< back to all forums
 
no image

PPI Response Discrepancies

Aaron Stroud-Romero
• OneWorld Health
• United States
• 12/23/12
• 1 Comment
 

I just finished working with my organization to conduct a midline survey which included the Uganda PPI for a VSLA project we are working on.  As I reviewed data from the baseline I saw some discrepancies concerning the question related to the education level of the female head of the household.  Numerous respondents listed lower educational levels than they had stated during our baseline survey.  Thus even though there are improvement in other areas it is adversly affecting the poverty rates that we are seeing.  Has anyone else experienced similar issues and/or have any recommendations for addressing this problem?  As our survey participants are spread throughout a large, hard to reach area, re-surveying to confirm correct answers at this point is unfortunately not a plausable option.

1 Comment
cris lomboy says (02/14/2013):
I think this is not a big problem since the indicator in question is pertaining to the education of the female head or espouse since this is not so much liable to change over time (unless the female head attends some adult literacy and equivalency programs). If you are comfortable with your own data about the education level of this person then just adjust the responses accordingly. In this way, you don't have to go back and re-survey. You can then adjust the points and scores. However, you have to make sure that the baseline and midline data are the same otherwise you are going to have significant jump in the scores and can be misleading since the change in the over-all score is because of the adjustment made on this particluar indicator.
 
 
< back to all forums