Well-Being: Psychological Research for Philosophers



Well-being in the broadest sense is what we have when we are living lives that are not necessarily morally good, but good for us. In philosophy, well-being has been an important topic of inquiry for millennia. In psychology, well-being as a topic has been gathering steam very recently and this research is now at a stage that warrants the attention of philosophers. The most popular theories of well-being in the two fields are similar enough to suggest the possibility of interdisciplinary collaboration. In this essay I provide an overview of three of the main questions that arise from psychologists’ work on well-being, and highlight areas that invite philosophical input. Those questions center on the nature, measurement, and moral significance of well-being. I also argue that the life-satisfaction theory is particularly well suited to meet the various demands on a theory of well-being.

1. Introduction

If you have ever thought about whether it would be better for you to have a different career, get married, or make some other major life change, then you have wondered about your own well-being. If you have considered how to respond to the needs or wants of another person (a dependent, spouse, or friend) for his or her sake, then you have thought about the well-being of others. Well-being in the broadest sense is what we have when our lives are going well for us, when we are living lives that are not necessarily morally good, but good for us.

In philosophy, well-being has been an important topic of inquiry for millennia, though it has received a small share of attention in ethics recently as compared to the attention paid to it by the ancients. For a variety of reasons, the focus of research in psychology since its inception as a discipline has been on poor functioning, mental illness, and disease (Seligman 4). Perhaps because of the strides made in treating mental illnesses, or perhaps because of the freedom afforded by relative affluence in developed countries, well-being as a topic in psychology has been gathering steam in the last decade and is now flourishing.

The current spate of research on well-being in the social sciences has caught the attention of governments and public policy makers, especially in the UK and Europe, but increasingly also in North America. Given this and given the complexities in the science, this is an area that cries out for interdisciplinary work. In particular, philosophers can play an important role in shaping future well-being research and its application to policy. It is the goal of this entry to provide an overview of three of the main questions that arise from psychologists’ work on well-being, and to highlight areas that invite philosophical input. Those questions center on the nature, measurement, and moral significance of well-being.

Before turning to these topics, a note about terminology is in order. Well-being is one of a family of prudential values – goods for a person, as opposed to moral or aesthetic values – that also includes happiness, flourishing, and the good life. There is a good deal of controversy (among philosophers and psychologists) about the meanings of these various terms and the relationships among them. For the purposes of this short essay, I will stipulate the following definitions, which I believe are likely to find wide agreement. Happiness, though it has been taken to be a synonym for well-being, most often refers to a psychological state. Here the most popular views are hedonism, the life-satisfaction view, and the emotional state theory of happiness.1 Eudaimonia, often translated as “flourishing,” refers to an objective set of conditions of a person's life. The most popular conception in the current literature is the Aristotelian one according to which flourishing consists in fulfilling your nature as a human being (Hurka; Hursthouse; Nussbaum). Finally, the notion of “a good life” seems to include moral and prudential value so that a person would not be said to be living a good life, no matter how psychologically happy she was, unless her life met a certain moral standard.

For the purposes of this essay, well-being will designate the condition of life that is good for an individual creature in the broadest sense so that it is an open question whether well-being is identical with, or requires, happiness, eudaimonia, or a good life.

2. Well-being: What Is It?

Derek Parfit (493–502) distinguishes three philosophical approaches to defining well-being: hedonism, desire-based theories, and objective list theories. The first takes well-being to consist in pleasure. According to the second, well-being consists in the satisfaction of desires or preferences, whether real or idealized. And the third says that well-being consists in the achievement of certain objective values such as the perfection of one's nature or the realization of human capabilities. One important type of theory that this taxonomy leaves out, especially given an interest in comparison to the psychological theories, is life-satisfaction theory, an example of which is L. W. Sumner's authentic happiness theory (1996).2

The research programs in social and personality psychology correspond roughly to the divisions among philosophical theories. We find affect-based theories, according to which well-being consists in pleasure, positive affect, and positive emotions; life-satisfaction theories, according to which well-being consists in a positive overall evaluation or judgment about how life is going for you; and eudaimonistic theories according to which well-being consists in meeting certain objective psychological needs.

Eudaimonist accounts of well-being are the most theoretically laden of the approaches in psychology. According to these views, well-being consists in meeting core human needs, and lists of needs are derived from theories of human flourishing that are only partly based on experimental findings. Two such programs are the Self Determination Theory of well-being, which posits three basic human needs for autonomy, competence, and relatedness (Ryan and Deci), and Carol Ryff's multidimensional account, which posits six basic aspects of human actualization: autonomy, personal growth, self-acceptance, life purpose, mastery, and positive relatedness (Ryff and Singer).

