For this assignment you will be required to show how creative you are by developing and
analyzing your own fictitious data sets. Please read each question carefully and label and
submit all the appropriate SPSS materials required. Please realize that if you fail to provide
all the necessary SPSS information and/or explain it will be extremely difficult for us to
evaluate your answer.
Here are a few things that we’ll be looking for when we evaluate your answers:
1. An original research idea for each of the research scenarios below. In other words,
please provide us with the following for each statistical test you are asked to perform.
Note: your Assignment should be divided into 8 sections, each section detailing your
research scenario, the statistical test being perform, and an description/explanation of
the most relevant SPSS output.
a. A brief introduction and your research question. By brief we mean not more than 2 –
b. State your Null (Ho) and Research (Ha) hypotheses (we need this to determine
pvalues in some cases).
c. State and label the names of your independent and dependent variables.
2. Data. You’ll need to create reasonable fictitious data for each your research and enter
it into SPSS. By reasonable we mean data that might actually approximate the
variability one would expect to see in your variables if you were conducting an actual
3. SPSS outputs. Be sure to submit, in either Word or a PDF, only the relevant SPSS
outputs (Note: Do Not Submit the SPSS notes). By relevant we mean those outputs
that summarize the statistical evidence of your research for each question. NOTE: only
submit 1 document to the Assignment dropbox. So you’ll have to organize your
answer into its various parts. If you are having trouble formatting SPSS outputs, try
‘Exporting’ from SPSS.
4. SPSS explanations. In your own words, please explain what the crucial components of
the SPSS output actually means. Some things that come immediately to mind include:
a. Explanation of Graphs (e.g., interaction) and Tables (e.g. Chi-square) when
b. Explanation of checks for violations of statistical assumptions (e.g., Levene’s).
c. State and describe the meaning of the most important statistics (e.g., group means,
df, t statistics, F statistics, p-values).
5. Draw accurate conclusions based on your SPSS output in terms of your study
variables. In other words, do you reject the null hypothesis or is your study
inconclusive? And, what does that actually mean in terms of the variables contained in
6. Try to have fun with the Assignment – be creative. You will have ‘real’ data for your
honour theses, this is a good stepping stone to familiarize yourself with SPSS.
Part 1: T-Test
1. (10 Marks) Generate a research scenario where you create a data set of 30
participants, and use a t-test for a single-sample to analyze the data. You’ll need to
be very clear about why you are using this test – so your Introduction is really important.
2. (10 Marks) Generate a repeated measures research scenario where you create a data
set of 30 participants. Use the appropriate t-test to analyze your findings.
3. (10 Marks) Generate a research scenario of 48 participants that should be analyzed
using a t-test for independent means. See your class notes for an example.
Part 2: ANOVA
1. (10 Marks) Generate a research scenario of 40 participants where a one-way ANOVA
needs to be employed. Your data must produce a significant overall result, and there
should be a post-hoc test performed and interpreted.
2. (10 Marks) Generate a 2X2 factorial research scenario of 40 participants. Your
interaction must be significant. Be sure to produce all cell and marginal means and
explain all interactions and main effects. A graph of the interaction is expected too.
Part 3: Nonparametric test
1. (5 Marks) Generate a research scenario of 30 participants where a chi-square test for
goodness of fit must be used.
2. (5 Marks) Generate a research scenario of 30 participants where a chi-square test for
independence must be used.
3. (5 Marks) Generate a research scenario of 30 participants where a Mann-Whitney U
test needs to be used.
Part 1: T-Test
Part 2: ANOVA
Part 3: Nonparametric test
SPSS Assignment Example: One Possible Outline
Below is one possible outline for how to tackle the different scenarios of the SPSS Assignment. If you
follow it great, if not, and submit something that we can easily follow when marking, that is fine too.
The more detail you provide the better – in most instances.
Your research idea and design needs to match the statistical procedure you are using. To that end;
1. It is easiest to create 8 different scenarios and data sets.
2. We need to know your hypothesis – make sure they are easy to find.
3. You need to include the key SPSS outputs for each scenario (no SPSS notes) – and you need to
explain the most critical components of those outputs. You need to determine what those
outputs and components are.
4. We need to see a conclusion, both in terms of the hypotheses and in terms of your study. In
other words, saying you reject or retain Ho is fine, but you will NOT get full marks. State that
you reject or retain Ho, and then state what that means in terms of your study.
A Possible 5-Step Outline Example
Step 1: An Idea: The Canadian government estimates that the average debt load for a student who
recently graduates with an undergraduate degree is $20000. You believe that Covid-19 and its
economic impact will significantly increase this debt load. You randomly select 30, 4
and ask them about their debt load. Note: the design above, fits best for a One-Sample T-Test.
Step 2: Hypotheses and Alpha: Ho: µ = 20000; Ha: µ > 200000 ; Alpha = 0.05
Step 3: The SPSS Output
Step 4: Explain the main components of the SPSS output.
From the SPSS above, our sample size of 30 ensures approximate normality. Our sample mean was
22616.6667 with a standard deviation of 8122.6411. Our one-sample t-value (or t-test statistic) was
1.764 with 29 df. The 2-tail p-value is 0.088, since our research hypothesis is 1-tailed, we divide this p-
value in half, to obtain our p-value of 0.044.
Notes for Step 4:
1. Should you choose to explain the Effect sizes and/or Confidence Intervals to base your conclusion that
is perfectly fine too.
2. Should you choose to say, since the t-test statistic (t-value) is 1.764, and it lies beyond the t-cut score
of 1.699 we have evidence to support Ha, and reject Ho, that is fine too.
3. If you choose to do some of these by hand, in addition to SPSS, that is fine too, and really good
practice. But it is not required unless specifically stated in the question.
Step 5: Conclusion, and state (re-state) the p-value
Since our p-value is less than our set alpha level (0.044 < 0.05) we have evidence to support the research hypothesis (Ha), and thus reject the null hypothesis (Ho). Student debt following Covid-19 has increased.