Post
JA EN

Good Work Isn't Chosen—It's Crafted: The Evidence Behind the Seven Factors of Job Satisfaction

Posted
By
28 min read
Good Work Isn't Chosen—It's Crafted: The Evidence Behind the Seven Factors of Job Satisfaction
  • Intended readers: Software engineers torn between the standard career advice—”follow your passion,” “maximize salary,” “find a role that fits your personality”—and the reality that none of it quite delivers. Both early-career developers taking their first steps and managers supporting the career growth of their reports.
  • Prerequisites: None. Some intuition for correlation coefficients (explained in the text) will help.
  • Reading time: about 21 min

Overview

You landed a job doing something you loved. Six months in, you catch yourself wondering: “Is this really what I wanted?” It happens all the time. And yet some people land in jobs they weren’t particularly excited about and thrive.

What explains the difference?

“Follow your passion.” “Optimize for salary.” “Find a role that matches your personality type.” These are the three pieces of career advice almost everyone hears at the starting line. Research in psychology consistently shows that none of them reliably predicts a fulfilling career—though calling them flat-out wrong would also be too strong. Passion can be cultivated. Salary does matter, up to a point. The problem is that all three are entry-point decisions: they tell you almost nothing about how satisfied you’ll be once you’re inside a job.

So what does predict satisfaction? Decades of organizational psychology—drawing on meta-analyses involving hundreds of thousands of workers—have converged on roughly seven factors: Freedom, Achievement, Focus, Clarity, Variety, Companions, and Contribution. In effect-size terms, autonomy correlates with job satisfaction at ρ≈.48–.56; role ambiguity drags satisfaction down by r≈−.27. These aren’t properties of which job you chose. They describe the state of the work you’re doing right now.

That’s the core argument of this article: the seven factors aren’t fixed at the moment you accept an offer—they can be reshaped while you work. Researchers call this reshaping job crafting: the active editing of your tasks, relationships, and how you understand the meaning of your work. Longitudinal studies put its effect on engagement at d≈0.37—not dramatic, but real.

AI is now changing the conditions for that reshaping. As AI takes over the routine parts of implementation, engineers are recovering 40–60 minutes a day. But that margin won’t automatically convert into better work. Left unmanaged, it gets absorbed by more tasks; role boundaries blur and the freed time turns into anxiety rather than opportunity. Whether you can deliberately redirect that margin toward the seven factors is what will separate satisfying work from unsatisfying work in the AI era.

This article works through, in order: the limits of the three entry-point myths, the evidence behind the seven factors, the mechanics of job crafting, and practical moves for the AI era—including things you can do starting tomorrow. The conclusion stated upfront: good work isn’t something you choose your way into. It’s something you build from the inside.

Part 1: The Entry-Point Myths—What “Passion, Salary, and Personality Tests” Don’t Deliver

Career advice tends to focus almost entirely on picking the right door to walk through. Choose what you love. Choose the better paycheck. Choose a role that fits your personality type. These feel intuitive, and as starting heuristics they’re not unreasonable. The problem is that all three are poor predictors of satisfaction once you’re actually in the job. Here’s what the research says about each.

“Follow Your Passion”—Passion Is Grown, Not Found

The implicit assumption behind “follow your passion” is that the passion already exists inside you, fully formed, and your job is to find the career that matches it. O’Keefe and colleagues tested this directly. They showed that people who hold a “find your passion” mindset (fit theory) are significantly more likely to give up when an interest hits difficulty—compared to people who see passion as something you develop1. In the researchers’ framing, advising someone to “find their passion” is like telling them to put all their eggs in one basket and then drop the basket the moment it gets heavy.

There’s also an important distinction between two kinds of passion. Vallerand and colleagues showed that harmonious passion—where an activity is integrated into your sense of self on your own terms—predicts flow and well-being. Obsessive passion—where you feel compelled to do it, almost regardless of cost—connects directly to burnout2. The “follow your passion” rallying cry tends to cultivate the second kind.

One overlooked side effect: passion can be exploited. In a series of studies (N > 2,400), Kim and colleagues found that when a worker was described as “passionate,” observers judged unpaid overtime and menial task assignments as more legitimate for that person3. “They love what they do” becomes a rationalization for poor treatment.

