For the average modern worker, their working week is around 40 hours per week. So assuming that you work for 40 hours per week, from ages 20-60 before you retire, the total amount of hours that you will work in your life is roughly 80,000 hours (50 weeks x 40 years x 40 hours). That’s a lot of hours.
If you’re like the majority of humans, you will have no option but to work to generate income to provide yourself with food and shelter. Sure, you do have other options like being voluntarily homeless, opting for a life of extreme frugality - eg in a van), living off of welfare or leeching from others, etc. But to most, these are not seductive alternatives to simply abiding by the system which gives them a regular paycheck.
“The three most harmful addictions are heroin, carbohydrates, and a monthly salary.” - Nassim Nicholas Taleb
Fortunately, you have a lot of control over how you will spend this time. And given the disproportionate amount of your life that you will spend working, surpassed only by the amount of time spent sleeping, it is only logical to spend some time thinking strategically about how you will spend it, otherwise you risk living an unexamined life.
Not only do you spend a lot of time working, but how satisfied you are with your job is also one of the strongest determinants of how satisfied with your life you are. These concepts in the research are both referred to as job satisfaction and life satisfaction measures. There is a significant correlation of 0.44
1 between them. It is suggested that the relationship is reciprocal or bidirectional in cause - ie life satisfaction may have a positive influence on job satisfaction, and job satisfaction can also positively influence life satisfaction.
Funnily enough, retirement or unemployment is not the answer. Unemployment turns out to make people very unhappy
2 (see here for a full exploration). Other research supports that job satisfaction strongly correlates with happiness.
4. This fits our intuition about the retired getting less happy potentially due to boredom or losing social connections.
According to Gallup
5, in 2013 around 70% of American workers are either “not engaged” or “actively disengaged” with their work. Given what I’ve said above, obviously this is not an optimal way to spend a finite existence. The purpose of money is to be a flat medium that allows you to exchange it for things that you want - or more basically, the purpose of money is to solve problems. If the generation of money creates more problems than it solves, then it is no longer serving its purpose and its usefulness should be reevaluated.
Throughout this article, I will be sharing my independent research on which correlations make up job satisfaction, and hopefully by doing so I will increase your awareness of what the current research says about the determinants of a satisfying job, and possibly contribute to your future choices in some way.
If you want to skip straight to the results, see the research summary for the correlation matrix, the conclusion for the readable summary of the whole article, or the ultra compressed conclusion if you want a TL;DR, but note by doing so you will be losing a lot of granularity and context and the suggestions will bear little weight.
Job Satisfaction (JS henceforth) is a measurement taken in Psychology of an attitude that people have towards their job, measured in a variety of ways. It is linked to
6 worker productivity and performance, motivation, workplace civility, prosociality, absenteeism, lateness, stress
7, accidents, withdrawal intentions and employee turnover/retention, organizational commitment, mental and physical health, and most interestingly Life Satisfaction.
The most widely built upon definition of job satisfaction is that it is
“a pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences”.
There are three main ways to measure JS: evaluative, cognitive, and behavioural. Evaluative satisfaction is when the subject reports “I like/dislike my job”. Cognitive satisfaction regards beliefs about the nature of the subjects work - eg that their work is challenging / respectable / demanding / rewarding. And indirect behavioural measurements of JS are found via absenteeism, lateness, sickness.
There are five main schools of thought on what leads to JS:
Schools 1-3 Have fantastic data on them which we will look at to see how much they are related to job satisfaction. Schools 4 and 5 both have some data but not quite as robust. We will then look at a handful of other variables which also are less well-supported than the first 3 but nonetheless are relevant - such as income.
I’m going to cover what the research on these has to say, and then explore a list of variables that I think are noteworthy, and hopefully during that process we can discover what are the most valid determinants / antecedents / predictors of job satisfaction so that we can control them. Although the following may seem like a lot of information and correlational evidence, note that the main paper it is drawn from - Humphrey (2007)
8 originally found 6,333 unique correlations across 260 articles with a pooled sample size of 75,000 subjects, and those discussed here were the most noteworthy out of them all.
Throughout the article I will be posting a number that correlates with the amount of job satisfaction, eg “Noise correlates with job satisfaction by 0.33 amount”. If you are unsure what this means, it is this: Suppose you wanted to figure out what the likelihood of getting in a car accident whilst driving drunk was. The way to answer that question is to look at a sample of drivers, lets say 100 of them, and look at how many got in an accident in total over a certain period, lets say 12 months. If 20 of those drivers got into a car accident, and we found out at that every single one of them had alcohol in their system, the correlation between drinking and getting in an accident is 100% or 1.0. That is, every single driver that got in an accident also had alcohol in their system. When correlations are this strong, and with a large sample size, you can reasonably deduce that one thing is causal on the other, or that they correlate highly enough that they both overlap on some third causal factor which is still affected by it. If only 1/3rd of the drivers had alcohol in their system, then the correlation would be roughly 0.33.
How does this all relate to job satisfaction? Basically the more characteristics you have down, the more likely you are to have job satisfaction, even if they aren’t causal. Just like finding out that 33% of driving accidents are by drivers with alcohol in their system, and it being wise to avoid doing this thing that correlates with that risk, in the same vein it is wise to follow the correlations for JS. And note that in the same way not drinking doesn’t 100% prevent the risk of accidents - you can still get hit by another driver, you can have certain JS factors under control and other factors can still have the ability to raise or lower your JS.
Providing little insight, the absolute best measure we have of Job Satisfaction as a whole is done via the “Overall Job Satisfaction Measure”, a 5-item inventory that correlates extremely highly at 0.85 which is what you’d expect when it is directly asking you about JS:
All things considered, are you satisfied with your present job (circle one)? YES NO
How satisfied are you with your job in general (circle one)? 1 very dissatisfied; 2 somewhat dissatisfied; 3 neutral; 4 somewhat satisfied; 5 very satisfied.
Motivational characteristics encompass the characteristics of the job which are related to the nature of the work itself, and as a result, how that affects the worker. The original model is called the Job Characteristics Model (JCM henceforth) and is the most widely studied, oldest, and strongly supported theory in job satisfaction research.
“The most consistent predictor of job satisfaction is mentally challenging work (2) - derived via meeting the JCM”
There is widespread agreement that the job characteristics model below is the strongest predictor of what constitutes a satisfying job above and beyond all other factors. Of all the variables correlated with job satisfaction, each 5 of the job characteristics model criteria correlated with job satisfaction at a high .41.
