Writing up quantitative reports - a detailed general guide PDF

Preview

Summary

notes from lecture...

Description

Writing up psychology quantitative lab reports. A good book to look at, in addition to this handout, is: Harris, P. (2008, or earlier) Designing and Reporting Experiments in Psychology. 3rd edition (or earlier). Berkshire: Open University Press This should be available as an e-book via the library. It contains sample examples for the sections of the report, of how to write up. A good number of introductory textbooks on statistics and quantitive research methods will have chapters on writing up lab reports too. Also, do look at the quantitative lab marking criteria, which are organised by the relevant sections that are required for a report (in your Module handbook). Here, you will see (in brief) what markers are looking for, for each section. __________________________________________________________ Writing a quantitative lab report ______________________________________________________________ The subsections of a lab report appear in a standard order:    

   

Title Abstract Introduction Method o Participants o Design o Materials o Procedure Results Discussion References Appendices

You will find below details of what should be included in each of these sections. Part of the point of having sections is so that readers can look quickly to see the specific information they may be interested in, when 1

browsing the literature. Information should not really be repeated across sections, therefore (though naturally there will be some repetition/elaboration of concepts and theories across Introduction and Discussion).

Title (and title page) Your title should summarise the main idea of your report, making the key outcome result clear. One common and clear approach is to identify the variables under investigation and aim to also convey the results found. The recommended length for a title is 10-12 words. The title is presented on a page on its own.

(In two changes from standard APA style we would like you to use your student number and NOT your name on the title page, and the running head at the top of each page should also be your student number rather than your title).

Abstract The abstract acts as a brief, self-contained, summary of your report. Many journals do not ask for the abstract to be divided into subsections, but you can, if you prefer, structure it under sub-headings, as indeed some other journals do.

(Note that each section begins on a new line). Otherwise just

write it in a similar way, but without the subdivisions.

Introduction: Usually written in the past tense and about 2-3 sentences long, provides the context for the work by identifying the specific area/background

2

of the research and indicates where the question comes from. An effective Introduction includes a hypothesis, states the question to be answered, or the purpose of the study. Methods: Usually written in the past tense and 2 or 3 sentences long. Explain what procedures you followed to test the prediction(s) or answer the research question. This section should include brief details of participants and the principal techniques and design features that you used to carry out the study. Results: Usually written in the past tense and about 2-4 sentences long, and states the main results of the study. Discussion: Usually written in the present tense and about 2-3 sentences long, points out the most significant finding/s and explains the connection between the important finding and why it matters to the particular field of research.

The abstract needs to cover these four sections in approximately 180 words

3

Introduction- [use Insert\ Break \ Page Break to separate Introduction from Title Page\Abstract] The body of your report opens with the introduction on a new page. As the introduction is always the first section of a report there is no need for an ‘Introduction’ subheading, although this is optional. The role of the introduction is to introduce the problem, develop the background, provide a rationale for your study, give a brief overview of your study, and articulate the predictions/hypotheses, as follows.

Your introduction should begin with a brief broad introduction to the topic/issue that your study addresses in paragraph 1.

You should then move on from Paragraph 2 to discussing the existing literature. You are not, however, writing an essay review of all of the existing literature. It is important to remember that you are trying to demonstrate continuity between the existing literature and the study you are going to report. You are aiming to provide an appropriate background and rationale for your study.

A very important point for any of your psychology reports or essays is to make sure that you provide a citation for any assertion or claim that you make (see also marking criteria). So, if you wrote in your introduction “Mindfulness is related to anxiety”, then you need to back this up by referring to the authors of relevant published work i.e. a citation given in brackets (please see separate guidance on APA style). Or, if you wrote “Memory can be divided into long

4

and short term memory”, please provide a citation to back this up. Where’s the evidence, please!

If there are one or two key studies underlying your own study, do make sure you provide a comprehensive overview of the key points and results for these directly related studies, otherwise the rationale for your study will not be very clear.

If you look at the marking criteria, you will see that for a merit or

above, markers will be looking, amongst other things, for a clear understanding of theories and concepts, and a clear overview of any key studies, for example. Take care that when summarising results from other studies, don’t just describe/summarise - relate them to theory - spell out the implications of results e.g., how were they interpreted by the authors? Perhaps you can offer an alternative interpretation (this shows good critical ability).

Any research questions/issues raised by any past studies and/or theory which are to be addressed/tested by your study should be made clear. Or, if you are adopting a novel approach, clarify this. So, in short, make clear the rationale for your study, and its general research question.