That said, this isn’t an argument against caring about your work at all. Wrzesniewski and colleagues’ foundational research found that workers who experienced their job as a calling—not just a job or a career—reported the highest satisfaction4. The critical point is that the sense of calling doesn’t inhere in the occupation itself. It’s a meaning the worker constructs. Which means passion isn’t something to find at the entry point; it’s something you grow after you’re in.

“Optimize for Salary”—Money Works, Just Not the Way You Expect

The standard counterargument to salary-maximizing is the Kahneman & Deaton finding from 2010: emotional well-being plateaus around $75,000 per year5. But that finding has since been revised. Killingsworth, using 1.7 million real-time experience samples, showed that well-being continues to rise with income well past $75,0006.

In 2023, the two research teams resolved the conflict through an unusual adversarial collaboration7. The conclusion: for roughly 80% of people (the comparatively happy majority), income and well-being rise log-linearly even above $100,000. For the least happy ~20%, well-being does plateau around $100,000—because certain categories of unhappiness (grief, clinical depression, relationship breakdown) can’t be bought away.

So “money doesn’t matter” is wrong. Money works. But it comes with three caveats. First, the effect is logarithmic: doubling your salary from $50K to $100K buys about the same increment of well-being as doubling from $200K to $400K—each step up requires more money for the same return. Second, relative income matters: part of why raises feel good is that your position relative to your peers shifts, and if everyone gets a raise, the benefit cancels out8. Third, hedonic adaptation: the happiness boost from a higher income fades faster than people expect9.

The error isn’t caring about salary. It’s that in the pursuit of higher pay, people often trade away the factors—autonomy, meaningful work, quality relationships—that actually drive sustained satisfaction. And Killingsworth et al.’s study is observational: high-paying jobs tend to come with greater autonomy, so the money effect and the autonomy effect can’t be cleanly separated7.

“Find the Job That Fits Your Personality”—Useful for Self-Awareness, Not for Career Prediction

The personality-matching approach has historically leaned on tools like the MBTI. The problem: MBTI doesn’t work as a career predictor. Its test-retest reliability is low—39–76% of people receive a different four-letter type when retested five weeks later10. If your type isn’t stable, basing career choices on it is building on sand. A 2025 review synthesizing 25 years of literature concluded that studies validating MBTI’s structural validity and test-retest reliability are “nearly nonexistent,” and the instrument’s developer explicitly cautions against using it for hiring or career selection11.

Even the scientifically more robust Big Five—particularly conscientiousness—predicts job performance at a validity coefficient of only ρ≈.2012. Statistically significant, but accounting for roughly 4% of variance. Personality constrains the shape of work, but it doesn’t come close to determining it. Personality tests have genuine value for self-reflection and structured conversation; they simply don’t have the predictive power to issue strong prescriptions like “this type belongs in this career.”

So What Should You Look for at the Entry Point?

What the three myths share is that they’re all entry-point choices. Passion, salary, and personality type can give you enough direction to take a first step—and there’s nothing wrong with using them that way. But if you’re going to look for better signals, here’s what the evidence supports.

First: whether the job can supply what you actually want. This isn’t personality matching. It’s being explicit about what you need from work—autonomy, growth, stability, social impact—and checking whether a given role can actually deliver it. This needs-supplies fit correlates with job satisfaction at ρ≈.61, stronger than any single factor covered in this article13. Personality tests fail to predict career fit not because personality doesn’t matter, but because they measure type rather than the alignment between what you need and what the job provides.

Second: whether the seven factors can be built there. The seven factors don’t come pre-installed in any job—they’re built from the inside. But the underlying conditions for building them can be observed before you sign. Is there real discretion over how work gets done? Are there opportunities to stretch and grow? Is there someone you could genuinely trust and talk to? Can you see whose life your work improves? These are seeds of Freedom, Achievement, Companions, and Contribution—and fertile soil makes crafting more likely to take root.

Third: prioritize “what will I develop here?” over “do I love this?” Rare and valuable skills create the leverage—autonomy, mastery, options—that come later. There’s no study that tests this claim directly, but the supporting evidence is consistent: competence satisfaction (the sense that you’re actually getting better) correlates with job satisfaction14, and meta-analytic research on career success finds that proactive career management predicts satisfaction while mere tenure is essentially uncorrelated with it15.

