|
 |
Karnataka Assembly Election 2008
Methodological Details |
Karnataka Assembly Election Study 2008 used a mixed methodology of Post-Poll and Exit Poll. The survey was conducted in 75 out of 224 assembly constituencies in Karnataka. The post poll survey of 4104 respondents was conducted in constituencies, which went to poll in the first and the second phase, while the Exit poll of 4348 respondents was conducted in those sampled constituencies where elections were held during the last phase of election. All the 75 assembly constituencies were randomly selected using the probability Proportionate to Size (PPS) method. In each of the selected assembly constituencies, four polling booths were randomly selected using systematic random sampling method.
We preferred conducting Post poll survey in constituencies, which went to poll in the first two rounds of election, because it enables strict randomization of respondents, which is not possible in an Exit Poll situation. During the post poll survey, at each location, 30 respondents randomly selected from the updated electoral roll using the systematic random sampling technique were approached for interview. Post poll survey also enabled strict comparisons since these respondents were also interviewed during the pre-poll survey conducted a month earlier.
All the respondents were interviewed at their home using a structured interview schedule. The respondents were interviewed in face-to face interview situation in the language, which they spoke at home. To facilitate this, the interviewed schedule was translated in three languages, Kannad, Hindi and English. While majority of respondents were interviewed in the Kannad language, but there were few respondents especially in Bangalore city, who were interviewed using interview schedule in Hindi or English language.
In constituencies, which went to polls in the third phase, we conducted the Exit Poll at 92 polling booths.
A total of 150 field investigators, mostly students from different universities, carried out the hard work of data collection. All the field investigators were given training in survey research and data collection at workshops held in different regions of the state. A total of five workshops were held to train these field investigators. Three supervisors, Veena Devi, Javeed and Jayaprakash Mavinakulli, supervised the fieldwork for the survey. B.S Padmavati of International Academy for Creative Teaching coordinated the fieldwork for the survey. Sanjay Kumar of Centre for the Study of Developing Societies (CSDS) Delhi directed the survey.
The Central team, which designed and analysed the survey involved Yogendra Yadav, Dhananjai Joshi, Praveen Rai, Himanshu Bhattacharya, K.A.Q.A Hilal, Kanchan Malhotra and Vikas Gautam from CSDS and Sandeep Shastri from International Academy for Creative Teaching.
Table 1: Spread of the sample?
| |
Post Poll |
Exit Poll |
Locations |
208 |
92 |
Sample Size |
4104 |
4348 |
Table 2: How representative is the sample for Post Poll for Phase I & II
Category |
Population estimates in Karnataka* (Census 2001) (%) |
Survey Figure**
(%) |
Rural |
63 |
67 |
Women |
49 |
45 |
Hindu |
85 |
84 |
Muslim |
11 |
9 |
Dalits |
17 |
16 |
Aadivasis |
7 |
6 |
* Only for those districts which went to poll in first two phases and where post-poll survey was conducted
** Based on constituencies where Post Poll survey was conducted.
|
|
|
|
|
