Live blogged in Nashville. Any errors or bad jokes are my own. Any typos are the fault of iPad.
online data collection is the best methodology ever. except if you ask an open text question. snippets of conversation. tiny tidbits. some people will take the time to write out good responses but most answers are thin. why can’t the world be a long discussion like you would get if you just talked to people.
but wait – who would be able to make a video, who would choose to make a video, is the data comparable, are there biases involved.
You need a camera and a microphone. about half of people have a camera with a microphone in front of them.Men are slightly more likely to be able to make a video. massive age bias against older people being able to do it. men are more likely to agree to do it.
knowing that the video will be used a bit more widely doesn’t affect conversion rates a lot. the language around permissions needs to sound somewhat legalize so people know it’s serious. People expect you to ask permission to take video, that gets higher conversion.
You get far more content from speech versus typing, and it’s not just ums and ahs, it’s more codeable data. Much more verbal content even to difficult questions. The same top answers come up in both types of data but you also get many other answers, and you can figure out the second top answer much more accurately. YOu get more colour from the answers.
not everyone knows how to make the audio work or the video work. sound and lighting matter and not everyone is a filmmaker. if you can’t see or hear a video, you’ve lost data. Video gives you visual cues about whether the person cares or is trying or is searching for their own opinion.
about half CAN do a video. about half WILL do a video. about a third of a sample will end up producing a video. but it is demographically biased. be careful with permissions. consent will get better with experience. the age problem will go away.
do not use video along. don’t frighten off with excessive permissions. dont use plain english. check that the sound and video actually work. warn against videoing other people. don’t expect people who say yes to actually do it. do not expect it to instantly solve your problem. these are regular normal humans in their own home doing their own thing (e.g., might be naked!).
video open ends are a major step forward. [and some serious coding $$!]