Victoria Alexander May 2010, for Psychology 11-100 Bond University

Research Methods in Psychology Reporting a Psychology Experiment

The writing of laboratory reports is an essential part of any practical course in Psychology. Psychologists (and more generally most scientists) write accounts of their studies using a standard format, which makes explicit certain aspects of the study. There are two main reasons for doing this: 1. Ease of communication: it is easier to find what you want from a study if it is written in the standard format. 2. Provision of a precise and complete description: the format makes it clear what information is important for scientific communication and must be spelled out in detail. It should be added that many professions now require the skills of technical report writing, with its emphasis on clear, direct and concise expression. Learning to write laboratory reports will provide you with a valuable and transferable skill. This handout tells you about the structure and style that is required for a psychology laboratory report. Since journal articles are written in the same format, learning to write lab reports will help you in reading the literature. When you read journal articles, think about the formats used and why they have been adopted. Not all articles are perfect, so when you come across a section that you do not understand, think about why it is unclear. The fault may lie with the authors, who are not being as clear as they could be. If so, how could the section be improved? Although learning from your mistakes is an important part of the education process, learning from others mistakes is far less painful. The guides listed at the bottom of the page provide additional help on writing reports and general writing style. The purpose of a lab report is to communicate to others the important points of a piece of research: why you did it, how you did it, what you found and what you think it means. Readers will sometimes want the answer to very precise questions (e.g. Who were the participants? What exactly were the mean scores for the two groups?) and not want to read the whole report in order to find this information. For this reason it is essential to follow a standard format (with correct headings) which allows the reader to locate the information that he/she requires immediately without having to work through the entire text.

The simple rule for report writing is remember the reader. In journals, papers are intended to be read by someone who knows general background for a topic only and nothing about a particular experiment. People will usually see the title first, then perhaps read the abstract, and only then read the bulk of the report if their attention is caught. The format suggested below is the same as that used in most published papers. Fine details concerning exact format and required information will depend upon the nature of the study, but most of the studies should follow this format fairly closely. Particularly important is the use of separately headed sections (and sub-sections in the method section). If you do not use these sections correctly you will incur severe marking penalties. The numbers next to each heading are included here to structure these notes; they would not appear in the report itself. 1) Title The title should provide a single line summary of what you actually did. In many cases, the title will mention the independent and dependent variables and the relationship between them. Thus, The effect of sleep loss on the exploratory behaviour of gerbils would be a suitable format for a title; Keeping gerbils awake would not. Try to avoid using catchy newspaper style headlines as titles (Gerbil insomnia); a formal report is not supposed to be an exercise in journalism. Remember that your reader will initially see the title and nothing else, but wishes to know whether or not the report is relevant to his/her research interests. Your title should be a brief, but accurate reflection of the content of the report. It is not recommended that the title of the paper be longer than 12 words (APA 6th edition). Dont start a title with phrases like An investigation into... or An experiment to determine... or include the words results and method Dont start a title with Title: The reader will know that it is the title from its location. 2) Abstract The abstract is another short summary of the report, but with more detail than the title. It should contain a brief description of the rationale and of the method, results and discussion sections. Avoid fine details such as numbers and the names of statistical tests here. You should aim for an abstract, which is about 150 to 250 words long depending on the requirements for the paper. The abstract is the second thing any reader will see and it might be as much as they encounter. It should therefore be a comprehensive but concise summary of the whole report, which will enable readers to decide if they wish to read any further. A useful rule of thumb is to try to write four concise sentences describing: (1) Why you did it, (2) What you

