepathically at rates greater than chance. In the absence of methodological concerns about the studies Milton page 6 Chapter One: How to Read Sources Criticallyand Wiseman analyzed, it would be reasonable to conclude that preponderance of the evidence from published studies did not support the demonstration of ESP. Therefore, even though Bem and Honorton (1994) found evidence of ESP in one study, one cannot conclude that ESP exists on the basis of one study if many other studies found evidence against ESP. In evaluating evidence, you must ex- amine whether the finding has been replicated across more than one study. III. Consider the Quality of Evidence — Evidence Presented in Review Articles and Empirical Journal Articles In addition to considering the quantity of evidence that an author cites in support of his or her conclusion, you must consider the quality of that evidence. As readers of psycho- logical literature, you will often need to evaluate the quality of the evidence presented to you in review articles and empirical journal articles. Each type of article requires a different discerning eye; review articles require that you assess the authors' inten- tions and the way they use other empirical articles to support their argument, whereas empirical articles require that you assess hypothesis, methodology, conclusions and other issues more directly. When reading an empirical A. Review Articles article, ask yourself: Does An author writes a review article to provide a summary of the studies that have been the research hypothesis done in a given area and to advance a thesis/conclusion based on his or her reading of this literature (e.g. pointing out limitations, establishing the state of a field, etc). To sup- follow logically from the port a thesis in a review article, the author must substantiate it with evidence (i.e., data results of previous studies, from the pertinent studies reviewed). Otherwise, his or her thesis is an unsupported opinion. Ideally, an author should specify the type of evidence that would support the or is the author making conclusion as if it were a logic problem (for example: If thesis x is true, then the avail- assumptions? able research should show that a, b, and c are true, but d is false). If the available research proves inconclusive with respect to a, b, c, or d, the researcher's thesis is only partially substantiated, and the conclusions must be tentative. If a conclusion or any of its prem- ises are contradicted by the extant data, and the author of the review has not presented evidence to question these data, the conclusion is likely ill-conceived and inaccurate, and you should regard it as such. Some review articles venture into theoretical territory by presenting a framework that integrates previous research and makes predictions for future research. In this instance the author will also explain the logic of the problem and the expected outcomes of necessary studies. B. Empirical Studies An author writes an empirical article to present data that speak to a particular research question. Typically, the author will present evidence that either supports or challenges a specific hypothesis about the relationship between two or more variables.. Some ques- tions to ask yourself when reading an empirical article are: 1. Is the author's research question/hypothesis logical? 2. Do the methods adequately address the research question posed? Writing for Psychology page 73. Do the data support the author’s specific hypothesis? 4. Are the author's conclusions supported by the data he or she presents? 5. Can you think of OTHER possible explanations for the results? 1. Research Question/Hypothesis Empirical articles begin with a brief introduction that provides the conceptual basis of the study. The context provided is often a review of research studies that have been con- ducted on the topic. The studies cited should be directly relevant to the research ques- tion/hypothesis being addressed in the author's study. When reading an empirical article, you should consider the manner in which the researcher derived the research question or specific hypothesis. Does the research question/hypothesis really follow directly and logically from the results of previous studies, or is the author making assumptions, jump- ing to conclusions, or misinterpreting previous studies? If the researcher is testing a spe- cific hypothesis, then he or she should be able to clearly articulate both that hypothesis and an alternative hypothesis in the following form: " If Hypothesis A is true, then the study results should show B" " If Hypothesis A is not true, then study results should show C" If the hypothesis cannot be put into the above format, either by the researcher or the reader (and you will read many published articles that do not meet this ideal), then the When reading an empirical hypothesis is logically murky. As a result, the study methods will likely be inadequate, and the conclusions that can be drawn will likely be limited. article, ask yourself: Do Example: One team of researchers hypothesized that aggression and testoster- the methods employed in one would increase in men exposed to a violent object (Klinesmith et al., 2006). the research actually test The alternate hypothesis was that exposure to a violent object would not affect the hypothesis the author aggressive behavior or testosterone. The researchers measured the testoster- one of thirty men before and after playing with either a realistic toy gun or a presented? nonviolent game. To test aggressive behavior, the subjects were asked to add hot sauce to some water that the next subject would drink as part of a taste sensitivity test; the addition of more hot sauce was a proxy for more aggressive behavior. Both testosterone and the proxy for aggressive behavior increased significantly in those men who played with the gun (Klinesmith et al., 2006), which appears to support the researchers’ original hypothesis. This is an exam- ple where the authors did indeed present both their preferred hypothesis and acknowledged an alternate hypothesis. In this instance the alternative hypoth- esis is simply the opposite of the preferred hypothesis—that testosterone would not have an effect. However, at times an alternative hypothesis is not simply the opposite hypothesis but reflects a competing hypothesis that involves a differ - ent set of results, not just a lack of an effect. After ascertaining whether a hypothesis is conceptually sound, you should consider whether the author has done a good job designing an experiment to test the hypothesis. The study design is presented in the Method section of an empirical article. 2. Method The Method section of an empirical article describes exactly how the researcher carried page 8 Chapter One: How to Read Sources Critically out the study. Many students — indeed, many long-time scholars — are tempted to skip reading the Method section; often assuming that nothing useful will be gleaned from this section, or that reading it will be tedious. However, just because a research question or a specific hypothesis is conceptually and logically sound, you should not assume that the methods used to test the hypothesis were sound. One way you can better understand an empirical article’s methods is to give them a cursory look on the first read-through of the article, then go back more meticulously once you have seen the authors’ discussion and conclusions. When trying to understand the methods and how they fit the hypotheses being tested, ask yourself: • "If Hypothesis A is true, then is B the only possible result? If A is true, are any other results possible?" • "If Hypothesis A is NOT true, then is C the only possible result?” • “If B could result from factors other than those specified by Hypothesis A, has the researcher included appropriate methods to control for these other fac- tors?” When reading the Method section, you should consider whether the following methods are appropriate to answer the question posed by the researcher: • sample of participants (gender, ethnicity, education, clinical status, etc.) • materials (scales, questionnaires, scenarios written for the study, etc.) • apparatus (computer test batteries, EEG, neuroimaging, etc.) • procedures (what exactly the participants did…and in what order) a. Experimental Group Participants. If a researcher is interested in drawing conclu- sions about a population, he or she must test a sample group from that population (i.e., an experimental group). Always ask yourself: Are the type of people (or animals) who are participating in the researcher's study the types of people (or animals) about whom he or she is interested in drawing general conclusions? Or, alternatively, is the study sample limited in some way? Example: In the above example of men, aggression, and testosterone, clearly the researchers needed to conduct a study of men. The extent to which a researcher can generalize from the behaviors of the men who participate in a study to men in general depends upon the make-up of the researchers’ sample. In this case, Klinesmith et al. (2006) used a sample group limited at least by age and educational status — they were 18-22 year old male col- lege students — and also possibly by race and class. The question is whether these limits on the sample population limit generalizations to men as a whole. Though the only way to truly answer this question is to repeat this study on more populations, perhaps it could be argued that age and educational status should not affect the particular variables studied. These are determinations you get to make as a critical reader of psychology. b. Controls (Participants or Tasks). In order to test a hypothesis about the effect of a particular variable (i.e., independent variable), one must try to isolate the effects of that variable. To do this, a researcher must design a study that manipulates the inde- Writing for Psychology page 9pendent variable while keeping everything else constant. Only then can a researcher truly be confident that his or her study results are directly related to the manipulation of the independent variable. Often, however, it is not possible to keep “everything else” constant and more than one variable is manipulated at the same time. When this occurs, all is not lost if the researcher includes appropriate controls. A study design may require a control group, control tasks (or conditions in a task), or both. Example: In their study of violent objects and aggression, Klinesmith et al. (2006) knew they could not simply measure testosterone and aggression when their subjects played with a gun; they needed to have a control group do something similar but nonviolent and have their