One final note: you don’t need a perfect answer at the entry point. Planned Happenstance Theory argues that many satisfying careers grow from unexpected events rather than deliberate plans, and that the behaviors which capture those opportunities—curiosity, persistence, flexibility, optimism, risk tolerance—matter more than the initial choice16. The quantitative support for this theory is limited, but the practical implication is sound: the entry point is a hypothesis. Enter, start building the seven factors, and adjust as you go.

Part 2: The Seven Factors—What Primary Research Actually Shows

When you measure job satisfaction from the inside rather than from the entry point, certain factors show up reliably across studies. Scanning meta-analyses in psychology and organizational research, seven cluster together. Below, each factor with its representative effect sizes.

A quick note on reading the numbers: ρ (rho) and r are correlation coefficients, where .1 is small, .3 is moderate, and .5 is large. Negative values mean inverse relationships—as that factor goes up, satisfaction goes down.

FactorWhat It MeansRepresentative EvidenceHow It Shows Up for Engineers
FreedomAutonomy over how your work gets doneAutonomy → job satisfaction ρ≈.48–.56 (meta-analysis, 220K+ people) 1718Can you choose the technical approach, the debugging strategy, the architecture?
AchievementThe felt sense of progress and growth76% of best-mood days contained a small win 19Do you end most days with a concrete sense of something having moved forward?
FocusGoal type aligned with your motivational orientationPromotion-focused → satisfaction ρ≈.15–.45; prevention-focused → safety tasks ρ≈.51 (context-dependent) 202122Are you in offense mode (new features) or defense mode (quality, reliability)—and does it match your natural orientation?
ClarityClear purpose, role, and success criteriaRole ambiguity → satisfaction r≈−.27; clear, challenging goals raise performance d≈.5–.8 2324Do you know what’s expected and what “good work” looks like?
VarietyRange of tasks and skills usedSkill variety → growth satisfaction ρ≈.61 (stronger than job satisfaction ρ≈.42) 1825Can you move across domains and roles?
CompanionsRelationships with people who have your backRelatedness need satisfaction independently predicts well-being; lacking a best friend at work cuts engagement odds to ~1 in 12 1426Do you have colleagues who review your work seriously, and someone you can actually talk to?
ContributionKnowing your work helps real peopleContact with scholarship recipients boosted fundraisers’ monthly yield ~2.7× (+171%, randomized experiment) 27Can you see specifically whose life your code improves?

The strength of evidence varies across the seven. Freedom, Clarity, and Contribution have the strongest research base—large-scale meta-analyses and randomized field experiments. Achievement, Variety, and Companions show consistent directional effects, but each carries caveats: the Achievement finding comes from diary research without a control condition; Variety studies show that job rotation has modest effects25; Companions relies heavily on Gallup’s proprietary data. Focus is the weakest—two meta-analyses diverge (effect sizes of .15 and .45), with large context dependence2122.

A few factors warrant elaboration. Freedom is in a class of its own among the seven. It’s the centerpiece of Job Characteristics Theory and is treated as a fundamental human need in Self-Determination Theory1718. Achievement is supported by Amabile and Kramer’s Progress Principle: analyzing 12,000 diary entries, they found that 76% of best-mood days contained a small forward movement, and 67% of worst-mood days featured a setback19. Focus deserves caution—Higgins’ regulatory focus theory is intellectually compelling (promotion-oriented people flourish when pursuing gains; prevention-oriented people flourish when maintaining safety and security), but the workplace effect sizes vary significantly across meta-analyses202122. Contribution is illustrated by Grant’s famous experiment: at a university call center, fundraisers who spent just five minutes with a scholarship recipient they were helping raised 2.7× more money the following month (+171%) compared to a control group27. Seeing a face behind the work changes behavior.

The key takeaway: the seven factors describe the current state of your work—not the category of job you selected. Autonomy, meaning, clarity—none of these are automatically delivered by changing job titles. The question shifts from “which job will give me all seven?” to “how do I build more of each in the job I’m in?”

Part 3: The Seven Factors Aren’t Given—They’re Crafted

When workers actively reshape their own job descriptions—rather than passively executing whatever they were handed—organizational psychologists call it job crafting.

flowchart TB
    A["Entry point: passion, salary, personality tests"] --> B["Starting a job: an initial direction, nothing more"]
    B --> C["Reality: the seven factors aren't pre-installed"]
    C --> D["Job Crafting<br>Actively reshaping your work"]
    D --> E["Seven factors come alive:<br>Freedom · Achievement · Focus · Clarity<br>Variety · Companions · Contribution"]
    E --> F["Well-being and engagement at work"]
    F -.->|"invites the next challenge"| D

Two research traditions map the terrain. Wrzesniewski and Dutton identified three types of crafting28: changing the tasks you do (their scope, their nature), changing the relationships you build and how (who you interact with and how much), and changing how you cognitively frame what your work means. The clearest example of the third: a hospital janitor who reframes their role not as “cleaning” but as “maintaining the environment that helps patients recover.”

