Appendix 2
How to write your
experimental or survey report
Contents
Introduction 151
Sections of a report 152
Title 152
Abstract 152
Introduction 152
Method 154
Results 157
Discussion 158
Conclusion 159
References 159
Appendix 159
Introduction
It can be difficult when you come to write up your quantitative
(experiment or survey) project to know what information should go
where, so the aim of this appendix is to provide guidance on how to
structure your report. In terms of how much to write in each section,
there are no strict guidelines, but your introduction and discussion
should generally each use around one third of the available word
count. The abstract and results sections should be brief. The remainder
should be used on your method section.
Psychology demands a high standard of writing, and one of the key
skills that you have been developing during your degree is the ability to
write in a clear, scientific and concise manner. One of the key aims of
this project is to develop these skills even further.
Your audience will be your peers, therefore you need to write as if the
reader is uninformed of your research area and define any key terms
and concepts. Always try to write in the third person, for example, ‘It
was considered pertinent to …’ as opposed to ‘I thought that I should
…’. Also use the past tense – research that you have discussed is from
the past. Similarly, you are discussing procedures and analyses that you
have already conducted, and therefore these are also in the past.
Reports of experiments and surveys are very similar, but you must
remember that whereas experiments are concerned with causal
relationships, surveys look at correlations. An experiment allows you to
test whether changes in one variable cause changes in another, but
with surveys you can only hypothesise about where there might be
associations between your variables of interest.
Figure A.1 Correlation does not imply causation!
Introduction
151
Sections of a report
Title
Your title should be concise and give a clear indication of what you did
and/or discovered. For an experiment, your title should include some
indication of your independent and dependent variables. So to use our
prospective memory example, the title may be ‘An investigation on the
impact of emotional saliency on prospective memory: a comparison
between older and younger adults’. For a survey, a title such as ‘The
relationship between personality type and attitudes to online learning
among teachers’ could be used, You can use the title that you used in
your poster here, but if you received feedback that it could be
improved then it would be a good idea to amend it.
Abstract
Despite the abstract being the first part of your report, it is generally
written last. It should provide readers with sufficient information about
what you did in your study and what you found, without having to
read the whole report. An abstract must be concise (150–200 words)
and include: essential points stemming from the rationale; the method
used; results; and a conclusion relating to the rationale. Specific details
about gender or age of participants, method and design are not
required, unless the information is key to your study. State your
findings with an indication of what was significant, and end with a
sentence explaining your findings. Be aware that although you write
this as the last section, readers encounter it as the first!
Introduction
The introduction should be structured as a down-pointing triangle,
starting from a broad statement concerning the area you are
investigating and moving towards the more specific areas that you will
be exploring. This should include research that has been conducted,
what it revealed, and any specific questions, that is, the rationale
behind your own research followed by your research hypotheses.
Appendix 2 How to write your experimental or survey report
152
General introduction
Specific studies
Rationale
Hypotheses
Figure A.2 Outlining the stages of an introduction
Start with a general introduction spanning no more than two sentences;
don’t make it too vague or start with a life, the universe and everything
sentence (e.g. ‘It is people’s ability to comprehend sentences that
separates us from animals and enables us to be the most successful
species on the planet’). Begin with a clear explanation of the theoretical
models or claims that you are aiming to test. This should lead directly
on to research within the field that is relevant to your project. You
have spent a long time researching the literature, developing your ideas,
building a rationale, then designing and running your study, and this is
where you report what you did and why. Don’t be sidetracked into
other areas of research – keep your hypotheses in mind when you are
writing and use the notes that you have been taking throughout the
whole research process. Your introduction is built from the detailed
notes made from the articles in your literature review. Make sure that
these are in your own words and tell the story of how your research
developed, demonstrating your creative and critical thought processes.
Each new paragraph should introduce a new, but related, topic. You
can think of your introduction as a detective novel. Introduce the
characters (your variables) one by one, each in a new paragraph.
They’re all related to each other in some way, so make those links
explicit. Explain how these characters have behaved (what has the
literature said about them?), and how they are linked, or might be
linked. Comment on inconsistencies in the way these variables have
behaved (you’re building the plot – your argument), asking whether
other variables shed any light on this strange behaviour. Then at the
Sections of a report
153
end summarise all of this information in a substantial paragraph that
sums up your thinking, your rationale, for designing this study.
You do not need to go into the minutiae of each paper that you cite –
merely state what was investigated, what the method was (briefly) and
what was found. Ensure that you write in your own words, reporting
the gist of each study rather than relying on direct quotations. And
remember, only the research that is directly relevant to your experiment
or survey should be discussed in your introduction.