These accounts overlap significantly with the front runner in objective philosophical accounts, Martha Nussbaum's capabilities approach. Nussbaum (78–80) postulates ten human functionings that are essential to living well: life; bodily health; bodily integrity; senses, imagination, and thought; emotions; practical reason; affiliation; other species; play; and political and material control over one's environment.3

While objective accounts are mainstream in philosophy, they are somewhat on the fringe of psychological work on well-being. This may be due to the difficulty in finding an empirical basis for claims about needs and their relationships to well-being. Insofar as well-being is defined in terms of the satisfaction of needs, the notion of “need” seems to be normative. In order to defend a needs-based conception of well-being, empirical researchers are forced to correlate the satisfaction of needs with other well-being markers such as life-satisfaction and positive affect. But this strategy invites the question whether it is really life-satisfaction or positive affect that is the essence of well-being while the satisfaction of needs is a cause. I turn now to a discussion of subjective theories.

Desire or preference theories have been very influential in philosophy. Interestingly, there is no direct counterpart to preference or desire theories in the psychological literature. This may be because discoveries in psychological research have demonstrated good grounds for skepticism about the reliability of preference as a guide to what is good for us.4 In particular, poor affective forecasting is a well established phenomenon according to which we make inaccurate predictions about what will make us happy.5 Preference theories are the domain of economists and many psychologists see themselves as providing needed alternatives to this picture (Diener and Seligman).

While it is true that preference theories continue to be popular in philosophy, it should be noted that current philosophical theories have a different conception of the relevant kind of preferences from that of economists. Economists take preferences to be revealed by consumer behavior, whereas, most philosophers in this camp take the relevant preferences to be idealized, not actual. Idealized preferences are typically (at least) fully informed, and this means that the problem of affective forecasting is not obviously a problem for philosophical accounts.

In the psychology literature, life-satisfaction accounts are currently enjoying a lot of attention. These accounts are sometimes called evaluative or cognitive because they require subjects to make a retrospective judgment about how their lives are going overall. They correspond nicely to Sumner's philosophical account according to which life-satisfaction is a kind of endorsement that has both a cognitive and an affective dimension. The main difference between Sumner's account and the psychological life-satisfaction account is that the former takes the relevant kind of endorsement of one's life conditions to be subject to idealizing constraints. In particular, Sumner (138–83) argues that life-satisfaction must be authentic (informed and autonomous, on his analysis) to count as constituting a person's well-being. Psychologists, of course, cannot measure ideal life-satisfaction. That said, the notion that psychologists seem to be aiming at when constructing their measures might very well be somewhat idealized. Further, Sumner presumes that most people's actual assessments of life-satisfaction come close enough to the ideal that they are, in fact, authoritative.

An important challenge to life-satisfaction accounts has to do with the validity of self-reports. The problem here is that people's retrospective evaluations of their experience often do not correlate well with the experience itself because retrospective evaluations are biased by irrelevant features of memory and attention. This problem has been taken to be an argument for hedonistic accounts of well-being by those prominent in the field, notably Daniel Kahneman, who has argued that “objective happiness” is a function of instant utility over a period of time (Kahneman 5). Instant utility is a matter of how happy a person is at the current moment; it does not depend on retrospective evaluations or memory.

In philosophy, hedonistic accounts have been unpopular since Nozick's experience machine, a thought experiment designed to pump the intuition that there are values other than mental states. In his example, you are asked whether you would be willing to hook up to an extremely reliable and effective virtual reality machine that would guarantee you a greater balance of pleasure than you would experience in the real world. Nozick (42–4) argues that we would not use the machine because we value, for example, doing things and not merely having the experience of doing them.

The main philosophical defender of hedonism in recent years, Fred Feldman, has a conception of pleasure that he thinks is immune from the experience machine objection. Feldman makes a distinction between sensory pleasures, which are feelings or sensations, and attitudinal pleasures, which are positive attitudes to states of affairs including sensory experiences. According to Feldman, attitudinal pleasure, not sensory pleasure, is “the chief good for man” (55–7). Feldman's view here is quite similar, in fact, to Kahneman's who describes positive affect or pleasure as an experience you would rather have than not (7). Feldman thinks that his characterization of the relevant kind of pleasure allows him to introduce the notion of “truth-adjusted” pleasures. On his view, pleasure taken in true states of affairs get more weight than pleasures based on falsehoods. This is how he answers the experience machine objection (109–12).