Tims and Bakker later reframed crafting as adjusting your job demands and resources, organized into four moves29: increasing structural resources (broader autonomy, skills, professional development); increasing social resources (seeking feedback and support); increasing challenging demands (taking on harder work); and decreasing hindering demands (shedding draining, obstructive tasks).

How large are the effects? Cross-sectional studies make crafting look dramatic (d≈1.01 for engagement), but that’s an overestimate. Longitudinal and experimental designs that control for causal direction put the effect at d≈0.3730—moderate, but consistent. And the gains compound. Weekly diary studies find that crafting in one week predicts higher engagement the following week, which in turn predicts more crafting—a self-reinforcing loop31. The first move starts a flywheel.

The connection to the seven factors is direct. Increasing structural resources builds Freedom and Variety. Increasing social resources builds Companions. Increasing challenging demands builds Achievement. And the most powerful single lever is cognitive reframing: in one study, cognitive crafting correlated with work meaning at r≈.63—the highest of any dimension—and the path from meaning to engagement accounted for 92% of the mediated effect32. The sense of Contribution can be increased just by changing how you see the work, without changing the work itself.

Here are concrete moves, mapped to each factor, that an engineer can start tomorrow:

FactorA Move to Make Tomorrow
FreedomDecide on an implementation approach yourself, then state it up front: “I’m going with this design.” Stop waiting for permission; start by staking a position.
AchievementWrite one line at the end of each day: what moved forward. Break large goals into weekly forward steps.
FocusFigure out whether you’re offense-oriented (new features) or defense-oriented (quality, reliability), and actively claim the assignments that match.
ClarityWhen a request arrives ambiguous, start by locking the exit: “So we’re done when X is true—does that sound right?”
VarietyOnce a quarter, reach into an area outside your usual lane. Backend-only? Take a small frontend task, an ops ticket, an on-call rotation.
CompanionsShift from someone who receives reviews to someone who exchanges them. Give one thorough, thoughtful PR review per week.
ContributionPut a face on one user. Read the support ticket logs. Sit in on a user call. Know specifically whose problem your code solves.

These moves collapse into four broader patterns: expanding autonomy and range (Freedom, Variety), building relationships (Companions), taking on harder challenges (Achievement, Focus, Contribution), and reducing draining ambiguity (Clarity).

Three caveats deserve space. First, much of the research relies on self-report, and crafting scales frequently overlap with job-resource scales, inflating observed correlations33. Second, intervention studies (workshops and training programs) show small average effects (d≈0.15, non-significant)—the effect is concentrated among workers who were already doing little crafting (d≈0.33)34. If you’ve been in wait-for-instructions mode, the first move is likely to pay off most. Third, the dark side: when “reducing hindering demands” means quietly offloading work to colleagues without a conversation, it creates team friction and productivity losses35. Reducing your load is legitimate—but do it by automating or streamlining, not by passing the burden sideways. Crafting without transparency just purchases your own well-being at the expense of others.

Part 4: The AI Era—Margin Appears, and Disappears If Unmanaged

AI changes the conditions for job crafting on both sides of the ledger.

On the upside: as AI absorbs routine implementation and boilerplate, engineers are gaining back time. A 2025 Microsoft Research report found AI users saving an average of 40–60 minutes per day36. Telemetry analysis from Faros AI found individual task completion rates up 21–33.7% and epic completion rates up 66.2%37. In principle, that margin can be redirected toward Freedom, Achievement, and Variety.

But there’s a large asterisk. Freed time doesn’t automatically become better work. The same Faros AI analysis found that while individual productivity metrics rose, team-level delivery indicators didn’t improve—and bugs per developer increased 9–54%, with incidents per PR up 242.7%37. Recovered hours tend to be refilled with more tasks. Output volume goes up; the quality of the work experience doesn’t.