did, (3) What results you found, and (4) What you concluded. Write the Abstract after you have written the rest of the report. 3) Introduction (Why you did it.) The Introduction should present the reasoning behind the particular experiment which you are describing. This means that the reader, having read the introduction, should feel able to predict what your experiment will be. At the same time your introduction should allow someone who is not an expert to understand why you did this experiment. For this reason the introduction will begin at a general background level and progress through to the specific reasons for and aims of the experiment. This will normally include a review of past work in the area and an explanation of the theoretical or practical reasons for doing the study. A logical progression of content for an introduction would be as follows: Introduce the problem, describe and define the area that you wish to study, perhaps explaining why it is interesting and/or important if this is not obvious. Describe previous work by others (and perhaps yourself) that is relevant to the area. Explain why the previous work is not sufficient. It may have methodological problems, or perhaps there is plenty of scope for extending it, perhaps it has not been replicated before, or you may be comparing the adequacy of different theories. (If the previous work is complete and has no problems and has been replicated many times, or it is known which is the best theory, then there would be no point to any further study.) Outline the experiment that you have run. You dont need to give details here, but it should be clear how the present experiment addresses certain theoretical issues, the shortcomings of previous studies and/or how it extends our present knowledge. Given the results previously found, and your proposals, what would you predict the outcome of your study to be, and why? You should thus end this section with your research hypothesis (i.e. what you expect to happen and what this would mean). For example, it was hypothesised thatIt may be necessary to have more than one hypothesis. If you are doing a more exploratory experiment in which you are genuinely unsure as to the outcome, describe the aims of your experiment and what you hope to achieve. This final part of the introduction is the real key to understanding the study itself, and the report. If this part is clear then describing and interpreting the results becomes much easier.

Method (How you did it.) The method section is not a single section by itself, instead it consists only of the five or so sub-sections described below. In the method section, you describe the essentials of how you gathered your data. This section must contain enough information for the reader to be able to repeat the experiment, but should exclude any irrelevant details. For example, if you are studying the effect of word types on the ability to remember lists of such words, then the words themselves which make up the lists are extremely important. You would not be expected to explain in detail how participants were seated at a desk, say, unless you were specifically studying the effects of seating arrangements on memory. Unfortunately, the details which are relevant and which are irrelevant vary from experiment to experiment. When in doubt, consult a journal article that is related to your work and see which details have been included there. All text in the method section should only be given under one of the sub-headings below. The apparatus section or the materials section may not be necessary if the contents are trivial, other sub-sections will always be needed. A final general point. Your memory for the experiment should be clear when you write the method section. Therefore it is a good idea to write (or at least draft) this section as soon as you can after completing the experiment. 4a) Participants This should state how many participants were tested, who they were (i.e. from what population they were drawn), sampling procedures (e.g. randomly selected sample, volunteers, course requirement etc.) and any other important characteristics (e.g. mean or median age, composition in terms of males/females, educational level). Which characteristics are important will depend upon the task you are asking people to perform and the kinds of conclusions you wish to draw. If you study only undergraduate students, you may not be able to generalise to the elderly. If most of your participants are female (a common imbalance in psychology student populations) then you may not be able to generalise to male populations. Depending on the experiment, these details may be trivial or extremely important. You should provide as much demographic information that allows the reader to understand the population that has been utilized. 4b) Apparatus Apparatus consists of the equipment necessary to present stimuli and record dependent variables. Some experiments just involve trivial items (e.g. pencil,

paper, stop-watch etc.), and so an apparatus section is often not needed. This section is only required when more complex equipment is used (e.g. a computer running special software), and if purpose-built, you should describe it in sufficient detail, using a diagram if necessary, to allow equivalent apparatus to be constructed. 4c) Materials Words, puzzles, questionnaires etc. are materials, and this section should describe what these are and how you devised them (or who did devise them if you didnt). The criteria used to select the particular items which you used should be described. For example, if using words as your stimuli for a memory test you should tell the reader about any features of their selection, such as word length, word frequency (in the English Language), or their grammatical role (noun, verb, concrete, abstract, etc). For some materials it may also be useful to enumerate or provide a list of the items (e.g. the numbers from 1 to 7 inclusive) or provide examples of questions. Please note that listing the materials is no substitute for explaining how you selected them. For extensive materials, listing the items is an inappropriate way of describing the materials. If there is a list, it should be provided in an appendix. 4d) Design Here various aspects of the experimental design should be described. Unfortunately, these will vary according to the analysis that you use For simple experiments in the first year in which you are comparing a pair of means and are not using more complicated statistics than a t-test, state the design (within or between) and state what your independent variable (or classification variable) and dependent variable is. For example: This experiment utilised a between-subjects design. The independent variable was drug dosage (high or low dosage). The dependent variable was the number of problems successfully completed. For all experiments, you should also explain how you decided which experimental condition was performed by which participant (between-subjects designs/factors) randomly usually, or in what order the conditions were presented (within-subjects designs/factors or any other design where a set of tasks are being given) randomly or by counterbalancing usually. For correlational studies there is no experimental design and no independent variable(s). The aim is to measure the relationship between a number of different variables that you are studying. Say what the variables are and why they are included, and in what order the tests were given.