testosterone and aggres- sion measured as well. Thus, they had half the subjects play the game “Mouse Trap” as the control task, and the other half of the subjects played with the realistic, but toy gun. The researchers might also have considered studying responses to guns in a group of women, but as sex differences in the respon- siveness of testosterone have been well studied, it was probably safe for them to keep their sample to men only. This is an example of work where a control task was an important component of the empirical article. c. Confounds. In addition to considering the study procedures and the sample char- acteristics, you should consider factors not discussed in the article that may have in- When reading a fluenced the results. In particular, you should ask whether there were any confounds, factors correlated with both the independent and dependent variables, that might psychological report, you explain their relationship. should consider factors not Example: Klinesmith et al.’s (2006) hypothesis was that mere exposure to a discussed in the article that violent object would increase aggressive behavior, but there are many differ- may nevertheless influence ences between playing with a gun and playing with Mouse Trap besides the gun’s status as a violent object. To take just one: Perhaps Mouse Trap is more the results. fun and engaging than a gun, and thus leaves people in a better mood. Be- cause mood can affect aggressive behavior, mood may have been a confound in the study design. To rule out this alternative explanation—that those in the gun condition were more aggressive because they were in a worse mood, not because they had handled a violent object, per se—the researchers could have included a measure of mood and statistically controlled for any differences in mood between the two conditions to see if the difference in aggression remained significant. Thus, when reading an empirical article, you must make sure to ask yourself wheth- er the researcher has done a good job of setting up the study so that the effects of the independent variable are isolated. Only then can a researcher be confident that if two groups differ, they do so as a result of the independent variables, not as a result of confounding variable. That is, "if B occurs," one can be confident that "it is be- cause hypothesis A is correct." The types of factors that can confound study results will differ depending on the kind of independent variable(s) a researcher is examin- ing. For experiments, in which the independent variable is manipulated, common confounds include participants’ mood, interest level, and fatigue. In correlational studies, common confounds include gender, age, socioeconomic status, and IQ. page 10 Chapter One: How to Read Sources Critically3. Results The Results section is the place where researchers present the study results. Because the Results section is usually filled with statistics, it can appear intimidating. Resist the urge to skip over this section, though again, it may help to read the discussion section first, then return to the results section. Even those without any knowledge of statistics can garner important information from reading the results section. In this section you will find such information as group averages, percentages, data tables, and figures. Take the time to review this information and consider how it bears on the researcher’s ques- tion and/or whether it appears to support the researcher's hypothesis. In the discussion section, the researcher will tell you whether he or she believes the results support the hypothesis, but you should form your own opinions of the data. If you have not yet taken statistics, you will not be able to understand all that is presented in a Results section. This is okay to start, but the more you can educate yourself about statistical methods the savvier a reader of psychological literature you will become. In the beginning, it is helpful to know two things: that correlation does not equal causation, and the that a lower p-value indicates a lower probability that a relationship between two variables was detected by chance. Note on correlation vs causation: After reading that a researcher has found a significant re- lationship between two variables, you may be tempted to conclude that the independent variable causes changes or differences in the dependent variable. However, you should resist this temptation. Much of psychology research deals with correlations between When reading a research measures, but correlation doesn’t necessarily mean causation, even though causation is article, be sure to accompanied by correlation. distinguish between the Example: Klinesmith et al. (2006) showed that aggressive behavior and testos- empirical results and the terone increased when male college students played with a gun. Here, two vari- ables — aggressive behavior and testosterone — increased with exposure to a researcher's interpretations violent item. Did the gun cause increases in both aggressive behavior and tes- of them. tosterone, which were unrelated to each other? Did the gun cause an increase in testosterone, which then caused an increase in aggressive behavior? Or, did the gun cause an increase in aggressive behavior, which then caused an increase in testosterone? This study is unable to resolve the correlation versus causation