Writing Empirical Psychology Papers (for Beginners)
Please note: The below reviews the commonly accepted practices in writing an empirical APA-style paper (6th Ed).
However, it was originally written for a specific class. Remember to refer back to your own assignment and your professor for more information.
A Word about APA Style
Overall Shape of the Paper
Daryl Bem (download a PDF of his chapter here) suggests that crafting a well-written empirical article is like "telling a story." Your story has a beginning that sets up the primary problem to be solved (the Introduction). The primary problem is the gap in the existing research, or the question that is unanswered by existing research. You will be explaining that question, and why it is important, to the reader. Your story has a middle (the Method), a climax (the Results) and an end (the Discussion). Although your language will remain formal and you must attend to the proper content required by APA-style, your paper will probably turn out better if you keep this notion of story-telling in mind as you write. Never forget that you are communicating with an audience through your paper.
A well-written article should be shaped like an hourglass. That is, the Introduction begins very broadly by introducing the topic and defining terms, and then begins to narrow to more specifically focus on the variables in your study. At the end of the introduction, the paper is at its most specific (or "narrow") in that the Method and Results both provide extremely focused information about your study. The Discussion begins by reviewing your specific findings, but then starts to slowly broaden out again as the implications are discussed. By the end of the Discussion, the paper has become as broad in focus as it was at the beginning of the Introduction. Thus, an hourglass shape.
Use the following as a guideline for what to include in the introduction and how to organize it.
(1) Try to capture the reader’s interest right away in the first few statements. You want to introduce your topic. One way to do this is by posing an interesting question. You do not have to summarize your entire argument in this first paragraph. It is mostly a stylistic paragraph to orient the reader to the general topic and your question. It would be easiest to write this paragraph after you've finished the rest of the paper. (For more on this, see the section on "Opening Statements" at: http://dbem.ws/WritingArticle.pdf)
(2) The main goal in the introduction is to provide a review of the relevant psychological literature, providing definitions and past research findings that inform the reader on your topic. Your goal is for the reader to understand the need for more research in the area (i.e., your proposed study), and to be able to clearly see the reasoning for your hypothesis. You should organize this section of your paper in such a way that you logically build to your study. You aren't just citing research, you are crafting a line of reasoning which leads to your research question. Avoid simply summarizing each of the different studies you read in a “list” type format. Remember this is a paper and you need to present information in a coherent way that moves from the broad to the specific, and in a way that leads the reader to the gap or question in the literature that you’ve noticed. You can accomplish this goal in many ways (look through multiple published articles for ideas), but you might try the below as a starting point:
INTRODUCING YOUR TOPIC, REVIEWING BACKGROUND LITERATURE
| After your opening statement, choose one of the variables that is relevant to your study. If you wish, you can label this section of the paper with the name of that variable. |
Provide a transition to...
Review the next relevant variable from your hypothesis. If you used a section heading for the previous variable, provide a new heading for this one.
|BRINGING IT ALL TOGETHER: PRESENTING YOUR RESEARCH QUESTION AND LOGIC FOR HYPOTHESIS|
Present the logic for your conceptual hypothesis. Some papers put this entire section under a heading called "The Present Study." Other papers provide a different section heading for the logic, and use "The Present Study" heading for the information in the box below. This is a stylistic choice.
| In the last paragraph of your Introduction, name the specific variables you intend to study and generally what you will be asking your participants to do (e.g. “…we plan to administer a survey measuring both variables X and Y to determine if there is a correlation.”). Finally, state your hypothesis(-es) formally: refer to the specific variable names and use relevant "statistical" language (e.g., "We hypothesize that X will have a positive correlation with Y." or We hypothesize that participants in the experimental condition will score higher on Y than participants in the control condition.") |
Throughout the entire paper, remember to provide a citation for any information that is not your own idea (including information from textbooks). Put the citation at the very first mention of the cited material (not, for example, at the end of the paragraph). Read and follow the guidelines on in-text citations found at the web page: Finding, Reading, and Citing Sources and consult the APA Manual.
