Please take my survey (and spread the word!)

[yesthattom]

I’m curious about “recommendation engines”, the systems that enable Amazon and NetFlix to make recommendations like “People who bought _____ also bought ____”. I’m conducting experiments to build such a system. To do that, I need raw data. Would you help me by filling out my survey?

Click HERE to take the survey NOW (U.S. citizens only).

I’m doing this in my spare time as a hobby. I am a computer system administrator and author but this is a side-project. I’ll keep the data private and secure. I’m not collecting names or email addresses thus the data is useless to marketing companies and spammers. There is a link to a privacy policy on the survey.

All that I ask is that you answer the questions honestly.

For best results, I need to collect information from at least 10,000 people of a large variety of political, ethnic, and geographic backgrounds. Since most of my friends are in a similar political leaning, I really need help finding people outside my social circle to fill out the survey. If you could spread the word (blogs, etc.) between now and the end of June (I’m shutting off the survey on July 1, 2008).

Thanks!

Click HERE to take the survey NOW (U.S. citizens only).

Comments

[catwithbell.livejournal.com]

*blink*

I'll be fascinated to see how you correlate some of this information. Outhouse, indeed....

[yesthattom.livejournal.com]

By the end of the year I hope to publish my analysis.

[bikergeek.livejournal.com]

Sorry, but that survey really needs a "prefer not to say" option for "Household income". I stopped right there--I consider that a way-personal piece of information between me, my employer, and the IRS.

Also, when considering "household income", think of how many people we both know who live with roommates. If you live in a shared residence with a bunch of roommates, do you count their income as "household income" as well? What if you have no clue what your housemates make, and don't care, so long as they pay the rent and bills on time?

[yesthattom.livejournal.com]

I've added a "prefer not to say" option. Thanks for pointing that out. It's a good point. (Though, since I don't ask for email address or name, it isn't like I could figure out who you are).

The "house hold income" is a standard demographics question (I searched the web for "standard demographic questions" to see what experts do).

Do you have housemates or is this theoretical?

[bikergeek.livejournal.com]

It's theoretical, since I live alone. And yes, I'm aware that it's a pretty standard survey question. Has been for a long time. It's just that the concept of a "household" has become a lot more fluid in the last quarter-century or so. Maybe the "standard" survey question should evolve. :-) I mean, it was relatively easy for survey creators to modify the "ethnic group" question to include "I'm multiracial" and the "Gender" question to include "Transgender" and/or "I'm something that doesn't fit neatly into one of the other options".

[qwrrty.livejournal.com]

"Household income" traditionally means the total income for the pool of money shared by your household. If you live with two roommates, split the rent equally and pay for your own utilities and groceries, there's no shared household fund and your "household income" is just your personal income. While there may be more ways to configure a "household" nowadays, I don't actually think that the concept of "household income" has changed drastically in the last generation.

[ckd]

I said that I take a train to work. That's a lie. I walk, because the nearest subway station is about three times as far away from home than my office is, and in the same direction.

(I don't drive. The garage where people who do drive park is also farther from the office than my house is.)

I don't see the point

[2nd-story-story.livejournal.com]

I had no problem filling out the questionnaire. My question is, how does what you asked relate to recommendation engines. I would think the problem breaks into at least two parts -- how do people cluster (demographics like you have in your questionnaire) and what do people in a cluster purchase. You didn't ask about the purchasing part.

Incidentally, my complaint about Amazon's "engine" is that it doesn't distinguish things I buy for myself from things I buy for someone else from that person's "wish list". I have no interest in purchasing more of the latter unless and until my giftee adds it to her/his list. Since most of my Amazon purchases are for others, the Amazon's recommendations for me are off base.

Cheers,
Alan

Re: I don't see the point

[sweh.livejournal.com]

You can go into your Amazon purchase history and tell it that items were gifts for others. I think this takes it out of your recommendation preferences. If that doesn't there is also a "don't use this for preferences" option.

Re: I don't see the point

[whc.livejournal.com]

Yeah, I was expecting some questions about recommendation engines too!

Re: I don't see the point

[yesthattom.livejournal.com]

Nope. That's not the data I need for this current experiment.

I'll publish results by the end of the year.

Re: I don't see the point

[yesthattom.livejournal.com]

I'm examining clustering within a selection of demographic questions.

[vees.livejournal.com]

If I were an intolerant person I might have abandoned the survey mid-way through.

[yesthattom.livejournal.com]

What would have turned you off?

[vees.livejournal.com]

I think there was a run of alt lifestyle check-boxes where I thought to myself that if I had a problem with or was disgusted by that sort of thing on principle I might have abandoned the survey. I didn't think it was anything in particular.

[yesthattom.livejournal.com]

I thought there was only 1 alt lifestyle checkbox. Interesting.

[tn-grrl.livejournal.com]

i'm assuming that you're using this data for the experiments you'll be conducting, rather than this having anything to do with our use of recommendation engines.

personally, i always have a problem with this type of "choose one" question:
# I am married
# I am divorced
# I have never been married

technically, i'm divorced, but to use that makes me feel as if my 19+ year unmarried other sex relationship is invisible. it's lasted about 5 times longer than my marriage, so i left those boxes blank. some surveys do now ask if one is partnered in addition to other options.

[fengshui.livejournal.com]

The system was implemented as check boxes. Was there a technical prohibition from selecting all three?

[tn-grrl.livejournal.com]

i'm pretty sure i could have checked all three boxes if i had wanted to.

checking 2 wrong boxes to make up for the fact that i don't feel comfortable checking the only box that is technically correct simply isn't the choice i made.

[yesthattom.livejournal.com]

I decided to let each item be individually checked on purpose. Mark whatever you feel best describes your situation.

[fengshui.livejournal.com]

Fair enough.

Tried and failed...

[babbage.livejournal.com]

... to be honest.

Your nomenclature for tertiary education doesn't translate very well. The options appear to be completed "high school", completed "college" or completed "graduate school". AFAICT, that fails to distinguish those with undergrad degrees from those with very little education. I think americans just use different words for things than us folks here. Over here one leaves high school at 16, leaves college at 18, and then goes to university for an undergraduate degree and then a Masters/Doctorate. We just don't say "grad school" (and in fact never use "school" to describe any kind of tertiary education) and "college" seems at best ambiguous. If you're looking to get lots of responses, you should reword this. As it is I can't complete the survey as I can't be sure I'm correctly honouring your request for honesty.

Other nits:

• "I plan on voting in the 2008 Presidential election" fails to specify which country we're talking about (and FWIW, you can't use country of residence to determine if the respondent is eligible to vote in US presidential elections, in case you were using the first answer for that). "I am registered to vote....." ... where, for what?
• I'm also a pedestrian commuter.
• You might consider more ethnicity options. For example, an "other, non-multiracial". Natives of Yemen might not be able to categorise themselves, as one example.

Re: Tried and failed...

[sweh.livejournal.com]

And even if the survey is meant to be limited to those living in America (which is says at the top; dunno if that was added later), similar failings still apply to resident aliens (can't vote, different schooling) or naturalized citizens (can vote, still maybe different schooling).

Re: Tried and failed...

[yesthattom.livejournal.com]

Actually, I'll be eliminating people from outside the U.S., so that's ok. Maybe I should be more clear.