The JS research is one of most well researched approaches and identifies that it is the nature of the job itself, not anything else, that determines how satisfying the job is. The original and withstanding model was developed in 1975 by Hackman and Oldham
11 and proposed five factors which make work satisfying for workers and that make up a satisfaction job:
These were all originally measured by what is called the Job Diagnostic Survey (JDS)
11, a series of questions with a 1-7 scale. The survey that is used to test it is most recently called the Work Design Questionnaire, and you can see the questions asked here.
All of these factors together result in high work engagement. The more of them you have, the more likely the job is to be satisfying, irrespective of who is doing the job, where it is, how much it pays, etc. You want a job that engages you because challenge gets you into flow
The right job according to Mihaly C is a job that challenges us to an optimal level - one that’s neither so hard that we give up, nor so easy that we get bored. Finding a job that engages your natural talents and give you constant feedback is a sure way to build happiness.
What’s the most interesting about the job characteristics model is that none of the criteria are uniquely related to the subject matter of any specific job, a failed bias we are typically drawn to. Instead, what the content job is, or is about, doesn’t matter so much as what characteristics it has.
Autonomy is the freedom that an individual has in carrying out their work, or the degree to which one they control over how to do a job. In the JDS, it was measured by asking
How much autonomy is there in your job? That is, to what extent does your job permit you to decide on your own how to go about doing the work?
It was later argued
13 that autonomy wasn’t a singular construct as it was originally being measured, but rather it was made up of multiple factors that each had their own predictive values. Most important being:
So roughly speaking, job autonomy is how much the job allows you to decide yourself how and when to do your work. A paper also calls it simply “absence of close supervision”
Skill Variety is the extent to which an individual must use different skills to perform at their job. Separate but related is Task Variety which is the extent to which an individual performs different tasks on the job. Task Identity wasn’t originally measured by Hackman
11 but was instead measured in a separate survey called the Job Characteristics Inventory in 1976
Both constructs are very similar but have slightly different predictive power. Skill variety focuses on the skills of the worker, whereas the task variety focuses on the job itself which is more consistent with the other 4 factors of the JCM. As a result of the similarity between skill variety and task variety, they have often been combined and measured as one construct in many studies.
Task Identity was measured by asking:
How much variety is there in your job? That is, to what extent does the job require you to do many different things at work, using a variety of skills and talents?
Skill Identity was measured by asking:
How much variety is there in your job?
How similar are the tasks you perform in a typical work day?
The opportunity to do a number of different things
The amount of variety in my job
So roughly speaking, task and skill variety are together the extent to which a job allows an individual to engage in different tasks utilizing different skills. A paper also calls it simply nonrepetitive work
As an example of what work looks like that has low skill variety or task variety, consider driving a bus in circles around specific routes.
Task Identity is about the extent to which the outcome of your work is connected to the tasks that you do, or about the extent to which an individual can complete a whole piece of work from start to finish.
Task identity was measured in the JCM by asking:
To what extent does your job involve doing a “whole” and identifiable piece of work? That is, is the job a complete piece of work that has an obvious beginning and end? Or is it only a small part of the overall piece of work, which is finished by other people or by automatic machines?
As an example, piecemeal work where you are handling one piece of a whole product in a factory repeatedly is obviously bad for many reasons, most majorly likely being repetitiveness, but another reason it is bad is because the outcome (the final product) is abstracted away from your tasks.
Task Identity implies that it is better to see work from the beginning stage to the end, and be a part of that whole process, for whatever it is.
Task significance (TS) is about helping others. It is the extent to which your job impacts others lives, contributes to their wellbeing, or how important you see your job tasks. It’s related to how meaningful your work is, or more concretely, how directly and intensely it has an impact on other’s wellbeing.
TS was measured in the JDS by asking:
In general, how significant or important is your job? That is, How much are the results of your work likely to significantly affect the lives or well-being of other people? and
Grant, et al.
17 later re-evaluated TS and updated the JDS survey question to:
> My job provides opportunities to substantially improve the welfare of other people
> My job enhances the welfare of others.
He then extended his measurements beyond questions directly about TS and looked at four others indirectly, again on a 1-7 scale:
1 and 2 are outcome measures and demonstrated a causal link between TS and JS / job performance. What is noteworthy is that the JCM assumes TS is a job characteristic, but
17‘s experiments demonstrate that it’s a social characteristic instead. How much TS affects job performance and JS is mediated by how much an individual subjectively evaluates their job for 3 and 4.
Work that helps others is also often linked to meaning, and vice versa. An extreme view which potentially captures the dissatisfaction that results from meaningless work is by the psychologist Erich Fromm:
I think [mans]work is to a large extent meaningless. Because he is not related to it, he is increasingly part of a big social machinery governed by a big bureaucracy, and I think American man subconsciously hates his work very often, because he feels trapped by it, imprisoned by it, because he feels as if he is spending most of his energy for something which has no meaning in itself. He’s not dignified in making a life with his work because that’s not quite enough to make one happy, if one spends 8 hours a day in something which in itself has no meaning and interest except that one gains money from it.
This is a little extreme, but it highlights the concept. Interestingly, there is a great organization called 80000 hours which focuses on effective altruism, and they think that one of the best ways to go about this is to use your career effectively. As a part of that, they also do a lot of JS research and career advice, some of which I recommend here. One of the main arguments that they make is that JS is primarily driven by task significance, and that you should design your career to have the largest social impact that you can to increase your job satisfaction and improve the world at the same time. Although they have gained a lot of popularity recently for various reasons, their argument hinges on how significant task significance really is.
Firstly, in regards to motivation, the motivation to do prosocial work is based on the extent to which the nature of your work allows you to interact with those who benefit from it
18 - by frequency, proximity, duration, depth, and breadth of contact. This means that the 80k argument to benefit others indirectly (ie. by donating to effective charities) won’t have as large of an effect on your JS as they advise.
But most critical of 80k’s mission is that the relationship between Perceived Social Impact, and Perceived Social Worth is moderated by conscientiousness, and pro-social values (see below). conscientiousness negatively mediates the relationship - ie the higher the conscientiousness, the lower the effect of task significance on job performance (due to effective performance and high effort being a reward in and of itself). And the higher the pro-social values, the higher the effect of task significance on job performance (due to them being more influenced by perceived social value and worth).
To measure prosocial values, use the benevolence subscale of the Portrait Values Questionnaire
- It’s very important to him to help the people around him. He wants to care for other people.
- It is important to him to be loyal to his friends. He wants to devote himself to people close to him.
- It is important to him to respond to the needs of others. He tries to support those he knows.
- Forgiving people who might have wronged him is important to him. He tries to see what is good in them and not to hold a grudge.