All method sections need three basic categories of information:
Participants – who was in your study and did they volunteer or get some sort of course credit.
Materials – what were your measured variables (a.k.a. operational definitions)
Procedure – what exactly did you do (literally during the study session)
You may choose to use three different heading for this information (as presented in the example below), or you might want to combine procedures and materials into one section. Format the method section however it works best for you – but be sure to put participant information first and in its own section. Write in the past tense.
The below example should help provide you with some commonly used (conventional) ways of writing out this information.
Participants were __N___ college students enrolled in introductory psychology classes. The students received extra credit in exchange for their participation. This sample consisted of _N_ women and _N_ men. Participants were generally college-aged (M = XX, SD = XX), and most selected "White" (XX%) when asked their race/ethnicity (XX% selected "Black" and XX% selected "Latino/a").
Note: In the participants section, gender, age (or year in school), and ethnicity are typical standard demographic statistics to include . You should also report any other demographic statistic that relates to your hypothesis.Materials
Contingencies of Self-Worth Scale. Selected subscales from the contingencies of worth scale (Crocker & Wolfe, 2001) were used. The subscale of interest for this study was the school competency scale. The measure of school competency as a basis of worth consisted of XX# items. An example item is: “xxxxxxxx.” Participants indicated the extent to which they endorsed each statement using a 7-point Likert type scale (1=strongly disagree, 7=strongly agree). After reverse coding the appropriate items, the scale was created by averaging across items. The internal consistency of the scale for our sample was adequate/high/low (alpha = XX).
Sexual Orientation Prime Manipulation. We designed a PowerPoint slide show in order to prime either heterosexual or homosexual relationships. Both slide shows consisted of 20 slides, ten of which showed photographs of neutral objects (e.g., trees, tables, houses) and ten of which showed photographs of two people hugging, holding hands, or touching one another's faces with obvious affection. Participants in the heterosexual condition saw images of male/female couples. Participants in the homosexual condition saw five images of male/male couples and five images of female/female couples. All images were free-use photographs downloaded from various internet sites.
Another Scale or Variable. Continue in a new paragraph with a new heading for any other scales or manipulations that are relevant to your hypothesis. (If you are reporting data from a larger data set collected by several researchers, you do not need to report scales that you are not relevant to your hypothesis.)
Note: For scales/questionnaires, be sure to always include a reference (unless you wrote all the items), the number of items on the scale, the response format, the internal consistency, and an example question.)
In this section, include what participants were initially told about the intent of study, how they run (e.g., in groups or individually? in what order were the questionnaires administered?), and also mention that an informed consent statement was administered and a debriefing session was conducted. (If there was no informed consent or debriefing, you should explain why in this section). You’ll probably want to combine Materials & Procedure for studies with simple procedures (like a short survey). For a more complex study (for example, one which uses a manipulation like the one described above), you will probably want to keep the Procedure section separate from the Materials section.
The results section is where you tell the reader several things about your data and data analysis. First, provide basic descriptive information about the scales you used (report the mean and standard deviation for each scale). If you have more than 3 or 4 variables in your paper, you might want to put this descriptive information in a table to keep the text from being too choppy and bogged down (see the APA manual for ideas on creating good tables). Most central to the Results section, you tell the reader what statistics you conducted to test your hypothesis (-es) and what the results indicated.
Include the following, in this order, in your results section:
- In some papers, the Results section begins with the descriptive statistics for the relevant variables (e.g., mean, standard deviation). The statistics you present in this first paragraph should be overall means for the entire sample rather than means broken down by condition. Again, you want a single mean and standard deviation for each (continuous) variable. You do not want a mean for each condition for each variable. You will provide the means by condition later. Also, remember that you cannot claim one number is higher or lower than another without a significance test. Save those types of comparisons for later sections.
- Provide a brief rephrasing of your hypothesis(es) (avoid exact restatement). Then tell the reader what statistical test you used to test your hypothesis and what you found. If you did more than one test, report each test in its entirety (what you did and the results), one at a time.
- For each test:
- Explain which findings were in the predicted direction, and which were not (if any). Were differences statistically significant (i.e., p =.05 or below)? Don't merely give the statistical numbers without a supporting sentence.
