Why tell pollsters the truth?

Great Debates
1

Okay someone phones me up from a a political pollster organization. Asks me who I support in the next election.

Am I better off telling them the truth, so that the poll shows more support for my party , and folks who just go along with the herd will vote my way.

Or should I lie, so that people who wouldn’t otherwise vote, or who would vote for a fringe party are more likely to vote for my party. And those who would vote for the other party, feeling more secure in their lead, are less likely to get out of bed to vote.

Am I missing something. Is there any reason why I shouldn’t loudly support the party that I am against?

2

As a member of this board, aren’t you dedicated to “fighting ignorance?”

If so, why would you deliberately misinform anyone?

3

Good question. I never have lied to a pollster, but I guess I see this as more of an all’s-fair in love and politics, game theory kind-of question. Is there a political advantage to telling the truth or lying?

4

Welcome to the wonderful world of statistics, where you learn that taking the poll and making the graphs are child’s play, and analyzing whether the statistics are valid, what factors may have changed the results (even the phrasing of the questions), what correlates and what causates, etc. This is also why some statistics firms are more trusted than others - generally, AFAIK (I’ve never actually been polled), they don’t call and ask “who will you vote for to be president” - or “who do you support for president” getting a very different answer - jot down your answer, and hang up. So lying is a little bit harder than just saying the opposite name. While they have you on the phone, they’ll probably also ask other questions about policy, which would be harder to lie about.

In any case, statistics are like TNT. Effective, but dangerous in the wrong hands. O_o

5

Well, here in the UK we’ve been promised a referendum on Europe at some point. Thing is, you can bet that the incumbent party (if it’s Labour) won’t give us one until the polls tell them they can win it. There’s quite an incentive for people who want to vote <no> to tell any pollsters that they’d vote <yes>, wouldn’t you agree?

6

I’ve often wondered about the validity of polls in general, for a few reasons:

1 - the sample is biased right off the bat to people who are willing to answer the damned thing. I usually refuse to participate in any poll for anything, on the general grounds that somebody who wants to detain me for some inane poll can simply blow it out their ass. I suspect that you would find a significant difference in the honest opinions of the populations of people willing to participate, and the people who tell the pollster to get lost (who may, nevertheless, eventually vote, purchase the type of product the survey is concerned with, etc.). Unless you could compel people to participate in the poll, I don’t see how you get an unbiased sample.

2 - whenever I hear one of those polls with absolutely preposterous results like “90% of Americans think Fidel Castro wrote the Declaration of Independence”, I wonder how much of it stems from people messing with the pollster’s heads. Another reaction to being annoyed by a pollster can be to simply give idiotic answers to amuse yourself. I’ve been known to do that a couple times, too - “Yes, I believe the recent floods are God’s retribution for our sinfulness” (snerk, snerk).

3 - you might lie on the damned thing just to get through it in the quickest way possible, if you know that certain answers are going to elicit more detailed questions. I’ve been taking my car into the dealer regularly for its scheduled free maintenance, and I absolutely can not keep both the local dealer and the national organization from conducting two separate “customer satisfaction” surveys afterwards, and there doesn’t seem to be a tactful way to tell them “no, I don’t want to answer the survey, goodbye”. I’ve learned to just answer all “scale of 1 to 5” questions with “5” to avoid being asked a followup question about what I didn’t like about the service (“Finally, is there anything we can do to make our service better for you?” - “Yeah, don’t conduct these surveys.” usually nets a very defensive answer, too).

4 - I don’t care how anonymous a poll is, people are going to lie simply because they are ashamed to tell the truth, rather than reveal their dirtiest secrets. In political terms, you will have people saying that they are going to vote for the candidate reflecting their professed ideals, say the Green Party candidate, then get into the poll come election day and vote for the incumbent Republican congressman because he delivers great pork barrel.

Add these to the OP’s concerns about strategic lying on the part of the poll recipients, and one has to wonder.

7

Hundreds of millions of dollars and decades of manhours have gone into the field of statistics, and verifying statistics, and creating reputations of accurate statistics. Additionally, not all statistics are collected through polling. Additionally additionally, they can usually tell when you are BSing them. :-p

8

Why should someone deliberately inform someone when that someone is the phone spamming them?

9

Since there’d likely be several hundred people (at least) sampled, a few goofy answers are not going to make much of an impact on the results, particularly if some of the goofiness is in one direction and some is in another direction. They’ll just raise the margin of error a little bit.