Awhile ago, Yvain released the results of the 2012 survey (that I participated in), which contained responses from over 1000 readers. So I uploaded the data into STATA and fooled around looking for cool things.
Mainly, I've heard some people argue that the moral theory you hold has little to no impact on your actual day-to-day behavior. I want to use these survey results to see if this is true -- are consequentialists more likely to donate money than deontologists? Are virtue ethics more likely to be vegetarian? We'll see!
While I can't make any claims to the representativeness of this survey or the external validity of drawing conclusions from its results, at least among the people who did take the survey, ethical theories people endorse seem to have little impact on their actual self-reported behavior.
The Ethical Theories
First, a breakdown of the actual ethical theories people endorse. Respondents were asked to categorize themselves into "accept / lean towards consequentialism", "accept / lean towards deontology", "accept / lean towards virtue ethics", or "other / no answer" (N = 1055):
Consequentialism | 63.41% |
Deontology | 3.98% |
Virtue Ethics | 14.41% |
Other / No Answer | 18.20% |
(Note that "no answer" are the people who specifically chose to note they had no answer. Other people ignored the question and didn't choose any answer at all, truly having no answer. Those people are not included in this analysis.)
Ethical Theories and Donation
So first up, I want to see if your ethical theory predicts how much money you are willing to donate, if any. Given the famous connection between utilitarianism and arguments like Peter Singer's who suggest you should donate all your money until it really hurts (you give so much money, you yourself become as poor as the people you're trying to help). Certainly consequentialism is not utilitarianism, but I would expect most consequentialists would endorse donation more than deontologists or virtue ethicists where donations aren't as mandatory.
So I took the charity data and dropped all the non-numerical answers, and LessWrongers have donated an average of $445.15 (N = 879, SD = 1167.095, min = 0, max = 9000) to charity over the past year. To get a better proxy for "effective charity", LessWrongers donate $331.05 (N = 884, SD = 4087.303, min = 0, max = 110000) on average to SIAI and CFAR. (I don't know why the max is higher on the SIAI/CFAR question, but not the inclusive charity question...)
Breaking down generic donations by ethical theory, we get this:
Mean | T-Test p | |
Consequentialism | $479.41 | 0.266 |
Deontology | $333.89 | 0.561 |
Virtue Ethics | $433.25 | 0.887 |
Other / No Answer | $358.54 | 0.323 |
Breaking down SIAI/CFAR donations by ethical theory, we get this:
Mean | T-Test p | |
Consequentialism | $420.47 | 0.391 |
Deontology | $0.00 | 0.629 |
Virtue Ethics | $86.85 | 0.459 |
Other / No Answer | $288.44 | 0.886 |
What we're seeing is that there are clear differences in the mean amount of money donated, with consequentialists giving the most. However, when we do t-tests (group compared to all not in the group), we find that none of these differences are statistically significant, which indicates the differences are probably due to chance.
This would lead us to suspect that ethical theory has no influence on the amount of money donated...
Ethical Theories and Percent of Income Donated
However, I have one more trick in the bag -- these donations don't take into account the income people earn. Many LessWrongers are students, and therefore can't donate much, even if they wanted to. What if we adjusted the donation totals by income, and instead looked at percent of income donated?
Overall, LessWrongers donate 1.75% of their income on average to generic charity (N = 523, SD = 5.70%, min = 0%, max = .88.24%) and 0.49% of their income to SIAI/CFAR (N = 523, SD = 3.11%, min = 0%, max = 52.38%).
(For those keeping score at home, the average income was $49563.76, N = 602, SD = $59358.34, min = $0, max = $700000.)
Breaking down percent of income spent on generic donations by ethical theory, we get this:
Mean | T-Test p | |
Consequentialism | 2.09% | 0.069 |
Deontology | 1.09% | 0.568 |
Virtue Ethics | 1.36% | 0.490 |
Other / No Answer | 0.91% | 0.161 |
Breaking down percent of income spent on SIAI/CFAR donations by ethical theory, we get this:
Mean | T-Test p | |
Consequentialism | 0.60% | 0.261 |
Deontology | 0.00% | 0.454 |
Virtue Ethics | 0.15% | 0.289 |
Other / No Answer | 0.49% | 0.993 |
A bit more can be made out of these results -- specifically, there's weak evidence that consequentialists actually donate more of their income (M = 2.09%) than non-consequentialists (M = 1.14%) with a p-value of 0.069, which is fairly significant. However, there are no differences across any other ethical theories or across SIAI/CFAR donations.
(Though Unnamed concludes that this might just be in-group bias. Unnamed also does a more thorough analysis and finds more correlations between consequentialism and donation.)
Vegetarianism
However, Peter Singer isn't just famous for consequentialist arguments for charity... he's also famous for consequentialist arguments for animal rights, which he argues necessitate veganism, or at least vegetarianism. Are consequentialists more likely to be vegetarian?