- You cannot use statistics as though they were parts of speech (i.e., nouns). For exampledo notwrite “The correlation between private self-consciousness and college adjustment was r(60) = - .26, p = .01.” Instead, translate important data for the reader into words and provide the statistics as evidence for your reported results.
- For example, “The negative correlation between private self-consciousness and college adjustment indicated that increased self-consciousness, predicted poor adjustment, r(60) = - .26, p = .01."
- However, don't try to interpret why you got the results you did. Leave that to the Discussion.
- You should also include an effect size statistic (if you know what that is and how to compute it). When comparing means (t-test or ANOVA), the "d" is usually appropriate. When reporting correlations, Pearson's "r" is the effect size.
- Note that for t-test and ANOVA findings, the "result" consists of the following in the following order: (1) the t (or F), degrees of freedom, the p value, and the effect size; and, if significant, (2) the means. In other words, keep the means together with the related significance tests. In addition, for two or three way ANOVA results, you should first provide all of the relevant information (including means if significant) for each of the main effects (including indicating non-significant effects), and then report on interactions (including indicating non-significant interactions).
- Examine this document for more information about how to report statistics in a result section or consult the APA Manual.
- If your result was non- significant, but p < .10, it is commonly accepted to still talk about the results. You might write something like the following text in your paper: “While the correlation was not significant using the standard alpha level of .05, the p-value was less than .10; r (49) = .23, p = .08.” BUT, you must provide a rationale for why you should still be able to discuss this non-significant correlation (e.g., power, effect size issues). You may cautiously interpret such a correlation. Don’t make grand conclusions or use strong language based on the existence of a marginally significant finding. Also, you should indicate that a marginal finding is non-significant in a table; only refer to the statistic as “approaching significance” in the text of the paper.
- If putting your statistics in the body of your results section seems to make the section difficult to read (i.e., if you feel the reader is distracted from your results by too many numbers and statistics), consider putting the statistics in a table. For example, with simple bivariate correlations, you should create a correlation matrix (see the link below under Tables for an example). If you include a table, you should, in the text of the result section, refer readers to your table instead of typing out the statistics for each finding. (See below for Tables.)
- You need to report the actual statistics in some way in your result section, but regardless of whether you use a table or type the statistics in the text, you should put sentences describing the results in this section.
- E.g. “As expected, college adjustment was positively correlated with the amount of contact with friends and family members (see Table 1).”
- E.g. “No significant relationship was found between the importance of one's social life and social adjustment to college, r = -.11, n.s.
- E.g. “As shown in Table 1, some of my predictions were supported. There was a significant correlation between extroversion and life satisfaction. However, life satisfaction was not significantly related to college adjustment.”
- It is helpful to write the words of the results section first, and then go back to insert the numbers and statistical information. Really - write the words only first. Then go back and add numbers.
If you are using Microsoft Word as your word processor, create the table, then you can adjust the "borders and shading" for each cell/row/column to get the table formatted properly. Another tip is to play around with double-spacing and/or using the "Format" "Paragraph" "Spacing before/after" features. Other word processors, PowerPoint and Excel can produce similar tables. Do not use Figures or Tables produced by SPSS.
Click on this link to see 2 examples (PDF).
See the APA Manual for more specific instructions for certain kinds of tables.
DiscussionIn your discussion section, relate the results back to your initial hypotheses. Do they support or disconfirm them? Remember: Results do not prove hypotheses right or wrong, they support them or fail to provide support for them.
Include the following information in (roughly) the following order:
- Provide a very brief summary of the most important parts of the introduction and then the results sections. In doing so, you should relate the results to the theories you introduced in the Introduction. Your findings are just one piece among many -- resist the tendency to make your results the final story about the phenomenon or theory of interest. Integrate the results and try to make sense of the pattern of the findings.
- In the case of a correlational project, be careful to not use causal language to discuss your results – unless you did an experiment you cannot infer causality. However, it would be impossible to fully discuss the implications of your results without making reference to causality. That is fine. Just don't claim that your results themselves are demonstrating causality.
- Talk about any limitations relevant to the interpretation of your findings (all studies have weaknesses/qualifications).
- If your results did support your hypothesis, the limitations section often includes a discussion of possible "third variable" explanations, unmeasured mediators, and/or issues with the generalizability of your results.
- If your results did not support your hypothesis, the section on limitations often includes discussion of various features of the study which might be responsible (e.g., operational definitions, self-report biases, unmeasured moderator variables, the size or composition of the sample). Where possible, support your speculation with references.
- BE SPECIFIC when discussing limitations. For example, if you claim that a third variable might affect your correlation, tell the reader what that third variable is and how it affect the results. If you think that the use of a convenience sample (and thus, a non-representative/random sample) is a limitation, you must explain what segment of the population might respond differently than did the participants in your sample and why.
- Discuss about future directions that research could take to further investigate your question. This might relate back to any weaknesses you’ve mentioned above (or reasons why the results didn’t turn out as expected). Future directions may also include interesting next steps in the research.
- A discussion section is about “what we have learned so far”; and “where we should go next”; Your final conclusion should talk briefly about the broader significance of your findings. What do they imply about human nature or some aspect of it? (Don't wildly speculate, however!) Leave the reader feeling like this is an important topic. You will likely refer back to your opening paragraph of the introduction here and have partial answers or more specific responses to the questions you posed.
Other Parts of the Paper
Title page - Try to write a title that maximally informs the reader about the topic, without being ridiculously long; 10-12 words is the maximum recommended by APA. Use titles of articles you've read as examples of form. Also provide the RUNNING HEAD (an abbreviated title that appears in the header of each page along with the page number). Provide your name and institutional affiliation (e.g., Muhlenberg College). Format as per APA style.
Abstract: Write the abstract LAST. An abstract is a super-short summary and is difficult to write.
In 250 words or less your abstract should describe:
--the topic of research (an "introduction" type sentence)
--the specific question and method of doing so (a "method" type sentence)
--the results (no numbers, just words)
--a hint about the general direction the discussion section takes
References: Use APA style. See your APA manual, textbook and/or a sample paper for examples of how to cite and how to make a reference list. Make sure that all references mentioned in the text are also mentioned in the reference list and vice versa. Also see Finding, Reading and Citing psychology sources for information.
Tables and/or Figures: Use APA style. Tables go at the very end of your paper. Make sure you refer to the table or figure in the text of your paper.
APA Style Guidelines
The American Psychological Association publishes a book containing the standards and rules all psychologists typically follow when writing a manuscript. There are many guidelines and some are more obscure than others. Check with your professor to see which APA Guidelines he or she would like you to learn and follow.
Below is a list of the guidelines most commonly required for undergraduates learning APA style. The numbers refer to the relevant section in the 6th Edition of the manual.
Headings (3.02, 3.03)
Clarity of Expression (3.05 - 3.11)
Reducing Biased Language (pp. 72-73, and sections 3.12 - 3.17)
Basic Grammar (3.18 - 3.23; see Chapter 4 for more on mechanics)
Numerals vs. Words (4.31 - 4.32)
Proper Statistical Notation (4.44 - 4.46)
Tables and Figures (5.01 - 5.25)
Tables - basic information (5.05, 5.08, 5.10, 5.13, 5.16)
- Figures - basic information (5.05, 5.20 - 5.23)
Using Quotes (6.03)
Citing References in the text (6.11 - 6.21; see Chapter 7 for examples)
Reference list (6.25; 6.27 - 6.31)
Typing your manuscript: spacing, margins, font, and so forth (8.03)
Other useful links:
Finding, Reading and Citing Psychology Articles- help with PSYCHINFO, reference pages and other citation issues
Writing Empirical Papers: Advanced (PDF)
General Writing Tips on Writing Psychology Papers
Avoiding Inappropriate Paraphrasing
More Writing and APA Style Links
Information about purchasing an APA Manual
Web site maintained (such as it is) by Connie Wolfe.