Notice that Feldman's concession moves his account away from a pure mental state account by imposing a condition that is external to the agent's experience. This move away from actual mental states is a common pattern in subjective philosophical accounts and for a common reason: theories that identify well-being with actual mental states do not seem to capture what is normatively significant about well-being because our desires, satisfactions, and pleasures may be entirely misinformed or manipulated. Hence philosophers who favor subjective accounts tend to identify well-being with informed desire satisfaction, authentic life-satisfaction, or truth-adjusted pleasure.

We can now see that there are several constraints on a theory of well-being. First, because well-being is a paradigm case of a prudential value, the right account must explain why well-being is good for the person who has it. This criterion favors subjective accounts, though objective accounts such as Nussbaum's can appeal to universal human nature to meet it. Second, a theory of well-being must be normative, that is, it must capture the sense in which well-being is a valuable goal worth promoting. Objective theories seem to have the edge here, but subjective theories that identify well-being with idealized desires or satisfactions can go a long way toward meeting this criterion. These two criteria are generally agreed upon in the literature. I would add a third. Wide reflective equilibrium, the most widely accepted method for theory construction in ethics, requires that we aim for coherence among our considered judgments, principles, and scientific theories (Daniels 1979). Insofar as we accept this method, there is another demand on theories of well-being, which is that they ought to be compatible with psychological research on well-being. I suggest that this compatibility means not only that psychological findings should not contradict the assumptions of our theories, but further, that our philosophical theories should have implications for social scientific research programs and their application to the real world. These are the two issues we will consider in the remaining two sections.

Before turning to these issues, however, I want to suggest that a life-satisfaction account of well-being meets the above three criteria quite well. First of all, life-satisfaction accounts are designed to reflect the individual subject's point of view. On the philosophical end, Sumner defends his version of a life-satisfaction theory precisely in order to capture the subjective point of view that he thinks is central to well-being. On the psychological side, Ed Diener, one of the leading scientists in well-being research, argues that the life-satisfaction view is to be preferred over eudaimonist accounts because it is democratic and gives authority to individual subjects (Diener and Suh 4). Second, a philosophically sophisticated life-satisfaction account can account for the normative aspect of well-being either by imposing idealizing standards that rule out problematically adaptive or misinformed assessments of life-satisfaction (Sumner's solution), or by specifying a certain normative perspective from which assessments of life-satisfaction are to be made.6 Finally, the life-satisfaction program in psychology is robust and productive. Life-satisfaction research is still in its early stages, but there is good evidence that life-satisfaction can be measured, that it correlates well with other intuitively compelling values, and that there are things we can do to increase it in ourselves or others.7

3. Well-Being and Measurement

The most common method for measuring overall or global life-satisfaction is Ed Diener's Satisfaction With Life Scale (SWLS).8 The SWLS is a five-item instrument, which asks subjects to indicate their level of agreement (on a 7-point scale from “strongly agree” to “strongly disagree”) with the following five items:

  • • In most ways my life is close to my ideal.
  • • The conditions of my life are excellent.
  • • I am satisfied with my life.
  • • So far I have gotten the important things I want in life.
  • • If I could live my life over, I would change almost nothing.

Critics of the life-satisfaction approach in psychology have worried that retrospective self-reports of the kind assessed by this questionnaire are too subject to irrelevant changes in context to be valid reports of overall life-satisfaction. For example, a person's overall satisfaction with her life should not vary with the weather, but some studies have shown that weather influences the degree of life-satisfaction people report.9

These critics tend to favor affect-based theories of well-being and measures that do not rely on subjects’ memories. Research on affect, pleasure, and pain, has shown that people's retrospective assessment of these states does not correspond well to the real time experience. It turns out that in assessing past painful experiences, for example, we tend to follow the Peak End Rule. That is, in retrospective assessments of pain we put more weight on the worst part and the very end of the experience (Redelmeier and Kahneman; Kahneman 20). To counteract the distorting effects of memory, some social scientists favor a type of measurement known as Ecological Momentary Assessment (EMA) to get at people's actual experiences, as they happen.10 One way of accomplishing this is with palm pilot studies in which subjects are beeped at random intervals throughout the day and asked to report their mood, affect, or level of satisfaction. Such methods are well suited to affect-based accounts of well-being, but have also been used by life-satisfaction researchers.11

The worry that self-reports of life-satisfaction are so variable that they are not an accurate report of anything beyond the moment is a serious worry for both psychological and philosophical accounts. This worry, however, seems to be exaggerated. In spite of contextual influences, there is good support for the claim that life-satisfaction is moderately stable across situations and across the life span (Diener and Lucas 214–15). Further, there is increasing evidence for high retest stability in life-satisfaction judgments and for the claim that this stability is due to the fact that people rely on relevant and chronically accessible information in order to form the judgments (Schimmack et al.; Pavot and Diener).

As far as philosophical life-satisfaction accounts are concerned, this much stability seems to be sufficient: overall life-satisfaction is psychologically real and usually based on assessments in important domains of life.12 But there is another problem for accounts that identify well-being with ideal life-satisfaction, which arises when we take seriously the idea that our philosophical account should be compatible with empirical research, as noted above. Ideal life-satisfaction, it would seem, cannot be measured; it is not what we experience and, therefore, we cannot report on it.

One way to operationalize ideal life-satisfaction would be to make subjects more similar to our philosophically ideal person before they answer the survey questions. For example, if we think that only informed judgments count, we could give people as much relevant information as practically feasible before asking them to report their life-satisfaction. But this strategy has some problems. First of all, people may be difficult to change in this way. Telling people some facts may not actually make their assessments informed because we do not always take the relevant facts into account in our judgments.13 Second, even if we could bring people closer to our philosophical ideal in order to assess their ideal life-satisfaction, some ways of changing people might undermine one of the advantages of the life-satisfaction account, namely, the weight it gives to subjective experience.

Fortunately, these problems are not insurmountable, for two reasons. First, we can bring people closer to an ideal without fundamentally changing them or divorcing their assessments from their own experience. Social scientists have already experimented with ways to change the perspective of the responding subjects in ways that make their assessments more normatively significant. For example, in Ulrich Schimmack's online version of the SWLS he instructs those about to take the test that the SWLS

allows people to consider the aspects that are most important to them and to evaluate how satisfied they are with them. Please think of the aspects that are most central to your life while answering the questions.14

The effect of these instructions is to get subjects to focus on the right things when they evaluate how their lives are going, thus bringing them closer to a reflective point of view.

Another strategy that might be used is “deliberative polling,” which requires that polled subjects engage in structured group discussion about the relevant questions. The assumption behind this tool is that such discussion “can overcome informational, cognitive, motivational and interpretive problems that afflict individual polls” (Adler 42). Most of the experiments with deliberative polls have been done in the area of policy choice, but similar techniques could be used in well-being questionnaires.

Second, bringing people closer to the philosophical ideal is not the only way to compensate for distortions from the ideal. Given an idealized subjective theory, if we know something about how people's judgments tend to be distorted, we can compensate by asking questions tailored to the purpose. For example, if we knew that people tend to take their health for granted, even though it is something they value, we could ask specific questions about health in order to find out how satisfied people are with respect to this value. Further, information about what people do value would be relevant to what objective measures should be used. For example, if we know that people care about feeling happy or being in a good mood, but we have reason to expect that selective memory distorts retrospective evaluations of these feelings, then we ought to consider using independent measures of positive affect even if we have a life-satisfaction account of well-being. In this way, the life-satisfaction account provides an overarching justification for using multiple measures. Rival accounts may be able to do this too, but for the life-satisfaction view the move is natural, not ad hoc.

4. Well-Being: How It Matters

According to welfarists in moral philosophy the promotion of well-being is the ultimate aim of all moral action; well-being is the fundamental notion in moral theory.15 But even if we do not agree that well-being is the only aim of moral action, it certainly seems to be one important aim. Non-utilitarian ethicists who believe that we have duties of beneficence or that we ought to develop the virtue of benevolence, should think that well-being is important even if they do not think it is the ground for all of morality.

