Science fair help

[Mythestopheles]

Seems that in my school we have to take part in a science fair. For those unfamiliar with the idea, it's basically where a bunch of students make science projects which are then judged. Generally they involve answering a question.

Anyways, My Project idea is: How does the music genre you listen to relate to your personality. I would appreciate any advice people had on which music genres I should use. I'm thinking that if I use subgenres though, it will get a bit long.

Here's my current list:
Blues
Classical
Country
Indie
Jazz
Metal
Opera
Pop
Rap
Rock
Soul

Any advice on what to add would be nice, thanks.

[Miklus]

You could compare music preference with hair length. :smallsmile: Make a big diagram where the genres are sorted by average lenght.

[jlvm4]

Seems that in my school we have to take part in a science fair. For those unfamiliar with the idea, it's basically where a bunch of students make science projects which are then judged. Generally they involve answering a question.

Anyways, My Project idea is: How does the music genre you listen to relate to your personality. I would appreciate any advice people had on which music genres I should use. I'm thinking that if I use subgenres though, it will get a bit long.

Here's my current list:
Blues
Classical
Country
Indie
Jazz
Metal
Opera
Pop
Rap
Rock
Soul

Any advice on what to add would be nice, thanks.

I'm not sure adding is what you need to be doing here. You are going to have to set up a testable scenario with controlled variables. The fewer options, the less complicated the project and the more opportunity to repeat it enough to get a sample size that might actually mean something.

More importantly, you need to consider not just music choices but personality category and how you will assign each subject. This strikes me as the more difficult design problem rather than music choices.

In fact, you might want to deliberately pick only 3-4 music types that are so different that the subjects are less likely to answer 'I don't know--both?' or some variation on that theme.

It sound fascinating, but also long, so be sure you can do what you need in the amount of time you have.

Good Luck

[Mythestopheles]

Thanks, what I meant by a bunch of music genres was so nothing was left out. I could probably group similar music genres together though.

For the personality category I was thinking I might be able to use the Meyers Briggs personality types.

[Quincunx]

What aspect of the personality does music taste predict? This allows you to give a little test to people to determine the trait instead of relying on self-reported personality traits. Will you test linguistic skills and organization by timing how long they take to alphabetize a list of 10 uncommon words which share an initial letter? Will you test visual and spatial skill by timing how long they take to "sort" cut-outs of different shapes and colors, and get additional insight by how different musical types interpreted the instruction "sort"? Will you sit outside the supermarket with a folding table, getting a cross-section of the population who needed to buy groceries yet was unhurried enough to heed the "Science in Progress--Please Give Three Minutes of your Time" sign? Will you confine your results to people within one year of your own age?

Self-reported music taste is good enough data, although if you can use something like their top five most-played tracks on last.fm* or the like, that is less subjective, and superior data. jlvm4 has an excellent thought about making the genres distinct and the resulting data easier to analyze, although Miklus' suggestion would require the richer variety of categories. As for the initial list, I scratch my head over where to mark down my tendency to listen to "Prog", although I suppose "Indie" will do, don't know if "Regional" or "World" would be selected even if included, and struggle to distinguish the people who would listen to Classical but not Opera (or vice versa).

*if that's available in Canada

[EDIT: Forgot the most important part. . .if your science fairs were anything like my science fairs, there'll be enough "How in the world is this considered a project?!" entries that your initial proposal will look pretty good by comparison. Heck, you can even come in with the exact same project as other participants provided you all get the data from different people--after all, Real Science repeats its tests many times to check their validity. Science fairs remove a lot of the pressures of other presentation-style projects. Stress not.]

[Mythestopheles]

Thanks for taking the time to help me out.

As for what to measure, the only idea I have is to maybe link certain parts of the personality types (Extroverted vs. Introverted; Intuitive vs. Sensing; Feeling vs. Thinking and Perceiving vs. Judging) to say, genres of music, or vice versa.

Thanks for the reassurance. I actually probably had one of those not-so-project-projects in past years. :P

[Viera Champion]

Almost everyone has to do a science fair sometime in their life.

Also, you should add alternative to that list.

[Shyftir]

Yeah, if alternative was possible to define...

Anyway I think your major problem might be that a lot of science fairs discourage any involvement of human subjects.

Otherwise, you might want to use broader music categories, like:

Intense: Hardcore, Grindcore, Metal, Industrial, Hardstyle, True Punk, Screamo etc.
Easy listening: soft rock, country, modern-folk, softer pop, trance etc.
Medium: Alternative, Modern rock, pop-punk, Emo (ugh), Dance pop etc.
Traditional: classical, world, celtic, big band, etc.
Unusual: Ska, niche rock, filk, zydeico, etc.

[Viera Champion]

Oh. Also, I think you should count Beatles as there own genre. Firstly, because they're awesome. Secondly, They practically redefined rock.

[Dvil]

Oh. Also, I think you should count Beatles as there own genre. Firstly, because they're awesome. Secondly, They practically redefined rock.

I'm not sure. Reason one is clearly a joke, so I won't insult you by taking it seriously, but their own genre just for redefining rock? Maybe have pre-beatles and post-beatles rock as separate genres, but not Beatles as their own.

Unless 2 was equally joking, in which case disregard everything I just said.

[Viera Champion]

You actually insulted me by thinking reason one was a joke. The Beatles are awesome.

[Rutskarn]

Yeah, uh, that seems a bit odd. What's weird about liking the Beatles, again? Has the music scene really gotten to such a state that we assume anyone claiming to like the goddamned Beatles is being sarcastic?

[Mythestopheles]

Umm, I think he was talking about how claiming that the beatles being awsome wasn't a good enough reason for them to have their own genre.

[Rutskarn]

Hm. I suppose that's possible--reading Viera's post first probably clouded my interpretation.

If so, apologies.

[Crimmy]

I'm not sure. Reason one is clearly a joke, so I won't insult you by taking it seriously, but their own genre just for redefining rock? Maybe have pre-beatles and post-beatles rock as separate genres, but not Beatles as their own.

Unless 2 was equally joking, in which case disregard everything I just said.

First Bolding: It's no joke, dude.

Second Bolding: Why not?

[Don Julio Anejo]

Okay, this is right up my alley :smile: I'm a psych major (and I'm actually working on an independent study project related to personality right now) :tongue:

Anyway, my comments on the topic:

1. Use less categories than you had. Try to get them as mutually exclusive as possible. E.g. pop, dance and rnb would not be different enough.
- One way is to group them by speed like Shyftir did (although trance is not easy listening, even by definition).
- Another way is to group them by similarity. E.g. Rock/metal, dance/pop/rnb, electronic, rap/hip-hop, folk/country. Make them as broad and mutually exclusive as possible so an average person can easily say "this" without hesitation. Maybe even put pop, electronic and non-gangsta rap/rnb into the same category.

2. How exactly are you going to measure personality? MBTI in itself is great, but what the test is something like 300 questions and even a simplified version would be something like 50-70 questions. I simply don't see how you would get enough people to answer them unless you have 50-60 really good friends who would do that for you, and even then you'd suffer from bias because they would, well, be your friends and would have tastes closer to yours, leaving other groups underrepresented.*

It also has the disadvantage of people trying to answer questions based on how they want to appear instead of how they actually are. If your school is anything like mine was, 90% of people would try to seem extroverted even when they're obviously, loners (I mean people who don't seek company of others and prefer to be alone, not those who do but are shy or whatnot). Even though a typical INFP would be much more fun and outgoing than a typical ESTJ.

* - I know it's a science fair and no-one really cares about proper control or anything, but if you bring that up, you can actually score yourself bonus points from the teacher for even thinking about it.

For personality, I would suggest something simpler. Maybe 10-15 true-false questions that measure a few personality aspects instead of the whole test. I also suggest you use the Five Factor Model (http://en.wikipedia.org/wiki/Five_factor_model) (Big 5). It has a better theoretical base than MBTI (which is based almost entirely on Jung's works) and more utility in general. It's also probably easier to do a short test on and correlate music tastes to specific dimensions, because, well, they're dimensions and not dichotomies (dimensions = you can be anywhere on the scale, dichotomy = you can only be one or the other).

