Multiple Visits and Data Quality in Household Surveys

IZA Logo
   

IZA Seminar

Place: Schaumburg-Lippe-Str. 9, 53113 Bonn

Date: 19.11.2013, 12:15 - 13:30

   

Presentation by 

Matthias Schündeln (Goethe University Frankfurt)
   

Abstract:

We study the question how repeated visits to the same household affect data quality in household surveys. In order to increase data quality, especially with respect to frequently purchased consumption items, some household surveys visit the respondent households several times. For example, in the Ghanaian household survey that we analyze, households are visited up to 11 times over a period of one month. In the Ghanaian data the level of the estimated annual consumption and, consequently, poverty and inequality measures are a function of the number of visits, with consumption decreasing over survey rounds. To study whether early or later survey rounds are of higher quality, we employ a strategy based on Benford's law. Preliminary results suggest that the consumption data from earlier survey visits are of higher quality than data from later visits. This effect is most severe for consumption items that have on average a higher reported value and for items that are towards the end of the questionnaire.

   
   
For more information, please contact seminar@iza.org