Not a Vegetarian | Yes, Vegetarian | |
Consequentialism | 84.39% | 15.61% |
Deontology | 83.78% | 16.22% |
Virtue Ethics | 88.97% | 11.03% |
Other / No Answer | 90.00% | 10.00% |
Perhaps surprisingly, there is no statistically significant relationship (N = 958, chi2 = 4.73, p = 0.192). Your choice in ethical theory has no correlation with your choice to eat or not eat meat.
(Though Unnamed finds somewhat contrary results that are consistent with a connection -- consequentialists are more likely than non-consequentialists to be vegetarian [p=0.03], this effect holds up when looking at men [p<0.01] but not for women, and as a single analysis, sex and consequentialism both predict vegetarianism [p=0.007 and p=0.02 respectively].)
Dust Specks
We can also take it into a more abstract realm of theory. Eliezer Yudkowsky in "Torture vs. Dust Specks" outlines a thought experiment:
[H]ere's the moral dilemma. If neither event is going to happen to you personally, but you still had to choose one or the other:
Would you prefer that one person be horribly tortured for fifty years without hope or rest, or that 3^^^3 [an obnoxiously and unfathomably large number] people get dust specks in their eyes?
I think the answer is obvious. How about you?
According to Yudkowsky, consequentialists (at least of the utilitarian variety) should choose torture over dust specks, since less total harm occurs. Does this turn to actually happen?
Torture | Dust Specks | |
Consequentialism | 45.19% | 54.81% |
Deontology | 11.11% | 88.89% |
Virtue Ethics | 13.79% | 86.21% |
Other / No Answer | 35.85% | 64.15% |
Here, we see another statistically significant relationship (N = 636, chi2 = 39.31, p < 0.001), and it goes exactly as expected. People's ethical theories seem to influence their choice of theory in this scenario (or the other way around, or a third variable).
Politics
Going back into the practical realm, let's look at politics.
Communist | Conservative | Liberal | Libertarian | Socialist | |
Consequentialism | 0.30% | 1.82% | 41.03% | 29.64% | 27.20% |
Deontology | 2.38% | 16.67% | 21.43% | 26.19% | 33.33% |
Virtue Ethics | 1.99% | 5.30% | 34.44% | 28.48% | 29.80% |
Other / No Answer | 0.00% | 2.69% | 31.18% | 34.41% | 31.72% |
Here, there is a statistically significant relationship (N = 1037, Chi2 = 50.91, p < 0.001) between ethical theories and political beliefs -- the plurality of consequentialists and virtue ethicists are liberal, the plurality of deontologists are socialist, and the plurality of others or no answers are libertarians.
Religion
Continuing along a similar path, next let's look at religion:
Agnostic | Atheist, nonpiritual | Atheist, spiritual | Theist, committed | Deist/Pantheist | Theist, lukewarm | |
Consequentialism | 5.71% | 81.35% | 8.72% | 1.35% | 1.35% | 1.50% |
Deontology | 11.90% | 54.76% | 7.14% | 11.90% | 7.14% | 7.14% |
Virtue Ethics | 12.50% | 61.18% | 13.16% | 5.26% | 3.95% | 3.95% |
Other / No Answer | 10.00% | 75.79% | 6.84% | 3.16% | 1.05% | 3.16% |
Again, another statistically significant relationship (N = 1049, Chi2 = 64.74, p < 0.001) between choice of ethical theory and religious beliefs -- consequentialists were more likely than deontologists and virtue ethics to not be religious.
Sequences and Meetups
And for a bit of bonus material, here's another interesting finding -- there is also a statistically significant relationship (N = 1052, chi2 = 128.43, p < 0.001) between ethical theory endorsed and amount of sequences read. Whether sequences convince people to adopt more consequentialist theories, whether people with consequentialist theories are more likely to enjoy and therefore keep reading the sequences, or some hidden third variable at work, I cannot figure out with the current data.
~1/4 of Sequences Read | 1/2 of Sequences | 3/4 of Sequences | Never looked | Never heard of 'em | Nearly all | Some, but <1/4 | |
Consequentialism | 11.83% | 14.82% | 21.71% | 0.60% | 31.44% | 4.49% | 15.12% |
Deontology | 14.29% | 7.14% | 9.52% | 4.76% | 19.05% | 11.90% | 33.33% |
Virtue Ethics | 11.26% | 12.58% | 10.60% | 1.99% | 15.23% | 20.53% | 27.81% |
Other / No Answer | 14.66% | 14.14% | 10.99% | 6.28% | 16.75% | 11.52% | 25.65% |
And this trend continues among those who have been to a LessWrong meetup (N = 1044, chi2 = 34.27, p < 0.001):
Never Been to a Meetup | Been to a Meetup | |
Consequentialism | 66.52% | 33.48% |
Deontology | 92.68% | 7.32% |
Virtue Ethics | 84.00% | 16.00% |
Other / No Answer | 79.14% | 20.86% |
Conclusion
In meta-ethics, there is a distinction between moral internalism, which is the theory that moral beliefs must be motivating, and moral externalism, which is the theory that you can have a moral belief and not be motivated to follow it. For instance, if moral externalism is true, you can legitimately think that eating meat is morally wrong but still eat meat.