However, the choice is yours, I can quite understand MBTI being more interesting to do a project on (well, it is more interesting), and if you're familiar with one but not the other, there's no point reading a bunch of stuff on something you don't know much about since it won't help you that much. But if you want, I'd be willing to help you.

PS: one thing you should keep in mind is that you're in high school. Chances are, there's a third factor causing a link between music and personality: social groups, which to a large extent are music-based. If most of your friends are metalheads, chances are you listen to metal. And even though you also like dance pop, you'd probably never admit it because, well, your friends are metalheads. It'd be like coming in a tootoo to a gun club. Don't necessarily need to do anthing about it, it would be pretty hard to control for a factor like this even if you were doing a 2000-people randomized survey across the nation. But again, you could score bonus points from the teacher if you mention it.

[scsimodem]

I will preface the bulk of the reply by saying that I went to the Arkansas School for Math and Sciences (now the Arkansas School for Math, Science, and the Arts). When I attended, every student was required to do a science fair project, and science fair was a 3 hour class once a week.

Project Feasibility: Low, project not recommended for non-professional scientists without serious alterations

Factor 1: Quantifiability: Classifications of music are largely grouped based on stylistic differences and have very few measurable factors of classification. Second, personalities and personality types are so varied and nuanced that Ph.D.s have spent their collective fortunes and brainpower just trying to figure out what personality traits cause people to be attracted to each other, romantically or platonically, with little to show for it.

My suggestion is to try to limit one or both measured attributes to something more easily quantified. Try limiting music to a single, measurable attribute, such as tempo or intensity or the psychological aspect to such things as tendency toward aggression or even something as specific as tendency to speed.

Factor 2: Control variables: Everything tends to affect personality in some way. Controlling for simple differences in musical taste would be near impossible. Your sample would have to come from similar backgrounds, have similar home lives, etc.

My suggestion is to try to focus on the reverse. Treat music selection as a conscious choice that is more a function of personality, rather than vice versa. That way, it doesn't matter what caused the personality, as the personality caused the music.

Factor 3: Difficulty in performing tests. Your data points are all humans with complicated psyches. People lie. Some are very strange. To control for a behavior change from known observation (somebody lying about his love of Mozart because he's afraid some of his friends will think it's lame is a good example), you would have to find some way to either coerce honesty or observe the subjects undetected. Then there's the randomness of personality tests. Acquiring a credible test would be expensive, as it takes a complex test to account for lying on the test.

I have no suggestions for this step. This is a major obstacle.

Overall, I don't think this is a good idea for a science fair project. It's a fascinating subject, but I'd save it for a doctoral dissertation rather than a high school science fair project. Here are the hallmarks of a good project.

Ease of control: Most good projects attempt to link a change in one variable to a change in another variable (the independent and dependent variables, respectively). However, to link the two definitively, you must ensure that the only appreciable change is the one you are testing. Let's say you're testing a theory that high levels of CO2 cause faster plant growth. If you grow your experimental group in a sealed aquarium with fresh CO2 pumped in from a container dry ice by a fan, then leave your control group outside, you can't definitively state that the difference is caused by the CO2, since it could be the difference in temperature, exposure to sunlight, or ventilation. Sealing both in aquariums of the same temperature and level of ventilation would be ideal. Find something you can control for.

Quantifiable variables: You need to be able to point to your data and show that everything was appropriately and concretely classified. Attaching numbers is the easiest way to do so, and is something you can't do to more abstract or poorly documented attributes, like style and personality.

Measurable results: When done with the study, you need to be able to definitively conclude that your assertion is true, false, or indeterminate. To do that, you need to be able to measure your results against something. This relates somewhat to having quantifiable variables, but is worth its own mention.

Once again, I'm not dissing the idea in general. I just think you're biting off more than you can chew. This would be a good dissertation for a Ph.D. in psychology.

[Mythestopheles]

Thanks for the input.

Don Julio Anejo: Yeah, I think I probably should simplify it by quite a bit. I'll check out the big five personality traits in a bit.

scsimodem: Thanks for the realistic approach, I think I will keep this sort of idea, though I will definitely have to simplify this quite a bit.