A second trap is role boundary erosion. As AI takes over more of the implementation, “what is my value here?” becomes harder to answer. As shown in Part 2, role ambiguity drags satisfaction down by r≈−.27. And this ambiguity splits people’s crafting responses in two directions. Workers who experience AI as an expansion of their autonomy move toward approach crafting: seeking new challenges and resources. Workers who experience AI as a threat retreat into avoidance crafting: shrinking their scope. What separates the two groups is not personality—it’s AI knowledge and organizational support38. Engineers who invest in understanding and using AI tools tend to experience them as expanded capability, not encroachment.

There’s also a skill atrophy risk. Overreliance on AI is associated with measurable skill decline after three months, and in high-AI-exposure fields, employment for workers aged 22–25 dropped roughly 13%36. Job crafting can broaden your surface area, but if the foundational skills underneath are thinning, the breadth can’t hold.

This connects skill shape to well-being. If AI replaces the implementation layer where you’ve been working, a skill set that stops at the surface—a shallow point without depth—loses its market value along with its ability to generate Freedom, Achievement, and Contribution. The prerequisite for sustained well-being in the AI era is depth that reaches the layer AI can’t easily replace: verification, judgment, design. Breadth (T-shaped, π-shaped) hedges against any single axis being commoditized, but breadth-without-depth is the most fragile position—it’s what AI substitutes most easily. The order matters: depth first, then breadth. For more on skill shape and why depth matters in the AI era, see the posts on the I/T/π skill matrix and AI-era verification depth.

What, then, should engineers do with the recovered margin? The clearest signal from the evidence is that increasing challenging demands is the crafting move most strongly associated with outcomes (rc≈.42)33. Use the time AI frees up to take on challenges one level higher: design the quality gates for AI-generated code, take ownership of architecture reviews, own a feature end-to-end from design through operations. This aligns with the “from implementation to orchestration” role shift that is increasingly advocated for the AI era39.

One important distinction: whether this shift is crafting you choose or reorganization imposed from above determines its effect on meaning and engagement. The same movement—from writing code to orchestrating systems—generates well-being when you’re steering it and anxiety when it’s being done to you. The locus of control is what matters. Note also that research on job crafting in AI-specific contexts is still early: most samples to date are from Chinese financial services and service industries; evidence specific to software engineers remains thin4038.

Starting Small

Intervention research on job crafting points to something encouraging: the largest gains go to people who were doing the least crafting to begin with34. If you’ve been primarily in receive-and-execute mode, the first deliberate move is where the return is highest.

You don’t need to rebuild all seven factors at once. Pick one item from the table above and try it this week. Because crafting and engagement mutually reinforce each other31, once one thing starts moving, the next move becomes easier to make.

You may not be able to change where you started. But the work you’re doing right now—you can start reshaping it today.

Summary

“Follow your passion.” “Optimize for salary.” “Find a job that fits your personality type.” These are entry-point heuristics, not blueprints for a fulfilling career. Passion is something you cultivate, not discover. Salary’s effect on well-being is logarithmic and subject to adaptation. Personality tests can’t predict career fit. At best, the three give you a direction for the first step—and looking instead at “can this job supply what I need,” “is there room to build the seven factors,” and “will skills accumulate here” will serve you significantly better.

What ultimately determines satisfaction is not the entry point, but the state of the seven factors once you’re inside: Freedom, Achievement, Focus, Clarity, Variety, Companions, and Contribution. These don’t get assigned when you accept an offer. They get built while you work. Job crafting—actively editing your tasks, relationships, and the meaning you make of your work—shows a moderate but real effect in longitudinal studies (d≈0.37), with the strongest impact coming from deliberately increasing challenging demands.

AI creates the margin to do this reshaping. But that margin, left unmanaged, refills with more tasks and blurs role boundaries until it turns into anxiety rather than opportunity. Whether you can intentionally convert that margin into the seven factors—whether you hold the steering wheel—is what divides satisfying from unsatisfying work in the AI era.

Stop mourning the entry point you didn’t choose. Start building the work you’re already in. Good work isn’t something you choose your way into. It’s something you craft from the inside.