When you have written your rationale, your introduction should move
on to state your predictions.
For an experiment with a two-way design, there will be three
hypotheses: one for each main effect and one for an interaction.
However, you should only state those relevant to your predictions. Do
not state the null hypothesis.
For a survey, your hypothesis should refer to what you will measure
and how you expect the measures to relate to each other. Chapter 13
suggested this hypothesis: ‘Teachers who have an introverted
personality type have significantly more positive attitudes to online
learning than teachers who have an extroverted personality type’. This
reflects that the survey would measure personality type and attitudes.
Method
This is where you describe everything you did. Your study should be
able to be replicated exactly as you ran it, based on the information
provided in the method section.
The method section for surveys and experiments are similar but there
are a few key differences. For survey work you would usually report
the following subsections: participants, procedures, measures, and
treatment of data. In experiments, the normal order would be: design,
participants, materials, procedure. To avoid confusion we’ve described
how to construct methods sections for surveys and experiments
separately.
Appendix 2 How to write your experimental or survey report
154
For survey projects
One common mistake students make is to try to report a design
section for survey work and then struggle to define which variables
were independent (IVs) and which were dependent (DVs). To be
clear, survey research is correlational and you do not manipulate
any variables; we are not assessing causation. Taking the research
aim from Chapter 13, ‘to investigate whether personality type is
associated with attitudes to online learning among teachers’, many
students might say that the IV is personality type. For the purposes
of running the analysis in the statistics software where an IV is
required, this would be correct. However, given that an IV is
something that you manipulate, it would not be possible to allocate
people to a personality type. As you have not assigned this
category to anyone you would not report it in terms of being an IV.
Report only the details of the data you have collected; do not state
whether any variables were the IV or DV as these terms are not
applicable. Finally, you would not usually report in terms of between
or within groups for a survey methodology. You might state that you
used repeated measures if you collected survey data from the
same participant on subsequent occasions.
Participants: This is where you report what the sampling method
was, the power calculation if you did one, and any inclusion/
exclusion criteria.
Procedure: Here you report exactly what you did; this includes
what survey software you used and details of how participants were
‘approached’ to participate. For email invitations, report how
participants were invited to participate (by whom, how long they
had to respond, whether any emails bounced back, if anyone asked
to be withdrawn, when reminders were sent out). For freely
available web surveys, report how long the survey was live, where
it was advertised and/or how respondents would find the survey.
You should report how you gained consent from the participants,
and state who gave ethical approval for the study.
Measures: This is where you describe what was in the
questionnaire. You should briefly list any demographic questions
you asked. For other questions, you might summarise the question/
topic area and then provide some detail about the response option.
For key variables this should include reporting what the response
options were, for example, ‘Smoking status was assessed by self-
Sections of a report
155
S
report of “never smoked”, “used to smoke but have now quit”, or
“current smoker”; or details of the anchors and labels if a Likert
scale was used, and an indication of direction of effect, for
example, ‘Respondents were asked to rate agreement with a
number of statements, such as “I try to see the good in people” on
a scale of “strongly disagree”, “disagree”, “neither agree nor
disagree”, “agree”, “strongly agree”, with higher scores indicating a
more optimistic personality’.
If you used a pre-existing scale you should reference it. Include
basic details if it is a more commonly used scale, state whether you
made any changes to the scale, what the Cronbach’s alpha has
been in previous studies (usually take the value from the paper that
contains the psychometric validation of the instrument), and what
the direction of effect is.
Treatment of data: Always state which software package you used
and reference it. For pre-existing scales, report whether you
followed published scoring procedures or describe how you scored
the data. Give detail about any items that were reverse scored.
Give some detail about your data-cleaning processes: did you
identify and remove duplicates? Did you delete any data or remove
outliers? Did you calculate any new scores or transform variables
(e.g. z-score, log-transformation)? You report which tests you
carried out on which variables and why; sometimes it helps to do
this under the subheading of your hypothesis if there is more than
one hypothesis. Finally, report the alpha level that you used to
determine statistical significance.
For experimental projects
An experimental method section starts with a design subsection.
Design:This should begin by describing the experimental structure,
explaining whether a repeated-measures, independent or mixed
design was used. It should specify how many IVs there were and
how many levels each had, and it should explain rather than simply
label variables and levels. Thus, for the prospective memory
example you would explain what was meant by ‘emotional salience’
and how age was defined. Always use meaningful labels rather
than referring to Condition 1 or Condition 2, etc. You also need to
clearly state your DV(s), describing how you measured them. Any
Appendix 2 How to write your experimental or survey report
156
E
measures that were taken, such as counterbalancing or random
allocation, should also be described here.
