Hello,
I'm hoping someone might be able to steer me in the right direction.
I'm undertaking a research project looking at parents/caregiver's attitudes towards telemental health (TMH). Specifically, comparing 3 groups:
1. parents/caregivers who have a child/adolescent with a mental health diagnosis
2. parents/caregivers who have a child/adolescent without a mental health diagnosis
3. adults with no dependents
In addition to this, I want to understand how parents/caregiver's culture (CALD background) influences their attitudes towards future use of telemental health.
My survey questions will be based on parents/caregivers future attitudes toward using TMH.
I am stuck on selecting the correct statistical analysis to undertake. I am thinking some form of regression analysis, but not certain which one or if a regression is appropriate.
Thank you :)
Sonya
As you are studying attitudes I attach two of our papers on attitudes to give you some ideas:
It seems to me that you have two binary predictor variables
1. Does the person have children?
2. Do the have a child with a mental health diagnosis.
Note that these two binary variables describe the three groups
0-0 = No children
1-0 Children but no diagnosis
1-1 Children with diagnosis
Since I don't know what scale your outcome variable(s) are measured on, I can't comment on the actual type of regression.
Following on Ronán Michael Conroy questions, your analysis is PROBABLY an ANOVA with the grouping variable as the independent variable and the background variables as covariates. But a multiple regression with the group variables entered as dummy variables would work too. The former is simpler, but would not yield a regression equation if that's what interests you.
Rhianon Allen makes a very useful point. With a regression, you can define the effects that interest you. In this case, a) what is the effect of having children and b) does the presence of mental health issues have an additional effect over and above that of having children.
With an ANOVA, you end up asking if there's "some kinda" difference between the groups. You rarely encounter a situation where the's the best hypothesis that the current state of theory and evidence can come up with. Looking back on it, I don't think that in my entire life I've ever used an ANOVA in a published paper.
- Badges
- Science topic
- Similar topics
- Developmental Psychology
- Attachment