I stratified a dataset based on a categorical variable to get 10 strata. The purpose of the stratification is to increase observation homogeneity and remove potentially confounding unobserved effects. The observations are independent. Within each strata, I fit a weighted multiple regression with the same outcome and dependent variables. I am interested in meta-analyzing the coefficients for one of the dependent variables. Specifically, I want an estimate of the coefficient across all strata and the corresponding confidence interval and p-value. Any methods which are easily implementable in R would be appreciated.
I am confused with your "a categorical variable to get 10 strata", is it a continous variable and catergorized? How is about the regression over all range of data? Could you please show the figure of that regression?
Hi Huy, the categorical variable is states of the U.S. My unit of observation are counties within states. For each county, I have Yi (the outcome) and several predictors (let's label them X1 through X5). We can call the state variable, Xs. I am interested in B1 - the regression coefficient for X1 - and stratified the analysis by state for greater observation homogeneity. The attached figure shows the estimate and 95% confidence interval of B1 for each state-specific model. Given that I would like to fit separate models for each state, what options do I have to meta-analyze B1?
I think you can consider each state as an individual study and perform a meta-analysis of B1. Based on your figure, I think the data is highly heterogeneity and you need to use to random model. The pooled B1 would be a good fit.
Hope it helps
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics