Another Reason to Be Skeptical of Poll Results:

Here's a question from a Pew Research Center, Pew Forum on Religion and Public Life survey, Oct. 15-19, 2003:

Society should not put any restrictions on sex between consenting adults in the privacy of their own home. Do you completely agree, somewhat agree, somewhat disagree, or completely disagree with this statement?

Completely agree - 61%
Somewhat agree - 19
Somewhat disagree - 5
Completely disagree - 8
Mixed opinion (vol.) - 1 Don't know/Refused - 5

So can we confidently say that 61% really do completely agree with the statement -- to the point of supporting the legalization of prostitution? I'm pretty sure the answer is no: I doubt that most respondents would have realized that prostitution is covered (though it clearly is).

OK, can we at least that 61% is the rough fraction of people who completely agree with the statement as to those matters about which the statement is most often made -- homosexual activity, nongenital sex, and the like? Not really: The number may well be higher, but some people who believe that might have said "somewhat agree" or even "somewhat disagree" precisely because they realized that prostitution is covered. What that number is, we have no idea.

I do think we can guess that probably at least about 60% of the public oppose bans on homosexual activity and nongenital sex; I'd think that when most respondents heard the "consenting adult" language, they'd probably have thought of those things. But we can't even be confident of that, it seems to me.

Surveytaking is necessarily imprecise, partly because people don't accurately report what they really believe (or how they'll really vote). But questions like this, it seems to me, just exacerbate the likely inaccuracy.

Downtown Lad (mail) (www):
Why is prostitution covered by this? It's the monetary transaction that is prohibited by laws against prostitution - not the sex.
8.22.2005 5:00pm
Chris Lansdown (mail) (www):
Actually, I don't think that this includes prostitution unless part of the deal with the prostitute is selling a portion of the ownership of the home to the prostitute prior to the sex.

Moreover, it doesn't cover what restrictions might be placed on the sex that people can have in apartments, hotel rooms, etc.

This does suggest an interesting slogan, "To do the deed, you need the deed."
8.22.2005 5:01pm
Chris Lansdown (mail) (www):

While I don't think that this covers prostitution because its rare that the sex takes place in a home which belongs both to the payer and to the prositute, your objection is pretty easy to overcome — just declare the exchange of money to be part of the sex act, rather than something ancillary to it. If lesbians use dildos it's still considered sex (I think); so why not if two people happen to use a rolled up wad of bills?
8.22.2005 5:03pm
NickM (mail) (www):
I too think that the reference to "the privacy of their own home" implies that the persons involved are living together, which falls outside the normal conception of prostitution.

Since adult siblings can share ownership of real property (not uncommon if parents died intestate or had a simple will), this question would cover adult sibling incest as well, though I doubt many people would think of that when answering the question.

8.22.2005 5:06pm
Chris Lansdown (mail) (www):
Actually, this raises another point — this question doesn't say anything about whether restrictions on casual sex are permissible. I.e. John owns his home. John brings home Joan. John &Joan engage in some behavior defined as "sex".

The survey question doesn't seem to cover this either, though I have no idea whether people's opinions of what sex should be punishable by years in prison differs in cases like this.
8.22.2005 5:08pm
Downtown Lad (mail) (www):
If an "exchange" of money takes place, the transaction can be considered commerce. And let's face it. If the government can regulate rents, or the price of wheat, or have laws against medical marijuana since they are considered "commerce" then they can certainly outlaw prostitution.
8.22.2005 5:08pm
Caliban Darklock (www):
I think prostitution requires a public transaction by nature. The prostitute must advertise her services somehow, and some form of negotiation with the potential john must take place. These aren't likely to happen over private channels without establishing a red light district of some sort.
8.22.2005 5:17pm
Craig Oren (mail):
what about brother-sister incest in the privacy of the home? Perhaps that's a better example than prostitution. Are the 61% voting for that?
8.22.2005 5:18pm
just me (mail):
This survey Q hits my pet peeve - the failure to distinguish between "society" and "the state/government." Many people treat them as equivalent, but they're not. This Q merely asks whether "Society should not put any restrictions" on sex between consulting adults. I think most survey listeners/readers probably heard that to mean "should we have laws against . . ." But I think it's equally plausible to read that as asking whether "society" should do ANYTHING, e.g., if you do XYZ and tell us about it, we won't invite you to our cocktail parties.

This same blurring usually occurs whenever someone wants some government "activism," but doesn't want to make the case for why the GOVT has to be the agent. So they just trumpet that "society should do more for the poor/children/whatever"--with which I might agree-- and then try to glide effortlessly into "THEREFORE, we must pass the Money For A New Agency Act,"--and the "therefore" is the problem.

I seriously doubt that such a high percentage of people think there ought to be no societal scorn for ANYTHING consensual. If I let it be known that I had a big old orgy last week, I might not be invited to certain parties (but might receive increased invitations from other sectors!). I think that the group that would "scorn" me might be higher than these #s show. Same for adultery -- if everyone knows that I cheated on my spouse, I might lose social standing.

But again, we don't know from this survey whether people really understood the question in terms of government action or of private action, as the question did not go out of its way to clarify.
8.22.2005 5:21pm
Eh Nonymous (mail) (www):
I think you have to be a little more careful with wording of the question to get everyone on board and not include certain taboo or criminal or immoral acts, including family incest and so on.

But it's not far off; most people think that

Consensual private acts between mature adults not acting involuntarily or under duress, in a home they own (as Chris pointed out, this post was wildly incorrect when it suggested the prostitution was covered, and Downtown was missing the real reason why; it's not the cash, it's the commitment)

is not something the gov't ought to regulate.

For that matter, remember the Onion article about a woman being caught in a sex-for-security scam, written up as if it were another suburban madam scandal? She gets consumer goods, he gets his illicit needs met?
8.22.2005 5:24pm
just me (mail):
Let me add that in my above example, I understand that the info-leakage might be considered the "public" aspect, that is, some might say it's OK that you did the act privately, but the telling is the problem, as such things must be kept quiet. To close that loophole, I suggest that if someone made best efforts to keep it quiet, but the info leaked out -- e.g., because a partners (or partners!) told, or because of the hidden press stakeout at my house -- that attempted concealment would probably not absolve me the eyes of all those who'd condemn the behavior. That is, of those who say "I won't socialize with someone who does X," I doubt that there's a very large subset of them who say, "but of course, if he tried to keep the orgy quiet but was ratted out, then he gets full forgiveness and stays on the invite list."

So I think my point still stands.
8.22.2005 5:28pm
Jeffrey King (mail) (www):
I would also point out that the question is one of policy, not constitutionality. So while legislators might want to take heed, this poll has no direct applicability to, say, Roberts' qualification for SCOTUS based upon (a theoretical) take on Romer... although technicalities like this wouldn't stop most politicians from making such a claim.

For this, and the reasons stated above, we math majors often define statistics as the art of lying with numbers.
8.22.2005 5:30pm
Chris W (mail):
As "just me" noted, you're pulling a bait-and-switch between the question and your interpretation. Any Libertarian would answer "disagree" to the question, but if the word "legislation" was used in place of "society", they would answer "strongly agree".
8.22.2005 5:30pm
Goober (mail):
John Zaller, of course, has the classic work on this. Prof. V. is correct; because no one's had a serious debate on legalizing prostitution in a while, respondents hear "consenting adults" and what comes to mind is homosexuality, or heterosexual sodomy.

Also, note it's "society" and not "government" that shouldn't put any restrictions on behavior, according to the poll. I'm guessing that would include moral restrictions, etc.
8.22.2005 5:30pm
Gabriel Rossman (www):
professor volokh raises a valid point, people will readily agree to broad principles even when they would actually disagree with specific applications. For instance, in the 1994 GSS, 36% of respondents claim that there are no circumstances under which they would approve of a man hitting another man. but then when they ask whether they would approve of a man hitting a man he saw beating a woman, only 16% claim pacifism. so about 20% of respondents give a different answer to an abstract principle than to a hard case. it's very easy to imagine people who think they believe in the "consenting adults" principle but then scoff at prostitution, incest, or some particularly obscene act.

this yet another reasons why the marginals of opinion surveys are very unreliable (fortunately the covariance is usually ok and that's what social scientists care about -- in other words, you may not be able to say that 80% of Americans think something or other, but you can confidently say that more men than women or blacks than whites think something or other).

there's also a specific practical problem in that in order to save time and money, many survey researchers like to use "skips," where only someone who answers yes to an initial question is asked subsequent, more specific questions. (for instance, in early years of GSS only people who approve of violence in the abstract were asked the follow up questions). unfortunately if you can't trust the answers to screening questions then it is very difficult to interpret the module, which is why skips should be avoided.
8.22.2005 5:32pm
Shelby (mail):
Then again, maybe 61% really DO support legalizing prostitution. Wouldn't be the first time the government was out of step with its constituents.

To the "in their own home" diehards: Note that "their" is ambiguous in this context. Perhaps the implication should be that if you entertain a prostitute in a house you, but not she, owns, 61% would let you off scot-free, but would still send her to jail. ;-)
8.22.2005 5:41pm
akiva eisenberg (mail):
A) Indeed, "privacy of their own home" actually implies subtly that prostitution is excluded.

B) I understand that your article is primarily a discussion of the imprecision/dishonesty of some surveys. However, the secondary content leads me to this issue...I am a clergyman of what some would call a fundamentalist persuasion, and (believe it or not) have never participated in extra-marital sex. However, if I divorce religion from my response to this survey, I have to ask, what exactly is wrong with prostitution?
Granted, there is an increased risk of disease, but the medical-supervision model used in many countries where prostitution is legal obviates this, and indeed makes prostitution preferable to unsupervised casual sex in this regard.
Since modern birth-control effectiveness (even excluding abortion, which I personally consider murder for convenience, but which is an entirely different issue), means that there would probably be no innocent victims of the act, what ethical contraindications are there?
Other "victimless crimes" may involve possibly valid societal interest such as preventing self-destructive behavior (although this leads to inherent paradoxes in jurisdictions where suicide is legal). There may also be societal interest in preventing behavior that might damage participants to the point that they becomes a "burden on the state" either directly or by consuming limited economic/medical resources. This logic has been used to justify the requirement for wearing seat belts.
On the other hand, the only self-destructive behavior in prostitution from a secular persepective seems to be purely on the emotional level. If society wants to prevent emotional self-destruction, it should probably ban a large proportion of marriages also.
Furthermore, if society tolerates (or more honestly, glorifies) casual sex between consenting adults, why should the involvement of cash remuneration make the identical act criminal or immoral?
I suspect that in polls of this type, respondents tend to divorce their personal theological viewpoints from their responses. Americans respect freedom of religion to the extent that generally they do not impose their religious views on others except in extreme cases such as abortion and assisted suicide. If indeed the above perspective is shared by others, I see no a priori reason to think that the high "agree" results from the poll is inaccurate.

I would like to see the results of a follow-up poll with these questions:
1) Do you think regulated prostitution should be legal?
2) (For men) Have you or would you personally use a prostitute? [I forsee great difficulty in getting honest answers to this one.]
3) Do you see a difference between casual non-marital sex and non-marital sex for remuneration (i.e., prostitution)?
8.22.2005 5:43pm
Eugene Volokh (www):
Isn't reading "their home" to exclude prostitution a little too much subtlety? Would anyone really interpret the statement as being limited to sex partners who co-own the home (or who even live in the home together)? "I'm a big believer in consenting adults' rights to have sex in their own home -- but you and your girlfriend aren't living together, and the home in which you were having sex was only yours and not hers, so off to jail you go"?

It seems to me more likely that people will understand "privacy of their own home" to mean "privacy of the private place in which they're lawfully staying" -- or that at least many people will understand it that way, so we can have no confidence about whether they're seeing prostitution as covered or excluded.
8.22.2005 6:10pm
Gabriel Rossman alludes to what I claim is a "grownup effect." Namely that a lot of us have figured out that the only place that always/never applies is in the statement that "I never completely agree or disagree with anything!"

But I think that you don't have to go to prostitution, adultery (the "own home" thing is a little vague, but if a tenant committed adultery with his/her landlord that probably qualifies) or adult sibling incest. "Restrictions" by "society" would include the preacher preaching a sermon where he says that some sexual circumstance is against God's rules. And it's pretty hard to believe that 61% of the survey respondants are in favor of repealing the 1st amendment.

cathy :-)
8.22.2005 6:23pm
Poll or no poll, juries still regularly support stiff penalties in adultery cases in North Carolina, at least where there is an aggrieved spouse who claims to be hurt by it.
8.22.2005 6:36pm
The problem with this poll question is less what it does or doesn't cover, and more with the lack of balance in the question itself. The lack of a balancing response in the question biases the question in favor of the implicit assertion in the question. This is a common problem in polling, even among sophisticated professional polling operations, and it creates some of the most insidious biasing problems in public opinion research.

To get a more accurate response to this question, the question should be phrased something like "Society should not put any restrictions on sex between consenting adults in the privacy of their own home, including restrictions on birth control, homosexuality, or sexual practices not accepted by mainstream society." Writing this question well is difficult, because using terms like "incest" or even "adultery" carry negative connotations that will bias the outcome.

Ultimately, this question is a helpful reminder that polling generally results in an accurate assessments of people's answers to the question asked, but not necessarily an accurate measure of people's opinions of the subject asked about.
8.22.2005 6:48pm
just me (mail):
This is just another example of how adding "reminders" to a question can radically change a poll result, even if those reminders are not CHANGING a factor in the initial question, but are simply clarifying something that is, or arguably is, in the initial statement. For example, in the popular poll area of abortion, this is a commone phenomenon. Please forgive the seeming digression into "another topic," but I think it's still withing the topic of poll-distortion, and this is the best example because it's so big that many respondents add their own assumed "baggage" to the question. For example, surveys will often ask

1. Do you believe that a woman should be able to obtain an abortion without government interference? (agree/dis, strongly/somewhat, etc.)

2. Do you believe that a woman should be able to obtain an abortion without government interference, AT ANY TIME FOR ANY REASON?

The first might garner a 58% majority, while the second might drop to only 22%. That's because the second conjures up negative connotations on TIME -- e.g., at twenty-four weeks?? and on REASON -- just because it's a girl, not a boy?? Oh my, then I'll say no, even after saying YES to Q-1. But some might argue that questions 1 and 2 mean the same darn thing -- that unless there's a qualifier, such as "for rape or incest," or "in some circumstances," then the generic "should be able to obtain without govt interference" should implicitly mean "on demand, any time, any reason." But a dropoff shows that some people hear them differently, unless there's an option to say, "oh, that reminds me, I now want to go back and change my earlier answer."

But if a person has heard the debate many times, she might fill in her best-case/worst-case language, in either direction, giving skewed results. Here, for example, a person who wants abortion to be legal ONLY for rape/incest/life of mother should answer "disagree" to Q1 (in my view), as she'd add those "only when . . ." qualifiers. But she might answer "agree," if she READS INTO the question her concerns about keeping it an option for rape/incest, i.e., if she hears the question as asking for a different answer only if she'd block ALL abortions, no matter what. So such a respondent might get listed as functionally pro-choice, at least to a degree, and become part of a headline's "pro-choice" majority, even if she wants to ban (by law, not just social pressure) about 99% of abortions, and would have the same position as many who call themselves pro-life.

