Re: music popularity experiment

Not at all: although I should recognise straight away that the link is to a secondary source and this may misrepresent the primary research. On the face of the report, the first group selected their favourites based upon a positive reaction to what they heard thus giving a measure of 'quality'. We do not know the selection criteria of the second group but are led to the inference that their selection was based in whole or in part upon the new variable introduced, namely knowledge of the selections made by the first group.

However, at least as reported, this doesn't give any insight in its own terms into real world behaviours since it is not apparent that there is a community of listeners exercising choice based solely on perception of musical quality which is in any position to influence the choice of the putative second group. Nor does it account for potential distortions in the market caused by operators with a commercial interest in the outcome - such as record companies.

More fundamentally though, since the first group were provided with the names of the artists, there was not, in terms of this report, any control group who's choices were made in the absence of knowledge of the artist. Therefore one cannot, as one of the researchers is reported to claim, state that the choice of 'quality' was made in the absence of social influence since there was no control of the social influence of pre-existing perceptions of the quality attaching to the particular artists within that group. If the report in Live Science is correct (no doubnt a big if), then there appears to be a methodological flaw in the experiment.

It follows that the experiment, as reported, can not be said to measure the effect of peer group influence since there was no control group which effectively removed that influence (assuming that such a thing is possible). One could therefore read the report as saying that songs selected on the basis of peer-group perceptions of quality are chosen on the basis of peer-group perceptions of quality, which is tautologous. Quad Erat Demonstrandum

Cheers

Ian



Jim Lawton wrote:
On Fri, 10 Feb 2006 12:50:41 GMT, Ian Burdon
<ian.burdon@xxxxxxxxxxxxxxxx> wrote:


The conclusion was, apparently, that:

"Researchers found that popular songs were popular and unpopular songs were unpopular, [...]. They also found that as a particular songs' popularity increased, participants selected it more often".

Next up in New Tautologist comes research showing that the woods are a popular area for ursine defecation and that the majority of Popes have professed Roman Catholicism as their denomination of choice ;)



I'm sorry - you just snipped that for effect. The experiment was about
peer-group influence :-

.

Relevant Pages

  • Re: Clear and Requery combo boxes
    ... Q3 - when a selection is made for the project team, ... When I close the report and go back ... >> The choices then generate a report based on the selection. ... >> Exit Sub ...
    (microsoft.public.access.formscoding)
  • Re: Option Groups
    ... Can't find "Command button report selection" even though I did a search. ... 'This procedure used inh Preview_Click and Print_Click Sub procedures. ...
    (microsoft.public.access.tablesdbdesign)
  • Re: OpenReport action cancelled because of Printer
    ... You should be able to import the toolbar from the sample database with: ... setting my report toolbars to 'ReportToolbar' caused ... printer (I just clicked 'OK' with the 'use default printer' selection, ...
    (microsoft.public.access.reports)
  • Need help with multi threading, got error "Cross-thread operation.
    ... SortFields sortFields = rdLog.DataDefinition.SortFields; ... selection.", MessageBoxButtons.OK, MessageBoxIcon.Information); ... goto CleanUp; ... or clik the OK button to continue creating the report? ...
    (microsoft.public.dotnet.languages.csharp)
  • Re: What Liberal Bias?
    ... The real truth is that you are a pedantic ... Because every newspapers circulation is going down. ... want to influence the outcome rather than report the outcome. ...
    (rec.sport.golf)