The popular media often highlight studies purporting to show that some (usually very large) proportion of people are happy, whereas some other (usually very small) proportion are unhappy (e.g., here and here). I suspect that not many people really believe such assertions, and that they are a major source of skepticism about the science of happiness.
Such claims seem mainly to rest on three kinds of studies: self-reports of happiness (which I critiqued in my last post), life satisfaction surveys, and measures of affect such as the balance of positive versus negative affect. Today I want to focus on this last source of evidence, assuming for the sake of argument a hedonistic or emotional state theory of happiness. Affect balance measures can only support claims that people are “happy” given some view about the threshold for being happy: what balance of positive to negative is needed to count as happy?
Traditionally, the answer has been this: a bare majority of positive affect suffices for being happy. Greater than 50% PA, you’re happy; less and you’re unhappy. (Note that you must have precisely 50% PA vs. NA to fall in the category you might have thought many of us fit: neither happy nor unhappy.) If you are enraged for several hours a day, or cry yourself to sleep every night, you may still be happy if your negative affect doesn’t hit 50%. Perhaps you could be depressed and count as happy on this sort of view.
This seems deeply implausible to me: life probably has to be pretty awful for negative affect to literally be in the majority. When a family member dies, we usually aren’t happy, yet the laughter sometimes outweighs the tears (the unhappiness perhaps revealed more by the ease with which the tears come rather than by their frequency). In informal surveys of students in my classes, a majority refused to ascribe happiness in cases where the percent of NA was significantly less than 50%. Recent work by Fredrickson et al. claims that a 3:1 ratio of PA to NA is needed for people to “flourish”; much less and they “languish.”
I think the correct standard for happiness, even on hedonistic or emotional state views, is less than obvious. Researchers should not just assume the 50% threshold; it needs some defense. (How can we determine the right threshold?) But if the threshold for hedonistic/emotional state happiness is an open question, then we have no basis for saying whether people are happy given such views. This seems to me correct: the science of happiness has taught us a lot, but we really have no idea, except for our hunches and obvious cases like depression, whether people are happy or not. (My hunch: people probably aren’t as miserable as intellectuals tend to think, but most people still probably aren’t happy.)
Perhaps researchers should drop absolute claims about whether people are happy, and focus on relative levels of happiness: who is happier and why? Isn’t this what we mainly care about?
Didn't Brandstatter (2001?) find about 50-50 for postive/negative affect in experience sampling? What do you make of that? Am I remembering wrong (I can't access it now)? Do you think his methodology is off? Or do you think that that implies that many people are seriously unhappy?
ReplyDeleteIs that in his 2001 book? I just found it on Amazon and will have to get a copy! I've read his 1991 experience sampling study, I think about 200 diverse European subjects, which found 66% positive affect vs. 34% negative. This is usually taken as evidence that most people are happy, but I think it probably indicates they aren't! (It certainly doesn't meet the 3:1 ratio suggested by Fredrickson et al.)
ReplyDeleteI've been meaning to look at more recent studies to see what they found--a 50/50 breakdown would be very interesting. That's a lot of negative affect! (Although I wonder if these ESM studies might skew positive in some ways and neg in others--positive b/c of positivity biases, neg b/c some pleasant states, like tranquility, may not get reported b/c they don't obviously involve occurrent feelings.)