(Yes, my bias is evident in the above. Oh well. If you like, flip everything around to match your preferences; the point is that distortion can and does happen.)

So the interesting Q, to me, is not whether such things get twisted, but whether they're accidental -- because the pollster or the media entity is just dumb -- or whether they're intended, e.g., when the Society For Funding Protection of Weasels trumpets its poll showing that "85% of Americans want more protection for weasels!"

Remember -- 91.4% of statistics are worthless.
8.22.2005 6:49pm
Eh Nonymous (mail) (www):
Prof. V:

Your amphigory and attempts at cocktail-party linguistics and sociology are to no avail.

The question excludes prostitution, which I will define for these purposes as "sex for money between people not living in the same location." See my point? "In their home." "In his home" or "In one of their homes." It's a simple question, using simple English words, and you're attempting to mangle it to support a proposition for which it cannot reasonably be interpreted to stand.
8.22.2005 7:25pm
Lawrence Hayes (mail):
Almost everyone seems to agree that your basic point about such questions is correct. There is disagreement on issues of definition. Here's another one. I would say that to most people "consenting adults" implies that the parties desire to engage in sex with each other. Some prostitutes no doubt have this desire and like their work. Others may do it purely because of a need for money and would definitely not do it at, least with with certain persons, in the absence of pay. Some, not masochists, may find it painful and humiliating, but still do it for the money. Arguably, those without desire should not be regarded as "consenting adults;" further, because the payment is essential to prostitution, no prostitute should be included in that category.
8.22.2005 7:30pm
James Taranto (mail) (www):
Quoting Eugene:
Isn't reading "their home" to exclude prostitution a little too much subtlety? Would anyone really interpret the statement as being limited to sex partners who co-own the home (or who even live in the home together)? "I'm a big believer in consenting adults' rights to have sex in their own home--but you and your girlfriend aren't living together, and the home in which you were having sex was only yours and not hers, so off to jail you go"?
This misses the point. Although the hypothetical statement here is logically consistent with an affirmative answer to the poll question, it is in no way required by it. Eugene is surely right that few if any people would make a moral distinction between a cohabiting couple and a couple with separate residences but a serious relationship. Most of us would draw the line somewhere else, and my guess is that most people who are more or less liberal-minded about such things would draw it to include most noncommercial sex but not prostitution.

The question, however, does not ask the polling subject to draw a line. It draws a line arbitrarily and asks the subject for his opinion about one side of the line only--a side that does not include prostitution and also does not include some circumstances in which presumably many respondents would oppose restrictions.
8.22.2005 7:32pm
Craig Oren (mail):
"amphrigory"? ah, now I see. . . "a nonsense verse or composition."

learn something every day on this site!
8.22.2005 8:00pm
arbitraryaardvark (mail) (www):
The Pew Center is in the business of manipulating public policy by creating scientific-looking studies which are then used in press releases, lobbying or lawsuits, notably McConnell. Pew is a special interest group that spent 100 million to buy campaign finance "reform." I'm guessing the question was asked that way in order to produce the results desired. Spin and counterspin.
"In a study by the Pew Foundation, 80% of Americans surveyed opposed restrictions on consensual sex between people and their dogs, in the privacy of the home." This might be literally true, but isn't what people meant. (Is a dog adult at 18, or at 3?) When the question is deliberately ambigulous, as here, people will tend to respond with how they feel rather than what they think. The population of Volokh readers is more compulsively literal than the average poll respondent.
8.22.2005 9:03pm
Carol Anne:
"just me" actually hints at a bigger problem in polling: The technique called "push polling." Reputable polling organization work hard to wordsmith questions to avoid "push" questions...that is, questions that "push" the respondent toward a particular outcome.

Given the reputation of the Pew Trusts, I'd be surprised they chose "...privacy of their own home..." deliberately, because it does (as akiva eisenberg points out) include the presupposition of "their" (i.e., conjointly owned) "home" (as in "in my neighborhood?!?"). I would have preferred "in private." If it was deliberate, it would be "push" polling.

