When discussing political campaigns where voting is optional, it is often advocated that everyone (who can legally do so) should vote:

It's reasonable to conclude that the people pushing such campaigns must believe that there is a significant difference between the voting preferences of those who already intend voters and those who intend not to vote. Otherwise, such a campaign will have no effect on the result, and hence be a waste of time.

I firmly believe that voters and non-voters do not have an identical distribution of voting preferences; the opposite is statistically very unlikely. But, I'm not sure whether the discrepancy is significant to ever affect the outcome of an election, since I had never seen any scientific evidence of this.

Has it been shown that reluctant voters tend to vote differently that keen voters?

Note: Given such voter-encouragement campaigns can be targeted at certain demographics (age groups, races, certain cities and neighbourhoods, etc.), showing that there is a difference in reluctant voters/keen voter preferences in defined subgroups should also suffice.

  • [removed all comments pertaining to closure and reopening]
    – Sklivvz
    Feb 16, 2012 at 22:04
  • The argument for compulsory voting is that it tends to draw the parties towards the centre. Anecdotally, this is what happens in Australia (which has compulsory voting. Both major parties are much closer to each other than they are in the US.
    – Stephen
    Nov 8, 2016 at 0:34
  • 1
    @Stephen The Republican and Democrat parties have had periods of both being much more centric. We are not in one of those periods.
    – user11643
    Apr 26, 2019 at 3:06
  • 1
    It's far more reasonable to think the "voting preferences" of those who choose to not vote is that they have no voting preferences at all, hence, they don't vote. The reasonable assumption is that people with preferences do take the time to vote.
    – user11643
    Apr 26, 2019 at 3:09
  • 1
    The one preference we can be fairly sure about is that non-voters don't care enough to vote. Except in cases where the reason why they don't vote is because there are barriers to voting or registration (which do exist even in fairly democratic countries like the USA).
    – matt_black
    Apr 30, 2019 at 10:07

1 Answer 1


First, I'm just going to say that there are several of studies on this topic from J.E. Leighley and J. Nagler that breaks down differences in voting turnout on several different factors (as well as dozens more by other authors). I'm providing this information for comprehensiveness, rather than to answer this specific question directly, because the issue is certainly multifaceted. Below is a small selection of these articles for the interested reader.

Individual and systemic influences on turnout: Who votes? Leighley & Nagler, 1992

Socioeconomic class bias in turnout, 1964-1988: The voters remain the same. Leighley & Nagler, 1992

Unions, voter turnout, and class bias in the US electorate, 1964–2004. Leighley & Nagler, 2007

Specific to your question though, the best resource I can find is this: Who Votes Now? And Does It Matter? Leighley & Nagler, 1007

The authors analyzed the NAES data from 1972 and 2004 to detect changes in voter turnout. They do not provide data on statistical significance, but does display mean differences for various beliefs, attitudes, and associations between voters and non-voters. Given the sample size of most of the analyses (N >= 1000 for most analyses), we can assume that most mathematical differences greater than 3-5% are also statistically significant in the population (if one feels this assumption is inappropriate, I would be happy to run some analyses when I have a bit more time). This assumes, of course, that the NAES sample is representative of the population which I cannot verify conclusively (but its longevity supports this assumption).

As for the findings provided, there does appear to be differences and they can be seen in their entirety in the tables from page 24 through 28 in the linked pdf. To summarize these tables indicate a difference between political affiliation and voting behaviors such that voters tended to have strong republican or democratic affiliations (e.g., 20% of voters were strong republican v. 4% of non-voters, 6.5% of voters were independent v. 21% of non-voters). As far as attitudes on specific issues, there do appear to be some systematic differences but the directions are varied. Some examples include:

  • 51% of non-voters v. 44% of voters prefer government run health insurance
  • 32% v. 19% of non-voters believe that abortion should always v. never be legal, while the distribution for voters is 39% v. 12%
  • 73% of non-voters v. 60% of voters favor unions.
  • 68% of non-voters v. 50% of voters favor recalling troops from Iraq.

So to answer your question in the most general sense. No, non-voters and voters do not have identical distributions of viewpoints.

  • 1
    Please note: all the above applies to the United States and is not a general answer -- the question seems to be a bit more general than that. +1 though.
    – Sklivvz
    Feb 16, 2012 at 17:37
  • Interestingly every difference where non-voter's favored something more then voters it was a political view that would usually be seen as the democratic view. That suggests to me that non-voters are more likely to be democratic then republican. You didn't specify what percentage of non-voters tended to identify as democratic vs voters, but I'm now curious.
    – dsollen
    Apr 26, 2019 at 15:43
  • 1
    @dsollen: Your statement is only 75% correct over the sample... as far as I understand, "abortion should never be legal" isn't exactly a Democrat POV.
    – DevSolar
    Apr 29, 2019 at 6:36

You must log in to answer this question.

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