I need to find out whether "gender influences the effectiveness of the two intervention programmes designed to increase participants confidence in being able to cope with statistics training?".
I have not used SPSS for quite some time, and I am unsure which test on SPSS to carry out to answer the question above. Thanks.
Holly Woods, assuming that you have a continuous variable to represent confidence, there are several ways you could analyse your data. At the simplest level, you could simply conduct an independent-samples t-test in which you compared the final scores on confidence for males versus females - assuming the two groups' confidence levels were identical/similar to begin with.
Using a more sophisticated approach, you could (or, rather, should) also conduct an independent-samples t-test on the original confidence scores to explore whether the two groups' confidence levels were similar at the start of your study.
You could also calculate, for each person, a before and after confidence score, then conduct an independent-samples t-test (again males versus females) on those "difference" scores.
There could even be justification for conducting two paired-samples t-tests (one for males, the other for females) and seeing what kind of results eventuated.
I recommend you conduct an ANCOVA in which you use the baseline level of each person's confidence score as the covariate.
Which of the above you choose might depend on the level of statistical and methodological sophistication expected of you as well as your own confidence with stats.
And, of course, you would need to consider something other than any of the above if your data need to be regarded as ordinal or categorical in nature. You really need to have provided that kind of information if you'd like others to help you.
Robert Trevethan has described all aspects of the question in quite detail leaving not much for us to say. I would go for two way ANOVA. I would take two groups of gender (Male and Female), and take two groups of intervention program (A, B), I would take confidence as a DV. To simplify things for my self, I would take the difference of confidence before and after the intervention.
So, in fact I would be answering whether confidence change is different across Male and female. Additionally, I would be answering whether confidence change is different for two intervention groups.
Holly Woods It is critical to evaluate the type of variables you have while determining which test to utilize (i.e., whether your variables are categorical, ordinal or interval and whether they are normally distributed)
However, the selection of an acceptable statistical approach is dependent on three factors: the study's aim and purpose, the kind, and distribution of the data utilized, and the nature of the observations (paired/unpaired).
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS