How To Live A Happy Life Series
* Part 0 - Introduction
* Part 1 - History of the 3 Greeks
* Part 2 - Remembered Self (The Ingredients of Happiness)
* Part 3 - Experienced Self
Note: This series of posts can largely be read independent from one another, but it may be of benefit to read them in sequence.

How To Live A Happy Life (Part 2) — Self Reports (The Ingredients of Happiness)

So as concluded in part 1, the basic critique against hedonism is that pleasure is not equal to happiness and that pain is not equal to unhappiness, arguably. Let’s take this reasoning to its logical conclusion and assume it’s true. If we assume that to be the case, how do we go about finding the causes of happiness so that we may attempt to replicate the behaviours, beliefs or environments which lead to happiness?

If you believe the assumption that happiness can be studied and quantified, then there are three four main that ways we can go about uncovering the causes of happiness, for us to replicate ourselves:

  • (1) We can study what others believe, by reading books written by others, or listening to others opinions.
  • (2) We can do studies with large numbers of people where we ask them how they feel about their lives as a whole — a psychological measure of what’s called life satisfaction.
  • (3) We can study how people feel about their lives from moment to moment.
  • (4) We can do experiments on ourselves by trying out a lot of things and trying to discern what does and doesn’t work.

Method 1: Studying What Others Believe

There is wisdom in the crowd, in history, and in endless quotes which resonate with us personally. I personally believe quotes are quite underrated as a method of communicating value.

The problem with believing to be true merely what others believe to be true is that logically you can inverse their beliefs to be the opposite of what they are and make them false. Take for instance the belief that fruits give you an upset stomach. If everybody believes that this is true, but then you eat fruit and it turns out that this isn’t true for you, what good is it to believe what others do?

People have also believed a lot of things throughout history that have turned out to be outright false.

But let’s be a bit more charitable and assume you are replicating the beliefs and behaviours of people that you view as being successful or happy. The problem with this is that for all of them that do this thing there is going to be another subset of people who also did it and didn’t get the desired effect. Take for instance meditation. Let’s say that you admire and respect the Buddhists and ascetics and lead yourself to believe that what they say is true and that dedicating your life to meditation is the single behaviour which will make you happy, there is going to be a large part of that population which is still “working towards nirvana” and is still unhappy.

This individuality and uniqueness of opinions and preferences is one of the big reasons we can’t just base our lives on the opinions and advice of others, unsystematically. If we go about it systematically using the scientific method though, we might be able to give some rigor to the opinions of others. That brings us to the scientific study of subjective wellbeing with self-reports.


Method 2: Self Reports

Thanks in part to the positive psychology movement, recent years have brought a surge of interest in the correlates of subjective well-being. Self Reporting is a technical way of saying ‘the measurement of asking people’. It is purported to measure subjective well-being, which is the technical name for self-reported happiness. SWB is individuals perceptions of their overall happiness and life satisfaction. If you ask people how happy they are on a scale of 1–10, they can self-report a 7 for instance, and this can give a somewhat accurate picture of the state of affairs if you assume that people know how they feel (have the self awareness) and have the ability to attach a number to it. There’s no reason to think that we can’t do this since we can do a thing called intensity matching. Intensity matching is converting between different unrelated mediums. For example, if I ask you “If John 6'2 and Peter is 5'2, and John is as angry as he is tall, how angry is Peter?”, you’re likely to answer “less angry than John” or something of the sort. It turns out that humans can and do utilise this for all sorts of thinking. People can do a similar thing with happiness.

This has largely been the method of scientists over the last 50 years, doing what they call life satisfaction research, where they will basically ask people a question very roughly similar to “all things considered, how happy are you with your life at the moment on a scale of 1–10?” and they will give an overall rating. This has been the fruit of the new science of happiness starting in the 90s (ie positive psychology). What happens is that you ask enough people this question and correlate it with other questions such as how much money they make, whether they are happy with their relationships, how much they sleep, etc. You then find items that correlate highly with people saying they are satisfied with their life and you assume there is some link between the two when there are enough people, via statistical analysis. The factor may not cause happiness, but some sort of association is usually enough to suggest that there is a real relationship we are measuring. If we find out that two things correlate highly, then we can do randomised studies where we compare the two groups and find out whether the thing causes the happiness or whether happiness causes the thing. eg do people who smoke become unhappier or do unhappier people smoke?

There have been a lot of findings in this field of research. The findings of these studies are quite interesting, and many commonsense notions about happiness appear to be inaccurate. Here is a list of things that do or don’t correlate with happiness:

Happiness Predictors

  • An Optimistic mindset in explaining negative setbacks predicts happiness (1)
  • The absence of smoking cigarettes/quitting smoking improves happiness significantly (ref)
  • There is a strong link between love and marriage and happiness (ref). Romantic relationships can be stressful. Nevertheless, people consistently rate being in love as one of the most critical ingredients of happiness. Married people are happier than people who are single or divorced. However, the casual relations are unclear — it may be that happiness causes marital satisfaction more than martial satisfaction promotes happiness. Perhaps people who are happy tend to have better intimate relationships and more stable marriages, while people who are unhappy have more difficulty finding and keeping mates.
  • There is a strong link between job satisfaction and happiness (ref). People often complain about their work. Therefore, one might not expect work to be a key source of happiness. But it is. It's not as critical as love or marriage, but job satisfaction has a substantial association with general happiness. Studies also show that unemployment has strong negative effects on SWB. It's difficult to sort out whether job satisfaction causes happiness or vice versa, but evidence suggests that causation flows both ways.
  • There is a strong link between genetics and happiness (ref) — accounts for about 50% variance (which is a lot in happiness research — more than anything else). The best predictor of individuals future happiness is their past happiness. Some people seem destined to be happy and others unhappy, regardless of their triumphs or setbacks. The limited influence of life events is demonstrated with the hedonic treadmill (which I’ll talk about further below).
  • There is a strong link between personality and happiness (ref, ref) — with the largest effect being due to extroversion, with agreeableness, conscientiousness, self-esteem, and optimism also being strongly related. (ref) Peoples genetic predispositions account for a substantial portion of the variance in happiness. How? By shaping one's personality and temperament, which are known to be heritable.
  • There is a moderate link between health and happiness (~.32) (ref). Good health, may not by itself produce happiness because people tend to take good health for granted. May explain why researchers only find a moderate positive correlation between health status and SWB.
  • There is a moderate link between social activity and happiness (ref). Humans are social animals. As such, interpersonal relations do appear to contribute to people’s happiness. Those who are satisfied with their social support and friendship networks and those who are socially active report above-average levels of happiness. Furthermore, people who are exceptionally happy tend to report greater satisfaction with their social relations than those who are average or low in SWB. People who had deep, substantive convos were happier than those who mostly engaged in small talk.
  • Another moderate link between religion and happiness (ref), though this may be due to religious attendance and not religious belief (ref)
  • Video game use in young people improves well being slightly (ref)
  • Part-time workers are very slightly happier than full-time (ibid)
  • There is a positive correlation between income and SWB, but its weak (between .12 and .20) (ref). Though it’s high between very low income and unhappiness. In other words, money doesn’t make you happier unless you’re below the poverty line (can't pay for food, shelter, clothing, and sanitation). For a full exploration see here.
  • Home Ownership is positively correlated with life satisfaction (ref) (but more important for females than males) — however, its effect on well-being seems to depend on income level (less important for lower income), potentially due to aspiration preferences (ref). Some seemingly better research suggests that homeownership improves housing satisfaction but not life satisfaction (ref) and that they might actually not be happier at all due to spending their time in worse ways (ref).

The list of factors that turn out to have fairly strong associations with happiness is surprisingly short. The key ingredients of happiness appear to involve love, work, and genetic predisposition expressed through personality.

Kahneman (2006) Developments in The measurement of subjective wellbeing

Neutral Predictors

  • Physical Attractiveness is unrelated to happiness (ref)
  • Intelligence is unrelated to happiness (ref)
  • Parenthood is unrelated to happiness (ref)
  • Gender does not correlate with happiness (ref) — men and women are equally happy. (this is a separate measurement from depression incidence)
  • Age is not correlated with happiness (ref) (less than 1% of the variance) and has been found to be consistently unrelated.
  • Parenthood: Compared to childless couples, parents worry more and experience more marital problems. Apparently, the good and bad aspects of parenthood balance each other out. The evidence indicates that people who have children are neither more nor less happy than people without children (Argyle, 2001).

As you can see, quite a number of factors that you might expect to be influential appear to bear little or no relationship to general happiness.

Unhappiness Predictors

  • Facebook use predicts decline in happiness (ref)
  • Being sick/disabled highly correlates with unhappiness (ibid)
  • There is a strong link between debt and unhappiness (ref1, ref2)
  • Unemployment makes people unhappy (ref)
  • Having a home loan has a negative effect on life satisfaction (ref)

As you can imagine, you can find correlations for all sorts of interesting things. If you’re interested in digging a little bit deeper, Ed Diener has done a lot of seminal work in this area. For a macro perspective, The World Happiness Report 2018 covers international and rural migration and happiness by country. Another great survey is the Sainsbury’s Living Well Index which measures a lot of factors which correlate with self-reports of happiness in the UK. It’s worth reading the full report, but they found factors separating those living unwell vs those average, and found 8 factors that separated the average person from those living very well:

Conclusions about SWB

We must be cautious about drawing inferences about the causes of happiness. The available data are correlational, not causal. I don’t think the causal direction is the main offender for this field of research though.

The first minor takedown comes from hedonic adaptation. Research on SWB indicates that people often adapt to their circumstances. Hedonic adaption occurs when the mental scale that people use to judge the pleasantness-unpleasantness of their experiences shifts so that their neutral point, or baseline for comparison, changes. Evidence suggests people adapt more slowly to negative events than to positive events. The idea is that the effect of big events on wellbeing is quite fleeting as humans adaptable, suggesting that we have to depend on the non-event correlates like genes, relationships, etc, mentioned above.

I think the major takedown of the research is the following.

Two types of measures: global evaluations (“all things considered, how do you feel about your life?”), and local evaluations (“have you experienced specific positive or negative emotions the previous day?”) are used to measure happiness, based on self report. In both situations, the person being questioned is making a single, retrospective judgement/report about the past (whether it’s a day ago, year, or life). This essentially depends on the individual to firstly accurately remember, remember in sufficient detail, and then calculate their experiences accurately to make a correct summary — what’s called remembered utility. Psychology has shown that there are a lot of quirks in humans ability to do this, and we are terrible at framing. These are called Framing Effects (see dan gilbert’s research for all the issues).

The idea that you can make a single retrospective judgement from a single point in time, and do it accurately, is obviously faulty. A much better alternative would be if we could tell how people actually feel moment to moment, and then we can see whether what they report is truly how they felt. This was (3) mentioned at the start of this article — We can study how people feel about their lives moment to moment. This is called experienced utility. This will be the subject of part 3 of this series.


←Previous …………………………………………………………… Next Up →