References

Additional References (not cited by number in the text)

  1. Implicit Theories of Interest: Finding Your Passion or Developing It? - O’Keefe, P. A., Dweck, C. S., & Walton, G. M., Psychological Science (2018). 【Reliability: High】 ↩︎

  2. Les passions de l’âme: On obsessive and harmonious passion - Vallerand, R. J., et al., Journal of Personality and Social Psychology (2003). 【Reliability: High】 ↩︎

  3. Understanding contemporary forms of exploitation: Attributions of passion serve to legitimize the poor treatment of workers - Kim, J. Y., Campbell, T. H., Shepherd, S., & Kay, A. C., Journal of Personality and Social Psychology (2020). 【Reliability: High】 ↩︎

  4. Jobs, careers, and callings: People’s relations to their work - Wrzesniewski, A., McCauley, C., Rozin, P., & Schwartz, B., Journal of Research in Personality (1997). 【Reliability: Medium-High】 ↩︎

  5. High income improves evaluation of life but not emotional well-being - Kahneman, D., & Deaton, A., PNAS (2010). 【Reliability: Medium-High (original finding revised by the 2023 adversarial collaboration)】 ↩︎

  6. Experienced well-being rises with income, even above $75,000 per year - Killingsworth, M. A., PNAS (2021). 【Reliability: High】 ↩︎

  7. Income and emotional well-being: A conflict resolved - Killingsworth, M. A., Kahneman, D., & Mellers, B., PNAS (2023). 【Reliability: High】 ↩︎ ↩︎2

  8. Relative income, happiness, and utility: An explanation for the Easterlin paradox and other puzzles - Clark, A. E., Frijters, P., & Shields, M. A., Journal of Economic Literature (2008). 【Reliability: Medium-High】 ↩︎

  9. Pursuing happiness: The architecture of sustainable change - Lyubomirsky, S., Sheldon, K. M., & Schkade, D., Review of General Psychology (2005). 【Reliability: Medium-High】 ↩︎

  10. Cautionary comments regarding the Myers-Briggs Type Indicator - Pittenger, D. J., Consulting Psychology Journal: Practice and Research (2005). 【Reliability: High】 ↩︎

  11. A 25-year review and psychometric synthesis of the Myers–Briggs Type Indicator (MBTI) – Form M - Erford, B. T., et al., Journal of Counseling & Development (2025). 【Reliability: High】 ↩︎

  12. The Big Five personality dimensions and job performance: A meta-analysis - Barrick, M. R., & Mount, M. K., Personnel Psychology (1991). 【Reliability: High】 ↩︎

  13. Consequences of individuals’ fit at work: A meta-analysis of person-job, person-organization, person-group, and person-supervisor fit - Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C., Personnel Psychology (2005). 【Reliability: High】 ↩︎

  14. A review of self-determination theory’s basic psychological needs at work - Van den Broeck, A., Ferris, D. L., Chang, C., & Rosen, C. C., Journal of Management (2016). 【Reliability: High】 ↩︎ ↩︎2

  15. Predictors of objective and subjective career success: A meta-analysis - Ng, T. W. H., Eby, L. T., Sorensen, K. L., & Feldman, D. C., Personnel Psychology (2005). 【Reliability: High】 ↩︎

  16. Planned happenstance: Constructing unexpected career opportunities - Mitchell, K. E., Levin, A. S., & Krumboltz, J. D., Journal of Counseling & Development (1999). 【Reliability: Medium (theoretical proposal; limited quantitative support)】 ↩︎

  17. Leader autonomy support in the workplace: A meta-analytic review - Slemp, G. R., Kern, M. L., Patrick, K. J., & Ryan, R. M., Motivation and Emotion (2018). 【Reliability: High】 ↩︎ ↩︎2

  18. Integrating motivational, social, and contextual work design features: A meta-analytic summary - Humphrey, S. E., Nahrgang, J. D., & Morgeson, F. P., Journal of Applied Psychology (2007). 【Reliability: High】 ↩︎ ↩︎2 ↩︎3

  19. The power of small wins - Amabile, T. M., & Kramer, S. J., Harvard Business Review (2011) (diary research underlying the book The Progress Principle). 【Reliability: Medium-High (naturalistic observation without control condition)】 ↩︎ ↩︎2

  20. Beyond pleasure and pain - Higgins, E. T., American Psychologist (1997) (original statement of regulatory focus theory). 【Reliability: High (theoretical)】 ↩︎ ↩︎2

  21. Regulatory focus and work-related outcomes: A review and meta-analysis - Lanaj, K., Chang, C., & Johnson, R. E., Psychological Bulletin (2012). 【Reliability: Medium-High】 ↩︎ ↩︎2 ↩︎3

