Job satisfaction research

Last updated: March 2016.

This page outlines our research into the predictors of job satisfaction.

Research process

To survey the literature, we familiarised ourselves with the latest work on positive psychology by reading all of ‘Flourish‘ and ‘Learned Optimism‘ by Seligman, ‘Stumbling on Happiness‘ by Gilbert, ‘Drive‘ by Pink, and several review papers. We also did a Google Scholar search for relevant terms, read two textbooks on organisational psychology (‘Work Psychology‘ by Arnold and Randall, and ‘The Handbook Principles of Organisational Behaviour’ by Locke (find a copy in this folder)), and two summaries of the job satisfaction literature in the OSH wiki.

We weighed the messages of the literature against our impression of common sense, placing more weight on meta-analyses and consensus positions among psychologists. For more on our general research process, see our research principles.

How robust are these findings?

There appears to be broad consensus that the job characteristics model is a good predictor of job satisfaction. The other factors in our list don’t have as wide consensus, but this page reflects our current synthesis of the evidence that we’ve read. With further research we can imagine adding or subtracting a factor or changing a factor’s relative importance. For example, we previously put less weight on personal fit as important for job satisfaction.

Predictors of job satisfaction

To find out what the predictors of job satisfaction are, we started by looking at general theories of life satisfaction and human needs. We then investigated how these general predictors translate into the world of work and combined this with research that directly focuses on job satisfaction, to come up with six key factors for fulfilling work.

The PERMA theory, created by the founder of positive psychology Prof. Seligman, summarises the key ingredients of living a fulfilling life as follows:

• Positive emotion – feeling happy day-to-day.
• Engagement – challenging, absorbing tasks.
• Meaning – having a purpose higher than yourself.
• Relationships – connecting with others.
• Achievement – being good at something.

It isn’t consensus that these factors constitute well-being, but it is widely agreed that they are important determinants of well-being.

You can read more about PERMA in the book Flourish.

Self-determination theory claims there are three fundemental parts to human well-being:

1. Competence – experiencing control and mastery
2. Relatedness – connecting with others
3. Autonomy – having choice and control

We now outline each factor and the main evidence for each.

1. Engaging work

The job characteristics model claims that job satisfaction is largely determined by how engaging the job itself is. The model breaks engagingness into five main variables: variety, sense of completion, autonomy, feedback from the content of the work, and sense of contribution. Note that the last factor is essentially ‘do you think you’re making a difference?’, and we cover it separately in the next section.

The job characteristics model has been studied extensively, including by several meta-analyses.¹ The most recent meta-analysis of 259 studies showed that each of the five characteristics correlates with job satisfaction (mean p=0.41).7 It’s widely thought to be the best single predictor of job satisfaction. For instance, in the Handbook of Principles of Organisational Behaviour, Judge and Klinger claim:²

There are many possible influences on how favorable one appraises one’s job, and numerous theories of job satisfaction have attempted to delineate these influences. Empirical evidence, however, has suggested only one clear attribute of the work itself that consistently influences job satisfaction – the cognitive challenge of the work.
The empirical data suggest that intrinsic job characteristics are the mostly consistently significant situational predictor of job satisfaction.

The effect of job characteristics on job satisfaction has been found to be moderated by an individual’s need for growth. A meta-analysis found that Growth Need Strength (GNS) moderated the relationship between the five job characteristics and job satisfaction. For those with high GNS the relationship was on average r=. 57, whereas for those with low GNS the relationship was on average r = .32. Note that even this is still a moderate relationship.³

There’s also some indirect support for engaging work being important for job satisfaction. Employee surveys of job satisfaction have shown that “interesting work” turns out to be what’s most important to people, as opposed to pay or status.⁴ Also, sense of completion and feedback are similar to the conditions needed to enter a state of flow – the pleasurable state of being so immersed in an activity that you’re completely free of distractions and lose track of time. Csikszentmihalyi has argued that being able to enter states of flow is key to having genuinely satisfying experiences, so we’d expect this link.⁵