Back to the JCM - Task feedback, also referred to as Feedback From The job, is the extent to which a job imparts information about an individuals performance. Or roughly, the degree to which the work itself provides feedback of how well one is performing on the job.
Task feedback was measured by asking:
To what extent does doing the job itself provide you with information about your work performance? That is, does the actual work itself provide clues about how well you are doing - aside from any ‘feedback’ co-workers or supervisors may provide?
Though the JCM is a strong predictor (one of the strongest) of JS, the effect of them on JS was found to be moderated by an individuals needs for growth in what’s called the Growth Needs Strength (GNS). For those high in GNS, the correlation between JCM and JS is very high at an average of 0.57, which is the strongest correlation that we have, but for those low in GNS it is .32, which is still quite a strong relationship, but less strong. You can measure your GNS quickly and easily here. The scale ranges from 12-60, with low growth needs being 12-29, moderate being 30-42, and high being 43-60. My GNS is 32, which is moderate. This means that the correlation is around .44-.53. Interestingly, I did the test around one year later, forgetting my previous results, and scored 38 again which is still in moderate, demonstrating its stability.
What this shows is that although the JCM is one of the most important factors in determining what type of job you should seek out, you should weigh it slightly less than other factors if your GNS is low. Put roughly speaking - you need a less engaging job to be satisfied, as other factors are likely to be just as or more important.
The JCM comprises the 5 core motivational characteristics, but 5 more were further identified later on as being potentially important, and discussed in the literature:
The JCM and task variety are concerned with the nature of the job itself, whereas these other 4 are focused with the knowledge demands of the work.
Information Processing is the extent to which a job requires the individual to focus on and manage information. Jobs differ in their required levels of monitoring and processing of information. Jobs that have higher processing requirements are expected to be jobs with more knowledge required in order to complete their work.
Job complexity is the extent to which a job is multifaceted and difficult to perform. As a result they tend to be mentally demanding. Oddly, although high complexity is expected to increase JS, it is also more likely to promote stress and feelings of work overload, even if engaging.
Specialization is the extent to which a job involves the performance of tasks requiring specific knowledge and skill, and the depth of them (rather than skill variety).
Problem solving is the extent to which a job requires the production of unique solutions or ideas - by innovating, solving nonroutine problems, and dealing with (or preventing) errors.
The third school of thought on what constitutes JS is that people primarily derive their evaluations from the people that they surround themselves with. It’s established that humans are social animals and are generally positively affected by socialising, making us more energetic, happier, excited, enthusiastic, and it is a fundamental motivation for most behaviour.
Lately, researchers have thought social characteristics via the relationships between workers as being among the most important causes of JS. This fits with our intuition that working with people you dislike is terrible, or vice versa. The ways in which social characteristics may influence JS are many, but the ones that have been noted as potentially important are:
80000hours simply refers to this school of thought as “work with supportive colleagues”, which accurately captures the meaning.
Social information processing is the theory that employees look to their coworkers to make sense of and develop attitudes about their work. As a rivaling theory to JCM, Jex (2002)
20 argues that employees look to coworkers to see what their levels of job satisfaction are, and the attitudes they have about their work. If their coworkers attitudes are positive, they will more likely be satisfied, and the converse. Put simply - if your coworkers are (dis)satisfied, so will you be. This is interesting because new hires can be ‘tainted’ during the socialization process if placed around dissatisfied employees.
This school of thought is less about what job or criteria help in promoting JS and moreso about the social influences on the perceptions of the job characteristics themselves.
Generally the research supports social information processing as having an effect on job satisfaction, but the methodology of the studies is questionable, and there doesn’t seem to be a clear consensus on the correlation with JS.
Interdependence is the extent to which your work relies on interacting with others in a team to complete. The more interaction it requires, the higher the interdependence. It is the extent to which a job is contingent on the work of others, and other jobs are dependent on the work of yours.
This is about the extent to whether other organizational members provide performance information/feedback to an individual, separate from the JCM criteria of ‘Task Feedback’.
This is the extent to which a job provides opportunities for getting assistance and advice from supervisors or coworkers.
As simple as this criteria is, it turns out to be one of the best predictors of JS.
This is the extent to which a job requires an individual to communicate with people (eg suppliers or customers) external to the organization, also referred to as Serving The Public.
You would think that being able to chat with coworkers on the job would be quite important for job satisfaction, but it only has a mere .04 correlation, but a .11 correlation with net emotional state while on the job
Environmental Characteristics, or ‘Work Context Characteristics’ as they are more oddly referred to in the literature, are a relatively neglected area of research, but which involve the features of the working environment itself that may influence JS, the three most noteworthy being:
These are expected to influence JS via being more related to biological concerns for safety and survival, as a fundamental motivation.
This is concerned with the extent to which a job involves physical strength, endurance and the activity requirements of a job.
Work conditions capture aspects of the environment generally, such as health hazards, temperature, and noise.
Being exposed to offensive noise while at work only affects JS by a mere -.04, but it turns out that it has a bigger effect than any other factor on net emotional state while doing the job, at a whopping -.90 correlation
21. This means that noise pollution while working basically makes a job more dissatisfying than any other factor.
I couldn’t seem to find any research done directly on job satisfaction unfortunately, but it is exactly what you would expect - the effect of temperature on JS.
Ergonomics involves the extent to which work permits appropriate posture and movement.
Although there is a dizzying array of research on the effect of CO2 on job performance (nice primer here), I couldn’t seem to find any research done directly on JS.
Worker Characteristics (or Dispositional Caracteristics as they are referred to in the lit), are the extent to which JS is the result of internal traits of the worker. Some people may just be optimists or pessimists in regards to their work. The idea is that certain personality traits (eg how neurotic you are) can influence your evaluation of any job. Though the school of thought doesn’t aim to invalidate JCM or social characteristics, but rather to explain why those general principles don’t account for all of the variation / correlation in JS.
It is one of the latest methods of explaining JS, arguing that JS hinges on personality and individual traits, and that some people are inclined to be satisfied or dissatisfied with their work irrespective of the job characteristics or other factors
The support for this idea is a few folds:
There are a handful of dispositional and personality characteristics which are potentially relevant to JS, but the three most important aspects of personality are:
And fourthly, considering the hereditary nature of JS, these personality traits are believed to be partly genetic, with genes explaining up to 30% of the variation in JS
26. That is, the genes of genetically identical / monozygotic twins explained .29 correlation with JS, whereas genetically dissimilar / dizygotic twins explained -0.02
27. So it’s clear that there’s a genetic relationship, but interestingly, the relationship only moderated JS, and not external characteristics such as pay.
