These frequently asked questions and their answers were inspired by, and draw from the Pew Research Center for the People and the Press in an effort to keep the readers of Pack Poll informed about our practices, as well as the newest information in public opinion polling.
A reasonable question, perhaps the most frequently asked one, is how can 1,000 respondents accurately stand for the entire student body of around 25,000? The Pew has a good answer!
- Why am I never emailed for Pack Poll?
- Can I volunteer to be polled?
- Why don’t your surveys ever reflect the opinion of people I know?
- Why should I participate in surveys/what good are they?
- Do pollsters have a code of ethics? If so, what is the code?
Collecting Survey Data
- How are people selected for your polls?
- Don’t you have trouble getting people to answer your polls?
- Why don’t you just conduct surveys on your website instead of emailing me?
Why am I never emailed for the Pack Poll?
We sample from the entire undergraduate population of NC State students, totaling approximately 25,000 students according to the Office of Institutional Research and Planning, and contact anywhere between 2,000 and 4,000 people with each survey.
Can I volunteer to be polled?
To ensure our polling results accurately reflect the population, we strive for probability samples so we do not accept volunteers in these specific surveys, although we have created an opt-in panel of students that like to take our surveys. Historically, “non-probability samples” cannot be used to generalize to the broader population, which in this case is the student body. Unless we say the data come from the panel of volunteers, we utilize a randomized sample so that every type of person has an equal chance of having their viewpoint represented. Allowing someone to volunteer, while much easier on us, would violate that principle in a very serious way. Not to mention, your willingness to be surveyed means you are very likely far more likely to be politically engaged than other students, and thus not representative of the larger student body.
Why don’t your surveys ever reflect the opinion of people I know?
Odds are good your friend group doesn’t represent the entirety of the student body that was surveyed. It’s very likely that your friends and classmates are relatively similar to you in many ways. If you were to have a group of friends that represents the entire student body, you would have acquaintances who are black, white, Asian, rich, poor, Muslim, Catholic, from the coast and the mountains, cities and farms, and many combinations of the above, which few people do.
Why should I participate in surveys/what good are they?
Polls are a great way to make your thoughts and opinions heard about a wide variety of topics currently affecting the school, city, state, nation, and world. They can also be influential in decisions being made around you, and without your voice, your opinion may not be represented, and thus a valuable contribution to the conversation will not be made.
Do pollsters have a code of ethics? If so, what is the code?
Any professional organization of survey researches has very clear codes of ethics for their researchers. These codes cover the obligations to respondents, respondent’s privacy, and any potential relationships/partnerships with clients as well as their responsibilities when it comes to reporting poll findings to the public.
At Pack Poll, we attempt to adhere to AAPOR standards as much as possible in our work, and make certain to ensure the privacy of your survey responses.
How are people selected for your polls?
Our selection process is relatively straightforward. A random sampling list is provided to us by the university of all undergraduate students, containing a specified number of students to be surveyed, which can range between 1000 and 5000 depending on the particular poll, the exact number of those asked to participate and those that actually responded are recorded on every survey report.
Why don’t you just conduct surveys on your website instead of emailing me?
Conveniently, the fact that everyone in our potential survey pool has access to internet via the university, and we are reaching out to them via the university email system, which they presumably check on the regular, allows us to survey people online. However, if we were to simply post it to our website, instead of reaching out to you, we would run into issues with ensuring our survey truly represented the student body of NC State. For more information, see the “So how does it all work?” section.
Are election polls accurate?
Polling has become an increasingly large part of our election year discourse, and with it comes skepticism towards polling, as well as pollsters of ill repute. However, among those that continue to work towards quality polling, the final forecasts of most major national polls prove to be quality predictors of the final vote.
How do you know who is really going to vote?
That’s the hard question. Every polling group has their preferred set of questions to weed out those unlikely to show up on election day and cast their ballot, and we often draw on national polling organizations like Pew or Gallup, as well as our prior work, in order to craft our questions to best identify those we can expect to be likely voters.
For instance, we usually include questions regarding whether someone is already registered to vote, as well as whether they have voted previously, or in the primary, all three of which can be valuable pieces of information to fit into the larger puzzle.
Do people lie on surveys?
As we survey individuals, we invariably will encounter some false information, either intentionally provided, or simply because someone misremembered on a topic we asked. We have the luxury of not using in-person interviews with Pack Poll, which has its own set of drawbacks, but minimizes the effect the interviewer can unintentionally have on the individual being surveyed, particularly when it comes to sensitive topics someone is more likely to lie about when asked in person by a stranger, compared to the anonymous form submitted to us when you take our surveys.
This is a particularly large problem with more taboo topics like illegal drug use and sexual behavior, which are often underreported, and topics like charity work and voting, which are often over-reported. Some polling groups will take the effort to verify voting records of those surveyed, and others will use more neutrally-phrased questions to try to avoid this pitfall.
Do people really have opinions on all of those questions?
People have opinions and attitudes on almost everything we are asking about, which is often why we ask the questions that we do. That being said, we try to provide options such as “don’t know” or “no opinion” on many of our questions in order to allow people to most accurately represent their attitudes, including a lack thereof, or an unfamiliarity with a topic at hand.