Come to think about it, in our school we can do reaserch projects instead. Would that make things more doable do you think?

[Don Julio Anejo]

If you don't mind, will you keep us updated on your ideas? :smile: I'm actually quite curious how this turns out.

Also, I had an idea on something you could do: whether the tempo of music a person prefers is related to either their extraversion or their neuroticism trait on the Big 5. For example, I'm both neurotic and extroverted according to NEO-PI-R (the MBTI of Big 5), in the 80%ish and I like really energetic music like house, trance and hard rock. For comparison, good friend of mine is pretty phlegmatic and his iTunes playlist literally puts me to sleep...

Just an idea I came up with in the last 5 minutes that's doable without too many problems. 10-12 TF questions with 2-4 control questions will be accurate enough for your purposes and it'll offer pretty good insight, although you'd need to somehow quantify the tempo of music.

[FirebirdFlying]

If you have time, read 'This is Your Brain on Music', by Daniel Levitin. It's about the psychology of music - not exactly directly related to the genre=>personality thing, but it has some interesting stuff about the music people tend to like - such as the study that showed that, if a song was played repeatedly in the last trimester of gestation, a year later the baby preferred that song over another similar one, while the control group showed no overall preference.

[Don Julio Anejo]

such as the study that showed that, if a song was played repeatedly in the last trimester of gestation, a year later the baby preferred that song over another similar one, while the control group showed no overall preference.
Actually that's not related to music but rather our LAD (language acquisition device). Babies still hear stuff even though they're in the womb and after they're born, they show preference for familiar sounds. It helps in learning to talk as they concentrate more of their attention on sounds they hear more often (which are usually speech). Most of the research has been on language, but the same principle applies to music - babies simply find the way the song sounds familiar and so they like it better than unfamiliar ones.

[FirebirdFlying]

Right, but since the principle applies to musical preferences, I'd say it's still relevant to the topic.

[Lioness]

If it helps at all, I'm INFP and I like classical music

[Mythestopheles]

If you don't mind, will you keep us updated on your ideas? :smile: I'm actually quite curious how this turns out.

Also, I had an idea on something you could do: whether the tempo of music a person prefers is related to either their extraversion or their neuroticism trait on the Big 5. For example, I'm both neurotic and extroverted according to NEO-PI-R (the MBTI of Big 5), in the 80%ish and I like really energetic music like house, trance and hard rock. For comparison, good friend of mine is pretty phlegmatic and his iTunes playlist literally puts me to sleep...

Just an idea I came up with in the last 5 minutes that's doable without too many problems. 10-12 TF questions with 2-4 control questions will be accurate enough for your purposes and it'll offer pretty good insight, although you'd need to somehow quantify the tempo of music.

Sure, I'll keep you guys updated.:smallamused:

I haven't heard of the big five personality traits before. From a brief look, they do seem fairly interesting, and easier to do a project. Once again, I only looked briefly, and there was a lot of text (it is wikipedia after all) so I will accept your offer of help with understanding it. One question is, how would I make up a test for these types?

As for music grouping, I think I'll go by similarities. (Music style, lirical style and whatnot).

Lemme see, in no particular order...
Rock/Hard Rock/Metal
Country/Western
Pop/Dance
Hip Hop/Rap

Genres I don't know too much about, or about what category they would be under:
Blues
Classical
Indie
Jazz
Opera
Electrical

Edit: Oh, yeah and Alternative... Not really sure what this genre is. Seems a bit broad.

Once again, thanks for taking the time to assist me.

[Don Julio Anejo]

As for music grouping, I think I'll go by similarities. (Music style, lirical style and whatnot).

Lemme see, in no particular order...
Rock/Hard Rock/Metal
Country/Western
Pop/Dance
Hip Hop/Rap

Genres I don't know too much about, or about what category they would be under:
Blues
Classical
Indie
Jazz
Opera
Electrical

Edit: Oh, yeah and Alternative... Not really sure what this genre is. Seems a bit broad.
I'll post more on the test later as I'm at school right now (I think I may have a really good pdf overview on my other computer), so I'll deal with the styles first.