Participants: Here you should include the number of participants
and basic demographic information, for example, Open University
(OU) students, proportion of male and female participants, etc. Also
include any additional information you collected that was important
to your study, for example, handedness, normal or corrected vision,
native language, etc.
Materials: This describes what you used in the experiment. Don’t
include very basic information like a computer or a pen, but do
include the experimental design software you used, what materials/
stimuli you used, and where these were obtained. A single example
of your stimuli may be included, with the others in an appendix. You
should also state the specific characteristics of the stimuli, such as
height, contrast, brightness of images.
Procedure: This section describes what each participant did in the
experiment in sufficient detail to permit it to be replicated. So,
summarise the initial instructions that were given to the participant
and describe what the participant had to do during the experiment.
Information that relates to how your stimuli were presented should
also be given here; for example, the number of blocks and trials in
the experiment, the structure of the trials, their spacing, the nature
of the practice block, etc. If your pilot study led to you making
procedural changes, for instance, due to ceiling or floor effects,
describe these modifications. End the section by describing the
debriefing procedure you used.
Results
In this section you present the results, but don’t interpret them here.
Begin with a sentence describing what you found.
For surveys, report the total number of respondents and show a table
of demographic and other salient features, reporting N, percent/
frequency, and where appropriate means and standard deviations. If
you have split your sample by a key variable (such as personality type),
you may wish to display demographic data broken down by that
variable, and indicate if there are any significant differences between
groups. Ensure that the table is clearly labelled with conditions, and
that graph axes are clearly and appropriately labelled. Also report the
Sections of a report
157
Cronbach’s alpha (internal reliability) of any scale that you used along
with descriptive statistics for each domain (or total scale score if
domains are not used), including number of items (questions) in the
domain/scale, number of respondents, the score range, mean, and
standard deviation.
For an experiment, present descriptive statistics, mean, and standard
deviation of conditions in a graph or table.
For both surveys and experiments you should then note which
analysis you conducted and report the results of your inferential
statistics for each of your hypotheses. Remember to include the
probability value and whether the test was one- or two-tailed if
appropriate. For ANOVA, it is usual to present the main effects before
the interaction and to comment on the direction of any significant
effects. If you carried out post-hoc statistical tests to explore the main
effects or interaction, you should report the post-hoc test results after
this. Remember to give titles and labels to any tables, graphs or figures
that you include. You should resist the temptation to cut and paste
from the SPSS output as the test results are not presented in the
conventional format for reports; the table’s axes are likely to be
inappropriate for your needs; and numbers may be reported to five
decimal places. It is far better to draw your own table and use Excel
for graphs. You can refer your reader to the SPSS output tables in the
appendix.
Discussion
In contrast to the introduction, this section begins with a narrow focus
and broadens out to general issues. Start by restating your results,
relating them to your research question, and stating whether they
support your hypotheses. For example, thinking back to the
prospective memory example in Chapter 12, you might report:
The results are consistent with the findings of researchers who
claimed that elderly individuals are more likely to act upon an
intended event than their younger counterparts.
Then interpret your findings in terms of the theoretical issues raised in
the introduction, noting how these might be further investigated.
If your results do not support your hypothesis, discuss possible
explanations why. Don’t criticise your method, or the participants,
unless you are sure that they have contributed to results lacking in
Appendix 2 How to write your experimental or survey report
158
authenticity. Bear in mind that it’s when your results don’t match other
results that you have a golden opportunity to really discuss the issues
and provide an alternative explanation of what is going on in a
particular situation. It may be that the theory needs modification, so
you need be clear about how your explanation accounts for your
results. Avoid overstating the importance of your findings; they have
not ‘proved’ anything – you can only illustrate or support, given certain
conditions. Once your explanations are explained to the best of your
ability, extend the discussion to contemplate implications for future
studies, considering current theories or controversies. Make sure you
don’t fall into the trap of briefly describing your results then
embarking on a general discussion of the topic; this section is primarily
a discussion of what you discovered.
Conclusion
Try to end on a high note. Summarise the strengths of your findings,
conclusions and implications or ideas for the future.
References
The end of your report must have a full list of all the references (in
alphabetical order) that you have utilised in your report. Use OU
Harvard style for the references – full details of the format can be
found on the OU Library website.
Appendix
Your appendices should be at the very end of the report and include
any supporting evidence that you have referred to in your main report.
Number or list them as Appendix A or Appendix 1, etc.
Sections of a report
159