You could challenge the evidence for the job characteristics model by coming up with alternative hypotheses to explain the correlation between job characteristics and job satisfaction. For example, mood could be a third variable that explains the correlation. If you are in a good mood that might cause you to rate your job satisfaction as high, but also to rate your job highly on the five characteristics of the model (as is suggested by Judge & Klinger 2009).⁶ Or you might think that the causal relationship actually works in reverse (as was suggested in Caldwell (1982) — high job satisfaction causes you to rate your job highly on the five characteristics in the model.⁷

However studies that use objective measures of job characteristics still find a correlation between the five job characteristics and job satisfaction (Glick, Jenkins, and Gupta, 1986; Judge, Bono, and Locke, 2000), so these alternative explanations don’t seem to hold.⁸

The more general worry that the evidence is correlational does reduce our confidence in the job characteristic model. Nevertheless, the model is grounded in well-supported theories of motivation and well-being; it provides causal explanations of how the factors lead to job satisfaction; and it has been subject to a high degree of scrutiny for several decades — so if there were obvious confounders or problems with it, it is likely that these would have been uncovered by now.

2. Work that helps others

We’d expect that helping others in your job would increase your sense of meaning and purpose, and there’s a fair amount of empirical evidence that helping others in your job increases job satisfaction.

Theoretical evidence
If you help others through your job, you’re likely to experience a greater sense of meaning and purpose. Meaning and purpose is one of the components of Seligman’s theory of well-being PERMA as we saw above.⁹

Correlational evidence
As mentioned in the previous section, the job characteristics model is widely agreed to be the best predictor of job satisfaction. One of the factors in the model is task significance – the degree to which your job affects other people’s lives. It is measured using this question:

The empirical evidence shows that high task significance is one of the best predictors of job satisfaction. The meta-analysis we mentioned previously found that task significance is strongly correlated with job satisfaction (p=0.41¹⁰).

Another finding which you’d expect if helping others through your work increases your job satisfaction is that jobs which involve helping others would do well in job satisfaction rankings. Jobs that involve helping others do in fact score very well on job satisfaction rankings, as was found in the General Social Surveys conducted in the US from 1972-2006 (with 50,313 respondents):¹¹

If we move to the broader relationship between helping others (whether through your job or not) and well-being, Post (2005) overviews a wide range of studies which find a correlation between altruism and well-being.¹² He also gives plausible causal mechanisms through which altruism could lead to increased happiness:

Altruism results in deeper and more positive social integration, distraction from personal problems and the anxiety of self-preoccupation, enhanced meaning and purpose as related to well-being, a more active lifestyle that counters cultural pressures toward isolated passivity, and the presence of positive emotions such as kindness that displace harmful negative emotional states. It is entirely plausible, then, to assert that altruism enhances mental and physical health.

Experimental evidence
There is evidence from randomized controlled trials that performing acts of kindness increases your happiness.¹³ If there’s a causal relationship between performing acts of kindness and being happier, you might also expect that there is a causal relationship between helping others through your job and increased happiness and job satisfaction.

3. Work you’re good at

We expect that being good at your job leads to higher job satisfaction.
There is a fair amount of theoretical support for the claim that being good at your job will lead to higher job satisfaction. Most theories of human well-being and needs have achievement as a key component (including Self-Determination Theory and PERMA). The job characteristics model also has knowledge and skills as a moderator of the effect of job characteristics on job satisfaction.¹⁴

Also, believing that you are able to do well in a job is associated with higher motivation, whilst not having the knowledge and skills to do your job well is likely to lead to stress.¹⁵

This is confirmed by an analysis of 2,460 individuals which found a strong negative relationship between skill mismatch and job satisfaction.¹⁶

Finally, that being good at your job is important for job satisfaction is also the common sense view, and it is highly intuitive.

4. Work with people you like

Like having a sense of meaning and achievement, satisfying personal relationships are also a key component for a fully satisfying life in almost every theory of human needs and well-being (such as Self-Determination Theory and PERMA). Therefore we would expect social support and liking your colleagues to lead to job satisfaction. This is also confirmed by empirical research.

A meta-analysis of 259 studies found that social support was strongly positively correlated with job satisfaction (p=.56).¹⁷ Social support is the extent to which a job provides opportunities for getting help and advice from others and the opportunities to make friends on the job.