First question - where are you exactly in Canada? You don't have to tell me, but do keep it in mind when it comes to country music. If you're in a major city that's not on the prairies, I doubt you will find that many people who listen to country, and if they do, it's usually a cultural choice rather than a personal one (e.g. they grew up in a small town somewhere and moved when they were 10/13/15/whatever). If, on the other hand, you're in, say, Dog River, Saskatchewan, it's going to be a pretty big proportion of people who like it.

Now, alternative. Group it with rock and metal. Yes, they will grumble and complain that it's completely different. However, your goal is not to accurately account for every musical style out there (that's nigh impossible). Your goal is to accurately categorize people based on their musical preferences. And there's usually a very large overlap between people who listen to rock, metal and alternative. In contrast, there is virtually no overlap between alternative and, say, trance or gangsta rap.

Indie. This one is harder and I'll leave it up to you to decide. On the one hand, musically most of it is best described as "mellow rock." On the other, there is surprisingly little overlap between indie listeners and most other genres except folk (funny festival songs folk, not country cowboy music folk) and classic rock (i.e. Eagles, Guns n' Roses, The Rolling Stones). Also, a lot of people listen to indie to make a statement ("look at me, I like great unknown to anyone but me music unlike your commercialized crap"), which is directly related to personality. Hence, it makes quite a lot of sense to put it as a separate category, even though musically it's mostly rock.

Classical and opera can be safely excluded. Unless you're in a musical school or a school for the gifted, I doubt you'd find more than 5 people who can truthfully say it's their preferred genre. While you're at it, I would take out jazz as well. Everyone seems to like it but no-one seems to listen to it (may be my personal experience however). Leave it if people listen to it at your school however, but I honestly have no idea where to put it.

Blues - do you mean blues from the 50's or do you mean Rn'B? If it's the first one, with jazz. That is, group them together if you have jazz and not put either in if you don't. If it's the latter, either with pop or with hip-hop (see next section).

Now, this part you can do it two ways. What's left: pop, electronic, rap/hip-hop, Rn'B and dance. First, electronic vs. dance. Musically, there's a lot of overlap (most electronic music is dance music anyway). Most pop is also dance by the way. On the other hand, while pop overlaps easily with trance and electro, it doesn't do so with house (which is much closer to rap and hip-hop) or with techno (which, while dance music, is rarely liked by people who listen to a lot of pop), and that's without even going into the mess that is goa, jungle, psychedelic and other weird stuff. Fortunately, unless your school is populated by ravers, 90% of people will only really have heard of Tiesto and if they're lucky, AvB. So as much as it pains me to say it, it makes sense to make a category "pop/dance" and tell people who ask about any kind of electronic to put it under dance.

Rap/hip-hop deserves it's own category, obviously. But keep in mind that a lot of girls tend to listen to both pop, hip-hop and Rn'B at the same time.

As for Rn'B... Ask someone more knowledgeable than me. I'm still fuzzy as to what it is. I mean seriously, people tell me Beyonce and Ciara are Rn'B while at the same time Rihanna is pure pop and Black Eyed Peas are hip-hop. The only real pattern I've noticed is that black pop artists are considered to do Rn'B while white pop-artists are considered to do, well, pop. Even if the beat and the melody are frickin' identical.*

Why I'm bringing it up? Well, you put it with hip-hop and half the girls will be confused as to which category to pick - pop or hip-hop. You put it with pop and they will whine more than any alternative metalhead ever would about how it's you know, totally different and separate from pop.

Now, obviously, this is based on my high school experience and is how I personally would do it. It may or may not be applicable to you. And my school is weird since there were maybe 10 goths in the entire school (this was before emos displaced goths as the dominant teen angst subculture) while at the same time half the Asians (40% of the school population) listened to punk rock and all of the brown people (East Indians/Pakistanis, 20% of school population) pretended to be black and would be classified as wiggers if they didn't actually sell weed.