  22. A meta-analysis of the regulatory focus nomological network: Work-related antecedents and consequences - Gorman, C. A., et al., Journal of Vocational Behavior (2012). 【Reliability: Medium】 ↩︎ ↩︎2 ↩︎3

  23. Work role ambiguity, job satisfaction, and job performance: Meta-analyses and review - Abramis, D. J., Psychological Reports (1994). 【Reliability: Medium-High】 ↩︎

  24. Building a practically useful theory of goal setting and task motivation: A 35-year odyssey - Locke, E. A., & Latham, G. P., American Psychologist (2002). 【Reliability: High】 ↩︎

  25. More hype than substance? A meta-analysis on job and task rotation - Mlekus, L., & Maier, G. W., Frontiers in Psychology (2021). 【Reliability: Medium】 ↩︎ ↩︎2

  26. Q12 Meta-Analysis: The Relationship Between Engagement at Work and Organizational Outcomes (11th ed.) - Gallup (2020). 【Reliability: Medium-High (proprietary research, large sample)】 ↩︎

  27. Impact and the art of motivation maintenance: The effects of contact with beneficiaries on persistence behavior - Grant, A. M., et al., Organizational Behavior and Human Decision Processes (2007). 【Reliability: High (randomized field experiment)】 ↩︎ ↩︎2

  28. Crafting a job: Revisioning employees as active crafters of their work - Wrzesniewski, A., & Dutton, J. E., Academy of Management Review (2001). 【Reliability: High】 ↩︎

  29. Development and validation of the job crafting scale - Tims, M., Bakker, A. B., & Derks, D., Journal of Vocational Behavior (2012). 【Reliability: High】 ↩︎

  30. Longitudinal meta-analysis of job crafting shows positive association with work engagement - van den Heuvel, M., Demerouti, E., & Bakker, A. B., Cogent Psychology (2020). 【Reliability: High】 ↩︎

  31. Weekly reciprocal relationships between job crafting, work engagement, and performance - Lopper, E., Milius, M., Reis, D., Nitz, S., & Hoppe, A., Frontiers in Organizational Psychology (2023). 【Reliability: High】 ↩︎ ↩︎2

  32. Job crafting and work engagement: The mediating role of work meaning - Letona-Ibañez, O., Martinez-Rodriguez, S., Ortiz-Marques, N., Carrasco, M., & Amillano, A., International Journal of Environmental Research and Public Health (2021). 【Reliability: Medium-High】 ↩︎

  33. Job crafting: A meta-analysis of relationships with individual differences, job characteristics, and work outcomes - Rudolph, C. W., Katz, I. M., Lavigne, K. N., & Zacher, H., Journal of Vocational Behavior (2017). 【Reliability: High】 ↩︎ ↩︎2

  34. Effects of a job crafting intervention program on work engagement among Japanese employees: A randomized controlled trial - Sakuraya, A., et al., Frontiers in Psychology (2020). 【Reliability: High】 ↩︎ ↩︎2

  35. The dark side of job crafting (Academy of Management Proceedings) - Abukhait, R., et al. (2019). 【Reliability: Medium】 ↩︎

  36. New Future of Work Report 2025 - Microsoft Research (2025). 【Reliability: Medium-High】 ↩︎ ↩︎2

  37. Key takeaways from the DORA 2025 report (telemetry analysis) - Faros AI (2025) (telemetry analysis of productivity and quality metrics related to the DORA report). 【Reliability: Medium】 ↩︎ ↩︎2

  38. Organizational AI adoption and approach vs. avoidance crafting - Liu, Tian, Li, & Tan, Frontiers in Psychology (2025, published online January 2026). 【Reliability: Medium-High (Chinese sample, 3-wave longitudinal)】 ↩︎ ↩︎2

  39. When Code Becomes Abundant (ICSE-FoSE 2026) - Kohl, & Carro (2026). 【Reliability: Medium (theoretical/normative paper)】 ↩︎

  40. Digital-AI transformation and job crafting - Sha, & Chai, Frontiers in Psychology (2025). 【Reliability: Medium-High (Chinese sample, longitudinal)】 ↩︎

Career & Learning
Career Job Satisfaction Job Crafting Motivation Well-being at Work Self-Determination Theory Software Engineers AI Era Careers
This post is licensed under CC BY 4.0 by the author.