Thursday, April 12, 2012

Interrogating the media for the nest best thing

Now can you read this. Joe Dubois' daughter passed with flying colours. But something was still wrong. His daughter continued squinting at home when reading, so he brought her back to the doctor, who was not assumed having already diagnosed her as having 2020 vision.

This time Joe had the doctor turn away and observe as her dad asked her to identify letters on the chart. She squinted, couldn't and the Doctor was fascinated.

Somehow he was transferring to the little girl what he wanted her to tell him, because seemingly she'd memorised the grid.

This scene comes from the US series Medium starring Patricia Arquette as  detective Allsion DuBois who sees the future etc, but it's poignancy is vast.

You're often hear the following:

Video on the web should be below 2 minutes, and people don't read online, they skim. Ask yourself how you know this.

Because someone knowledgeable told you, and sometimes depending on your speaker, he or she found out in that great repository of knowledge - the web.

But here's the crunch. The web, indeed even research can tell you anything you want to hear. Let's leave research out for the moment.

The web says so, with a link to some firm or company, with no methodology of how they know - is one of the enduring myths of erroneous information on the web.

So how do you know videos should be so long? Because teenagers are telling you so. Well teenagers say a lot of things, as do non-teenagers, so the next best thing you need to do is undertake an empirical study. That means test the hypothesis you have.

Now depending on your questions you can also influence the answers. A subtle shift is giving times in ascending order 2 minutes, 4 minutes etc. Also if you're in a hurry to finish a survey how does your own time needs influence the outcome.

Research needn't be boring. Here goes. Firstly you can trend extrapolate - examining past trends. Why do we like short videos? Well if we look hard enough we'll find there's a trend in diminishing length.

But why does it apply to the web? Here you could examine features and lifestyles associated with the web - lean forward, you watching at work when you shouldn't, the brightness of the screens that makes you tired and so on.

Then you test those.

One of the things as a videojournalist/film maker I like doing is using film to examine live-studies. It's a practise that has a long lineage in ethnographic films, but given our level of Social Network activity has renewed purpose.

And from what I would call these films of inside records, we can begin to understand not just by hyperbole or more pointedly interviews, but also letting the camera observe in an impressionistic way - that's small details.

Details about what we do and how will come fast and furious at us, and even though the web is a gem, it doesn't have all the answers. Rigour and knowing how to devise and extract data will undoubtedly become a bigger industry than it has so far.

Because rightly we really truly want to know what's going on and that often emerges from deeper analysis than ..because she told me.