Now, moral interalism and externalism are more about the semantics of moral statements, what moral statements refer to, and less about actual behavior. Even if people claim that eating meat is morally wrong while still eating meat, it's easy enough for the internalist to deny that the meat eater was telling the truth or making a coherent statement.
However, when it comes to the results of the LessWrong 2012 survey, there are very mixed results on whether choice of ethical theory influences actual behavior -- there's weak evidence that consequentialists are more likely to donate, and even those who do donate are certainly not giving up everything and becoming poor themselves, let alone giving 10% like Giving What We Can advocates.
Furthermore, there's evidence that vegetarianism does not depend upon ethical theory, but there is evidence that it influences people's choice of dust specks vs. torture in the right direction, at least for slightly less than half of the consequentialist sample. Additionally, there's a relationship between politics and ethics, and relationships between religion and ethics.
This essay has no intention to make a normative point. Certainly there are all sorts of consequentialism theories that don't require you to donate all your money and never eat meat again, and I'm making no accusations of hypocrisy. However, this survey does seem to confirm what Chris Hallquist has previously noted -- moral beliefs don't seem to motivate much, at least for the average person.
I realize that I didn't do a very thorough job of looking at total charitable giving, so I did some more analyses.
Summary: People tend to give more if they are older, richer, or more religious. Consequentialists tend to be younger, but after controlling for age they do tend to give more than non-consequentialists (p=.02). Consequentialists also tend to be less religious, and if you control statistically for that as well then the relationship between consequentialism and giving is even stronger (p=.006).
log(charity+1) seems like the best variable to look at for total charitable giving - it has roughly a normal distribution plus a big point mass off in the left tail at zero giving (which stays at zero after adding 1 & taking the log), n=952. There is a weak & nonsignificant trend for consequentialists to have higher log(charity+1), p=.30. Average log(charity+1) is 3.21 vs. 3.01 for the 2 groups, which means that the geometric means are e^3.21 = $25 vs. e^3.01 = $20, a 1.2x ratio.
There are other variables which have stronger, clearer relationships with total charitable giving: income, age, and religion. People with more money give more, older people give more, and more religious people give more. All three variables have independent effects - in a multiple linear regression, age, income, and religiosity are all significant predictors of charitable giving.
(Details on those effects: for income, I used log(income+1000). About a third of the data points are lost with analyses that include income because a lot of people left it blank, and that's a non-random subset as they have lower giving than those who reported their income. The religion effect shows up on p(God) and p(Religion), and on the religious views question treated as categories, and on the religious views question treated as a quantitative scale from 1 (atheist & not spiritual) to 6 (committed theist). I combined the 3 quantitative questions into a single religiosity scale by standardizing each & averaging them.)
What does this tell us about the relationship between consequentialism and giving? Well, consequentialism is negative correlated with religiosity (consequentialists are less religious) and age (consequentialists are younger), which could hide a positive effect of consequentialism on giving. (It is uncorrelated with income).
And in fact, in a multiple regression using all four variables as predictors (age, income, religiosity, and consequentialism), all four are statistically significant predictors of charitable giving. The effect of consequentialism is significant at p=.006, and the effect size suggests that consequentialists give 1.9x as much as nonconsequentialists (of the same age, income, and religiosity); $42 vs. $22 geometric means (these are higher than the means before, since it excludes the people who left income blank, who tended to give less). So consequentialism does predict higher giving, controlling for age, income, and religiosity.
It's not clear if we should be controlling for religiosity in this way, though - if someone buys into the idea package which includes atheism & consequentialism then maybe they shouldn't get credit for being more generous than those who accept atheism but not consequentialism (especially when other idea packages, like theistic ones, are associated with higher generosity). But controlling only for age & income still leaves the effect of consequentialism statistically significant, p=.02, with consequentialists giving 1.7x as much as non-consequentialists of the same age & income ($40 vs. $23).
There is also some concern with controlling for income, especially because of the missing data (from a non-random third of the sample). There are also possible issues with endogeneity (e.g., people who choose a high income job so they can give more money to charity), although that suggests that controlling for income could lead to understating the effect of consequentialism rather than to overstating it. But it turns out to not be worth worrying about; running an analysis which controls only for age, consequentialism is still a significant predictor of giving, p=.02, 1.6x the giving ($27 vs. $17).
And there aren't similar concerns about statistically controlling for age, since age is an exogenous variable - the causal arrows can only go in one direction there. Another way to account for age is to just exclude everyone under the age of 25 (since many people in that age range aren't financially independent, so their giving rates aren't that informative). Of those aged 25+, consequentialists give more than non-consequentialists, p = .04, 1.7x the giving ($56 vs. $32), n = 526.
One more analysis, inspired in part by Gwern's comment here.
In my first analysis I broke people into two groups using a cutoff at the bottom of the distribution, giving anything vs. giving nothing. But if we care about how much money charities receive, then we should care more about the top of the distribution, because that's where most of the money is coming from (80-20 rule and all).
So, let's focus on the top of the distribution by putting a cutoff up there. $1000 is an especially convenient Schelling point, since it is the 80-20 point among donors: 21... (read more)