issue, and thus only a relationship between these variables testosterone and ag- gression can be reported, not cause and effect. Note on p values: When a relationship is observed by a researcher, it means one of two things: either (1) the effect is a "real" effect that you would observe time and time again if you repeated the study, or (2) the effect is a fluke, merely the result of chance, and would not be observed if you repeated the study. A researcher uses "inferential" statistics to ascertain whether an observed effect is real. Specifically, with the use of inferential statistics, a researcher calculates the probability that an observed effect is due to chance. This probability that a given effect is merely due to chance is presented as a p-value, expressed in the form p = .xxx (or p .xxx). According to statistical theory, the smaller the p-value, the less likely it is that an observed effect is due to chance. Usually, when p .05, researchers feel relatively confident that an effect is real, and not the result of a fluke. Usually, results associated with ps .05 are called "significant" results. “Significance” doesn't mean that the result is necessarily conceptually or pragmatically important (that depends on how novel, big, or relevant to real-world problems the effect is), but it does mean that the result should be taken seriously in the sense that it is probably detecting Writing for Psychology page 11something real and replicable. That said, the only way to be really confident that your effect is "real" is to conduct the study again and again, and see whether the result shows up again and again. Example: The results of Klinesmith et al (2006) were statistically significant, yet they were not able to parse out causation of the variables observed. Future research with a study design that can test the mechanisms behind the relation- ships of these variables may get at causation, and repeated studies will help determine whether the relationships observed were “real.” 4. Discussion The Discussion section is where the researcher interprets the results. The researcher not only must explain how the results bear on the research question being asked (or support or challenge any specific hypotheses that were made), but also should offer tentative ex- planations for unexpected results. Because the Discussion section is mainly interpretive, it is where the authors explain their reasoning behind the inferences and conclusions they make based on their results. Much of actual empirical work requires simplification in order to operationalize and isolate variables, but the Discussion section is where the researchers return to the broader concepts and make conclusions about the relationships among these concepts. Read the Discussion section carefully. Consider all the author has to say, but continue to ask yourself: Do these results really support the researcher’s conclusions? Ask yourself: Is the author sticking closely to the facts (i.e., the data) or is he or she jumping to conclusions, making assumptions, or viewing her own speculation as fact when it is really opinion? Be sure to distinguish between the empirical results and the researcher’s interpretations of them. When attempting to distinguish between the two, bear in mind the hypothesis, the study methods, the actual data, and all limitations you noted and their implications. Example: In our final look at the study of guns, aggression and testosterone, it’s important to note that in addition to testing increases in aggressive behavior and testosterone with exposure to a gun, the authors of the study also hypothe- sized that testosterone mediates the relationship between exposure to violence and aggression (Klinesmith et al. 2006). That is, exposure to violence causes an increase in testosterone, which in turn causes an increase in aggressive behav- ior. As evidence for this hypothesis, they pointed to results showing that guns and testosterone were correlated, guns and aggression were correlated, and testosterone and aggression were correlated. But is this the correct interpreta- tion of the results? If exposure to a gun was correlated with both testosterone and aggression, it is likely that testosterone and aggression would be correlated with each other, and this correlation would not imply that testosterone caused an increase in aggression (in fact, the reverse might be true). Familiarizing your- self with the statistics involved will help you to determine whether the authors interpreted their results appropriately. page 12 Chapter One: How to Read Sources Critically chapter two Writing a Conceptually Coherent Paper In Expos, you were directed to use a number of basic elements in the construction of your essay: presenting an arguable thesis and the use of evidence, situating on a thesis in terms of a question or problem, analysis, structure, transitions, orienting, and key terms. Y ou also had to write pre-drafts, drafts and revisions of your papers. Although most psychol- ogy courses will not assign pre-drafts and drafts, it is essential that you do similar kinds of prep work before writing the final version of any psychology essay: freewrites, litera- ture searches, annotated bibliographies, outlines, and drafts are still essential to putting together a piece of writing that is coherent and concise. All good researchers in psychol- ogy