Well-being is practically relevant to decisions we make about how to live our own lives, and also to other regarding decisions about how to help other people. Psychologists working on well-being have begun to appreciate the influence that their work can have in both contexts. In the former case, psychologists have begun to think about positive interventions that might increase positive affect, life-satisfaction, and other aspects of well-being. For example, there is evidence that cultivating gratitude by counting your blessings every day has significant effects on depression (Emmons and McCullough). In the case of other-regarding action, the significance of well-being research to public policy has been the focus. For example, psychologists Ed Diener and Rich Lucas are part of a research group whose aim is to devise national indicators of well-being that will go beyond economic measures and that will be used in shaping government policy.16

Using well-being research to inform our own choices about how to live might seem relatively unproblematic. If we read that counting our blessings will make us happier, and we want to be happier, then we should (other things being equal) count our blessings. If we learn that good social relations are the only thing that is actually necessary for well-being, then we might put more energy into our friendships and less into other activities, such as making more money, that do not have as sure a benefit.17 But, as philosophers well know, getting from facts (even psychological facts) to reasons for action is a tricky business. We can distinguish two types of complication in the personal case.

The first has to do with the nature of the correlations that psychologists establish. These correlations are always based on distributional averages. For example, well-being and income are poorly correlated if most people are not made much happier by increases in income. But there are always people who fall in the tails of these distributions. In other words, even though most people are not made much happier by increases in income, some people are. A person reading the psychological research might like to know where she in particular falls on the distribution before making significant changes to her way of life.

This is not the most serious problem; after all, many things we do are based on findings about the average person (e.g., taking aspirin to decrease the risk of heart disease, quitting smoking to prevent cancer, and so on). It does, however, reveal a gap between empirical research and action that is worth noticing. Further, this point also indicates the importance of studying sub-populations. I am much more likely to differ from “the average American” than I am to differ from “the average academic woman.”

The second and more difficult complication has to do with normative authority: Who gets to say what the good for a person is? Who has authority about the matter, the individual or the expert? Given that psychologists cannot measure the good itself, a person may not think that what psychologists are measuring is worth attaining. For example, a person who identifies strongly with the welfare of her children might not take information about what would increase her own life-satisfaction to be relevant to how she ought to live because she thinks that whether she is getting everything she wants out of life is not what's important.

This example allows us to see how a philosophical account of ideal life-satisfaction might help to shape scientific inquiry into well-being. If we think that life-satisfaction is normatively significant when it is a judgment made on the basis of a person's important values, and if we think that some people have important values that are not individualistic, then we might doubt that the SWLS is the best measure of normatively significant life-satisfaction. In other words, if the SWLS asks people about individualistic values, and if some people's values are community- or family-focused, then promoting life-satisfaction would seem to impose individualistic values on community-oriented people. If this were correct, it may make sense to alter the SWLS to accommodate a greater set of values.18 For example, instead of asking people whether they have gotten the important things they want in life, people could be asked whether they are satisfied with their life, given all the things they think are important.

Let's turn now to the case of public policy application. Here there are two concerns that pull in opposing directions, which give rise to versions of the problems just discussed in the personal case. First, there is the worry about paternalism; this is the concern that making people happy will mean using coercion to impose alien values on them: we will be acting for their own good, but against their wills. Second is the worry about endogenous or adaptive preferences, which is that making people happy by their own lights might result in reinforcing the oppressive patterns that cause people to be happy with their unhappy lot.19 We can see both of these concerns as having to do with normative authority. Regarding paternalism, the worry is that if the notion of well-being we promote on behalf of others does not accord with their conceptions of their own well-being, then our actions will undermine subjective authority in a morally objectionable way. Regarding endogenous preferences, the worry is that people's own sense of what is normative or valuable cannot be entirely authoritative because it may have been the result of oppression or injustice, or it may simply be misinformed. There are, in other words, recognized difficulties with taking an objective or external perspective to define what the goal of life is for people, and with taking people to be entirely authoritative about the matter for themselves. The normative seems not to be entirely a matter of the facts about what is good for us as agreed upon by experts, nor as decided upon by us for ourselves.

The most reasonable response to being pulled in these two directions, I would argue, is to try to find a middle path. Some objective accounts in philosophy do steer a middle path because of the emphasis they place on the role of practical reasoning in the best life for human beings (Nussbaum 59–60, 78–96). The fact that deliberation, choice, and planning are part of human well-being, according to such accounts, means that coercive state intervention into people's lives is not likely to make these lives better.

Idealized subjective accounts (such as the philosophical version of the life-satisfaction theory) also take a middle path. Recall that according to idealized subjective accounts, well-being consists in a positive response (this might be endorsement, positive judgment, or some other pro-attitude) to the conditions of your life that meets certain constraints (such as information or reflectiveness). These theories acknowledge the importance of individuals’ attitudes toward their own lives, but do not take actual attitudes to be definitive. Of course, there are many difficulties involved in making these accounts precise and in thinking about how to apply them; nevertheless, they seem to be a promising strategy given the various constraints on theories of well-being.