In Psychology, the spillover-crossover model examines the relationship in how the work domain affects the home domain, and vice versa. Specifically, it concerns how well-being is transmitted from one to the other. For example, having family conflict at home can spillover and affect your relationships at work.
Related to this is the link between life satisfaction and job satisfaction. Interestingly, it turns out that there is a stronger causal relationship between your happiness on your job, than from your job on your happiness
28. Research suggests that life satisfaction often precedes and is a good predictor of job satisfaction.
The Big 5 Personality Traits are Openness, conscientiousness, extroversion/Introversion, Agreeableness/Disagreeableness, and Neuroticism (OCEAN). The Big Five is an extremely respected area of study in Psychology and carries a huge amount of support and widespread consensus as a model. It has predictive power for many many fields, only superseded by the power of IQ. There is a lot to say about it, but a basic intro can be found on wikipedia.
Again, we’re curious as to how this relates to JS. In summary, conscientiousness has a consistently significant impact on JS, whereas openness has a consistently insignificant one. The other 3 - neuroticism, extroversion, and agreeableness, have inconsistent relationships.
Openness is characterized by intellectual curiosity, imagination, and non-conformity. Although important for many other fields, it has little effect on JS - correlating at 0.0
conscientiousness is an individuals level of organization, hard work, and motivation in the pursuit of established goals. Its a strong perdictor of job performance and success, more than any other personality or individual trait besides IQ.
Given that, it may explain why conscientiousness is positively related to JS. It seems to have a consistently significant impact on JS, and low conscientiousness results in low JS.
The Correlation is about .26
29, and it’s also highly predictive of work performance.
extroversion is the extent to which individuals are assertive, active, enthusiastic, energetic, and dominant. Roughly speaking they are more socially oriented than their introverted counterparts.
There appears to be a very strong relation between extroversion and JS, at about .24
29 - .25
Agreeableness is the level of cooperation (trusting of others) and likability of an individual, usually resulting in positive social relationships with others. It’s unclear whether how strongly it affects JS
29, but the correlatation seems to be around .26.
Neuroticism is the extent to which an individual lacks positive psychological adjustment and emotional stability. ie high neuroticism = low emotional stability. It’s a little more subtle than that, but the gist is that they are more liekly to experience negative emotions - including anxiety, depression, hostility, etc
Neuroticism is believed to have the strongest effect on JS out of all the big 5 traits, but in a negative manner, and emotional stability has an inversely positive effect. Though some other research contradicts this.
It correlates at -.26
29 to -.29
Core Self-Evaluations (CSE) are a broad personality concept that encapsulate 4 specific personality traits:
The mechanism on JS is believed to be mediated by the effect that these personality traits have on the perception of the job characteristics themselves. If we have a positive self-regard, we are likely to see our jobs positively and undertake ones that are challenging, and vice versa. The various processes that potentially mediate the relationship are that positive core self-evaluations lead us to both seek out and attain more challenging work, and then further perceive the work as more challenging and interesting thereafter
30. The individuals with positive CSE are also more likely to work towards goals for reasons consistent with their values
31 which would make them more satisfied.
Overall, the 4 core traits combined correlate highly with job satisfaction at .41
30, or .37
3, though the individual traits vary. The measures were all originally devised by Judge et al 1997
32, and including a 1-5 scale (strongly disagree - strongly agree).
Self esteem was measured by using Rosebnbergs 1965 10-item scale, including items such as:
I feel that I am a person of worth, at least on an equal basis with others
At times I think I am no good at all
Correlates .43 with job satisfaction
Measured by asking 7 questions, such as:
I am strong enough to overcome life’s struggles
I often feel that there is nothing that I can do well
Correlates .33 with job satisfaction
If employees don’t believe that they can successfully perform a job, this may results in low job satisfaction, and an organization might want to consider techniques designed to increase employee self-efficacy (see Chapter 10 of the handbook here
Locus of control is the extent to which an individual believes that he or she has control over the events in their life, positive and negative.
Locus of control is measured using 4 items, including:
When I get what I want, it’s usually because I am lucky
My life is determined by my own actions
Neuroticism is the degree to which an individual experiences feelings of irritability, nervousness worry, embarrassment, or guilt. It is essentially the same as the big 5 neuroticism trait, but measured in a slightly different way.
It’s measured by asking the individual to rate their level of agreement with statements such as:
I am a nervous person
I am a worrier
Affective Disposition is how likely people experienced positive and negative emotions. If you experience more negative than positive emotions on whole, you have higher trait negative affectivity, and positive.
Trait Negative Affectivity is somewhat more strongly correlated (.-37) to JS than trait Positive Affectivity at .29
Though when compared to Core Self-Evaluations, Affective Disposition correlates less strongly to JS, explaining less variance.
Having a career which is a good ‘fit’ for their personality is usually extremely high priority for most people. Person-Environment Fit is the most intuitive of theories around what causes JS. It is a high-level superset of a few other theories - such as Personality-Job Fit.
It states that JS is the result of an individual being in the correct environment - one that matches their interests, preferences, personality, “passions”, etc. This is the most intuitive because it fits with a lot of the career advice that is out there - namely that the way to find a satisfying job is to firstly discover your interests and passions, and then to find a job that allows you to do those. Basically, to make sure that you put the right person in the right job.
There are a few main tests worth mentioning based on their popularity, and merits:
Although these are not perfect approaches, I think they are better than systematically going through an uncategorized list of jobs and trying to forecast what you would and wouldn’t enjoy, as people have a poor ability at predicting what will make them happy.
Although person-environment fit doesn’t have the strongest predictive power, it could be potentially worth exploring in the chance of finding a career full of tasks that you love to do.
A very crude approach of how this could be done is something like the extremely popular Myers Briggs Type Indicator (MBTI) test which most of us have heard of or taken. It gives you your 4 letter personality type, such as “INTJ”, and then lists a bunch of careers at which those personality traits excel, like this or this.
Although the MBTI jives with a lot of our intuitions, it has very low validity and theres a 50% chance that you will receive a completely different result if you retake the test 3 months later
33. Also, if you are a teenager, research suggests
34 that you have a lack of insight into some of your abilities of what you are good and bad at, and this can have implications for trusting your intuitions over the career that you pick.
Stronger and much more supported is the Holland Code Test, aka the RIASEC which tries to categorize people into 6 different types, being:
It is again a similar concept to the MBTI where you divide people up into 6 neat types and then you find jobs that those types like to do. Although it seems almost too convenient, there is research that it works to some extent - with an established correlation of 0.21 with job performance
35. And that correlation makes sense if you consider that jobs can be grouped into what type of tasks they involve, and that perhaps people can be divided into the types of tasks that they love to do or enjoy doing.
