The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Fertility » About Binary logistic regression Analysis
About Binary logistic regression Analysis Mon, 06 February 2017 00:40
 Mohammad Nazmul Hoq Messages: 4Registered: May 2016 Location: Chitagong Member
Hello
I am working on BDHS-2011 data set. My topic is fertility analysis. For this I consider Number of living children as dependent variable. There are so many independent variables like as, respondents education, wealth index, husbands education, access to mass media etc. One of my analysis is binary logistic regression analysis. For this I prepare the data set in different parity (Parity one(1) indicates those respondents having one child, parity two(2) indicates those having two child's and so on). My question is whether the analysis is correct if I consider in binary logistic analysis (in parity 1) o for those having no child and 1 those having one child. Again in Parity 2, 0 for those having no child and 1 those having two child. Similarly for other parity.
Thank you

Md. Nazmul Hoq
Re: About Binary logistic regression Analysis [message #11745 is a reply to message #11738] Mon, 06 February 2017 09:52
 Liz-DHS Messages: 1516Registered: February 2013 Senior Member
A response from Dr. Tom Pullum,
Quote:

This is potentially a more difficult question than you may realize. There are several different possibilities. You could perhaps construct a set of binary variables Yk, for k=0,1,2,3,4..... as follows. Yk=0 for women with k children; Yk=1 for women with more than k children; Yk="." for women with fewer than k children. You would then be describing the probability of going from parity k to parity k+1. This would be better for number of children ever born than for number of living children. Logit regression is not well suited for an outcome that is a count. Another possibility would be to use poisson regression or negative binomial regression. There is not a general agreement on a single way to analyze this outcome.
Re: About Binary logistic regression Analysis [message #11746 is a reply to message #11745] Mon, 06 February 2017 11:28
 Mohammad Nazmul Hoq Messages: 4Registered: May 2016 Location: Chitagong Member
Dear sir
Sir i want to make the binary logistic analysis on fertility like the article of following link.

The Desire for Sons and Excess Fertility: A Household-Level Analysis of Parity Progression in India
Author(s): Sanjukta Chaudhuri
Source: International Perspectives on Sexual and Reproductive Health,Vol. 38, No. 4 (DECEMBER 2012), pp. 178-186

Also, Estimated odds ratios from binary logistic analysis (Table 6 and page-183)
What the author consider dependent variable in binary logistic analysis? What the author consider 0 and 1 in binary logistic analysis in each parity?

Sir, I am not clear the dependent variable consider at this article in fertility analysis. (((Page-183)))

Best Regards

Md. Nazmul Hoq
Re: About Binary logistic regression Analysis [message #11749 is a reply to message #11746] Mon, 06 February 2017 13:37
 Liz-DHS Messages: 1516Registered: February 2013 Senior Member
Dr. Tom Pullum responded:
Quote:

I hope another user of the forum can help you with this question, or you can directly contact the author, Sanjukta Chaudhuri, whose contact information must be provided in the article.

Re: About Binary logistic regression Analysis [message #11757 is a reply to message #11746] Mon, 06 February 2017 22:38
 Mohammad Nazmul Hoq Messages: 4Registered: May 2016 Location: Chitagong Member
Thanks sir
I sent email several times to the author Sanjukta Chaudhuri but she did not reply me.
Sir, Can you give me some suggestive analysis on the title given below:
"The Effects of Son Preference and Gender Composition of Surviving Children on Fertility in Bangladesh: Regional Differentials"

Data: BDHS-2011
I want to do binary logistic regression analysis on the above title. For the analysis of gender composition I construct different parity ( Parity-1 (those have one living children), parity-2 ( those have two living children) and so on ).

Sir, Is the binary logistic analysis appropriate for identifying different demographical factors in each parity ? If so, What should i consider 'o' and '1' in the dependent variable number of living children (as title is on surviving children)in each parity?

Sir, can you give me some article reference on the above title?

Sir, I will be grateful to you.
Regards

Md. Nazmul Hoq
Re: About Binary logistic regression Analysis [message #11766 is a reply to message #11757] Tue, 07 February 2017 08:53
 Liz-DHS Messages: 1516Registered: February 2013 Senior Member
Dear User,
We cannot assist you with this type of question. Perhaps someone else on the forum can offer some guidance. Thank you!
Re: About Binary logistic regression Analysis [message #14870 is a reply to message #11766] Thu, 10 May 2018 10:18
 Hassen Messages: 121Registered: April 2018 Location: Ethiopia,Africa Senior Member
Dear Dr.Tom Pullum,Liz DHS and Trevor-DHS Thank you very much. I miss you!!
With Best Wishes,Hassen

Hassen Ali(Chief Public Health Professional Specialist)
 Previous Topic: Son Preference and Infertility Next Topic: Factors affecting the duration of breastfeeding
Goto Forum:

Current Time: Mon Aug 15 11:05:33 Coordinated Universal Time 2022