13.07.2015 Views

Human Development in India - NCAER

Human Development in India - NCAER

Human Development in India - NCAER

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

216 human development <strong>in</strong> <strong>in</strong>diaFigure AI.3 <strong>India</strong> <strong>Human</strong> <strong>Development</strong> Survey 2005, District Coverage—Urban SampleSource: IHDS 2004–5.Survey. 2 However, given the potential for significant sampl<strong>in</strong>gand non-sampl<strong>in</strong>g errors, we present a detailed analysisof the quality of IHDS data below.COMPARATIVE RESULTSIHDS was not <strong>in</strong>tended to provide national nor, certa<strong>in</strong>ly,statewise estimates of levels of human development outcomes.There are already many excellent <strong>India</strong>n surveys that fill thatmission. The ma<strong>in</strong> purpose of IHDS is to provide a meansfor ga<strong>in</strong><strong>in</strong>g <strong>in</strong>sight by analys<strong>in</strong>g the relationships amongthese human development outcomes and the connectionsbetween human development and its background causes.Nevertheless, it is useful to compare IHDS estimates ofhuman development levels with estimates from other morenarrowly focused surveys that usually have larger samplesizes and smaller sampl<strong>in</strong>g errors. The NSSs and the NFHSsare obvious comparisons because of their excellent qualityand wide use. The <strong>India</strong>n Census provides another usefulreference. The Census and these surveys differ not only <strong>in</strong> theirobjectives and design, but their question word<strong>in</strong>g, sampl<strong>in</strong>gdesign, cod<strong>in</strong>g decisions, and government sponsorship, allof which should be expected to provoke somewhat differentanswers from respondents, and yield different frequencies(Table AI.3).2Leslie Kish and Alastair Scott were the first to describe the probability sampl<strong>in</strong>g procedures which are designed to optimize the reselection orretention of sample units dur<strong>in</strong>g a transition from an old to a new sample design. A description of this can be found <strong>in</strong> ’Reta<strong>in</strong><strong>in</strong>g units after chang<strong>in</strong>gstrata and probabilities’, <strong>in</strong> the Journal of the American Statistical Association, Vol. 667, Number 335, Applications Section, September 1971.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!