Unfortunately the paper referenced above is only interested in job performance and not job satisfaction, so we cannot see what we are interested in directly. Though there is a strong correlation between JS and job performance, so we can potentially use job performance as a proxy. The paper
Lastly, more research is needed on the relations between vocational interests and other important criteria (e.g., job satisfaction)
Although there is a large amount of research supporting the idea that the 6 different types are valid and do exist, its relation to JS is more sparse and inconsistent. But the findings are that its relationship can range anywhere from .15 all the way up to .54
36! How much you can trust this is questionable though.
A more robust meta analysis of 9000 individuals found only a 0.10 - 0.20 correlation with JS
Holland did establish that people are already motivationally drawn into work environments that are compatible with their interests. The question is how that relates to JS. He also argued that JS is unlikely to come merely from making the work more mentally challenging without being personally interested in it, eg studyin the law is mentally challenging, but many people have no interest in it at all
I think it is worth doing the free test or the more accurate one here and perusing the jobs that they recommend here out of curiosity to help widen your options. Self directed search is another name for the Holland test. These tests are basically aptitude tests that take in self-reported personality factors and output a list of careers that you might be well suited for. Then you have between 6-20 options to research more in depth for job caracteristics, etc.
Though there is a stronger version of the Holland codes called the Strong interest Inventory which correlates 0.27 with job performance
35, and I recommend doing that test instead if you are going to do one.
The test attempts to measure 24 strengths which comprise 6 virtues (bolded):
80000hours think that the strength of the evidence as it relates to JS isn’t too high, and may be more important for job crafting than selection - that is, it may not be useful for picking a career but may increase your job satisfaction slightly if you utilize your strengths in your current job.
Though I found some more recent research which directly looked at the relationships
39. First, it turns out that different strength correlate differently with JS, with the most highly being :
This is a similar list to which correlate most highly with life satisfaction.
The 24 character strengths together accounted for 12% of the variance in job satisfaction beyond gender, age, and education.
Another paper which looked at the correlation of all 24 strengths at work found that they only correlate 0.07
Most of the other person-environment fit approaches are focused on your interests and preferences and matching them with the job. But a later school of thought
41 is that fit is about how likely you are to exceed at your job, and this could potentially be important for JS as performing highly in your job can lead to JS.
See the Job Performance section below.
Beyond the five main schools of thought above, there has been a lot of separate research into individual variables which affect JS. I’ve included research on those below that I believe are the most noteworthy:
Income level is obviously one of the top factors that people weigh highly when career planning or predicting what jobs are satisfying. Obviously this has been recognized by JS researchers. Warr (1999)
4 simply referred to it as the “availability of money for needs”.
To start off with, it used to be thought that the social norm to work was the cause by which unemployment made people unhappy, but it turns out that it is is not affected by the social norm to work, but instead it was found that affluence and possessing money in and of itself was what mattered
If high income turned out to be important, what you could do is look at base rates for starting and lifetime income such as here or here
43 and try to optimize your career based on that. This is a more efficient approach than what most do. But the question is: is it smart to do so, and does it increase JS?
Hackman originally noted in the JCM
11 that if pay is low, job redesign by improving the JCM might fail. He noted that pay might also be more important if you have lower Growth Needs Strength.
Ok so what’s the effect on JS? First, as you would expect, high income does not actively make you happy, but low income actively makes you unhappy. So aim for the absence of a low income rather than the presence of a high income. How much is high/low?
Firstly, a lot of pay satisfaction comes down to it being Pay That You Think Is Fair, which is quite subjective. There are two components to that:
Both of these moderate whether you think a pay is fair or not
44 which influences your satisfaction with it.
As you’ve probably heard, research shows diminishing returns to increased income levels.
As the income level increases, its effect on happiness flattens out.
The effect of income on an individuals happiness (maximizing positive emotions, minimizing negative emotions, and minimizing stress) starts to flatten out at:
And completely peaks, with the effect becoming non-existent at:
This is much lower than most people expect, because all the studies are done on household incomes which people conflate with individual incomes. But, the effect of this is mediated by which country you grew up in and your cultural context.
And finally, related specifically to JS itself, pay level was found to correlate only .15 with job satisfaction
So the conclusion is not to aim for income to increase job satisfaction.
Job Status is to the extent that your position is socially valued by others, which in turn influences your perceived social worth. A paper also calls it simply self respect from the job
4. For example, lawyers and doctors are two conventionally high status positions. This is one of the top criteria that younger people select by (interesting piece about that here).
A low status job affects job satisfaction with a -.23 correlation, but only affects net emotional state on the job by -.12
As noted many times throughout here, job performance and JS are tightly linked.
How well you do at your job has a bidirectional effect on job satisfaction - meaning that the more satisfied you are with your job the better you perform, and also the better you perform, the more satisfied you are with your job. Both job satisfaction improves job performance, and job performance improves job satisfaction
80000hours weighs this quite heavily and recommends selecting for work that you could excel at. There may be some rationale to this argument - see here and ref
Research shows that low stress jobs don’t make you happy, but rather a challenging job. Some stress is bad, some is good. If you have few demands placed on you, then its more likely that you’ll be bored.
Moderate stress levels are good for JS
47. Resources on improving occupational stress here.
Being overqualified for ones job affects JS with a -.20 correlation, and affects net emotional state on the job by -.11
If the job that you are doing requires specialized education or training, it positively affects JS with a .20 correlation, and net emotional state on the job by 0.10
organization size also impacts job satisfaction
On average, employees who worked in smaller enterprises with 5 to 19 employees reported higher levels of satisfaction with their current job (5.54 / 7) than employees who worked for medium sized (20-199 people) (5.39 / 7) and larger enterprises (200+ people) (5.32 / 7).
Also, slightly unrelated to job satisfaction, but quite noteworthy is a set of research called the whitehall studies, and what was found was that the larger the organization, the more stratification there is between classes of workers, and the greater the job stress for those lower down.
Females are more satisfied with their job than male employees on average
49. Male employees are more affected by the level and autonomy and control over their work, whereas females are more affected by work-life balance (balancing work and non-work commitments).
If your job requires that you are under constant pressure to work fast, it negatively affects job satisfaction with a -.19 correlation, and affects net emotional state on the job by -.28
21. Does this make hospitality a bad choice? Potentially.
If your job required constant attention to avoid mistakes, it affects job satisfaction negatively by only -.04 correlation, but it affects net emotional state on the job by -.13
Recall that life satisfaction is a global measure of how happy people are with their lives as a whole. It turns out that whether you are employed or not has a strong relation to this. You would think that being able to louse around all day would make you happy, but it turns out that unemployed people are less satisfied with their lives and unhappier than employed
50. People on government bucks also have double the life dissatisfaction of the employed.
So employment makes you happier than unemployment, but self-employment happens to make you very slightly less satisfied than regular old employment.