utilize at least some of these prewriting exercises, as well as all of the elements of the essay you learned in Expos. I. Before You Begin to Write: Reviewing the Literature and Generating a Thesis Statement Generating a thesis statement for the psychology essay is similar in many ways to generat- ing a thesis for an Expos paper. Although the type of evidence required in Expos essays often differs from that required in psychology essays, you must rely on evidence when generating your thesis for either type of writing. In some cases, you will have a thesis idea before you begin your review of the literature; in other cases, reviewing the literature will suggest a thesis statement to you. It can be great to have tentative ideas about a topic, because they can drive your motive or help direct your literature searches. But remember not to stop there; your thesis must be an empirically justifiable position that is supported by the evidence you gather from the studies that you review. So, you will need to gather evidence by conducting a thorough review of the articles that have been written about your topic before you generate your final thesis statement. A. Reviewing the Literature: The Lit Search The purpose of the literature search is to find out what is currently known about a particular topic in a given field and to determine the sequence in which discoveries on the topic were made. Without reviewing the current state of knowledge on your topic, you cannot determine whether opposing theories have been tested, whether new theories need to be generated, or where future research needs to be directed. Whenever you wish to gather evidence to answer a question in psychology, whether your goal is to write a course paper, to gain background information prior to designing an experiment, or simply to increase your own personal knowledge in a subject area, you will begin the process with a lit search. A lit search is a survey of the published literature (journal articles, reports, books, and dissertations) in a given field of study. You may conduct a literature search by using one of the electronic databases available to Harvard students or you may use the reference sections of course materials, such as Writing for Psychology page 13textbooks or assigned articles. We suggest that you conduct both types of search in order to cover the full spectrum of possible references for your topic. Using Online Databases Two databases widely used in the field of psychology, PsycINFO and PubMed, are avail- able to you online through the Harvard Library system website (see Appendix). Both provide publication information and abstracts (or short summaries) of articles and book chapters. In addition, general academic data bases, such as Academic Search Premier and Google Scholar, may provide articles of interest and are also available through the Harvard Library system. Using Reference Sections You may begin a literature search by reading the section of a course textbook (or other assigned reading) that pertains to your topic. Note the citations on your topic and look them up in the reference section of the textbook. Acquire these referenced works and then check their reference sections for further articles on your topic. You will note that many articles cite the same sources. Sources that are cited by many authors are usu- ally central to the topic, and thus you should read them in the original. (Note that you should not use textbooks themselves as references in your paper. Textbooks are not pri- mary source materials; however, they can provide an informative list of primary source Once you have read your references on your topic as a starting point.) Most of the articles that you will cite in sources, ask yourself: your paper should be empirical journal articles or review articles published in a peer- review journal. For information on locating sources in the Harvard Library System, see What conclusions does the Appendix. existing research support? It is likely that one of B. Generating a Thesis Statement these conclusions What is a thesis and where does a thesis come from? A thesis is not a "topic." A topic is an area of study, whereas a thesis is a specific point of view that makes sense of data will be your thesis. about the topic. One way to develop a thesis is to select a topic area of interest. Then ask a research question about that topic area that could be answered by examining the current literature. Your answer to that question will be your thesis. Example of topic area: Creativity and psychopathology Example of research question: Are artists and creative writers more often depressed than less creative individuals? Example of thesis: Artists and creative writers are at greater risk for mood disorders than members of the general population. Before you begin to write, you first should locate and read the sources in your chosen topic area. Even if you have not generated a research question before you begin to read, an examination of the source materials may suggest one. Once you have read your sources, ask yourself: What conclusions does the existing research support? It is likely that one of these conclusions will be your thesis. Before arriving at this thesis, you will need to consider the importance of each article that you have read. Ask yourself: What page 14 Chapter Two: Writing a Conceptually Coherent Paper