My research is about alzhimer's disease caregiver stress. I want to implement a 4 week long quasi- experimental study on one intervention group (no control) using pre-test post-test. I want to collect demographic data, Vital signs, salivary cortisol levels, perceived stress scale (PSS), and Kingston caregiver stress scale (KCSS). Then I will do the intervention (daily lavender aromatherapy via essential oil diffuser & 15 mins music therapy sessions for 4 weeks). Lastly, I will remeasure vital signs, salivary cortisol levels, PSS, and KCSS.
Are my variables correct?
Nominal: Marital status, number of household members, occupation, race, gender, and relationship to the care recipient.
Ratio: age, salivary cortisol level, blood pressure, heart rate, and respiratory rate.
Ordinal: duration of providing caregiving, Kingston Caregiver Stress Scale (KCSS) and A Global Measure of Perceived Stress scale (PSS).
Is my research design appropriate? What is the next step? DO I need null hypothesis?
All variables sound okay except relationship to care recipient because you do not know the question. Best design would be 3 groups: one with music, one with lavender, and one with nothing. The do a MANOVA and you see if music or lavender had an effect and you see if they interact on stress.
Your null hypothesis is that music or lavender have no effect.
The Influence of Music on the Surgical Task Performance
read our paper, I attach another one with pre-post design:
Béatrice Marianne Ewalds-Kvist Thank you so much for your help! I will check out your paper! thank you for sharing!
The theoretical framework I am using is Caregiver Stress middle range theory from Roy's adaptation model! the theory states the relationship to the caregiving recipient (spouse, parents, in-laws) may influence the stress levels. That is why I included it as a variable! do you think I should remove it?
I like the three group model! it give stronger level of evidence for sure! would be an RCT at that point since there are three groups and they're randomly selected for the study? or would it remain Quasi-experimental since there is no original control group with no intervention?
Yes, but the relationship to different persons does not sound like a nominal scale.
There should be a group with no intervention, this group I called a group with nothing. I did this many time with mice on the open field and one group just ambulated on the open field but did not get any intervention. Then we could see if the groups with interventions differed from the one group which did not get anything, e.g. was not subject to maze learning and so on. With MANOVA you see eventual interactions between lavender and music for example. You can also use
linear regression analysis where you see how much of stress, music explains or is reduced and how much lavender explains and so on.
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress