8 Concluding remarks

This paper has outlined the appropriate steps to import, prepare and analyze datasets from CLSA using \(\texttt{R}\), \(\texttt{SAS}\), \(\texttt{SPSS}\) and \(\texttt{Stata}\). The data manipulation and analysis codes described in the paper can be applied to other datasets from surveys with a sampling scheme similar to the CLSA. From the comparisons presented in the paper, we see that \(\texttt{R}\) provides accurate and reliable estimates and standard errors as an attractive statistical package.

This paper highlights the comparison of the codes between \(\texttt{R}\) and other commercial statistical packages for different statistical procedures. The \(\texttt{survey}\) packages in \(\texttt{R}\) provides most of the procedures in which health policy researchers are interested. Compared to other packages, \(\texttt{R}\) is easy to use, flexible and open-source. We recommend that health researchers choose \(\texttt{R}\) as one of the statistical packages for data analyses.

This paper can also serve as a guide for the health policy analysts who can check the corresponding codes for {preforming and replicating } various statistical analyses with different statistical packages. This paper would be part of the groundwork for health-care survey administration organizations to include instructions and sample \(\texttt{R}\) codes in their technical documentation.