I recently stumbled upon an article while reading BBC News, claiming that a psychology professor in Scotland has linked a person's musical taste (by genre) to aspects of their personality.
Unfortunately for this prof, I play a lot of music, like a lot of different music, have a psychology degree, and immediately see multiple problems with his work. Before I drop some scientific method in your face, let's take a look at Professor Adrian North's weighty claim.
The study in question consisted of surveying 36,000 individuals with 104 questions about different types of music and aspects of their personality. The results state, among other things, that classical music fans are introverts, while jazz listeners are outgoing; reggae audiences are gentle, while those who listen to dance music are not gentle; hard rock fans have low self-esteem, while those who enjoy Top 40 Pop have high self esteem. Pretty interesting stuff which makes sense, initially. I do have to hand it to professor North; 36,000 participants in a pysc test is a very large number, and should provide a high level of reliability in the study. But for those of us who think critically about psychology, reliability is not the same as validity. In other words, a study can be very reliable, and wrong. Reliably wrong.
The first problem with our study here, is that over-generalized claims are being made, without any specific data to support them. This may be a bigger problem with the media than with science, but it is incorrect to say that "If you like heavy metal, you have low-self esteem". One could claim that 96.8% of heavy metal fans have low self esteem, but blatant over-generalizations are bad for science and people's perceptions alike.
Second, the personality aspect of the study is relying on
self-report data. If I am taking a survey the day after getting fired, losing my mortgage, and finding out that my wife wants a divorce: of course I will self-report that I have low self-esteem! Similarly, on a normal day, most people want to portray themselves as outgoing and positive, regardless of the study's implications. A participant will either make themselves look good, or find reasons to make themselves look bad, very rarely being objective and honest about their condition. This is known as
participant bias, or the
demand characteristics of a test.
Third and most importantly, no correlative direction can be assumed in any study such as this. In a statement that probably urged me to write this post, prof. North says:
"If you know a person's music preference, you can tell what kind of person they are, who to sell to." This is simply outrageous. To "prove" that a person likes a certain type of music because of their personality, or that they have certain personality traits because of the music they listen to, is preposterous. As scientists, all we can say is that X percentage of rap fans are X personality, and refine our hypothesis for the next study. That is the scientific method itself. The fact that North then brings in music marketing, claiming record labels could use his questionable claims to target consumers, leaves me stunned and slightly sickened.
Psychology aside, North's results don't make much musical sense, either. I know plenty of people who like jazz and classical, but according to North, these reflect opposite personality types. If reggae fans are gentle, dance fans not gentle, then what personality traits are present in fans of "dance-hall", a music that combines reggae
and dance influences? Top 40 fans apparently have high self esteem, heavy metal fans low self-esteem... but I know plenty of hard rock songs that make it to the Top 40 charts!!!
The problem is, I want to agree with professor North. In all areas of life, it is much easier to come up with stereotypes because it helps us define and navigate the world we live in without really thinking about it. In reality, all music fans are individually different and while we can come close to generalizing, we must provide statistics based on unbiased data. Until Mr. North can do that, I suggest he take a hard look at the science he is distributing to the media for popular release.
I'll close with some unrelated statistics, and unlike North, let you interpret the results:
We depend on wind and solar research to show us how to break away from our "addiction to oil". Congress has voted 8 times this year to keep giving tax subsidies to the companies involved with this research. Barack Obama showed up and voted yes, 3 of these 8 times. John McCain showed up exactly zero times, including one occasion where he was in Washington, and the motion failed, by a count of exactly 1 vote...
What kind of music do you think John McCain listens to???