Wednesday, June 27, 2007

Why life satisfaction is (and isn’t) worth measuring (by guest blogger Dan Haybron)

A lot of people think of happiness in terms of life satisfaction, and take life satisfaction measures to tell us about how happy people are. There is something to this. But no one ever said “I just want my kids to be satisfied with their lives,” and for good reason: life satisfaction is very easy to come by. To be satisfied with your life, you don’t even have to see it as a good life: it just has to be good enough, and what counts as good enough can be pretty modest. If you assess life satisfaction in Tiny Timsylvania, where everyone is crippled and mildly depressed but likes to count their blessings, you may find very high levels of life satisfaction. This may even be reasonable on their part: your life may stink, but so does everyone’s, so be grateful for what you’ve got. Things could be a lot worse.

Many people would find it odd to call the folks of Tiny Timsylvania happy. At least, you would be surprised to pick up the paper and read about a study claiming that the depressed residents of that world are happy. If that’s happiness, who needs it? For this and other reasons, I think that life satisfaction does not have the sort of value we normally think happiness has, and that researchers should avoid couching life satisfaction studies as findings on “happiness.” To do so is misleading about their significance.

So are life satisfaction measures are pointless? No: we might still regard them as useful measures of how well people’s lives are going relative to their priorities. Even if they don’t tell you whether people’s lives are going well, for reasons just noted, they might still tell you who’s doing better and worse on this count: namely, if people whose lives are going better by their standards tend to report higher life satisfaction than those whose lives are going worse. This might well be the case, even in Tiny Timsylvania. (Though caution may be in order when comparing life satisfaction between that nation and Archie Bunkerton, where people like to kvetch no matter how well things are going.)

This kind of measure may be important, either because we think well-being includes success relative to your priorities, or because respect for persons requires considering their opinions about their lives when making decisions on their behalf. The government of Wittgensteinia, populated entirely by dysthymic philosophers who don’t mind being melancholy as long as they get to do philosophy, should take into account the fact that its citizens are satisfied with their lives, even if they aren’t happy.

Note that the present rationale for life satisfaction as a social indicator takes it to be potentially important, but not as a mental state good. Rather, it matters as an indicator of conditions in people’s lives. Concern for life satisfaction is not, primarily, concern about people’s mental states. So rejecting mentalistic views of well-being is no reason for skepticism about life satisfaction.


Brad C said...

Hi Dan,

Nice, persuasive post which brought to mind one of my favorite church signs. When living in Milwaukee, I saw one that read, "Good enough / never is"

Some comments:

(1) Is there any good work on the way that people's level of life satisfaction varies with the opinions other have of them and the levels of life satisfaction of those who surround them?

(2) It seems to me that *resilient* life satisfaction, esp in the face of the changes mentioned and increasing awareness of the facts, is something that has (or can have) significant value. I am not sure it has the "sort of value we think happiness has" but do not, off hand, see why not.

- Brad

dan haybron said...

Hi Brad,

Thanks for your comments!
1. I'm not aware of any such work but it would be very interesting to see some. Perhaps some of the "positive illusions" literature dealing with self-enhancement biases would discuss the correlation btwn LS and others' opinions (in particular, the stuff looking at discrepancies btwn self-opinions and others' opinions of the individual). LS measures might be used to assess the well-being impact of the biases.

Regarding the levels of LS vs those around us, maybe work on "emotional contagion" would touch on it...

2. I'm not sure what you mean here, but I would agree that a robust sense of satisfaction with one's life can be valuable (besides often being a *sign* of things going well). Eg, it can promote emotional well-being (ie, more or less, happiness as I understand it). I just don't think it's nearly *as* important as people tend to think happiness is. Eg, H arguably serves as a rough proxy for well-being in ordinary thought--so if Smedley is happy, we guess he's doing well, conversely if he's unhappy. I think this practice holds up decently for emotional state or hedonistic views of happiness, but not for LS views.

Brad C said...

Hi Dan,

Thanks for the pointers.