Last update: 08/16/09
As a student at the University of La Verne, faculty may instruct you to read and analyze empirical articles when writing a research paper, a senior or master's project, or a doctoral dissertation. How can you recognize an empirical article in an academic discipline? An empirical research article is an article which reports research based on actual observations or experiments. The research may use quantitative research methods, which generate numerical data and seek to establish causal relationships between two or more variables.(1) Empirical research articles may use qualitative research methods, which objectively and critically analyze behaviors, beliefs, feelings, or values with few or no numerical data available for analysis.(2)
How can I determine if I have found an empirical article?
When looking at an article or the abstract of an article, here are some guidelines to use to decide if an article is an empirical article.
- Is the article published in an academic, scholarly, or professional journal? Popular magazines such as Business Week or Newsweek do not publish empirical research articles; academic journals such as Business Communication Quarterly or Journal of Psychology may publish empirical articles. Some professional journals, such as JAMA: Journal of the American Medical Association publish empirical research. Other professional journals, such as Coach & Athletic Director publish articles of professional interest, but they do not publish research articles.
- Does the abstract of the article mention a study, an observation, an analysis or a number of participants or subjects? Was data collected, a survey or questionnaire administered, an assessment or measurement used, an interview conducted? All of these terms indicate possible methodologies used in empirical research.
- Empirical articles normally contain these sections:
- Introduction-The introduction provides a very brief summary of the research.
- Methodology-The method section describes how the research was conducted, including who the participants were, the design of the study, what the participants did, and what measures were used.
- Results-The results section describes the outcomes of the measures of the study.
- Discussion-The discussion section contains the interpretations and implications of the study.
- References-A reference section contains information about the articles and books cited in the report and should be substantial.
- How long is the article? An empirical article is usually substantial; it is normally seven or more pages long.
When in doubt if an article is an empirical research article, share the article citation and abstract with your professor or a librarian so that we can help you become better at recognizing the differences between empirical research and other types of scholarly articles.
How can I search for empirical research articles using the electronic databases available through Wilson Library?
- A quick and somewhat superficial way to look for empirical research is to type your search terms into the database's search boxes, then type STUDY OR STUDIES in the final search box to look for studies on your topic area. Be certain to use the ability to limit your search to scholarly/professional journals if that is available on the database. Evaluate the results of your search using the guidelines above to determine if any of the articles are empirical research articles.
- In EbscoHost databases, such as Education Source, on the Advanced Search page you should see a PUBLICATION TYPE field; highlight the appropriate entry. Empirical research may not be the term used; look for a term that may be a synonym for empirical research. ERIC uses REPORTS-RESEARCH. Also find the field for INTENDED AUDIENCE and highlight RESEARCHER. PsycArticles and Psycinfo include a field for METHODOLOGY where you can highlight EMPIRICAL STUDY. National Criminal Justice Reference Service Abstracts has a field for DOCUMENT TYPE; highlight STUDIES/RESEARCH REPORTS. Then evaluate the articles you find using the guidelines above to determine if an article is empirical.
- In ProQuest databases, such as ProQuest Psychology Journals, on the Advanced Search page look under MORE SEARCH OPTIONS and click on the pull down menu for DOCUMENT TYPE and highlight an appropriate type, such as REPORT or EVIDENCE BASED. Also look for the SOURCE TYPE field and highlight SCHOLARLY JOURNALS. Evaluate the search results using the guidelines to determine if an article is empirical.
- Pub Med Central, Sage Premier, Science Direct, Wiley Interscience, and Wiley Interscience Humanities and Social Sciences consist of scholarly and professional journals which publish primarily empirical articles. After conducting a subject search in these databases, evaluate the items you find by using the guidelines above for deciding if an article is empirical.
- "Quantitative research"A Dictionary of Nursing. Oxford University Press, 2008. Oxford Reference Online. Oxford University Press. University of La Verne. 25 August 2009
- "Qualitative analysis" A Dictionary of Public Health. Ed. John M. Last, Oxford University Press, 2007. Oxford Reference Online. Oxford University Press. University of La Verne. 25 August 2009