For example: This was a correlational study in which AH4 test score was a dependent measure of intelligence and VA7 test score was a dependent measure of verbal ability. Half the participants were given the AH4 test first while the other half were given the VA7 test first. 4e) Procedure This section describes how the design was actually implemented and should describe exactly what took place during the testing session. You should write impersonally, slanting the description towards the events that happened to the subject during this. Be very careful to decide which details are necessary for replication and which are not. You do not begin description of the experiment from the beginning of the afternoon when the lecturer started describing the lab class for the day, only from when you started testing participants. Similarly, if the class data has been written on the blackboard, you do not need to describe this at all (you have already collected your data, and writing it on the blackboard should not have any effect on it!). This section should include a description of the instructions given to participants. You do not need to quote the entire instructions in the main text unless the exact wording was important for your results. Any particular emphasis (e.g. instructing participants to be as fast and as accurate as possible) should also be mentioned. You must specify in detail the events that occur on each trial, such as any warning signal, how long the stimuli are presented for, how the subject responds, etc. Other details could include the rate of presentation of trials (e.g. one every five seconds), maximum times allowed to come up with an answer and other times, such as lengths of rest periods. You also need to specify how much practice participants had before the main task (number of practice trials) and how many measurements were taken (the number of experimental trials in each condition). In this section you should particularly remember the needs of the readers. There should be enough information for them to repeat your experiment in every important respect. The only details which can be left out are those that do not matter (e.g., how you actually randomised the order of the experimental conditions; you can assume that your reader knows how to perform randomisations). How you treated your data after you collected it has nothing to do with the procedure. Likewise, the statistical tests that you use should not be described or included in the method section, as they are tools for analysing data, not gathering data.

As a final consideration, if you have written your method well, you should be able to determine, from reading the description, what it was like to be a participant in your experiment. (For example: What did you see on each trial? What decision did you make? Did you make the same kind of decision on each trial? Did you do the whole procedure several times with different stimuli or instructions? ). All such details are important if the experiment is to be replicated successfully. 5) Results (What you found.) Begin this section with a description of how you treated your data, e.g., if you discarded reaction times whenever a subject made a mistake, or if you discarded the entire data of any participants because they were not good enough at your task (in this type of instance, always make it clear as to the exact criterion used for exclusion). Follow the description of the treatment of the data with a clear, concise summary of the data using descriptive statistics. Usually, this will take the form of a table showing the mean (or median) score on the dependent variable(s), at each of the different levels of the independent variable(s), with some idea of the spread of performance associated with each mean (using standard deviations for example). A graph may also be helpful for more complicated designs. All tables and figures should be clearly numbered, titled and labelled (including units) - a figure is any picture or piece of diagrammatic information such as a graph. Figures and tables can never stand by themselves alone and they must be referred to in the text. That is, always describe your results in words. The reason for this is that the readers interpretation of your summary statistics may differ from yours. For example you may obtain scores of 5.1 and 5.3 items (out of a possible 10) in conditions 1 and 2 of a memory test. If you are expecting a change of 0.1, then this small difference in scores is really important. On the other hand you may be expecting a quite considerable difference of 2 or 3 items, in which case you would probably report that these means were very similar. Another reason for describing the summary statistics is that the outcome of statistical tests is often blind with respect to the direction of the effect. You must make explicit which conditions of the experiment generated better performance, and which yielded worse performance. NB: If you have provided the numbers in the table, there is no need to relist the numbers in text and vice versa. Next, present the results of the inferential statistics that you performed (t-tests, correlation coefficients, etc.. Do not present the working of tests (these should go into an appendix, with appropriate references).

