Thanks.

Michael]]>

I am trying to associate variables about mother and father to each child in primary school age who attended (or not attended) school, in this case for the latest Tanzania standard DHS.

I have taken info on school attendance from the PR file, and to find some mother and father variables I have done a merge with the BR file, by doing the following in BR:

rename v001 hv001

rename v002 hv002

rename b16 hvidx

sort hv001 hv002 hvidx

and then do the merge. The primary file i am working on is PR.

After the merge, a very high number of cases where i have valid info for children school attendance variable, do not seem to have information about mother and father, so i was wondering if i was doing something wrong, or if these info could be taken from some other files. If so, which files should I use? Can i please also have an example of how the merge should be done if using different files?

thank you! ]]>

I am working on Colombia DHS 2015 data and need some help regarding the following issue:

I hava data where my unit of analysis are children that contains child´s information such as age, gender, number of sibling, birth order. I want to have parental information(education, occupation of mother and father/partner and household characteristics( place of residence, years living in the same place etc)

What I am doing is:

use "COPR72FL.DTA", clear

keep.....

sort hv001 hv002 hvidx

save dhsPR.dta, replace

clear

use "COBR72FL.DTA", clear

keep...

keep if b16>0 & b16<19

gen line_number_of_mother=v003

ren v001 hv001

ren v002 hv002

ren b16 hvidx

sort hv001 hv002 hvidx

save dhsBR.dta, replace

clear

use dhsPR.dta

merge 1:m hv001 hv002 hvidx using dhsBR.dta

***the lines above I guess link mothers with children in the same household****

However, I also need to link the father/partner information to this dataset.

Could anybody help me pleas? Any ideas? Am I doing something wrong?

I will appreciate your help.

Thanks

]]>

Please send me the excel sheet for date conversion in my email umeth.th@gmail.com

Thanks!]]>

The questionnaire is attached for convenience.

Thanks,

Kripa]]>

For the variables labeled "Consent for additional tests", does anyone know what other tests this variable refers to. I cannot find any additional information on this in the recode manuals.

Thank you,

Brendan]]>

Is "weight_all = v005/1000000(Total number of households during each survey year/sample households in each survey)" a verbal description of a Stata command? If so, what command? It's not clear what you are doing to the weights, other than dividing by 1000000, which will not affect the results at all.

The "Population size" in the results has been distorted by svyset, particularly by the components other than the weight. You can ignore it.

]]>

I'm a PhD student and I'm currently working on the DHS dataset of 16 sub-Saharan African countries.

In regard, I would like to ask the following:

1) Could I please have the alphas and betas for these surveys:

Liberia, 2013

Côte d'Ivoire, 2005

Lesotho, 2014

Zimbabwe, 2015

Rwanda, 2014-15

Malawi, 2015-16

Tanzania, 2011-12

Ethiopia, 2016

2) In computing for the Comparative Wealth Index, do I need to divide v191 by 100,000 first?

Thank you in advance!!

Pearl]]>

I am trying to compute the Minimum Meal Frequency from the Indian DHS 2015-16, following the commands given by Nurnabi and Trevor's instructions. My computed percentage (28.66%) is not matching with the reported percentage (35.9%) in the DHS-4 report. However, the sample size (69,690) is matching,

Could you anyone point out my mistake in the commands (below)? Or if possible, share the correct syntax for Minimum Meal Frequency and Minimum Acceptable diet, please.

I am using the following commands:

* Last child in the last 2 years living with mother

* age in months

gen age = v008-b3

* drop if too old or not alive

keep if age<24 & b5==1

* keep only those children living with mother ...

keep if b9==0

* ... and keep the last born of those

drop if _n > 0 & caseid == caseid[_n-1]

drop if age<6

*Minimum meal frequency

tab m39, m

*Minimum meal frequency for breastfeed child age 6-8 months, 9-23 months

recode m39 (2/7=1 "2-7")(else=0), gen(mm1)

recode m39 (3/7=1 "3-7")(else=0), gen(mm2)

*Minimum meal frequency for non-breastfeed child age 6-23 months

recode m39 (4/7=1 "4-7")(else=0), gen(mm3)

*Age group

recode age (6/8=1 "6-8")(9/23=2 "9-23"), gen (age_g)

*Breastfeeding status

gen still_feeding=1 if m4==95

replace still_feeding=0 if still_feeding==.

*Minimum meal frequency applying all conditions

gen mmf1=1 if mm1==1 & age_g==1 & still_feeding==1

gen mmf2=1 if mm2==1 & age_g==2 & still_feeding==1

gen mmf3=1 if mm3==1 & still_feeding==0

*Weight varible

gen wgt=v005/1000000

* Minimum meal frequency for all children (6=23 months)

gen MMF=0

replace MMF=1 if mmf1==1 | mmf2==1 | mmf3==1

tab MMF [iw=wgt]

Thank you,

Sk]]>

With don't know as part of the denominator, the procedure I used is as follows,

. g iw=v005/1000000

. g h22_f=h22==1

. tab h22_f [iw=iw] if b5==1 & b19<60 with give 33.28% of fever as reported in the DHS report

With don't know considered as missing; the fever percentage is 33.96%

I would like to know which procedure I should follow.]]>

I'm interested in using the modeled surfaces that DHS makes available through the Spatial Data Repository. I was wondering, if an indicator has been modeled for one country but not for another, does that mean that you tried to model it but it didn't produce good results, that you haven't modeled it yet but will eventually, or that that particular indicator/country combination is not a priority? Are we allowed to request that you model particular indicators in particular countries? Or can we request the code used to produce one surface in a particular country so that we can use it with a different survey/country?

Thanks,

Josh]]>

I want to do a pooled analysis of BDHS 2007, 2011 and 2014 (KR file for children). As per forum discussion, during regression analysis of pooled data, I need to de-nomalize the sampling weight. I did this using

gen wgt = weight_all *** weight_all = v005/1000000(Total number of households during each survey year/sample households in each survey) and append three surveys data gen psu = cluster *** each survey clusters are unique eg. 2007_1, 2011_1 and so on svyset psu, weight(wgt) strata(strat) , singleunit(centered) || _n *** each survey strata are unique

When I fitted weighted logistic regression, after adjusting weight this way, I found:

svy: logit y x Number of strata = 63 Number of obs = 19,896 Number of PSUs = 1,561 Population size = 44,882,311

Could anyone please suggest that the process is correct? Is the population size reliable or not?

Thank you very much. ]]>

More specifically, variables from questions 502A (reason for the abortion) to 577 are not in the dataset made available on the DHS website. The variables on abortion in my dataset begin with q577_1 (ever heard of abortion) and end at q582_2 (conditions of legal abortion in Ghana) while all the variables on miscarriage are missing.

I would like to request for the full dataset. Please let me know how to proceed with this. Thanks for your help.]]>