Motivation and Coaching at Work — Research and Evidence

This page sets out what the research actually establishes about motivation and coaching at work — the mechanism and the theory, not the practice. It is the “science” face of the coaching material: how motivation works, and what is and is not known about the management behaviours that shape it.

This literature is unusually disciplined about not overclaiming, and that discipline is the point of this page. Self-determination theory is the spine, and it is rated verified. Each effect that hangs off it carries its own confidence label — verified, industry-consensus, or directional — because the strength of the evidence varies sharply from one claim to the next. Treating all of these findings as equally settled is the error this page is built to prevent. Where the evidence supports only the direction of an effect and not its magnitude, that distinction is preserved below rather than flattened.

What is self-determination theory and why does it explain motivation at work?

Self-determination theory (SDT), developed by Edward Deci and Richard Ryan, proposes that all humans have three innate basic psychological needs — autonomy, competence, and relatedness — and that when a social context such as a workplace and a manager satisfies these needs, people show more self-motivation, better performance, and greater well-being; when it frustrates them, motivation and mental health degrade.

The three needs are specific. Autonomy is experiencing one’s actions as self-endorsed rather than controlled. Competence is feeling effective and able to master challenges. Relatedness is feeling connected to and cared about by others.

SDT’s second core idea is that motivation sits on a continuum of internalization, running from amotivation, through external regulation (acting for reward or punishment), introjected regulation (internal pressure, guilt), identified regulation (valuing the goal), and integrated regulation, to intrinsic motivation (inherent interest). The practical insight from Gagné and Deci (2005), who brought SDT into organizational behaviour, is that autonomous motivation (intrinsic, identified, and integrated) predicts better outcomes than controlled motivation (external and introjected). What matters is the quality of motivation, not just the amount.

The evidence base is exceptionally large: thousands of studies across education, health, sport, and work, with the 2000 paper among the most-cited in psychology. What is verified here is the framework and its constructs — autonomy, competence, relatedness, and the motivation continuum. The size of any specific workplace effect is a separate question, handled in the sections that follow. Relatedness, understood as belonging and psychological safety, is developed in the belonging and psychological safety material rather than here.

For a 20–200-person Ontario employer, the takeaway is that the levers — autonomy, competence, relatedness — are interpersonal and low-cost. They live in how managers talk to people, rather than requiring large compensation or HR-systems budgets.

Confidence: verified. Source: Ryan & Deci, Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being, American Psychologist 55(1), pp. 68–78 (2000). Foundational and applied sources: Deci & Ryan, Intrinsic Motivation and Self-Determination in Human Behavior (1985); Gagné & Deci, Self-determination theory and work motivation, Journal of Organizational Behavior 26(4) (2005).

Does autonomy-supportive management beat controlling management?

Leaders who support autonomy — taking the employee’s perspective, offering choice, giving rationale, and minimizing coercive control — reliably have employees with more autonomous motivation, better need satisfaction, higher well-being, and modestly higher performance. The core mechanism is well supported, but most of the workplace evidence is correlational, so the direction can be relied on more than the magnitude.

The strongest synthesis is Slemp, Kern, Patrick and Ryan’s 2018 meta-analysis of leader autonomy support (LAS), pooling 754 correlations from 72 studies (N = 32,870). Leaders who acknowledge employee perspectives, encourage self-initiation, offer choice and input, communicate informationally rather than with pressure, and avoid coercive rewards and sanctions had employees with strongly higher autonomous motivation, better need satisfaction, higher well-being, and lower distress. Importantly, the correlation of LAS with performance was moderated by measurement type: stronger for self-rated performance (ρ = .35) than for other-rated performance (ρ = .15) — a clear signal that some of the headline relationship reflects shared-method (self-report) variance.

The field roots run to Deci, Connell and Ryan’s 1989 Xerox study — the first SDT work intervention — where managers’ autonomy-supportive orientation correlated with subordinates’ trust and positive attitudes, and a training intervention shifted managers’ orientations. Baard, Deci and Ryan (2004) found, in an investment bank, that perceived manager autonomy support predicted need satisfaction, which in turn predicted higher performance evaluations and better psychological adjustment.