Within those that have regular employment - part time workers have slightly higher life satisfaction and well-being than full-time workers, as many would think.
There would be a lot of relationships to tease out here - eg self-employed may be less happy due to optimizing for income, and employees for reduced responsibilities. But the results are interesting nonetheless: that there is a higher percentage of self employed people with very low life satisfaction than employed, and a higher percentage of part time workers with high life satisfaction than full-time workers, by 10%.
The conclusion is that being jobless is not the solution to the 80000hours of work problem. What you want is challenging work that increases JS and life satisfaction - to work on something that you love.
As mentioned above, part time workers (<38h/wk) have higher overall life satisfaction than full-time workers (38h+/wk)
48. But they also have slightly higher JS:
This fits with our intuition that having time affluence is better for our well-being, but also contradicts our desire for wealth. It is noted by a lot of JS Models as a ‘hygiene factor’ - that is, hours worked may only begin to affect JS when the hours become obsessively long.
For a full exploration of the effect of time scarcity on well-being, I recommend the very interesting book Happy Money.
Creative occupations (architecture, crafts, design, music, performing arts, and visual arts) report having above average levels of well-being, but also above average levels of anxiety.
Jobs in marketing and advertising, film, TV, video, radio and photography, IT,
and publishing are associated with lower levels of wellbeing than non-creative jobs
Unfortunately I couldn’t find anything relating to JS.
Quality I can’t seem to find any great studies of the effect of sleep quality on JS (send some over if you can find, I’d love to see), but the importance of sleep quality for life satisfaction and well-being is one of the most significant variables, so it would only make sense for it to spillover to JS as well.
Age does not appear to have any significant impact on JS
People who have long commutes are much more miserable than people with short commutes. The fact that people undervalue their commute to work even has it’s own name - The Commuting Paradox
In terms of experience, commuting is the least enjoyable of 19 daily activities measured in a well-designed study of how people spent their time and feel about it
Distance to work is established as one of the biggest determinants of well-being, but I can’t find anything directly testing its correlation to JS. Related to well-being, commuters have lower life-satisfaction, a lower sense that their daily activities are worthwhile, lower happiness and higher anxiety levels than non-commuters
Average happiness levels begin to fall after the first 15 minutes of the commute to work, then the negative effects peak at 61-90mins and remain stably detrimental after. The except is when you get to 3h+ each way, the negative effects begin to disappear - suggesting this subset of people might have different expectations and perceptions about their commute.
If you are traveling by car or private vehicle, those traveling less than 15 minutes each way are happier than those traveling 16min+ by the same method. If you are traveling by bus, those spending more than 30min+ commuting were less happy the latter. If you are traveling by train, there’s no adverse effects on well-being up to 30mins compared to those in cars and buses, but after 30 minutes of travel, happiness levels dropped for those traveling by any method. Interestingly, those cycling or walking to work btw 16-30mins had lower happiness than those traveling only 15 mins by any method, but people cycling 30mins or more to work had similar levels of those traveling only up to 15 mins by other methods.
The sad news is that almost everything in this article is based on self-report by asking people to rate 1-10 how satisfied they feel with their work - the whole job or certain aspects of it. Unfortunately there’s many issues with doing this that make the validity questionable, but fortunately a lot of independent measures corroborate the ratings that people give - eg behavioural effects of high and low JS such as turnover, absenteeism, workplace behaviours correlate highly with JS reports.
In saying that, the gold standard for measurement is the Experience Sampling Method (ESM) which pages/messages people many times throughout the day to get real-time reports on peoples moods rather than post-hoc ratings which are more tainted. The ESM measures now and there not later and elsewhere. For a few direct examples of the discrepancy between JS reports and ESM reports, have a look at small talking, noise, job status, qualifications, and working speed / pressure.
Unfortunately there doesn’t have been too much research done with this method directly on JS (though what has is interesting). From my elementary search, it seems like there’s barely anything done on non-teachers, or measuring which jobs are best, or measuring which characteristics of the job contributes to JS, only that personality/beliefs have a role. So basically what’s been done ignores most job satisfaction research. But I believe the ESM will be the future of how the studies are conducted and is promising to keep an eye on if you’re interested.
Daniel Kahneman has done large scale ESM studies though
54, which includes the moods of people at work:
> Mood at work, for example, are largely unaffected by the factors that influence general job satisfaction, including benefits and status. More important are situational factors such as an opportunity to socialize with coworkers, exposure to loud noise, time pressure (a significant source of negative affect), and the immediate presence of a boss (in our first study, the only thing that was worse than being alone). Attention is key. Our emotional state is largely determined by what we attend to, and we are normally focused on our current activity and immediate environment. There are exceptions, where the quality of subjective experience is dominated by recurrent thoughts rather than by the events of the moment.
This is quite interesting because it indicates that whats important to people moment-to-moment at work contradicts what some of our findings here are about JS - eg noise exposure, supervisor presence, and time-pressure all negatively affect JS, and coworker socializing positively influencing it, seemingly.
Though further work is to be done.
So there we have it, 80-90 total variables of potential interest to job satisfaction. That’s a whole lot of information to take into account when selecting for how you might spend 80,000 hours of your life, and a whole lot of complexity, contradictions, ambiguities, personal differences, and weights to take into account. How to even begin thinking about this? Hopefully this research summary was the end of your search, not just the start.
The best paper I came across in this whole process was the 2007 paper by Humphrey and Nahgrang
8 that meta-analysed 260 different studies, including 220k participants, and looked at 14 work characteristics (sections 1-3 here). I’ve listed them below and due to the large sample size and inter-comparisons, I would trust the validity of these more than most of the others. They found that the 14 characteristics explained beyond 55% of the variance in job satisfaction which is basically the most we can get at this point. What that means is that if you took 2 random people with different lives, personalities, and jobs, you could basically predict about 55% of what their job satisfaction would be by looking at these 14 variables.
Those 14 Work Characteristics (2.0 below) include the Motivational Characteristics (which explained 17% of JS variance), Social Characteristics (which explained 17%), and Work Context (mere 4%).
Here is a complete summary of all of the variables and correlations mentioned throughout this article, the most important part of this whole article:
See the sidebar for links
Overall Job Satisfaction Measure .85
Job Characteristics Model .41 (GNS: High .57; Mid .44; Low .32)
Job Complexity .52
Information Processing .42
Problem Solving ?
