1

I would like to conduct a study examining the correlation between the Democratic candidates people support (and the policies these people espouse) and the positions of the candidates themselves (based on voting records, public discussions, etc.). My issue is finding a platform/mechanism that would enable to conduct such a survey.

I don't have a particularly large budget (I'm currently in college), and all the avenues I've explored have several issues. I've looked into using Facebook and Twitter, but I don't have large presences on either site and I'm not convinced of the quality of answers that I would receive there anyway. I've given SurveyMonkey a cursory glance, but it seems to be more geared towards market/business research (although perhaps I'm reading that incorrectly).

I apologize if the subject material is not ideally suited for this exchange, but I could not realistically think of a better one given the subject material. Any suggestions would be immensely appreciated.

5

TL;DR: You should not expect to be able to do accurate polling on a college student's budget.

Live polling

A quick search finds this claim:

conventional surveys by full-service research firms can cost $10,000 or more

This would be a poll done by live operators and with careful attention to balancing demographics. I'm guessing that this would give around 500 responses at the $10,000 price point (because that's around the smallest sample polled).

Robo polling

It continues to say

Precision Polling says its average customer spends about $1,000. The company charges 10 cents per call for an automated poll like the one I conducted.

This is robo call polling. See also

Meanwhile, Rasmussen Reports — a firm that’s already controversial for its prolific use of automated political polls — now allows customers to create their own surveys through the website of a spinoff company called Pulse Opinion Research. Pulse charges as little as $600 for a single-question poll and $1,500 to $2,500 for more detailed ones. Unlike the surveys that carry Rasmussen’s name, Pulse’s questions are not reviewed or approved by the company.

Another example:

At an average price of $2,000 – $3,000 for most polls that we do for our clients, PPP is the high-value, low-cost poll you need.

and

PPP polls are generally one-tenth the cost of a traditional poll.

This suggests that polls typically cost $20,000 to $30,000 (may be exaggerated for marketing purposes). This is an example of a left-leaning political pollster.

It is worth noting that many consider robo polling to be less accurate than live polling. I don't want to argue for or against that claim here. I just want to note that it is made.

Online polls

SurveyMonkey will poll almost anything, since they allow you to specify the questions. (There is probably some kind of mechanism for eliminating offensive questions.)

The problem here is that online polls are somewhat self-selected. So you may find the responses to be biased. SurveyMonkey offers a version for pay that is supposed to be more accurate than the free version. It's unclear how much more accurate.

Again, many polling advocates order polling quality from live, robocall, and put online polls last. There are arguments against this, but it is a common position.

Poll design

I have mostly discussed the cost of polling itself. But an important part is the design of the poll. If you want a really good poll, you need an experienced pollster to design and test the questions. You should probably do a focus group with your questions to better understand how survey takers will understand them. Because if a poll taker thinks you mean one thing when you really meant something else, then the responses will be gobbledygook from your perspective. See How to treat illogical survey responses? for an example of a survey where the survey takers and the pollster apparently had different ideas about what the questions want.

If you just want to do a survey and aren't terribly worried about accuracy, use SurveyMonkey's free polling. If you want accuracy, you really need to work with a more experienced pollster. As you are a college student, you might be able to go to your college's political science department and find a professor willing to do independent study with you. Of course, the professor might tell you that you need more theory first.

Polling may seem simple. It's just asking people questions, right? But the truth is that polling often produces results that are difficult to interpret, even for professionals who do polling full time. Also, it is easy to do things that influence the results unintentionally. For example, some people prefer to always pick the first option. So many surveys rotate their options.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .