This task is to rewrite the analysis for two experiments to the standards listed below. I will provide the Word document containing a description of these experiments, as well as the data syntax, and output files used to generate these findings .
The following details the changes needed for two phases of two separate Experiments. However, I am requesting your help to transform the existing write-up and analysis into the format described below. I will also need this same analysis for two phases of a second experiment.
I can provide a paper that gives additional details about what is needed. The paper is on different data and uses a different type of ANOVA (multiple regression), but gives an idea of the format that is needed.
I am trying to see what you would charge to make the changes to the experiment analysis according to the directions below. I can provide the syntax, data, and output for the SPSS, as well as a template for the tables that will need to be generated. The SPSS analysis does not need to be redone, it just needs to be reformatted.
There are two phases of two experiments. Each phase examines three DVs. One person had attempted to perform the analysis, but was unable to complete the task. I can provide an example of the work done so far if needed.
I am looking for someone who:
1. knows how to communicate statistical findings in clear, concise terms and can interpret the instructions below
2. someone who is familar with SPSS, and can generate supporting figures
3. someone who can dedicate the time and effort to get these two phases of the two experiments done correctly in the next 10 days
If you do not fit the above criteria, please do not apply.
There is a strong possibility for follow-on work for an additional experiment - this data will be forthcoming in the next week or two.
First, you should include a comprehensive table of all possible means, SDs and ns for each ANOVA and a corresponding ANOVA summary table. A 2 x 2 x 2 , A x B x C ANOVA design would generate a grand mean and the following additional means: A1, A2, B1, B2, C1, C2; A1B1, A1B2, A2B1, A2B2, A1C1, A1C2, A2C1, A2C2, A1C1, B1C2, B2C1, B2C2; A1B1C2, A1B1C2, A1B2C1, A1B2C2, A2B1C2, A2B1C2, A2B2C1, A2B2C2. These categories should be described in words representing the experimental conditions, rather than letters. The ANOVA summary table would include main effects for A, B and C; the two-way interactions-AB,AC,BC; three-way interaction ABC, and within or error component. Again, state the conditions in words rather than letters. Columns would include the effect label, SS, df, MS, F and p-values. (Note: I will provide this table structures for these tables) In the experiments, for example, A will be interface, B will be participant and C will be collection. The second Experiment would be similar.
Second, when interpreting ANOVA effects, first describe in one sentence which effects are statistically significant. Then, interpret the results starting with the most complex statistically significant one. If you have a significant three-way interaction, focus your interpretation on that. There will be many ways to graph the interaction (e.g., C effects at each combination of A and B, B effects at each combination of A and C, and A effects at each combination of B and C). Typically, you choose just one of these and generally the one that highlights differences for your main variable of interest or the variable for which differences are greatest and/or most irregular. One can always graph all three ways and find the one that has the clearest interpretation. Any such figure should have 8 means in a 2 x 2 x 2 design.
Third, one typically does simple effect tests to interpret a three-way interaction, and there are also many ways to do this. One should do the tests that correspond to the way you graph the data. All of this stuff can be done with SPSS or SAS.
Fourth, you may not need to comment any further beyond the interpretation of your most complex interaction.
Fifth, you should not talk about differences between scores unless they are trustworthy (i.e., statistically significant).