* - heh, another project idea I had. Make up a story about some up and coming artist from some place far away you've just heard about on the internet. Find a typical top40 melody that no-one's heard of. Put it on your iPod. Now, the experiment. Play it to people while changing one variable - have a picture of either a white girl or a black girl on there. Then... ask what genre they think it is. A cool way of testing racial profiling reminiscent of my social psych prof's research.

PS: how good are you at math? It's possible to do the project another way, which could use MBTI types (or more likely a simplified version grouping people into 4 categories instead of 16) but it involves some fancy statistics, namely the chi-square test and more likely than not your science teacher will have no idea what it is.

[Vaynor]

If it helps at all, I'm INFP and I like classical music

This seems like a pretty good way of doing this, find out their personality type using a standardized test and ask them what their favorite genre of music is.

If you decide to do this, INTP and indie/folk/classical.

[Don Julio Anejo]

This seems like a pretty good way of doing this, find out their personality type using a standardized test and ask them what their favorite genre of music is.

Unfortunately, it's not. This forum is not representative of the general population, while in real life you'll never get enough people to do the MBTI or anything equally detailed, especially in a school setting. Also, most people like more than one genre which is what most of the discussion above is about.

[Vaynor]

Unfortunately, it's not. This forum is not representative of the general population, while in real life you'll never get enough people to do the MBTI or anything equally detailed, especially in a school setting. Also, most people like more than one genre which is what most of the discussion above is about.

Well I didn't mean to imply that the survey should be conducted only on this forum. Admittedly I did not read the thread completely, but I thought it might be interesting to see the personality types most commonly associated with differing personality types. The forum does not have to be representative of the general population in this instance as long as you have enough people representing different personality types, obviously not attempting to get an answer from every possible combination, but maybe a few from a number of simplified personality traits. As the OP stated, perhaps testing introversion vs. extroversion, and seeing what the percentage differences are between two personality types when referring to, say, jazz, or rock.

You don't even really have to refer to the Myers-Briggs Type Indicator test to determine this, I'm sure most people are aware if they're introverted or extroverted. I hope this clarifies my point.

[Don Julio Anejo]

Well I didn't mean to imply that the survey should be conducted only on this forum. Admittedly I did not read the thread completely, but I thought it might be interesting to see the personality types most commonly associated with differing personality types. The forum does not have to be representative of the general population in this instance as long as you have enough people representing different personality types, obviously not attempting to get an answer from every possible combination, but maybe a few from a number of simplified personality traits. As the OP stated, perhaps testing introversion vs. extroversion, and seeing what the percentage differences are between two personality types when referring to, say, jazz, or rock.
Well, that's the general idea :smile: We're just trying to figure out a way to do it so it's both scientifically valid and you get enough responses.

Unfortunately, this forum still wouldn't be very representative of anything even after controlling for personality.


You don't even really have to refer to the Myers-Briggs Type Indicator test to determine this, I'm sure most people are aware if they're introverted or extroverted. I hope this clarifies my point.
Want to know something funny? :smile: They're probably not aware if they're extroverted or introverted. In psychology (and Jung's work too but in different terms) it refers to cortical arousal - whether you need outside stimulation of your brain to increase your arousal to optimal (aka sensation seeking so you don't feel bored) or your arousal is already pretty high and increasing it from the outside actually makes you feel worse. The first one is extroversion and the second one is introversion. And it's only partially related to whether someone is outgoing or not. An extremely shy person that does, for example, skydiving or streetracing is probably an extrovert (barring a few neurological disorders) even though they've never started a conversation with someone in their life. On the other hand, a rock star could be an introvert even if they perform in front of sold out stadiums and have millions of fans if they do it not for the fame and being a rock star but for the creative output.

[Vaynor]

Want to know something funny? :smile: They're probably not aware if they're extroverted or introverted.
*snipped*

True, but that's less the term being ambiguous and more ignorance of the meaning of the terms themselves. Which is why I suggested the MBTI, being a generally sound way of finding out introversion vs. extroversion even if you're not consciously aware of which type you belong to.

[onthetown]

Well, if you're still looking for genres... Celtic, please. You know, the fiddle stuff. I listen to it all the time, I tend to think I'm calmer because of it. :smalltongue:

[Mythestopheles]

I'll post more on the test later as I'm at school right now (I think I may have a really good pdf overview on my other computer), so I'll deal with the styles first.