The honest caveats matter. The core mechanism is well supported (autonomy support → need satisfaction → autonomous motivation), but workplace effect sizes on hard performance are modest and mostly correlational, so causal direction is not nailed down by most studies. Effects on well-being and motivation are more robust than on objective output.

For an Ontario context, Slemp et al. found correlations were not moderated by country of sample, which supports transfer to Canada, and the behaviours are free and within a line manager’s control — a good fit for a resource-constrained 20–200-person firm. Generalize the direction with more confidence than the magnitude. This is the empirical backbone of “not telling, fixing, or saving”: autonomy support is the operational opposite of controlling management, and is developed on the coaching-in-practice page.

Confidence: industry-consensus; largely correlational. Source: Slemp, Kern, Patrick & Ryan, Leader autonomy support in the workplace: A meta-analytic review, Motivation and Emotion 42(5), pp. 706–724 (2018). Field-study sources: Deci, Connell & Ryan, Self-determination in a work organization (Xerox field study), Journal of Applied Psychology 74(4) (1989); Baard, Deci & Ryan, Intrinsic need satisfaction, Journal of Applied Social Psychology 34(10) (2004).

Do extrinsic rewards undermine intrinsic motivation?

Tangible, expected rewards given for an already-interesting task can modestly undermine intrinsic motivation, while verbal and positive feedback tends to enhance it — but whether this undermining effect is large and practically important has been genuinely contested for thirty years, so it should not be presented as settled science.

This is the one genuinely contested topic in the cluster, and it must be reported as such. For the undermining effect: Deci, Koestner and Ryan’s 1999 meta-analysis (128 experiments) found tangible rewards significantly undermined free-choice intrinsic motivation — engagement-contingent (d = −0.40), completion-contingent (d = −0.36), and performance-contingent (d = −0.28) rewards all reduced it. Crucially, positive verbal feedback (praise) enhanced intrinsic motivation (d ≈ +0.33), and unexpected or non-contingent rewards did not undermine. The mechanism (cognitive evaluation theory) is that controlling rewards shift the perceived locus of causality to external and reduce felt autonomy.

Against: Cameron and Pierce’s 1994 meta-analysis (96 experiments) concluded that reward does not overall decrease intrinsic motivation, that praise increases it, and that the only negative effect is a “minimal” one limited to expected tangible rewards for merely doing a task. Eisenberger and Cameron (1996) argued the detrimental effect is “largely a myth.” A sharp published exchange followed.

The nuanced consensus that survives the fight: (1) controlling, tangible, expected rewards for tasks that are already interesting can undermine intrinsic motivation; (2) verbal and informational feedback generally enhances it; (3) for dull tasks the undermining is weak or absent, and rewards can help; and (4) absolute effect sizes are small-to-moderate. Cerasoli, Nicklin and Ford’s 2014 meta-analysis (N = 212,468) reframes the bottom line: intrinsic motivation and incentives jointly predict performance — intrinsic motivation predicts quality more, incentives predict quantity more — and incentives only crowd out intrinsic motivation when directly tied to performance.

For an Ontario SMB, much of the evidence is lab- and education-based, which is a real external-validity gap. The defensible reading: pay people fairly (do not rely on under-paying), avoid micromanaging engaging work with petty contingent bonuses, and lean on informational feedback and recognition. Do not present “rewards always backfire” as settled science — it is not.

Confidence: industry-consensus; the magnitude has been contested for thirty years. Source: Deci, Koestner & Ryan, A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation, Psychological Bulletin 125(6), pp. 627–668 (1999). Opposing and reframing sources: Cameron & Pierce, Reinforcement, reward, and intrinsic motivation, Review of Educational Research 64(3) (1994); Eisenberger & Cameron, Detrimental effects of reward: Reality or myth?, American Psychologist 51(11) (1996); Cerasoli, Nicklin & Ford, Intrinsic motivation and extrinsic incentives jointly predict performance: A 40-year meta-analysis, Psychological Bulletin 140(4) (2014).