A "push" question, in its most egregious form, is like "If I told you Eugene Volokh was a pedophile, would you participate in the Volokh Conspiracy website?" (Note there's no claim he is, just an implication within the qeustion.) There are subtler examples, but in this case the "push" pollster would report the question having been posed as "When asked, 'would you participate in the Volokh Conspiracy website?', 89% of respondents reported 'No!'"

We have a "push" pollster here in the county where I live. When we find out who has engaged him, we avoid voting for that candidate based on any proffered polling data.
8.22.2005 11:20pm
Carol Anne:
Obviously, "arbitraryaardvark" and I have differing perceptions of the reputation of the Pew Trusts...but our points are similar.
8.22.2005 11:22pm
Bemac (mail):
If you had an infinite number of pollsters designing an infinite number of polls, it would never occur to any of them to simply ask the question they want answered in a clear, unambiguous fashion.

I don't know what Pew was after, but wording of the question is sufficiently poor to render the resulting data almost worthless for any purpose other than dishonest interpretation.
8.22.2005 11:37pm
Paul doson (mail) (www):
Nice blog.I like this.
8.23.2005 5:26am
Andrew McGuinness (mail) (www):
Of course, while it's arguable whether the wording includes prostitution, many adults share a home with their parents...
8.23.2005 8:46am
Phil (mail):
Consensual private acts between mature adults not acting involuntarily or under duress, in a home they own is not something the gov't ought to regulate.

A vast majority of Americans may agree with this, but it covers only a small fraction of the sexual conduct in which most people engage. Most of us have had sex in the home of another, in a hotel room, in a rented apartment, etc. Moreover, authentic disputes exist about the meaning of terms like "private" and "adults" (Having taught philosophy as wellas law, I do not even want to start down the "involuntarily" road.)
To agree with many of the other writers, when polls phrase things at a sufficiently generic level there is wide agreement, but it is impossible to determine about what.
8.23.2005 9:35am
Scott Wood (mail):
Eh Nonymous


To this native English speaker "sex between consenting adults" means, well, "sex between consenting adults." E.g.: Suppose I, as an adult man, pay an adult woman $X to have sex with me. I am a consenting adult. She is a consenting adult. We are consenting adults.

"Privacy of their own home" means not in public (say, on a beach), and not in a private place (say, an office building), in which we are allowed by the owner to be, but not for that purpose.

Those are the clear and simple English meaning of those words, to me.
8.23.2005 9:56am
Phil (mail):
Scott Wood

President Clinton, a Rhodes scholar with degrees from Yale and Georgetown, claimed to not be clear on whether oral sex was sex or not. He also could not figure out what "alone" meant. The current Supreme Court held that giving land to a private developer is a "public use."
"Privacy of their own home" is clear and simple. So they can do it in front of their picture window with the drapes open?
You do not have to be Ronald Dworkin to concede that words do not define themselves.
8.23.2005 10:42am
markm (mail):
As a non-lawyer, I find this whole discussion to be illustrative of why lawyers have trouble communicating with juries. Ordinary people don't chop words that finely. 99% of the respondents would not have interpreted "in the privacy of their own home" to require co-ownership - if you take your friend home to your rented apartment you're covered, in a stall in the men's room isn't, and people would see that a motel room isn't included in that if you asked them, but I doubt that very many would think the rule should be different for it.

"Consenting adults" seems simple: No children, no dogs, no rape. Prostitution is still a transaction between consenting adults (at least the buyer hopes the prostitute is of age...), but most people just wouldn't think of it in the context of this question. Ask the 61% about prostitution next, and then I think you'd see ordinary people chopping logic rather than going back and changing their first answer - the law isn't about the sex, but the exchange of money, right? (How could it be about the sex when giving her a $200 piece of jewelry beforehand is legal, but giving her $200 cash isn't?)
8.23.2005 6:48pm