(Update: Another meta-analysis found several measures of ‘organizational sponsorship’, such as being given supervisor support and training opportunities, were among the best predictors of career satisfaction of those studied, though interestingly, were weaker predictors of income and chance of promotion.)

One note is that people vary in who they like, so this factor of job satisfaction will mean different jobs are better for different people. One well-studied effect is that your degree of similarity with someone – the more similar you are (on almost any dimension — physical appearance, attitudes, personality, interpersonal style, cultural background) the more likely you are to like them.¹⁸

5. Meets your basic needs

1. Reasonable hours
Very long working hours crowd out your ability to take care of your non-work life, which could make you less happy, and make you more likely to perceive your job as conflicting with your other goals and needs. Not being able to meet your non-work obligations is also likely to lead to stress.
This claim receives some support from two large surveys. The British Household Panel Survey, of roughly 10,000 individuals, found that long hours were associated with lower job satisfaction.¹⁹ Another survey of 28,240 individuals found the same.²⁰

2. Job security
Lacking job security is likely to lead to greater stress as it is likely to decrease people’s confidence in being able to meet the demands of their lives. There is also theoretical support for this: job security is posited to be a moderator of the effect of job characteristics on job satisfaction by the job characteristics model.²¹ Also, a meta-analysis of 50 independent samples representing 28,885 individuals, found a strong negative correlation between job insecurity and job satisfaction (r=-.41).²²

3. Short commute
We would expect a long commute to be detrimental to life satisfaction for the same reasons as long hours are, and in addition, long commutes tend to be unpleasant in themselves and hard to use productively. This claim is supported by a study by the UK Office for National Statistics (sample size = 60,200 individuals) which found that long commutes were associated with lower life satisfaction. The worst effects were associated with journey times lasting between 61 and 90 minutes. Also taking a bus or coach was found to be the worst mode of transport for commuting.²³

4. Fair pay
Earning less than others doing the same or similar job seems to cause job dissatisfaction and lower happiness. This makes sense given that it’s likely to make you feel unfairly treated and possibly resent your employer.

This has been confirmed in correlational studies (Clark and Oswald, 1996; Hamermesh, 2001 for job dissatisfaction, Frey and Stutzer, 2002; Luttmer, 2005 for happiness).²⁴

You might worry about other potential explanations for the correlation. But a clever experimental design was used to control for confounding variables. It randomly assigned people to either be able to find out what their peers earn, or not (the control group). Those who were able to find out what their peers earned and found out that they earned less than their peers reported less job satisfaction and had higher intentions to leave their jobs than those in the control group.²⁵

Other factors we considered but that are poor predictors of job satisfaction

Income

In general, the salary of a job has only a weak effect on job satisfaction and happiness – see our review of the evidence.

Low demands

Moderate levels of stress are associated with higher job satisfaction. See our review of the evidence on the effects work demands and stress.

Interests match

1. There is lack of empirical evidence for interest-match, with a major attempt to find one (Holland-match). See our review of the evidence on the importance of Holland type match with occupation and job satisfaction.

2. Our interests change, and more than we expect.

3. Most people are likely to be interested in highly competitive areas with very few jobs, like sports, music or entertainment. See our page on why “follow your passion” isn’t helpful advice for having a satisfying career.

Signature strengths match

Another predictor we considered but didn’t end up recommending is finding a job which is matched to your unique signature strengths (a test created by Seligman and his colleague Peterson). From the evidence so far, it doesn’t seem that using signature strengths for choosing between jobs is useful. But it does seem that once you’re in a job, finding ways to use your signature strengths more does increase your happiness. See our review of the literature for more.

Occupational status

It is common sense that occupational status is something that matters. There’s some empirical evidence that it is correlated with job satisfaction (for example in the General Social Surveys conducted in the US.²⁶

However we didn’t include it in our list of factors because:

• The correlation is easily confounded by other variables such as the five factors of the job characteristics model, social support, etc, all of which have better empirical support.
• We don’t expect people to neglect this factor due to the strong social incentives to select jobs with higher status.

Directions for future research

In the future we would like to find out more about the base rates of job satisfaction in different jobs, whether the Big Five personality factors are helpful for finding a career that’s satisfying and investigate how large a difference individual factors, such as how good you are a job, make to job satisfaction.