Does a manager-as-coach style improve performance?

Managerial coaching — the everyday behaviour of a line manager who develops reports by asking questions, giving developmental feedback, and facilitating problem-solving rather than directing — is positively associated with employee satisfaction, commitment, and performance, but the evidence is largely correlational, cross-sectional, and self-report, so it is best treated as a well-supported working hypothesis rather than a proven causal lever.

“Manager-as-coach” is distinct from professional coaching: it is the everyday behaviour of a line manager who develops reports by asking questions, giving developmental feedback, and facilitating problem-solving rather than directing. Ellinger, Ellinger and Keller (2003) studied this dyadically in a US warehouse setting (438 employees, 67 supervisors) and found supervisory coaching behaviour positively associated with employees’ job satisfaction and supervisor-rated performance — but the study was cross-sectional with mixed self- and other-report, so it shows association, not causation.

Heslin, VandeWalle and Latham (2006) addressed why managers differ in coaching. Across three studies, managers holding an incremental implicit person theory (people’s attributes are malleable) coached more than those holding an entity theory (attributes are fixed). The experimental study is the strongest piece: a self-persuasion manipulation that induced a more incremental belief increased entity-theorist managers’ willingness to coach a poor performer — causal evidence that the propensity to coach can be shifted by changing managers’ beliefs.

Two honest weaknesses dominate the broader literature: (1) there is, as yet, no meta-analysis pooling correlations specifically for line-manager coaching — the quantitative synthesis is dominated by professional and executive coaching; and (2) at least one longitudinal study failed to find that managerial coaching predicted change in job satisfaction. The direction is robust and theory-consistent, but “manager-as-coach raises performance” should be treated as a well-supported working hypothesis — backed by correlational evidence plus one strong belief-change experiment — not a proven causal lever with a known effect size.

For an Ontario SMB, the most actionable, evidence-backed finding is Heslin’s: managers who believe people can grow are the ones who coach, so selecting for and cultivating that belief is a concrete intervention. Most studies are US or Asian, cross-sectional, and large-firm, so generalize direction, not magnitude. The related question of training managers to coach is handled separately.

Confidence: directional; correlational, cross-sectional, self-report. Source: Heslin, VandeWalle & Latham, Keen to help? Managers’ implicit person theories and their subsequent employee coaching, Personnel Psychology 59(4), pp. 871–902 (2006). Associated sources: Ellinger, Ellinger & Keller, Supervisory coaching behavior, employee satisfaction, and warehouse employee performance, Human Resource Development Quarterly 14(4) (2003); Kim, Assessing the influence of managerial coaching on employee outcomes, Human Resource Development Quarterly 25(1) (2014).

Does workplace coaching actually work — what do the meta-analyses show?

Professional, structured workplace coaching produces small-to-moderate positive effects on performance, well-being, attitudes, goal attainment, and self-regulation, but the evidence rests on small samples, mostly self-report outcomes, and few randomized trials.

Two meta-analyses anchor the evidence for coaching as a structured intervention (delivered by professional internal or external coaches, not everyday manager coaching). Theeboom, Beersma and van Vianen (2014) found significant positive effects across five outcome families, with an overall effect of g = 0.66, ranging from g = 0.43 (coping) to g = 0.74 (goal-directed self-regulation). Notably, the number of sessions did not predict effectiveness. A methodological caution: effects were strongly moderated by design (within-subjects g = 1.15 versus mixed g = 0.39), so weaker designs inflate the apparent benefit.

Jones, Woods and Guillaume (2016) deliberately excluded manager-subordinate and peer coaching to isolate professional coaches (k = 17), finding a positive overall effect (δ = 0.36): skill-based δ = 0.28, affective δ = 0.51, and results δ = 1.24 (from very few studies — treat with caution). Effects were stronger for internal coaches than external, and weaker when 360-degree feedback was bundled in.

