The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in SPSS » Variables on early initiation of breastfeeding (m34 and v426)
Variables on early initiation of breastfeeding (m34 and v426) [message #9827] Thu, 26 May 2016 18:42 Go to next message
WoodyWoodruff is currently offline  WoodyWoodruff
Messages: 5
Registered: September 2013
Location: Victoria, British Columbi...
I am having trouble understanding the relationship between variables m34 and v426, both of which contain the answers to a single question posed to women respondents about how soon after delivery breastfeeding was initiated. I am analyzing the child recode datasets from the 1999, 2005, and 2012 Guinea DHS. According to an answer received on this forum some time ago, v426 contains the response for only the most recent birth, while m34 contains the responses for all births for which this question was asked.

Therefore, it is understandable that in the 1999 dataset, there are many children with a value for m34, but missing for v426. These children are the penultimate or earlier births for responding women. However, in the 2012 DHS, this question was only asked about the most recent birth. Why then are there many children with a value for v426 but no value for m34? Shouldn't the values in these two variable be the same for all children?

The situation in the 2005 dataset is even more confusing. In this survey, the question was asked about the three most recent births, but the dataset contains: a) children with a value for m34 but missing for v426, b) children with missing for m34 but a value for v426, c) children with the same non-missing value for m34 and v426, and d) children with different non-missing values for m34 and v426. How should I determine which children in this dataset actually initiated breastfeeding within 1 hour of delivery?
Re: Variables on early initiation of breastfeeding (m34 and v426) [message #9980 is a reply to message #9827] Mon, 13 June 2016 13:53 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 799
Registered: January 2013
Senior Member
Variables m34 and v426 provide the same information, but from two different perspectives. m34 is from the child's perspective and v426 is from the woman's perspective and only about her last birth. v426 relates only to the last birth and when the woman put the last child to the breast. m34 and v426 will be the same for the last child (that is when midx==1)

Thus for the last birth, you can use either m34 or v426 as they are the same, but for earlier births, only use m34 as this relates to the child. Don't use v426 as this is a woman level variable about the last birth that is copied to each child's record for convenience (or, as it seems in this case, confusion).

Thus you will find some cases where m34 is missing (typically because the child was never breastfed), but v426 is not missing (because it refers to a different child), and vice versa.
Re: Variables on early initiation of breastfeeding (m34 and v426) [message #14016 is a reply to message #9980] Sat, 03 February 2018 03:01 Go to previous messageGo to next message
Radhika_DHS is currently offline  Radhika_DHS
Messages: 4
Registered: January 2018
Location: India
I am computing early initiation of breastfeeding for India and its states using NFHS data set, but my values are not matching with table number 10.5 given in NFHS4 India's report. I have used variable m4 to compute ever breastfed women and the from variable v426 initiation within 1 hr was computed, but after the crosstab, with states, my values are coming a bit different from the report. I have mentioned commands below, looking for your input.

***generating ever breastfed and initiation of breastfeeding variable)
recode m4 (98 94 .=0 "never_bf") (else=1 "breastfeeding"), gen(ever_BF)
ta ever_BF
tab1 v426, nol
recode v426 (0=1 "within 1 hour") (else=0 "more than 1 hour"), gen(initiation_bf)
ta initiation_bf

***cross tab with states (v024)
ta v024 Initiation_bf [iweight=v005/1000000] if ever_BF==1 , r nofr
ta v024 Initiation_bf if ever_BF==1 , r

Re: Variables on early initiation of breastfeeding (m34 and v426) [message #14067 is a reply to message #14016] Fri, 09 February 2018 07:11 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3087
Registered: February 2013
Senior Member
Following is a response from Senior Data Processing Specialist, Mianmian Yu:

Table 10.5 in the NFHS-4 Final Report has used M4 <> 94 for percentage of ever breastfed ( Column 1) and M34 with a value between 0 and 100 for percentage of starting breastfeeding within one hour of birth (Column 3).

Hope this helps.

Re: Variables on early initiation of breastfeeding (m34 and v426) [message #21877 is a reply to message #9980] Mon, 04 January 2021 19:47 Go to previous message
Shekhar_GS is currently offline  Shekhar_GS
Messages: 3
Registered: September 2020
Hi Trevor,

I am looking the frequencies of both the variables (M34, and V426) in IAKR74FL (India, NFHS-4) but they seem to be very different. V426 exclusively says that its for 'last born' but M34 doesn't say so. So I am assuming that M34 would be including all the children under 3/5 years, but the frequencies show that for M34 only 1,79,855 are available, while for V426 2,45,112 entries are available.

So my query is - IF M34 includes all the children wherin MIDX == 1 then it should be equal to OR more than V426 but its not the case here.

In continuation to your answer, I need to understand the following

1. Does M34 include only the earlier born children (i.e. all children under 3/5 except for the last born child), OR does it include the last born too?
2. Is V426 only for the last born child?
3. If I use V426 for estimating the breast feeding practices for the last born, how should I justify using this particular variable in view of the large difference in both of them.

Thank you.
Previous Topic: Complex Samples
Next Topic: Issue with duplicate mothers
Goto Forum:

Current Time: Wed Jun 19 17:18:56 Coordinated Universal Time 2024