My 454-Person Facebook Survey Before the 2014 Election

In the days before Kosovo’s 8 June 2014 election, there was almost no reliable public polling of the kind you would take for granted elsewhere. As the person handling social and digital for Partia e Fortë, that bothered me. We were making decisions about tone and targeting partly on instinct, and instinct is a bad thing to fly on when you can gather at least some data.

So on 30 May 2014 I ran my own survey. I put it out to my circle on Facebook and other social networks and asked people how they intended to vote. By the time I closed it, 454 people had responded.

I want to be very clear about what this was and was not, because I said the same thing at the time and I still believe it is the honest framing. This was not a representative national poll. The respondents skewed exactly the way you would expect a young, online, Facebook-based sample to skew: heavily male, heavily in the 18-25 range. That is the demographic of my network and of that platform in 2014, not the demographic of Kosovo. I never claimed otherwise. In the absence of any real field data, I wanted to at least measure the mood of the online, engaged, young slice of the electorate, because that slice was precisely who Partia e Fortë spoke to.

With that caveat carved in stone, here roughly is what it found:

• Vetëvendosje led the sample at around 39 percent.
• Partia e Fortë sat near 30 percent within this group. For a satirical party, in its own core online demographic, that is a striking number and it told me our reach was translating into stated intent, at least among the people most like our audience.
• On the prime-minister question, Albin Kurti led at around 41 percent, with Visar Arifaj, our Kryetar Legjendar, close behind at around 33 percent.

I found those results genuinely useful, not because they predicted the national outcome, they could not and did not, but because they confirmed something about our own audience. Inside the young, online electorate we were built to reach, the joke had landed hard enough to compete with serious movements. That is a real signal about message-market fit, even from a skewed sample, as long as you are disciplined about reading it as “here is my audience” and not “here is the country.”

I am glad I ran it, because it is a concrete piece of the actual work rather than a story about it. It is a small, imperfect, self-selected poll. It is also me doing, with the crude tools I have, the thing I am coming to believe in: get whatever data you can, be brutally honest about its limits, and let it sharpen your instinct instead of replacing it.

If you want the surrounding context of that election run, I also wrote about being candidate number ten on the party list and what the 71 votes I received actually meant.