Physical Demands -.08
Work Conditions .39
Dispositional / Worker Characteristics
The Big Five
Core Self-Evaluations .37
RIASEC / Hollands .10-20
1. Income .15
2. Low Job Status -.23
3. Job Performance ?
Moderate Stress > High/Low
5-19 employees > 20-199 > 200+
Females JS > Males
9. Fast Working Speed -.19
10. Constant Attention Required -.04
11. Employment Type ?
1. Part-Time > Full-Time > Self-Employed > Unemployed/Gov
12. Creative Occupations ?
1. Creative > Non-Creative
13. Sleep Quality ?
14. Age .0
15. Commute Time ?
So how can we take this dump of data and try to find actionable insights in it? Since as we discussed each variable has the possibility of facilitating job satisfaction or job dissatisfaction, our best bet statistically is to weigh variables which correlate highly heaver, and weigh the variables less which correlate less.
Let’s get the uncontrollable variables out of the way firstly. When trying to increase JS, your age is not at all relevant, so do not use that a criteria. If you are a female, you are lucky in that you are statistically slightly more likely than your male counterpart to be satisfied with your job. How much attention your job requires is also negligible, and that’s basically it. The rest is within your circle of control.
Although you would think your genetics are completely chance and outside of your control, which they are, they are expressing their affects on your JS predominantly through your personality, which I’ll discuss shortly.
As for unknowns - who knows whether the ergonomics, temperature, CO2 / air quality level, noise level, job specialization, or problem solving affects JS. The best thing we can do here is to experiment with it ourselves and make subjective ratings during those days, or to hold on for better studies in the future. It is unknown to what extent sleep affects JS but it almost certainly does in some way. The same for stress level, but good stress is different from bad stress. Also, your job satisfaction may be mediated by conforming to the evaluations around you (social information processing), but it’s hard to know by how much.
Also the correlation between small talking on the job, physical demands of the job, and interacting with people outside of the organization are so small that they are not worth paying attention to. Creative jobs are also slightly more satisfying than non, but not so much that you should weigh it highly.
Now for the cream of the crop: what actual job should you select for?
The most important take away that I haven’t highlighted yet is the strength of correlation that Social Support received - a whopping .59, even more than JCM. Recall that social support is a measurement of how assisting your supervisors and coworkers are. How this affects us is that making sure that you like the people that you work with will have the largest effect on whether you will likely find your job satisfying or not, so make sure to get this one right. Obviously you won’t know the people in your organization until you work there, but you should keep this knowledge in your mind to weigh each job that you try. This criteria is starting to be acknowledged by many orgs and they are referring to it as “culture fit”.
Okay so you know you should work with people you like, but in which type of job? As mentioned earlier, there are job listings out there that you could systematically go through, but it wouldn’t help for 3 reasons: they don’t list the criteria you care about for each job (like the JCM), you don’t know how much personal fit you have for each job, and it’s overwhelming. They usually only list pay scales and growth of the industry which is not very meaningful.
As we saw, the JCM are highly important for JS and will be the main metric by which you weigh different jobs. I’ll remind you that you should take the GNS to figure out how much you should weigh it for assessing different jobs. It only takes about 2 minutes.
I’ve tried my best to find research which lists different types of jobs according to their job characteristics (in the JCM) but to no avail. Ideally, job listings like the ones above would be ranked according to criteria of job satisfaction.
As far as I know, there is no comprehensive research done on job satisfaction and job characteristics by job type unfortunately.
I did find one paper
55 which measured global JS across 50k participants and found the following:
Which is interesting and indicative, but I wouldn’t weigh it too highly.
The other best and crudest thing that I could find was this which measures intention to change fields by industry, which you would assume somewhat correlates with JCM. You could potentially use these rates to inform your decision of how dissatisfied people are in different industries. At a high level, food and accommodation services, arts and recreation services, and admin and support services all have the highest turnover intentions. Whereas mining, professional, scientific and technical services, construction, public admin and safety, and IT have the lowest.
I think a better approach to filtering jobs is to start with your passions, interests, and strengths, even if not perfect, and then filter by JCM and job complexity. I would weigh job complexity and JCM about equally as highly. So the more complex the job, the better.
Don’t take the MBTI / Myers Briggs, unless you’re doing so for entertainment, as it’s basically useless for our purposes. I would instead take the RIASEC / Hollands and figure out what your type is, and then peruse the job listings, making a shortlist. This is not a be-all-end-all though, you should keep your options open for jobs not on here, as it only correlates .10-20 remember? But it’s a useful filter.
Then once you have a list, I would email, talk to, meet, interview, etc, as many people that you can from those jobs and try to assess to the best of your ability how they rank on the JCM - potentially even using the direct questions from the surveys, and also on job complexity and information processing. As for weighing the JCM - as you saw, weigh decision-making autonomy the highest, equal with task variety, and then skill variety/feedback from the job second. It might be worth doing the pros-social values test and its definitely worth doing the big 5 test, and use both the pros-social values and conscientiousness to weigh task significance of the job. If you are low in conscientiousness or pro-social values, then you shouldn’t weigh task significance highly at all, as having a social important won’t be as important to you as engaging work will be in and of itself. Task identity can be weighed pretty low.
Once you find jobs that seem like a good personal fit, and have the JCM, it might be worth trying to do an unpaid internship, volunteering, or participation in workshops just to try to get as close as you can to the day to day tasks of the job and assess how you feel doing it.
Hopefully the above gets you 90% of the way there. If not, the .20-.30 correlations may be worth cross checking, and make sure that:
Besides all of that, the only other thing you should weigh in your job is the size of the organization, which should ideally be <20 people.
Once you’ve decided on career, or are in one, there is a handful of things that are still within your control to maximize your job satisfaction.
If you are already in a job and it is low in the JCM, there’s a handful of things that you can do to increase them, and the mental challenge that results from doing so (also useful for managers):
First, change jobs or negotiate salary if your income is less than 700aud per week, non-negotiable, as this can invalidate everything else. Once it is over that, it is sufficient to optimize JS.
The second most non-job characteristic improvement that you can make after your income is above 700$/wk is to live as close to work as possible. The negative effect on this on job dissatisfaction is just as dramatic as the negative effects of low income. Ideally you are within 15 minutes of walking/cycling distance. Weigh distance over transport mode.
You might be slightly happier working part-time than full-time, but only if it doesn’t negatively affect your income too much. If you can find a work situation in a job that you fits all the other criteria, is close to work, and still pays well whilst allowing you to work part-time, you should definitely experiment with it as a means of improving your JS.
