I spent a good chunk of this week working on the April Konda Barometer survey. Unfortunately, I cannot disclose any of my results, but I can at least tell you what I did.
When I got hold of the survey, I was thinking on doing something on political polarization, but I found out, to my dismay, that Konda liked my take on that in the inaugural survey last month so much that they decided to do it themselves:) With them taking care of that, I had to find myself some other thing to work on.
First, there was a really interesting question on the survey asking about how respondents got to form their opinions on the constitutional amendments (TV, friends, newspapers, party opinions, etc). For an economist, that is a golden opportunity to do some interesting regressions, so I spent a day or so looking at the determinants of opinion formation.
Second, I noticed that it is possible to get unemployment rates from the survey. As the official statistics lag about a quarter, the Konda rates would be very valuable, if they were in line with the official statistics. While such a comparison is impossible at the moment, due to the fact that the latest official statistics are from January and the first Konda Barometer Survey was in March, I nevertheless did calculate TURKSTAT-comparable unemployment rates and used them, along with other recent data, to discuss the speed of the economic recovery and its asset price implications. I plan to do a complete comparison next month between Konda and TURKSTAT statistics.
This is all I can say about my work there, but if you'd like to find out more, feel free to contact them.
When I got hold of the survey, I was thinking on doing something on political polarization, but I found out, to my dismay, that Konda liked my take on that in the inaugural survey last month so much that they decided to do it themselves:) With them taking care of that, I had to find myself some other thing to work on.
First, there was a really interesting question on the survey asking about how respondents got to form their opinions on the constitutional amendments (TV, friends, newspapers, party opinions, etc). For an economist, that is a golden opportunity to do some interesting regressions, so I spent a day or so looking at the determinants of opinion formation.
Second, I noticed that it is possible to get unemployment rates from the survey. As the official statistics lag about a quarter, the Konda rates would be very valuable, if they were in line with the official statistics. While such a comparison is impossible at the moment, due to the fact that the latest official statistics are from January and the first Konda Barometer Survey was in March, I nevertheless did calculate TURKSTAT-comparable unemployment rates and used them, along with other recent data, to discuss the speed of the economic recovery and its asset price implications. I plan to do a complete comparison next month between Konda and TURKSTAT statistics.
This is all I can say about my work there, but if you'd like to find out more, feel free to contact them.
Posts
No comments:
Post a Comment