The DHS Program User Forum: Mortality » accounting clustering effects of women's data when using baby-based analysis

Home » Topics » Mortality » accounting clustering effects of women's data when using baby-based analysis

Show: Today's Messages :: Show Polls :: Message Navigator

Re: accounting clustering effects of women's data when using baby-based analysis [message #795 is a reply to message #794]

Sat, 28 September 2013 06:22

rkinoshita
Messages: 6
Registered: September 2013
Location: Nicaragua

Member

hi, and if you could look at below outputs from STATA and let me know what you think. When I used vce option, the results for vce (robust) and vce (cluster...) are very different. do you know why?
thanks

. regress IPVlife neonatal, vce (cluster HHCLUST)

Linear regression Number of obs = 32115
F( 1, 730) = 2.54
Prob > F = 0.1117
R-squared = 0.0001
Root MSE = .42334

(Std. Err. adjusted for 731 clusters in HHCLUST)
------------------------------------------------------------ ------------------
| Robust
IPVlife | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+---------------------------------------------- ------------------
neonatal | .0337656 .0212035 1.59 0.112 -.0078615 .0753928
_cons | 1.699111 .0427602 39.74 0.000 1.615164 1.783059
------------------------------------------------------------ ------------------

regress IPVlife neonatal, vce(robust)

Linear regression Number of obs = 32115
F( 1, 32113) = 3.34
Prob > F = 0.0674
R-squared = 0.0001
Root MSE = .42334

------------------------------------------------------------ ------------------
| Robust
IPVlife | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+---------------------------------------------- ------------------
neonatal | .0337656 .0184639 1.83 0.067 -.0024243 .0699556
_cons | 1.699111 .0366966 46.30 0.000 1.627184 1.771038
------------------------------------------------------------ ------------------

Report message to a moderator

[Message index]

		accounting clustering effects of women's data when using baby-based analysis By: rkinoshita on Fri, 27 September 2013 08:42
		Re: accounting clustering effects of women's data when using baby-based analysis By: Reduced-For(u)m on Fri, 27 September 2013 14:51
		Re: accounting clustering effects of women's data when using baby-based analysis By: rkinoshita on Sat, 28 September 2013 05:43
		Re: accounting clustering effects of women's data when using baby-based analysis By: rkinoshita on Sat, 28 September 2013 06:22
		Re: accounting clustering effects of women's data when using baby-based analysis By: Reduced-For(u)m on Sat, 28 September 2013 17:52
		Re: accounting clustering effects of women's data when using baby-based analysis By: rkinoshita on Sun, 29 September 2013 10:21
		Re: accounting clustering effects of women's data when using baby-based analysis By: Reduced-For(u)m on Sat, 28 September 2013 17:40
		Re: accounting clustering effects of women's data when using baby-based analysis By: rkinoshita on Sun, 29 September 2013 10:27
		Re: accounting clustering effects of women's data when using baby-based analysis By: Liz-DHS on Mon, 27 January 2014 17:24

Previous Topic:	Preceding Birth Inteval and child mortality
Next Topic:	SPSS file for U5 moratlity rate

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Nov 5 06:10:05 Coordinated Universal Time 2025