PAT & PPI results

Replies to This Discussion

Permalink Reply by Ole Dahl Rasmussen on April 7, 2011 at 4:43pm

I did. Here are info:
1) In Northern Malawi, 1700 households
2) As a first step in measuring impact
3) Compare with others in Malawi
4) Since the tools were made using data for all of Malawi, I'm not convinced of their applicability for my area. Prevalence is high as expected.

• ▶ Reply

Permalink Reply by POWG Facilitators on April 8, 2011 at 6:04am

Ole - Did you use the PAT or PPI? Why did you choose one over the other?  I spent time with PAT implementers in Malawi last summer and could try to get you in touch with them.  I also know at least 1 other PPI implementer there... I believe he had the same issue with regional specificity.

• ▶ Reply

Permalink Reply by Ole Dahl Rasmussen on April 8, 2011 at 12:26pm

I actually included both PPI and PAT as a part of a larger survey which also has a consumption module. I chose both of them to be sure to at least have something reliable. The consumption modules seems to turn out OK, so I can do some comparison. For my purpose, though, PAT is the better of the two since it actually predicts a level of consumption. Great if you could get in tought with Malawi PAT-interested folks.

• ▶ Reply

Permalink Reply by Muhammad Awais on April 23, 2011 at 10:14pm

1. PPI in 9 countries in Asia (Pakistan, India, Bangladesh, Nepal, Cambodia, Vietnam, Philippines, Indonesia and Sri Lanka)
2. Donor reporting, analysis of MFI targetting, product developments, redesigning expansion plans, etc
3. Increase transparency and accountability
4. Results are fine.

Awais

• ▶ Reply

Permalink Reply by Muhammad Awais on April 23, 2011 at 10:17pm

Ole: if the tool is developed using all Malawi data means data of your
area was also included. Where you see the disconnect?

 

Awais

Ole Dahl Rasmussen said:

I did. Here are info:
1) In Northern Malawi, 1700 households
2) As a first step in measuring impact
3) Compare with others in Malawi
4) Since the tools were made using data for all of Malawi, I'm not convinced of their applicability for my area. Prevalence is high as expected.

• ▶ Reply

Permalink Reply by POWG Facilitators on April 25, 2011 at 6:43am

Ole -- I'm working on getting in touch with these organizations -- the Project Directors I worked with have all (coincidentally) left in the last year so I need to find new contacts! 

Awais -- What guided your decision to use PPI over PAT?

Awais & Ole -- The level of regional specificity is something I've heard a number of people bring up as an issue with these tools.  In your experience, Ole, does the problem lay with not having faith in the results or in the survey tool itself (ie it's difficult to deliver in some regions because many of the questions seem irrelevant in that area)? Or both?  Or neither?

PAT includes regional and Urban/Rural indicators to add two levels of specificity... some people think this is sufficient (or at least helpful), some do not.  The biggest challenge to making tools that are region-specific is of course cost.  

Question 1: Is having a regional and/or rural/urban indicator sufficient?

Question 2: Does it even matter?  For example, would a PPI be 'better' with a regional indicator?

Question 3: Do you see any other solutions to improving how these tools handle regional differences?

• ▶ Reply

Permalink Reply by Muhammad Awais on May 10, 2011 at 6:05am

Simplicity of PPI, benchmarking against multiple poverty lines and great consistent technical support provided by Grameen PPI team and Mark Schreiner (very important) lead to our decision to use PPI.

 

In my opinion, there will always be some people who want to have very specific tool and regional specificity will not satisfy them. I think the regional specifictaion should significantly increase the accuracy of the results to make a case.

 

To handle regional differences some MFIs have developed gateway questions to improve targetting, inclusion or exclusion in terms of regional specificities.

 

Awais      

POWG Facilitators said:

Ole -- I'm working on getting in touch with these organizations -- the Project Directors I worked with have all (coincidentally) left in the last year so I need to find new contacts! 

Awais -- What guided your decision to use PPI over PAT?

Awais & Ole -- The level of regional specificity is something I've heard a number of people bring up as an issue with these tools.  In your experience, Ole, does the problem lay with not having faith in the results or in the survey tool itself (ie it's difficult to deliver in some regions because many of the questions seem irrelevant in that area)? Or both?  Or neither?

PAT includes regional and Urban/Rural indicators to add two levels of specificity... some people think this is sufficient (or at least helpful), some do not.  The biggest challenge to making tools that are region-specific is of course cost.  

Question 1: Is having a regional and/or rural/urban indicator sufficient?

Question 2: Does it even matter?  For example, would a PPI be 'better' with a regional indicator?

Question 3: Do you see any other solutions to improving how these tools handle regional differences?

• ▶ Reply

Permalink Reply by Brendan Janet on August 30, 2011 at 1:51pm

Greetings!

1. I used both the PPI and PAT in El Salvador and Guatemala.  I also used a formatted version of the Multidimensional Poverty Index (MPI).

2. I have used the results to compare 'single-dimension poverty' with 'multi-dimensional poverty'.  I also conducted an accuracy analysis of the PPI and PAT for El Salvador, using the EHPM 2008.  I would be happy to share my results with you when the report is finished. (I believe we have shared a few emails already, Margaret).

3. If the PAT were updated in Guatemala, I could compare the 'point-in-time' accuracy with the PPI.

4. The PPI and PAT produce very similar estimates of poverty.  As expected, the MPI produced a very different (smaller) estimate of poverty.  Also, the actual households that were estimated as poor by the PAT and PPI, were not the same as those estimated as poor by the MPI.  A bit of concern for organizations that want to paint a multifaceted view of poverty.

• ▶ Reply

▶ Reply to Discussion