First question - where are you exactly in Canada?

I'm from Alberta, so it's not too uncommon to see country fans among adults, and even, less likely some people my age.


Now, alternative

Yeah, sounds good, I'll group it with rock and metal.


Indie

Think I'll go with it's own group, as I'm looking for personalities rather than musical similarities.


Classical and opera can be safely excluded. While you're at it, I would take out jazz as well.

I agree with your reasoning. The only people I know of who listen to it tend to be band members and the like.


Blues - do you mean blues from the 50's or do you mean Rn'B? If it's the first one, with jazz.

Yeah, from the 50's. I'll probably take it out. Though I could put it and Jazz as their own category, thoughts?


Now, this part you can do it two ways. What's left: pop, electronic, rap/hip-hop, Rn'B and dance... So as much as it pains me to say it, it makes sense to make a category "pop/dance" and tell people who ask about any kind of electronic to put it under dance.

Sure, this is a science fair, and doesn't need to be perfectly accurate.


Rap/hip-hop deserves it's own category, obviously. But keep in mind that a lot of girls tend to listen to both pop, hip-hop and Rn'B at the same time.

Yep, thought so. As for people listening to multiples, I guess I ask them to say which one they listen to most?


As for Rn'B... Ask someone more knowledgeable than me. I'm still fuzzy as to what it is.

I've actually never heard of it before now. After a quick search, I think think I'll put it with hip-hop


* - heh, another project idea I had. Make up a story about some up and coming artist from some place far away you've just heard about on the internet. Find a typical top40 melody that no-one's heard of. Put it on your iPod. Now, the experiment. Play it to people while changing one variable - have a picture of either a white girl or a black girl on there. Then... ask what genre they think it is. A cool way of testing racial profiling reminiscent of my social psych prof's research.

Sounds fun. I'd guess it would affect it a fair bit. At least to musically knowledgable people.


PS: how good are you at math? It's possible to do the project another way, which could use MBTI types (or more likely a simplified version grouping people into 4 categories instead of 16) but it involves some fancy statistics, namely the chi-square test and more likely than not your science teacher will have no idea what it is.

I'm pretty good at math (at least I think so). And I have no problem with confusing my science teacher.:smallbiggrin:

Thanks. Taking your advice, I now have this list:
Alternative/Rock/Hard Rock/Metal
Country/Western
Pop/Dance
Hip Hop/R&B/Rap
Indie

Edit:
Riyoukaze: Celtic is folk music correct? It's traditional music, which I believe is counted as folk music. Let me know if I'm wrong.

[onthetown]

Kind of depends on your opinion. I've seen "folk" music from all different countries. Celtic is specifically from Ireland or Scotland (or both?). Cape Breton's rather famous for it after Natalie MacMaster.

I suppose it would be traditional, though. Still, a lot of countries have their own folk music, so I'm not sure if Celtic could be counted in there.

[purple gelatinous cube o' Doom]

Thanks for taking the time to help me out.

As for what to measure, the only idea I have is to maybe link certain parts of the personality types (Extroverted vs. Introverted; Intuitive vs. Sensing; Feeling vs. Thinking and Perceiving vs. Judging) to say, genres of music, or vice versa.

Thanks for the reassurance. I actually probably had one of those not-so-project-projects in past years. :P

Unless you're going to give everyone that does the study for you the Myers-Briggs test, I suggest scrapping that idea. The E v I is an easy one to determine, but unless you're well versed in behavioral sciences/consulting, the rest of them are quite difficult to determine who is actually what just based on the "eye" test so to speak.

[Mythestopheles]

Recently I discovered a site, Personality Cafe (personalitycafe.com) . It has a lot to do with MBTI. If I use MBTI, should I post the survey there? Because the people there seem to know their personality types, and it seems pretty varied. I'm just worried that an online survey wouldn't be a good representation of the average person.

Edit: Alternatively I could try out the Big Five Personality test.