Regarding the second point I was thinking that people's satisfaction with their life can be more or less responsive to, for example, changes in what other people think of them. Some people have satisfaction and would still have it if others thought worse of them and their life, while others might find that their satisfaction is more fragile in the sense that it is sensitive to what others think. By 'resilient' I meant satisfaction of the former sort.

So-called "Wisdom literature" often attributes resilient satisfaction to the wise. For example in the Dhammapada, one finds this:

"As a solid rock is not shaken by the wind, even so the wise remain unshaken amidst blame and praise.

Just as a deep lake is clear and still, even so, on hearing the teachings, the wise become exceedingly peaceful."

Usually, of course, resilient satisfaction like that is explained by appeal to wisdom or virtue, and that raises two questions:

(1) Can someone have resilient life satisfaction without wisdom or virtue.

(2) If they can, then can resilient life satisfaction really serve as a rough proxy for well-being.

I suspect that if we answer the first in the negative, then resilient LS is a good *proxy* for well-being but only because wisdom/virtue constitutes or reliably yields well-being.

I am not sure what to say about #2 if we answer #1 positively, because I am not sure what else could account for resiliency - but I suppose straight up denial might be thought to do the trick (Forest Gump II?). And if that would work than I agree that resilient LS is no proxy...


dan haybron said...

Interesting stuff, Brad. I could imagine that resilient LS, if it requires virtue or wisdom, will be a very reliable indicator of WB. But I don't think it is a type of state that could be used, in most people, as a very useful proxy for WB: it would identify a certain ideal type of well-being, but how could you use the concept to assess how most people are doing? (Other than to say that they aren't achieving that ideal.)

You might be interested in work on adaptation, particularly the "psychological immune system" and "sense-making" processes in Dan Gilbert et al's work. This research indicates that resilient LS may be pretty common, since people have a remarkable capacity to adapt to (esp) negative events. Eg, if everyone says you stink, you'll likely manage to restore your LS by rationalizing, say that no, *they* stink! But maybe there's a way of defining "resilience" to set aside cases of rank rationalization and the like...

Offhand, Dick Cheney seems to me like someone who is neither wise nor virtuous but probably has *very* resilient LS!

Brad C said...


(and thanks for more pointers)

Chase Wrenn said...


It would seem out of order to treat life satisfaction (as measured by the Satisfaction with Life Scale, for example) to measure happiness itslef. Happiness is too much of a moving target, for all the reasons you've mentioned.

But how frequently is that error really made? Ed Diener, at least, strikes me as pretty careful and clear in describing life satisfaction as the cognitive component of subjective well-being (with positive and negative affect constituting the conative component, I think).

Do you think subjective well-being is also a bad measure of happiness?

dan haybron said...

Hi Chase,

Good question--I agree that Ed, whose work I admire and respect a great deal, is generally pretty careful about this stuff in his research. But 'happiness' still turns up in his paper titles, and I'm not sure he's entirely consistent in its use (eg, sometimes for SWB, sometimes for the affect component, and sometimes for the LS component). Actually, I'd have to check on that to be sure he does this. (He does have a famous paper, "most people are happy.")

But looking at the literature as a whole, I'd say there's still a fair bit of trading in the term, esp in press announcements! Most investigators wisely avoid putting much weight on any definition of happiness in their work, so their actual results aren't much affect. But I think many of them happily trade in the term when describing and titling their work. This can mislead people about its significance. (To be clear: despite my criticisms, I think even LS studies are pretty important, more so than most of what happens in the academy (like, say, building big particle accelerators so they can discover still smaller particles, requiring still bigger machines... you could constantly monitor everyone's LS or stick all of Texas in an fMRI machine for that kind of cash).)

It'd be good to survey the lit and see how the word gets used, esp how often a given researcher will use it in different ways. Ie, do people exploit confusion about the meaning to lend their research the greatest possible aura of importance? I'd guess "yes!" (To be fair, it's hard to get decent funding, & calling your project a "happiness" study probably helps!)