Monday, October 5, 2015

Analyzing Election Analyses


The year 2015 started with a state election which continues to be talked about even today 2015 end.
The year 2015 will end with a state election which continues to be talked about since 2015 beginning.

The states may be different but the nature of election analyses by "experts" remain the same.

Prior to Delhi elections, the "experts" made us believe lots of things such as
'why BJP will win Delhi election',
'why inducting Kiran Bedi in BJP is a sign of Amit Shah's tactical genius',
'why Arvind Kejriwal will have a tough road ahead',
'why it will be a closely fought battle between the AAP and the BJP'.

Post elections, the "experts" again made us realize lots of things such as
'why BJP did not win Delhi election',
'why inducting Kiran Bedi in BJP was not a sign of Amit Shah's tactical genius',
'why Arvind Kejriwal did not have a tough road',
'why it was not a closely fought battle between the AAP and the BJP'.

The differences in these ex-ante (before the event) and ex-post (after the event) analyses are interesting to note. Even now in case of Bihar elections, such ex-ante analyses are finding their way to various media channels (televisions, newspapers, social media etc.) and many similar ex-post analyses will find their way post the elections.
The differences in the two analyses could be due to a lot of factors, one of which is the presence of 'bias' in the individual carrying out the analysis.

For an individual, it is normal to have sets of beliefs/prejudices/biases.
But for a researcher/political analyst, it is imperative to shed those biases before carrying out the research/analysis.
Juggling between the two facets is not easy and often individual biases find a way to affect the research design and/or the analysis, leading to incorrect interpretations of findings. Presence of biases can and does often lead analysts to present results which they want to present rather than what flows from the analysis.

In case of election/political analysis, absence of individual bias would require a person to be apolitical before starting with the analysis. Apolitical here is defined as someone who has similar (or zero) preference for all political ideologies/thoughts/entities.