Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Deepankar Basu <dbasu@econs.umass.edu> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: national sample survey data (India) |
Date | Fri, 23 Jul 2010 17:04:26 -0400 |
Thanks Abhimanyu for your message.I did go through the documentation to understand the layout. Using that I prepared a dictionary file. Then I used the dictionary file to read in the data. I appended the 8 data sets to get the master data set.
Using the information provided in the documentation I also generated the weights.
To make sure I had read the data correctly I then tried replicating some results given in the NSS reports. For instance, I tabulated the distribution of MPCE (monthly per capita expenditure) at the all-India level and compared the mean of the distribution with the corresponding figure for average MPCE that appears in Table P5 (Report 508, NSS). My figures are way off!
That is what indicated that I must be either reading in the data incorrectly or not specifying the sample design properly.
Deepankar Quoting Abhimanyu Arora <abhimanyu.arora1987@gmail.com>:
As far as I am aware of, NSS data is provided in notepad (.txt format), (around 8 in number) that cover the whole of India. I think there is an excel file provided with the data as well (apart from survey documentation) which gives details on how the data is organized (blocks and levels). The first step is to use the information in the excel file to create a dictionary file for loading the .txt data into stata (later one can append/merge the 8 datasets so obtained).On Fri, Jul 23, 2010 at 6:43 PM, Deepankar Basu <dbasu@econs.umass.edu> wrote:Hi Stata Users, I am using Stata 10 and trying to read in unit-level data from the 61st Round of National Sample Survey (Consumption Expenditure). I am having trouble converting the information about sample design provided in the documentation into the "svyset" command for Stata. I was wondering if anyone has used this data set earlier and could inform me how she/he provided the survey details to Stata. Thanks in advance. Deepankar * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/* * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/
* * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/