The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Other countries » Mali 5 Wealth Index classification errors?
Mali 5 Wealth Index classification errors? [message #3773] Mon, 09 February 2015 10:12 Go to next message
aligazan
Messages: 24
Registered: June 2014
Location: UK
Member
I am using Mali 5 dataset.

I was comparing variable V190 Wealth Index to variable V191 Wealth Index Factor Score, and it appears that there may have been some misclassification.

The range of V191 within V190 Poorest is -99968 to -71717 yet there are around 20 respondents with V191 between 100081 to 112373, which would be classified within Richer.

Please could you clarify whether this is correct? And if so, is the error in V190 (i.e. the miscoded variables should be recoded into the correct category within V190) or is the error in V191 (i.e. the miscoded variables are in the correct category for V190 but have an incorrect Wealth Index Factor Score)?

I would be interested to know also how the range of Poorest, Poor, Middle, Rich and Richest are categorised as there are not equal numbers in each category.

Many thanks

Re: Mali 5 Wealth Index classification errors? [message #3790 is a reply to message #3773] Wed, 11 February 2015 10:11 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 803
Registered: January 2013
Senior Member
Thanks for pointing this out. The values you see between 100081 to 112373 should all be negative values, i.e. -100081 to -112373. The problem is that the value was being written to a 6 character field and it chopped off the negative sign. Please just change all of these values to a negative value, e.g.
replace v191 = 0 - v191 if v191 > 100000 & v190==1

[Updated on: Wed, 11 February 2015 10:32]

Report message to a moderator

Re: Mali 5 Wealth Index classification errors? [message #3831 is a reply to message #3790] Fri, 20 February 2015 11:22 Go to previous messageGo to next message
aligazan
Messages: 24
Registered: June 2014
Location: UK
Member
Thanks Trevor

Just a further question, in SPSS 'REPLACE' is not a recognised command. Please could you give me syntax for this replacement that will work in SPSS, as I cannot seem to work it out myself?
Re: Mali 5 Wealth Index classification errors? [message #3832 is a reply to message #3773] Fri, 20 February 2015 15:30 Go to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 803
Registered: January 2013
Senior Member
if (v191 > 100000 & v190=1) v191 = 0 - v191
Previous Topic: Unique identifiers/duplicate entries in DHS Pakistan Data
Next Topic: Strata variables for Kenya 2008/9 - Rwanda 2011 - Ghana 2008
Goto Forum:
  


Current Time: Thu Nov 14 20:48:59 Coordinated Universal Time 2024