DJALCHEMI

Guardian digital music survey

Having previously said it would be useful to see more real data about people's listening behaviours, it's good to see that The Guardian is running quite an extensive online survey of digital music use. Fill it in if you feel like it. [Postscript, 2 October 02004: the survey has now been withdrawn — see my posting on the results].

However, the way a few of the survey questions are structured betrays some of the same myopia as is evident in the digital music 'debate' generally. And there's a strong argument that surveys are not the best approach to get evidence of trends.

The options the survey offers as sources "you go to find out what music to listen to" are interesting in terms of setting out the new 'gatekeepers' for new music (see my earlier posting on this topic). The list comprises: Ads in print, Music Weblogs, Specialist music websites, Online music stores, Ads on TV, TV programmes, General music websites, Ads on radio, Recommendations from friends/family, Newspaper music reviews, General music magazines, Radio programmes, Music download websites, Specialist music magazines, Peer-to-peer file-sharing websites, Music stores (such as HMV, Virgin and independent stores), Other, None of these.

Live music does not feature in this list, or anywhere in the survey, apart from a seemingly tangential question about intentions to visit Glastonbury or other festivals. Does no-one buy an album from a support band they stumble across, except me?

The position of radio in the survey is more ambiguous, not to say confused. In response to the question about how many hours I listen to music each week, I included time listening to music radio, which I thought was reasonable. But the question, "Which of the following formats do you currently use to listen to music?" does not include AM, FM, DAB or any streaming format other than Windows Media Player among its main choices.

Questions like "Where do you acquire your music for each of the following formats?" give the game away. This survey appears to be focused on a potentially short time-horizon (3-5 years?) where the convenience of cheap hard disk storage is greater than the convenience of cheap wireless (or wired) streaming. In terms of MP3, one of the formats covered by the question, the amount of interesting music legally available for MP3 stream may be approaching that legally available for MP3 download sooner rather than later.

Survey methods are better than nothing but they have limitations in an environment where formats are changing rapidly. Respondents are likely to be even more focused than the survey designers on technologies that are established in the market. Secondly, the habits they report may be influenced/constrained by current format wars (WMA vs. MP3 vs. AAC etc) more than their desires. And the opinions they express are likely to line up with the well-established positions of the piracy debate in the media, rather than deeper individual preferences.

The potential for streaming music services thus don't register on the opinion survey radar, partly because services to wireless portable devices are not yet well established, and partly because they're mostly legal so they don't whip up a media storm.

What would be more useful — if harder to get — would be, firstly, quantitative data about actual take-up of different listening habits. How do people divide their time between listening to

• 'stored' music (CD, MD, or digital downloads) vs.
• 'live' programming (radio, streaming, and even live performance) vs.
• hybrids (e.g. digital radios with pause and record, radio on demand).

Secondly, in order to predict trends across technology formats, it would be useful to back up the quantitative usage data with qualitative, ethnographic data to help understand why different people have different habits, and how this fits into their overall lifestyle and attitude to music.

Some related earlier postings:

• Doubts about models for listening to music in the future — covers more of the questions that ethnographic research could address;
• Preliminary sketch for online music listener research — gives a 'no-budget' example of some quantitative usage research.

Thanks to the excellent Five Eight free daily email service for the original link to the Guardian survey.