The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » Coefficients are large in multilevel analysis
Coefficients are large in multilevel analysis [message #16676] Fri, 15 February 2019 05:40
dga1n@soton.ac.uk is currently offline  dga1n@soton.ac.uk
Messages: 9
Registered: March 2018
Location: United Kingdom
Member

Dear DHS specialists,

When I have using a multilevel linear regression models to analyse birth weight (continuous), I am getting large coefficients for my results (see below) This can be seen even after applying for weights.Is this fine? or is there any way I can get change this? I checked the multicollinearity and also are fine..


b_interval |
First birth | 0 (base)
<24 | 115.4945 29.0165 3.98 0.000 58.59253 172.3965
24-47 | 137.8109 15.69198 8.78 0.000 107.0386 168.5832
48-59 | 122.9485 22.0587 5.57 0.000 79.69093 166.2061
60+ | 61.07513 14.89308 4.10 0.000 31.86946 90.2808
|

wealth index |
Lowest | 0 (base)
Secondary | 36.35094 19.18614 1.89 0.058 -1.273514 73.97539
Middle | 42.6571 20.92505 2.04 0.042 1.622606 83.69159
Fourth | 68.12738 21.27426 3.20 0.001 26.40808 109.8467
Highest | 103.369 25.5519 4.05 0.000 53.26114 153.4768
|
residence |
Urban | 0 (base)
Rural | -11.62815 17.74448 -0.66 0.512 -46.42547 23.16918
Estate | -74.43972 36.53593 -2.04 0.042 -146.0875 -2.791947
|


Could you kindly please explain this?

Thanks


gayathri

[Updated on: Fri, 15 February 2019 05:41]

Report message to a moderator

Previous Topic: Merging Household file to children file
Next Topic: Haiti treatment sources: more detail on country-specific labels?
Goto Forum:
  


Current Time: Tue Jan 19 09:54:24 Coordinated Universal Time 2021