Which method can I use to determine the trends in vaccination coverage of the six vaccines for a country such as Nigeria, using its 2006 and 2013 DHS datasets?

How can i build the multivariable logistic regression model to determine the factors associated with complete vaccination in nigeria using the 2013 dataset?

How can i determine the risk factors for Under- or non-immunization using the same dataset?

I am using stata for my analysis and I am looking at vaccination coverage in children aged 12-23 months.

]]>

Any help on the topic? It is urgent please.

Regards,

JK]]>

These questions are a bit outside the scope of this forum, but I'll try to help a little bit:

Which method can I use to determine the trends in vaccination coverage of the six vaccines for a country such as Nigeria, using its 2006 and 2013 DHS datasets?

The most obvious thing to do is just calculate the vaccination rates for each vaccine separately by each survey, and then subtract them. To do this in a single model, you could append the datasets together, and (for each vaccine separately) regress "having the vaccine (0/1 indicator variable)" on an indicator variable for "2013" and a constant. The coefficient on the 2013 dummy variable is the difference in coverage. You'd have to re-normalize your sampling weights to do this , and there is much guidance on that in forum. The simplest way, in this case, would be to just divide the original weight by the sum-of-weights within each survey (so weights in each round sum up to 1). That puts equal weight on each survey and preserves probability weighting.

How can i build the multivariable logistic regression model to determine the factors associated with complete vaccination in nigeria using the 2013 dataset?

This is basically your job as researcher. But in general, generate an indicator variable for "complete vaccination" and include as covariates all of the predictors you are interested in. See Stata documentation on logit (http://www.stata.com/manuals13/rlogit.pdf). It is not clear, though, why you want a "hierarchical' model here when using only 1 dataset. You can always include aggregate predictors at differently levels if you want. It is not clear what you want to accomplish with this, so I can't help much.

How can i determine the risk factors for Under- or non-immunization using the same dataset?

This is a research question and not a DHS question. But see above.

I MIGHT be able to help a bit more if you give more details - but most of these questions sound like substantive scientific questions, not DHS questions, and this is not the proper forum for that.

]]>