Finally, give one or two sentences to translate the statistical findings into English, drawing attention to the most important aspects of them where necessary. Do not interpret or discuss them (this comes later), only state them. This section could sometimes be as brief as: The mean scores for conditions A and B were 45 and 56 seconds respectively, a difference which was shown, by a related samples t-test, to be significant (t = 3.75, df = 32, p = 0.01). Thus, performance in condition B was slower than in condition A. The importance of this section is to locate everything clearly in its proper place. When embedded amidst discussion and opinion such information can become very difficult to find. One final point; in the above example, two conditions were named A and B. When you write reports, try to give more informative short names to your conditions, this will make the description of any design far easier to follow. For example, if you are looking at differences between smokers and non-smokers, call them the S and NS group rather than groups A and B. 6) Discussion (What you think it means.) This is the section in which you interpret the results of the experiment and discuss their meaning. It is important that your discussion relates to the issues raised in the introduction, since this presented the reasons for conducting the study and the results should provide more details about these issues. In particular, how do your results compare with your experimental hypothesis(es) and/or predictions? You then need to consider the implications of the findings both in terms of your hypotheses and also in relation to the previous work that you discussed in the introduction. Although you should always think about your results in terms of other work, there may have been methodological shortcomings that need to be discussed (with suggestions for their correction). Always think carefully about these. You need to consider sources of potential bias, threats to internal validity, imprecision of measures, the number of tests, effect sizes and other limitations, which may have confounded the study (APA 6th edition). Never simply list a series of possible shortcomings, and say that these could have affected your results in some (unspecified) way. Always try to think about the exact way in which a problem with the way the study was executed could potentially affect your interpretation (if you cant think of this, perhaps the shortcoming was not important). For example, if there was a departmental party the previous day, and all your participants were hungover, you might say that as all of our participants were hungover, this could have affected our results and therefore our study was invalid, this could be

corrected by repeating the experiment on non-hungover participants. But consider, how would hungover participants perform differently from normal participants? How exactly could this have changed things? If you think that the participants simply performed more slowly than would otherwise have been the case, and your experiment was looking for a difference between two conditions, then their current mental state would not have changed matters. If both conditions are being performed more slowly, then exactly the same difference would have been found regardless of the absolute times that the participants took. The same would be true in any well-controlled study. If the performance of all participants is likely to have been affected in an identical way regardless of their group, and you are looking for a difference between the groups, then your shortcoming has had no effect whatsoever. Instead, think about any problems with the experiment that could account for any anomalous or counter-intuitive findings. Any such comments should be supported by evidence to show why a particular feature of the methodology may have confounded the results of the experiment. For example, some tasks require lots of practice to attain a stable level of responding before measurements can be taken. In the limited time of a laboratory class it is not always possible to do this. Of course, there may be occasions when the lack of control of one or more variables means that the experiment is inconclusive. All that one can do here is to recommend that the experiment be re-run with appropriate changes. Even an experiment of this type is not a complete waste of time; it has at least made a contribution towards your education as an experimenter. Even with a well designed experiment, the results may not have led to clear cut answers to the questions raised initially, so your discussion might have to suggest further experiments which can now be seen to be necessary for answering the initial question. Finish by discussing the implications for future research. In other words, are there ways in which you could extend the present experiment to consider additional hypotheses? How could you improve on the (valid) methodological shortcomings you have mentioned? Are there any practical applications of your findings? etc.. Always be explicit as to what questions and problems your research raised, and how you would answer/solve them. Never conclude that further research is required, leaving your reader to guess what the further research possibly could be. 7) References References to previous studies by others in the text of the report should be made by citing the author(s) last name(s) followed by the year of publication and enclosing these in brackets. If the name of the author occurs in the text cite only the date within brackets. For example:

The assertion that people use mental models for reasoning (Johnson-Laird, 1983) has been criticised by Rips (1989). If more than two authors have contributed towards a paper, name all authors the first time that you introduce the paper. After the first time you name them you can then shorten any subsequent citation to the name of the first author with "et al." (but only if this is unambiguous, otherwise you must continue to name all of the authors) thus: Galotti, Baron, and Sabini (1986) have discussed the nature of individual differences in reasoning. However, one possible criticism of Galotti et al. is that If you reference in the references list, you must provide a citation in text and vice versa. For example, if Smith and Jones (2000) are a reference in text you would use either the following: IF you are saying Smith and Jones (2000) conducted a study which looked at..... you would use Smith and Jones (2000). HOWEVER, if you are just citing the reference in text you would provide the reference at the end of the sentence e.g. the study found that two out of three people suffered from stress (Smith & Jones, 2000). With two authors you will continually cite Smith and Jones throughout the paper. HOWEVER, if you have three to five authors you provide the first reference e.g. Smith, Jones, and Alexander (2000) and for all other references you would use Smith et al. (2000) or (Smith et al., 2000). When there six or more authors e.g. Smith, Jones, Alexander, Stevenson, Phillips, Riley you only cite the first author followed by et al. for first and all references. Citations will occur in both the Introduction and Discussion sections. You may need to provide reference to a questionnaire/material in the Method section as well. They are essential to providing a background for your experiment and for justifying the questions asked, methods used or techniques of analysis. You should get quickly into the habit of citing and referencing properly if you have not acquired this habit already. All the references must then be listed at the end of the report under the heading References, alphabetically by author and chronologically when several works by the same author appear. For journal references WITHOUT DOI NUMBER (usually found in the corner of the front page) give: last name, initials, year of publication, title, journal (italicised), volume, (italised), (issue in brackets if given), pages, and where the journal article was retrieved from (provide as much information as possible).

For example: Rips, L.J. (1989). The psychology of knights and knaves. Cognition, 31, 85-116. Retrieved from http://www.journalhomepage.com/full/url Galotti, K.M., Baron, J., & Sabini, J.P. (1986). Individual differences in syllogistic reasoning: Deduction rules or mental models? Journal of Experimental Psychology: General, 115(1), 16-25. Retrieved from http://www.journalhomepage.com/full/url For journal references WITH DOI NUMBER (that are electronic versions only) give: last name, initials, year of publication, title, journal (italicised), volume, (italised), (issue in brackets if given), pages. doi:. For example: Smith, S., & Jones, R. (2003). Outcome of psychotherapy on inpatients. Journal of Psychology, 27(1), 31-34. doi:.xx.xxxx/xxxxxxxxxxxx For both type of references if there are more than seven authors (Cited from APA 6th edition book) Gilbert, D.G., McClernon, J.F., Rabinovich, N.E., Sugai, C., Plath, L.C., Asgaard G., ... Bortros, N. (2004). Effects of quitting smoking on EEG activation last for more than 31 days and are more severe with stress, dependence, DRD A1 allele, and depressive traits. Nicotine and Tobacco Research, 6, 249-267. doi: 10.1080/1462220041000167305 For book references give: last name, initials, year, title (italised), place of publication, publisher. For example:

Johnson-Laird, P.N. (1983). Mental models. Cambridge: Cambridge University Press. If the citation is a chapter in an edited book only cite the names of the authors who wrote the chapter; Byrne and Johnson-Laird (1990) in the example below, and then use this format for the reference: last name, initials, year, chapter title (not italicised), "In", initials, last name, "(Ed(s))", book title (italicised), place of publication, publisher. Thus: Byrne, R.M.J., & Johnson-Laird, P.N. (1990). Models and deductive reasoning. In K.J. Gilhooly, M.T.G. Keane, R.H. Logie, & G.Erdos (Eds.), Lines of thinking: Reflections on the psychology of thought. Chichester: John Wiley. If the same authors have published more than once in the same year, disambiguate the reference by adding letters after the date in both the citation and the reference thus: The nature of the development of reasoning strategies is discussed by Wood (1969a) and Wood (1969b). Please consult the APA 6th edition manual or one of the guides posted at the bottom of the page for more details regarding referencing. 8) Appendices The final section of the report is the Appendix section (or Appendices). You should include here all material that would have been obtrusive or damaging to the flow of the report itself, and not just use it as a bin to contain things you wished to say but could not fit into the main report. Therefore, the contents of the Appendices usually consist of raw data, statistical formulae and computations, lengthy protocols, examples of stimuli and details of stimulus preparation, etc.. Have a separate appendix for each type of material, instead of just lumping everything together.

