As the saying goes… Hindsight is a bitch.
I had to spend time at home the last few days – a sort of self-imposed exile. And most of this was spent reading the reflections, rumination, reasonings post the US elections. Its been one hell of a year so far and this year, the polling industry has taken a double beating thanks to BREXIT and the US Elections.
Nearly two decades ago, a research agency in India had come out with a print ad announcing they were the only agency to successfully predict the winner of the last three national elections. Those were different times as it has been increasingly rare since then to correctly predict election results in India (and pretty much elsewhere).
I had a chance to talk to the Field Head (head of data collection) from that research agency then, and he had some great insights on how to carry out the fieldwork for elections. He’d explained each election needed a separate sampling framework. The term “complicated” was an under statement for the region-language-religion-caste-social class dynamics in India and it was pretty naive to assume that each province’s election could be predicted using a simple large-scale sample based on demographics alone.
For political polls this Field Head fitted the bill perfectly, as he used to be a student leader earlier and was passionate about political developments. I have seen a lot of discussions on television by political experts, but this Field Head was a totally different creature. He had an ability to understand which of the recent developments would be the key issues and who’d be affected and therefore, to be represented in the survey sample. He also had his ear on the ground and a fairly good idea of the kinship between segments and therefore which segments to cover/ ignore. Also, he’d recommend including questions about the various issues as well as fundamental ones like who you voted for last election, and how do you think others are likely to vote etc.
The critical point was that he spent a great deal of time on the ground interacting with the field staff as well as political parties and their workers and knew how canvassing worked. There would be ongoing feedback from the interviewers as they carried out face-to-face interviews and a reflection of sentiments, overclaims, non-response etc. would enrich the interpretation of results. There’s a fairly good chance he might have picked up on issues that the US polls missed out like the overall sentiments as well as the converse behavior from taken-for-granted segments.
In the case of the US elections, an excellent analysis from HBR talked about how the polling industry had an excessive over-reliance on phone interviews and therefore missed out non-response and those who were out of the sampling universe. It appears that the advantages that technology offered also brought along a bigger disadvantage. Would face-to-face interviews have fared better? I think they’d have easily covered a lot more of those missed out by phone polls.
Interestingly, some polls did actually predict Trump as the winner for quite sometime now. But these were ignored and clearly a minority opinion. (Read BBC Future’s article to help understand the biases in our minds also strengthened by social media feeds).
In an unrelated development, on the day of the US elections, the Indian Government decided to de-monitize currency bills of large denominations with immediate effect.
The result was chaos. I had the fortune & privilege of freezing all my discretionary spending as well as postponing all my travels (luckily my partners understood and agreed). Not so for the common man on the street who faced great difficulty and hardships, and this is yet to abate till banks return to normal functioning.
A big realization was that the electronic money that we take for granted was barely helpful or enough to take care of the day-to-day transactions. Most ATMs were closed, and a lot of electronic transactions simply failed at the check-out counters. Plus, in most day-to-day situations in India, only cash will do.
Not surprisingly, It was the cash stashed away in the cookie jars that actually saved the day. Hard cash was truly king
There seemed to be a common thread to these. Our technology might have advanced enabling a lot more electronic, digital communication and capture, but it cannot replace real-life face-to-face transaction (as also actual cash on hand).
Given the cynicism from the recent events, what role does the insights/ market research/ polling industry have in our future?
After being overwhelmed by recent events, our healing process is going to take some time (4 years?) and I am hoping the market research industry plays its role and contributes to this healing.
At the end of Oprah’s long career, her big learning had been that everyone from the celebrity to the non-celebrity guest yearned for validation and to be heard. Each of them asked her for re-assurance that their interviews with her had gone well. It was also an acknowledgement of their existence that they were seeking.
The instruction “Remember there are no right or wrong answers, and we are only interested in your opinion” is one of the most empowering sentences and an integral part of market research interviews.
IMHO Market Research fulfills this need for validation by representing the citizens, and voicing their beliefs, aspirations, concerns and dreams.
And I believe it is this – being ‘on the ground’ and listening patiently, that can make all the difference…
PS: Among those questioning the polls were a lot of advertising folk, and this is something I had addressed earlier. The simple and best response to them, is that market research helps them win awards 🙂