[TheThan]

I agree that this might not be possible, or at least not possible within the confines of a science fair project. Though it would be cool for a long-term science project. So how about you try something simple, like the effects of gasoline on fire (http://www.homestarrunner.com/sbemail91.html).

[Flarp]

If your science fairs are anything like those that I remember, you're going to need measurable results.

Personality, unless you're a psychologist (and maybe not even then - I don't think there's a "scale" for personality), cannot be measured in any objective manner.

I tried to do a similar thing in school, involving color. I think my eleven-year-old innocence died when the thick tome on pigmentation and optical theory were pulled out.

Music preference vs. GPA might be interesting. And funny.

Scientific method helps too, y'know.

[Don Julio Anejo]

Personality, unless you're a psychologist (and maybe not even then - I don't think there's a "scale" for personality), cannot be measured in any objective manner.
Yes it can. You just have to realize that what tests like MBTI or NEO-PI-R are measuring is not personality per se, but rather certain aspects of personality that are considered baseline by their respective theories and then comparing them to an average person.

For example, you can ask (using a number of questions) people about their assertiveness. Yes, whatever test you make up probably has bias. Yes, it won't be exact. But that's not the point. The point is that assertive and dominant people will, on average, score higher; timid, submissive people will, on average, score lower and average people will score somewhere in the middle. Letting you place test results on a normal curve and use them for fancy statistics to test your hypothesis.

What the tests are _not_ measuring, however, are individual people's personalities. It's not astrology, they won't tell someone what they didn't already know and they won't be able to accurately describe personality. Generally, the more accurate a test is, the more specific it is and the less utility it has. The whole point is statistics. You can find out, for example, that extroverted people like fast music and introverted people like slow music without ever pinpointing just how extroverted any given person is, just by seeing a correlation between tempo and average extroversion. It doesn't mean that any given extroverted person will like fast music, nor does it mean that any given extroverted person won't like fast music. But it does mean that when taken as a group, extroverts will probably like Guns n' Roses more than they will like Coldplay.

[Flarp]

Yes it can. You just have to realize that what tests like MBTI or NEO-PI-R are measuring is not personality per se, but rather certain aspects of personality that are considered baseline by their respective theories and then comparing them to an average person.

For example, you can ask (using a number of questions) people about their assertiveness. Yes, whatever test you make up probably has bias. Yes, it won't be exact. But that's not the point. The point is that assertive and dominant people will, on average, score higher; timid, submissive people will, on average, score lower and average people will score somewhere in the middle. Letting you place test results on a normal curve and use them for fancy statistics to test your hypothesis.

What the tests are _not_ measuring, however, are individual people's personalities. It's not astrology, they won't tell someone what they didn't already know and they won't be able to accurately describe personality. Generally, the more accurate a test is, the more specific it is and the less utility it has. The whole point is statistics. You can find out, for example, that extroverted people like fast music and introverted people like slow music without ever pinpointing just how extroverted any given person is, just by seeing a correlation between tempo and average extroversion. It doesn't mean that any given extroverted person will like fast music, nor does it mean that any given extroverted person won't like fast music. But it does mean that when taken as a group, extroverts will probably like Guns n' Roses more than they will like Coldplay.

Well, yes, you could measure aspects of personality, certainly. The problem is that the actual unmodified "numbers" of personality must be displayed, and there's really no way to make a nonarbitrary scale for that kind of thing.

Though, if raw data is not required, making a correlation between these statistics and music genre is a totally sound concept that I am completely on board with.

Actually, I think MBTI (http://en.wikipedia.org/wiki/MBTI) would be really useful here, as so many other posters have said.

[Mythestopheles]

Actually, I think MBTI (http://en.wikipedia.org/wiki/MBTI) would be really useful here, as so many other posters have said.

I agree with this. I'm kinda thinking, have a bunch of people of every personality type (Probably hard to pull off even online), ask them their favorite genre. Check out the results, and I'll see if there's any overlap between music types.

What do you guys think of this method?

[Mythestopheles]

Ok, later today, I'm planning to open up a thread here and on some other forum(s?). Where people will be able to say their genre of choice, and their personality type. I'll probably link it to an online test.