FEATURE1 June 2011

Diary: Polls, pulses and portaloos

A roundup of things that caught our eye in and around the research industry this month.


Wearing one’s heart on one’s sleeve is set to take on a new meaning, if a research innovation that Diary learned about this month takes off. According to Affectiva, which makes software for reading facial expressions, it is now possible to tell how fast someone’s heart is beating through their webcam. That’s right – no special equipment, simply by analysing a video of your face, Affectiva believes it can take your pulse. And your breathing rate. Founder Rana el Kaliouby explained that the so-called Cardiocam works by monitoring how light is reflected from the face. It all sounded a bit like science fiction to us, but Affectiva insists it’s very real, and claims that results have correlated highly with the standard finger sensors used to take people’s pulses.

This made Diary wonder what can’t be seen through a webcam these days. Can they tell what colour underwear I’ve got on? Or if I’m about to sneeze? Can they hear the voices inside my head?


The pollsters at ICM are feeling very pleased with themselves this month after correctly predicting the result of the UK’s Alternative Vote referendum, right down to one decimal place. Their survey, which said that AV would be rejected by 67.9% to 32.1% (and it was), was the “most accurate telephone poll on record”, ICM said.

The referendum result, Diary believes, offers lessons for research practitioners. Aside from the Nick Clegg factor, a major role seems to have been played by voters’ biases towards the status quo and the perceived social norm. In the face of controversy, uncertainty and mistrust over the two systems, voters seem to have fallen back on the one that is already in place, and that is more widely used in other countries.

On the other hand, the argument put forward by the No campaign that AV is just too complicated is in Diary’s view a red herring. All AV requires of the voter’s brain is to rank a few names in order of preference. Compared to this, most online surveys begin to look like advanced calculus.


Recruiting people for surveys is a constant struggle, but a new company in earthquake-hit Christchurch, New Zealand, has come up with an approach that goes beyond the bog-standard.

Texsys, founded by Adam Hutchinson, has been placing posters in the temporary toilets that are dotted around the city’s suburbs after the earthquake damaged its sewers. ‘Don’t spend the next wee while staring at this wall’, the signs say, inviting people to give feedback on the reconstruction effort by text message – all without leaving the loo.

Asked about his ambitions for the toilet-based approach, Hutchinson said he would love to become the country’s number one or number two research provider.


The Eurovision Song Contest is nearly impossible to explain to anyone from outside Europe. The kitsch, the costume changes, the weirdly reverential tone of the ceremony combined with the catty voiceover. Not to mention the music.

To make matters worse, as the number of competing countries has risen the scoring system has become absurdly complicated. On the other hand, it’s a data geek’s dream – full details of the points awarded to every country by every other one are available on the Eurovision website for anyone who wants to answer those burning questions: What was the secret to Azerbaijan’s triumph this year? What is the optimum number of costume changes? Do the Balkan countries spoil it by voting for their neighbours? There’s a huge opportunity here for data-driven europop, and Diary calls on the world of research to crunch those numbers and distil the formula for the perfect entry.