I ran a study to compare 2 different computer interfaces (say A and B) meant to be used in an unfamiliar environment.
I want to compare:
- How much effort participants took to complete a certain task when using the interface for the first time.
- How much easier or harder it is for participants to complete the same task the second time they use the interface.
The study procedure:
- Every participant used either interface A or interface B [Variable: type of interface, 2 levels: A or B, between-subjects]
- Every participant used the interface (either A or B) to complete the same task twice. (Most) participants did the two tasks on separate days. [Variable: time of task, 2 levels: T1 and T2, within-subjects]
- Since the environment needed to be unfamiliar, the environment participants did the task in changed each time they did it. The order of the environment was counterbalanced: Some participants did the task in environment 1 first (at T1) and then in environment 2 (at T2), and others did the task in environment 2 first (at T1) and then in environment 1 (at T2). [Variable: environment, 2 levels: E1 and E2, within-subjects, counterbalanced]
- The measure of "effort" is a continuous variable.
What statistical tests should I use to understand:
- The effect of the interface (A vs B) on effort?
- Which interface got easier with time (the difference in the effort of using a particular interface between T1 and T2)?
I'm primarily confused because I could consider this a mixed-factorial design with the interface and the environment as the independent variables. However, I don't really care for the effect of environments - the change in environment was just to ensure it stays unfamiliar during both tasks. What I really want to understand is the the impact of the interfaces and time.
Would I use a two-way mixed ANOVA and post-hoc tests? If so, which ones would give me what I'm looking for? If not, what should I be using?
I will be using SPSS for analysis.