Some Notes on Style: Using the correct style can be very difficult, even if you have written formal reports before. Hopefully, the conventions described below should become fairly automatic. Again, by browsing through journal papers you should remind yourself of what the usual practices are. Use of personal pronouns It is usually assumed that you, the experimenter, did not affect your results in any way that could be unique to yourself (e.g., personality, dialect, etc.). If other people followed your method, they should obtain exactly the same results. Thus, you should NEVER use personal pronouns (we, I etc.) in either the method, or the results section. The discussion/conclusions sections (perhaps also the introduction) are more open to personal interpretations and personal pronouns are more permissible. Even so, be sparing with these, as every one you use can make your arguments seem weaker, more anecdotal, and appearing dependent upon your subjective interpretation (as opposed to being obvious conclusions derived from your results). Despite this, try not to be frightened of putting forward your own views and interpretations, but do be careful not to go too far beyond what your data are telling you. Use of Tenses Tenses can be very difficult to use correctly. These guide-lines can only be very general rules of thumb. Basically, anything that is history should be written in the past-tense. When you write up your work, even your method and results will be history, and should be described in the past-tense. The conclusions of previous workers are history, however yours are still current and should be described in the present-tense. The theories and models that were derived from the results and conclusions still make predictions today (even if they are the wrong ones) and their predictions thus should be described using the present-tense. Thus, for a previous piece of work that you are describing: Smith et al. (1970) found that they concluded thatand developed the XYZ model. This predicts that If you were discussing the results of your experiment: It was found that and thus we conclude that. Our ABC model predicts that Other Points to Note Avoid contracting words (dont, cant, couldnt etc.). Procedure has only one E after the C.

Questionnaire has two Ns and one R. Data are plural. Affect (verb) to have an influence on something: "something has affected my experiment"; something has changed my experiment. Effect (verb) to cause something to happen: "something has effected my experiment"; something has done my experiment for me. Effect (noun) a consequence or an outcome: "this is a negative effect"; this is a bad outcome. Affect (noun) an emotional state: "this is a negative affect"; this is a bad mood. Most common usages are affect (verb) and effect (noun). E.g. "The problems described above affected the results by diminishing the size of the experimental effect." Some General Advice: On the whole, pieces of information should only occur once in the report, and therefore, if you find yourself repeating large chunks of material in different sections you have gone astray and either one of the occurrences is wrong or perhaps you have not planned the content of your sections properly. The exception to this rule is the abstract, which should only contain information reported elsewhere. Write the title and abstract last (once you know what is in the rest of the report), then add this to the front of your report. Keep the raw data and intermediate calculations but do not include them in the report (other than in an appendix). The following websites are useful in providing information on how to write lab reports and information regarding APA 6th edition
http://www.apastyle.org/ provides excellent information and tutorials on how to format using APA 6th edition as well as how to structure a lab report and providing sample papers.

http://www.vanguard.edu/faculty/ddegelman/index.aspx?doc_id=796 (downloaded 30 May 2010). Douglas Delgman Ph.D of the Vanguard University of Southern California has written an excellent guide of how to structure an APA style paper, as well as providing summaries of what should be included in each of the sections of the report.

http://psychology.about.com/b/2010/03/12/tips-for-writing-a-psychology-research-paper.htm provides some general tips and information of how to write a psychology research report

http://owl.english.purdue.edu/owl/resource/670/01/ provides some general information to how to write for psychology http://www2.gcc.edu/dept/psyc/Website%20Materials/APA%20style%20paper%20Sixth.pdf an example of an APA 6th edition paper http://writing.wisc.edu/Handbook/DocAPA.html excellent guide of how to format for APA

Other sites are available also. Take care with all sites that it is the revised 6th Edition that is being used.

M. J. Roberts Modified SEA Sept 1997/Oct 98/Oct 99/Oct 2000 Modified by Debi Roberson, Oct 2001 Modified 3 June 2010 by Victoria Alexander and Richard Hicks, for Psychology 11-100 Bond University.
The current version includes recent references in the place of the 1990s references in the original writing, and refers to the recently published 2010 revised 6th Edition of the APA Handbook. .