--- "Frederick \"FN\" Noronha" <[EMAIL PROTECTED]> wrote: > >In our study we found that children with high birth >order, Muslim religion, those residing in rural areas, >children with low parent education and socioeconomic >status and those from high household size had >significantly low immunization coverage levels >compared to children from other groups (P >0.05). Also >a trend analysis showed improvement in immunization >status with improvement in parent education, >socio-economic status and decreasing family size. >
The above conclusions are at best premature. The variables assessed are not at all independent of each other. In other words, for example, there may be significant differences in the socioeconomic status, parent educational level, household size, etc. between the Muslims, Christians and Hindus sampled in this study. In this case, the low immunization coverage may be explained by one or more of these other factors, and not the Muslim religion per se. The same rationale can be applied to each of the other variables. The study, as presented, is flawed. Either the whole study needs to be redone, or if the requisite data are available, they need to be properly re-analyzed to control for or assess the inter-dependencies or interactions between the different variables. Short of that, very limited meaningful conclusions can be drawn from this study. Cheers, Santosh _______________________________________________ Goanet mailing list Goanet@lists.goanet.org http://lists.goanet.org/listinfo.cgi/goanet-goanet.org