Hi,
I wish to assess blood dietary parameters of lactating mothers at three different times on the same respondents (at one month, three months and six months after delivery). The study has no intervention.
1. What is the best research design to use? Is it a longitudinal study, cohort? etc. (Kindly note, no intervention or controls)?
2. Which method/formula would I use to establish an adequate sample size for such a design with one arm? Most of the formulas are using two arms (intervention and controls or Exposed/Non-exposed)
Any assistance highly appreciated
Regards
I would suggest to adopt a two-level (measurement / mother) or even three-level (measurement type / measurement occasion / mother) design, and a mixed-effect (hierarchical) model to analyse it, in particular if you're interested in splitting the variance into these different levels.
Regarding the sample size you have to decide which model you're going to use, and which difference you want to test across the repeated design (growth in its broader sense). You will also need estimates of the variance at each design level. There is a nice presentation at the beginning of
Diggle, P. J.; Heagerty, P.; Liang, K.-Y. & Zeger, S.
Analysis of longitudinal data
Oxford University Press, 2002, 25, 379 p.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size