Now that your job and environment is taken care of, the last thing within your control is you and your personality.
It is worth taking the VIA on the off chance that Hope, Zest, or Curiosity are in your top 3 strengths. If they are, you might be slightly more satisfied if you find ways to actively use them each day - like this. If they aren’t, you shouldn’t weigh VIA at all.
Okay so we established that your personal life has spill-over effects on your work-life, but its unsure how much. What’s not uncertain and the most important of your personality traits is how neurotic you are and how much you experience negative emotions - ie trait negative affectivity. I highly recommend doing the BIG5 / Five Factor personality test here - as this is very useful information to have for self-knowledge and self-understanding, and quite actionable.
If it turns out that if you are a neurotic person and quite susceptible to negative emotions, there is a lot of different things that you can try, but you should know that many people worse off than yourself have found things that worked well for them, so don’t abandon this too soon. It’s worth starting here, and trying these evidenced-based treatments for depressive moods, or these for anxious.
If it turns out that you are not very neurotic, the next few things that you should give equal importance to from the big5 are increasing your extroversion, agreeableness, and conscientiousness, by trying these. Don’t focus on openness at all (at least not for JS).
Even more important than any of the BIG5 traits (including neuroticism) is your self-esteem. Your self-esteem is the biggest personality variable which can mediate JS, and you can attempt to improve it by trying these. Less important but still so are your self-efficacy and locus of control. A basic primer on improving locus of control is this, but for a more thorough treatment, CBT is basically a match for this - have a look at the book Learned Optimism by Martin Seligman. As for self-efficacy, it might be worth trying some of these or researching farther.
If everything all sounds overwhelming, 80000 hours again has some good advice which is: > Especially if you’re at the start of your career, try not to worry too much about what you’re going to do long-term. Instead, focus on making your next decision. | The best way to take account of the long-term is to focus on taking a step that will improve your career capital and help you learn more.
and William MacAskill (one of the EA founders), in his TED talk, recommends learning things to build skills if you’re unsure about what to do long-term.
In conclusion, the final take away from all of this to keep it simple is to do the following, in order of priority:
So basically to find a satisfying career: work a mentally challenging job, with people you like and receive support from, that ideally suits your personality strengths, that pays okay, close to home, without any obvious negative factors, in a small organization, preferably part-time, and make sure that you express positive personality traits whilst doing so. If you do this, you will have scientifically and commonsensically maximized your odds at a life of happy work.
1. Tait, M., Padgett, M. and Baldwin, T. (1989) Job and Life Satisfaction: A Reevaluation of the Strength of the Relationship and Gender Effects as a Function of the Date of the Study. Journal of Applied Psychology, 74, 502-507 (Sample size 20k)
2. Argyle (2001). The Psychology of Happiness (2nd ed.). New York: Routledge.
3. Judge & Klinger 2008. Job satisfaction: Subjective well-being at work. In Eid & Larsen (Eds.), The science of subjective well-being (pp. 393-413). New York: Guilford.
7. Humphrey, Nahrgang 2007. Integrating Motivational, Social, and Contextual Work Design Features: A Meta-Analytic Summary and Theoretical Extension of the Work Design Literature
9. http://www.timothy-judge.com/c06.pdf Judge, T. A., Klinger, R. (2009). “Promote Job Satisfaction through Mental Challenge”. Handbook of Principles of Organizational Behaviour, Second Ed, pp107-119. http://www.timothy-judge.com/c06.pdf
10. Hackman, Oldham (1975) Development of the Job Diagnostic Survey. An Instrument for the Diagnosis of Jobs and the Evaluation of Job Redesign Projects.#
12. Breaugh 1985, Morgeson and Humphrey 2006
13. Jackson, P. R., Wall, T. D., Martin, R. & Davids, K. (1993). New measures of job control, cognitive demand, and production responsibility. Journal of Applied Psychology, 78, 753-762.
14. The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics.
15. Sims, H. P., Szilagyi, A. D., & Keller, R. T. (1976). The Measurement of Job Characteristics. Academy of Management Journal, 19, 195-212.
16. Grant, A. M. (2008). The significance of task significance: Job performance effects, relational mechanisms, and boundary conditions. Journal of Applied Psychology, 93(1), pp.108-124.#
17. Task Significance and Performance: Meaningfulness as a Mediator https://journals.sagepub.com/doi/pdf/10.1177/1069072716680047
18. Schwartz, S., Melech, G., Lehmann, A., Burgess, S., Harris, M. and Owens, V. (2001). Extending the Cross-Cultural Validity of the Theory of Basic Human Values with a Different Method of Measurement. Journal of Cross-Cultural Psychology, 32(5), pp.519-542.
19. Jex, S.M. (2002) Organizational Psychology: A Scientist-Practitioner Approach. John Wiley & Sons, New York
20. Daniel Kahneman. Developments in the Measurement of Subjective Well-Being. Table 2
21. Staw, Bell, and Clausen, (1986) The dispositional approach to job attitudes: A lifetime longitudinal test.
22. Arvey, Carter, and Buerkley, (1991) Job satisfaction: Dispositional and situational influences.
24. Arvey, R. D., Bouchard, T. J. Jr, Segal, N. L. & Abraham, L. M. (1989). Job satisfaction: Environmental and genetic components. Journal of Applied P.yychology, 74, 187-I 92.
25. Are genetic and environmental influences on job satisfaction stable over time? A three-wave longitudinal twin study.
29. Judge (2000) Personality and job satisfaction, the mediating role of job characteristics*
30. Judge, Bono, Erez, and Locke, (2005). Core Self-Evaluations and Job and Life Satisfaction: The Role of Self-Concordance and Goal Attainment.
31. Judge, T. A., Locke, E. A., & Durham, C. C. (1997). The dispositional causes of job satisfaction: A core evaluations approach. Research in Organizational Behavior, 19, 151–188.
35. Arnold R. Spokane, Erik J. Luchetta, Matthew H. Richwine (1995). Holland’s Theory of Personalities in Work Environments
37. Holland, J.L. (1997) Making Vocational Choices: A Theory of Vocational Personalities and Work Environments. 3rd Edition, Psychological Assessment Resources, Odessa.
38. Heintz, S., & Ruch, W. (2019). Character strengths and job satisfaction: Differential relationships across occupational groups and adulthood. Applied Research in Quality of Life.
39. My Better Self: Using Strengths at Work and Work Productivity, Organizational Citizenship Behavior, and Satisfaction
46. https://80000hours.org/2016/02/should-you-look-for-a-low-stress-job/ #
53. See ‘experienced wellbeing’ chapter of Thinking Fast and Slow by Daniel Kahneman