In addition to the theoretical advantages, taking a middle path between objective and subjective accounts has desirable practical implications. Knowing that individuals’ experiences are important, we will be on guard against paternalism, and knowing that individuals’ perspectives can nevertheless be distorted, we will be more cautious about jumping to policy implications that will increase positive subjective experience. The middle path urges caution and requires that each case be carefully considered for its liability to the problems at either extreme. This kind of careful weighing and balancing of various normative concerns has been the business of practical (or applied) ethics for many years. As psychologists put their research onto the public agenda, philosophers would do well to pay attention and to help guide the process in whatever way we can. To date, philosophers have not paid much attention to this area of psychology.20 Surely this is, in part, because the area is so new. We can hope that as the psychology of well-being develops, philosophical interest will too.


I would like to thank Dan Haybron, Rich Lucas, and John D. Walker for helpful comments on previous versions of this paper.


  • *

    Correspondence address: Department of Philosophy, University of Minnesota, 831 Heller Hall, 271 19th Ave. S, Minneapolis, MN 55455-0310, USA. Email: tiberius@umn.edu.

  • 1

    For a defense of the emotional state conception see Haybron, “On Being Happy or Unhappy.”

  • 2

    Dan Haybron (“Philosophy”) agrees that Parfit's taxonomy should be expanded in order to include Sumner's authentic happiness theory, which is neither hedonistic nor desire-based.

  • 3

    Nussbaum herself does not take the capabilities approach to be an account of well-being, however, given the very broad notion of well-being employed here, it makes sense to include her view in this discussion.

  • 4

    It could also be that some psychologists do accept a preference-based theory of well-being but reject economic measures as the best way to assess our well-being in these terms. Thanks to Dan Haybron for suggesting this alternative explanation.

  • 5

    For an accessible overview see Wilson ch. 7. See also Loewenstein and Schkade.

  • 6

    For development of this position see my The Reflective Life (Oxford University Press, forthcoming).

  • 7

    While psychologists agree that well-being is partly genetically determined, there is enough evidence for an environmental contribution that it makes sense to think of well-being as something over which we have some control. On genetic determinants see Lykken and Tellegen. For a quick statement of Diener's view see the FAQ section on his website: http://www.psych.uiuc.edu/~ediener/faq.html.

  • 8

    The scale is in the public domain, available on Dr. Diener's website: http://www.psych.uiuc.edu/~ediener/hottopic/hottopic.html.

  • 9

    For an overview of studies on context effects see Schwarz and Strack. The effects of weather on life-satisfaction judgments are discussed on p. 75.

  • 10

    For an overview of EMA techniques see Stone et al. For an overview on affect measurement in general see Watson.

  • 11

    Diener and Suh, for example, favor multiple measures: “We will find ourselves standing on more firm ground if we find that our conclusions converge across measurement methods” (6).

  • 12

    For a more detailed discussion of the validity of self-reports see Tiberius and Plakias.

  • 13

    According to economist Paul Dolan, this kind of strategy has been tried with a preference-based account of well-being. For example, in one study, even after subjects were informed of the evidence showing that it is significantly worse (in terms of happiness and life-satisfaction) to lose your hearing than your sight, these subjects still reported that they would prefer to lose their hearing.

  • 14

    See Schimmack's website: http://www.erin.utoronto.ca/~w3psyuli/survey/swls/swls.htm.

  • 15

    Taking the broad conception of well-being, this is the view of Utilitarians from Bentham and Mill to L. W. Sumner and Peter Singer.

  • 16

    For more information on national well-being indicators see Diener and Seligman, “Beyond Money”; Diener.

  • 17

    Strong social relationships seem to be the best predictor of happiness and life-satisfaction (Diener and Seligman, “Very Happy People”).

  • 18

    Diener himself has suggested that the SWLS may need to be altered if it discourages people from collectivist cultures from reporting their true level of satisfaction (personal conversation, June 2006).

  • 19

    Adaptive preference formation is one important reason for rejecting subjective conceptions of well-being according to Amartya Sen (45–6).

  • 20

    An important exception is Dan Haybron's work on happiness and subjective well-being (see, in particular, Haybron, “On Being Happy or Unhappy”; “Philosophy”).