# Help with analysis of 2 x 2 crossover study

#### Belmo

##### New Member
I am having difficulty finding the correct statistical test to run for a 2 x 2 crossover design. The aim is to compare the exercise program delivery method (digital versus paper) on a range of balance measures (continuous measures). Participants are randomly assigned to one of two groups: Group A (digital / paper) or Group B (paper / digital). All participants are tested at baseline. Group A participants undertake 4 weeks of digital exercise program, group B undertake 4 weeks of paper exercise program. All participants are tested at 4 weeks. Then Group A participants undertake 4 weeks of paper exercise program and group B undertakes 4 weeks of digital exercise program. All participants are tested at the end of the 4 weeks. I know that I need to evaluate if order of delivery has an effect on the results but I can't work out which test to run. Any assistance would be greatly appreciated.

#### Karabiner

##### TS Contributor
I suppose that a mixed analysis of variance can do that. You need a variable
"condition A" and a variable "condition B" and a variable "group" (which
indicates whether a participant had the sequence A-B or B-A, respectively).
You can then perform the analysis with "group" and with the repeated-measures
factor "condition".

With kind regards

Karabiner

#### Belmo

##### New Member
I suppose that a mixed analysis of variance can do that. You need a variable
"condition A" and a variable "condition B" and a variable "group" (which
indicates whether a participant had the sequence A-B or B-A, respectively).
You can then perform the analysis with "group" and with the repeated-measures
factor "condition".

With kind regards

Karabiner
Thank you.