Hi, I'm very new to SPSS and stats so any help would be appreciated.
I am looking at the impact of deprivation on healthcare costs, so the dependent variable is cost, and the main independent variable is deprivation score.
I have cost data for 300,000 individuals, and because it is highly non-normal, I am using a generalized linear model for the regression.
The deprivation score for each of these individuals is based on the neighbourhood they live in. Each of the 900 neighbourhoods has a different deprivation score.
I want to perform regression (using individuals as the unit of analysis) that accounts for the fact that individuals from the same neighbourhood may have more similar costs. I have been shown in STATA how to add a cluster command: e.g. cluster(neighbourhood) to give robust standard errors that account for clustering based on another variable, in this case, neighbourhood.
Is there a way to do this in SPSS? or an alternative way of accounting for clustering?