If (as may well be in class

sessions) your study is simply a replication or extension of a previous study (which you will have summarised), then you can indicate this. It may be that you are adopting a slightly different approach in order to establish whether the results from a previous study generalise further (e.g., to a different population, or using slightly different stimuli or measures). Thus, although an appropriate

5

rationale is ideally a ‘gap in the literature’, in teaching exercises like this it can also be a clear underlying importance of the research question being studied.

Once you have provided an appropriate background and rationale, give a brief overview of your study, explaining the design, and relevant variables, making clear how it will address the research question raised. Include details of variables you intend to manipulate and/or measure. You may have more than one hypothesis (e.g. 3 predictions for different Mindfulness predictors), because you will be studying designs with more than one independent variable. After you have given the reader an understanding of your study, aim to articulate the more specific predictions clearly and succinctly (see the marking criteria for this final part of the Introduction).

Method

Your method section should describe in detail how your study was conducted. The method section is conventionally divided into three subsections: Participants, Materials/Measures or Stimuli, and then Procedure. As noted above, many textbooks on report writing still include also a Design section as part of the Method (e.g., Harris, 2008 - Designing and Reporting Experiments). This is not wrong, and journal articles often do this. However, as we noted above, APA style encourages the design features (e.g., IVs, DVs, conditions, type of design) to be given at the close of the Introduction, rather than have a Design section. 6

Method Participants

You should report the number of participants and who they are and how they were recruited. (This gives the reader information about your sample and the extent to which you can generalise to a wider population). Also, give the basic demographic details of your participants if they are available i.e. gender, age (the mean, and the range or standard deviation). It can sometimes be relevant, when you have groups of individuals, to know the number (and mean ages etc), in each group (e.g., 30 older adults, 28 younger adults), especially if unequal. In addition, where other demographic details are either variables used in your study, or relevant to interpreting your results (e.g., number of years of educational study), you should include them in this participant section. Design Some minor repetition is allowed between sections to reinforce the structure for the reader, but keep any design statements to only a sentence or two, specifying the in/dependent variables, and whether a quasi-experimental or correlational study.

Materials/Measures/Stimuli section,

if you have several types of stimuli for certain conditions of the study (e.g., a cognitive experimental study such as the Stroop study), then you will need to detail these, under a subsection Materials (or you could call it Stimuli). You should provide sufficient detail such that your reader would be able to duplicate the materials used in your study, and understand how they relate to any conditions (or other aspects of the design) you have outlined at the close of the introduction. For example, is there a control condition – if so, explain. You need to include appropriate examples within this subsection please, so 7

that your reader understands the stimuli in relation to the conditions fully. However, full lists of materials/stimuli can be placed in an Appendix. You could, after a few examples, write “See Appendix A for full details of the stimuli used”.

Thinking now about a questionnaire study (e.g., for a correlational nonexperimental design), you are likely to call this subsection |Measures”. If you have used well known standard questionnaires, you usually in this Measures section provide a reference to where it is available. If you know it, report the reliability and validity coefficients, if supplied by the authors of the questionnaire.

Also some indication of the nature and examples of the

questions used is helpful for the reader, and also the scale used (e.g., is it a scale of 1 to 5, for example). Useful also to give the reader the maximum and minimum score possible, for a particular scale. If, on the other hand, you have designed unique materials you will need to include full details of that process, and ideally provide reliability and validity coefficients. You will be given additional advice by the lecturer on how to write up materials/measures sections for questionnaire-type studies.

Procedure This section should provide the reader with sufficient information to allow them to replicate your study. You should thus take the reader through each step in your study including instructions to participants, any practice items, how any groups were formed (e.g., random allocation?), how any other variables were manipulated (if the study is an experiment). How is the DV measured. What about order of conditions, in a within-participants experimental design. Was there any counterbalancing of conditions or stimuli, or perhaps random presentation of stimuli, for example. If, on the other hand, the study is a correlational questionnaire study, what order were the questionnaires given in, for example.

Were participants

tested individually, and where? For example, how long were any materials/stimuli presented for? Be very clear about what participants had to 8

do. How long did the procedure take?

These are some examples of the

kind of information you need to provide, in order for another researcher to be able to replicate. As with the apparatus or materials section, if you have complex information in your procedure section which you think is distracting, you can move it to an appendix. You might thus summarise very briefly the instructions to participants, though still make sure the reader can understand what participants had to do in this Procedure section. However, you could then write “full details of the instructions to participants can be found in appendix B”.

It is appropriate and very important to refer to ethical procedures in this section.

So please have a brief paragraph on ethics.

Comment on method sections

You may well find Methods in journal articles which do not adhere exactly to the above. For example, you may find some information about stimuli (e.g., how many in each experimental condition) in the Procedure, whereas others might put this in a Stimuli section. It is OK to have a small degree of variation of this kind, as long as it is clearly not the wrong information for a section (e.g., it would be wrong to say what participants had to do, or how stimuli were presented, or the order of questionnaires, in the Participant section.). It is important to avoid unnecessary repetition across sections. DON’T LIST e g., pen, paper, computer!

Results section 9

This section summarises the data collected in your study and any statistical analysis carried out. It should be succinct, and clearly laid out. Discussing the implications of your results is not appropriate in this section. You should organise it so that you

•

have a brief summary of what the data consist of, and how they were collated/treated e.g., how scored, any missing data, then you

•

present descriptive statistics, then you

•

report the results from the inferential stats tests

How were the data collated/treated? (not actually a section heading in your report)

Your results should begin by explaining how you dealt with the raw data collected from participants. Students often forget to do this. For example, did the data consist of the mean reaction time for 20 trials in each condition? If so, tell the reader. Or did it consist of a total score after combining the response to 15 questions in a questionnaire scale? It can be appropriate to give details of scoring process here e.g., for questionnaires, though occasionally in some journal articles you see this at the close of the Method section. That would be ok too – just don’t repeat.

So, please tell the reader what the data consist of, and how they have been put together. What about any missing data – how was this dealt with? Was

10

any participant excluded (e.g., because misunderstood instructions, or left the study early)? If so, tell the reader briefly.

Have you done any exploratory data analysis, checking assumptions of parametric tests, for example? If so, tell the reader briefly. Were any outliers excluded – and if so indicate this briefly and on what basis they were identified and your reasons for excluding them. This information is often best presented by histogram distributions, which also give a useful set of overview of Descriptive statistics.

Descriptive statistics (not a separate section usually)

After you have done this short paragraph on what the data consist of and how they have been collated, then you go on to present descriptive statistics. That is, summary statistics. You nearly always need to present a measure of central tendency for the dependent variable (e.g., mean scores across all participants for each condition/variable). You also nearly always need a measure of variation (standard deviation usually, and/or range) as well.

For each measure in a questionnaire study, it is usual to present the means and standard deviations, and often the highest and lowest score. However, remember that it is very likely that different questionnaires have different scales, so don’t compare the means in this case! You could comment on the variability in the measures, and whether or not the full scale appears to have been used by participants, or where the mean is in relation to the ends of a

11

particular scale e.g., is the mean towards the top of the scale, and is there very little variation across the scale (this would be important to note).

If , say, you have just two conditions in an experimental study, then you could probably just present the means and standard deviations in the text. However, it is usual to present a Table, or a Figure e.g., graph. Graphs are particularly important when you have quite a number of conditions.

Scattergrams too can be very useful when you are exploring relationships between variables. If you have very many tables/figures, it may be appropriate to present just key ones in the Results section, and refer the reader to the Appendix for any additional figures or tables.

These tables or figures should be presented with appropriate labelling/numbering etc, and legends or titles , and APA style. As you read journal articles, you will see how figures and tables are presented, and also of course, within textbooks.

You will find that graphs generated by SPSS will not conform to APA style. NB DO NOT copy and paste tables and graphs from SPSS. Excel provides the flexibility you will need to produce correctly formatted graphs.

12

After presentation of descriptives, (Tables and/or Figures), you should make some brief comment about any clear pattern in the summary/descriptive data – draw the reader’s attention to anything of interest in the data (e.g., do the data in Table/Figure appear consistent with a hypothesis). But just a sentence or two, before you go on to the stats test. From looking at the descriptives, remember you will not know if any pattern in the data is significant.

Inferential statistics (again, not usually a subsection)

After you have succinctly presented the summary/descriptive statistics, it may be relevant to report briefly/summarise any “exploratory data analysis”. This is where you use software (SPSS again) to look at the distribution of the data, for example, to see if it is normally distributed, and/or whether there are any outliers. (if you do want to report figures/graphs from exploratory data analysis, probably better to do this in an appendix, otherwise it can detract from the key statistics).

The purpose of the exploratory data analysis is

usually to see whether or not the data are suitably distributed to allow “parametric” statistical analysis (you will learn about this in class).

After any exploratory data analysis, and commenting on whether or not paratmetric statistical analysis is suitable, you then clearly and concisely report the results of the inferential tests, as you will be taught in class. For example, the results to a multiple regression or ANOVA must be reported formally (see lecture material, also see examples in the Harris book).

13

You may also be taught about effect sizes and confidence intervals, and for some statistics, you will be able to find these on SPSS printouts. If so, you should present these too.

If there are any follow-up or further tests, then these would come after the main statistical analysis.

Remember to make clear the direction of any

significant effects, (e.g., is the mean response time in condition 1 longer than condition 2? Tell the reader, please. Or, for correlation and regression, is it a positive but weak cor...