The honest caveats, which the authors themselves state: small samples, heavy reliance on self-report outcomes, few RCTs, and publication-bias risk. RCT-only syntheses converge on a moderate effect (~g 0.5–0.6) but flag publication bias, so the true effect is probably somewhat smaller. “Coaching works” is best stated as “coaching has a reliable small-to-moderate benefit, strongest for goal attainment and self-regulation, with the quality of evidence still maturing.”

For a 20–200-person Ontario employer, the transferable signals are: brief coaching can be effective (session count is not the lever); internally delivered coaching performed at least as well as external, which favours building internal capability; and the strongest gains are in goal-directed self-regulation — exactly the manager-as-coach skill set. Coaching-industry ROI figures (for example, ICF) are vendor material, not the evidence base.

Confidence: industry-consensus; small samples, few RCTs. Source: Theeboom, Beersma & van Vianen, Does coaching work? A meta-analysis on the effects of coaching on individual level outcomes in an organizational context, Journal of Positive Psychology 9(1), pp. 1–18 (2014). Second meta-analysis: Jones, Woods & Guillaume, The effectiveness of workplace coaching: A meta-analysis of learning and performance outcomes from coaching, Journal of Occupational and Organizational Psychology 89(2) (2016).

Does a growth mindset reliably improve outcomes?

Growth-mindset interventions have much weaker and less reliable effects than their popular reputation suggests; the average effect on academic achievement is very small and often non-significant after correcting for bias, so it should not be treated as settled science, and there is essentially no rigorous evidence that growth-mindset training improves adult workplace performance.

Carol Dweck’s 2006 book Mindset popularized the idea that believing abilities are malleable (“growth mindset”) versus fixed drives achievement — the conceptual cousin of the “implicit person theory” in the manager-as-coach research. It is included here only to report it honestly alongside its strongest critique.

Sisk and colleagues (2018) ran two meta-analyses. The first (k = 273, N = 365,915) found only a weak correlation between mindset and achievement — about 1% of the variance. The second found growth-mindset interventions had a small overall effect (d ≈ 0.08), with most predicted moderators non-significant. A stricter analysis by Macnamara and Burgoyne (2023; 63 samples) reported an overall intervention effect of d̂ = 0.05 that was non-significant after correcting for publication bias, concluding that apparent effects are “likely attributable to inadequate study design, reporting flaws, and bias.” Some signal remained for specific subgroups (low-SES or at-risk students), and a parallel meta-analysis (Burnette et al. 2023) found a small positive effect (d = 0.14) — so the topic is itself contested.

The honest verdict: the effects are small, fragile, education-sector-specific (almost entirely students, not employees), and contingent on implementation. There is essentially no rigorous evidence that “growth-mindset training” improves adult workplace performance in an SMB.

For an Ontario SMB, do not build a development program on growth-mindset claims as if they were established. The transferable, better-evidenced kernel is narrow: managers who believe employees can develop are more likely to coach them — which lives in the manager-as-coach section above, not in mindset-training-for-everyone. This is the lowest-confidence item in the knowledge base.

Confidence: directional; weak and inflated effects, education-specific. Source: Sisk, Burgoyne, Sun, Butler & Macnamara, To what extent and under which circumstances are growth mind-sets important to academic achievement? Two meta-analyses, Psychological Science 29(4), pp. 549–571 (2018). Popular and critical sources: Dweck, Mindset: The New Psychology of Success (2006); Macnamara & Burgoyne, Do growth mindset interventions impact students’ academic achievement? A systematic review and meta-analysis, Psychological Bulletin (2023).

---

This page is general information drawn from the published research, not professional advice. The findings above are deliberately rated by confidence — verified, industry-consensus, or directional — because their evidence varies, and they should not be treated as equally settled. For the business stakes of developing people, see Employee Development and Business Outcomes; for how to put this into practice, see Coaching, Development, and Performance Conversations in Practice.