The Backbone · a case study of the middle

NCOs are the military's middle managers. Management research rarely mentions them.

I searched 24,443 OpenAlex abstracts of management and organization development research for NCOs and sergeants. Three mention them. None studies them as the layer that carries change.

3 of 24,443 OpenAlex research abstracts mention NCOsA grid of 24,443 small dots, one for each OpenAlex research abstract in the Unfrozen Research Index counting set, 28 management and organization development journals, ordered by publication year from 2007 at the top to 2026 at the bottom. Three dots are ringed in blue: the only abstracts that mention NCOs or sergeants, published in 2009, 2020 and 2026.20072010201520202026123

3of 24,443 abstracts

Every dot is one research article or review, oldest at the top. The three ringed in blue are the only ones whose title or abstract mentions NCOs or sergeants. Hover or tap a ring.

  1. 1A study of 109 NCOs and 421 recruits in the Singapore Armed Forces, rating their relationship with their platoon commander. 2009, Leadership & Organization Development Journal · paper
  2. 2An account of a university program that educated Army leaders, including NCOs. 2020, The Journal of Applied Behavioral Science · paper
  3. 3A theory paper on officer and NCO command teams as co-leaders. 2026, Leadership & Organization Development Journal · paper

Unfrozen Research Index counting set: 24,443 research articles and reviews with OpenAlex abstracts from 28 management and organization development journals, published 2007 to 2026 (2026 to date; index v4d, snapshot of 12 October 2026, corrected 1 November 2026). Title and abstract searched for "non-commissioned", "noncommissioned", "NCO", "NCOs" or "sergeant". Searching all 34,953 articles and reviews in the index finds the same 3. Within a year, dot order is arbitrary. Papers described from their abstracts only. A keyword search misses papers that use other words, and the index doesn't cover military journals such as Armed Forces & Society or Military Psychology, so this measures how often abstracts in these journals mention NCOs, not whether anyone has studied them.

Why NCOs are a case study of the middle

Management research keeps asking who carries change in the middle of an organization. The Army has a written answer. Read its leadership doctrine as a civilian job description and it describes the middle manager's work.

"Noncommissioned officers are the backbone of the Army and are responsible for maintaining Army standards and discipline."
ADP 6-22, Army Leadership and the Profession (31 July 2019), para 1-54.
NCO functions in doctrine and their civilian middle-management equivalents
In doctrine, NCOs…In a company, that's…
maintain standards and disciplineperformance management
train "individuals, crews, and small teams"coaching
make "commander's intent-driven decisions"turning strategy into today's work
advise and develop officersmanaging up

A conceptual mapping from doctrine paragraphs 1-54, 1-55 and 1-57. Doctrine says what NCOs should do; it isn't evidence that the skills transfer. Civilian supervisors also lack the legal authority military leaders have, so methods that depend on rank won't carry over.

Look at rank and level, not just "veteran"

Most studies we found treat veterans as one group, or split them only into enlisted and officer. Where studies do look at rank and level, the middle looks different.

Fewer than half of NCOs said the military prepared them well for civilian life

They sit in the middle of the rank structure. 48% of NCOs said the military prepared them well for the transition, against 67% of commissioned officers and 54% of enlisted personnel. Pew reports officers as more likely than both other groups to say so; it does not report a difference between NCOs and enlisted personnel.

Share of veterans who said the military prepared them well for the transition to civilian life, by rankOfficers 67%, NCOs 48%, enlisted 54%. Pew Research Center survey of 1,284 veterans, 2019.0%25%50%75%100%Officerscommissioned67%NCOsnoncommissioned48%Enlistedpersonnel54%
Pew Research Center, "The American Veteran Experience and the Post-9/11 Generation," September 10, 2019, chapter on readjusting to civilian life. Survey of 1,284 U.S. veterans (pre- and post-9/11) on two probability-based online panels, May 14 to June 3, 2019. Share in each rank group who said the military prepared them well for the transition to civilian life. Rank groups as Pew reports them: commissioned officers, NCOs, enlisted personnel.
Show the numbers
RankSaid the military prepared them well
Commissioned officers67%
NCOs48%
Enlisted personnel54%

In one Swedish study, the level conscripts were trained to lead changed how likely they were to become managers

about 75%more likely to become a civilian manager after platoon-level training: about 5 points on a 6.7% baseline

Squad-level training, the tier closest to a junior NCO, had no detectable effect. So "NCO" shouldn't be treated as one category.

This measures getting the manager job, not doing it well. There was no significant wage gain.

Authors' summary estimateRange across the paper's five main specifications

Effect of military leadership training on later becoming a civilian manager, in percentage pointsPlatoon-level training (leading about 20): estimates from 4.0 to 7.4 percentage points across the five main specifications, summarized by the authors as about 5 points on a 6.7% baseline, about 75%. Squad-level training (leading 7 or 8): estimates from minus 0.4 to plus 0.1 points, never significant, no detectable effect.0+2+4+6+8percentage pointsPlatoonled about 20about +5Squadled 7 or 8no detectable effect
Grönqvist & Lindqvist, "The Making of a Manager: Evidence from Military Officer Training," Journal of Labor Economics 34(4), 2016 (figures from IFN Working Paper 1069, 2015). Swedish men in enlistment cohorts 1974 to 1988 (1978 excluded); about 302,000 in the manager analysis. Natural experiment (fuzzy regression discontinuity on test-score thresholds). Platoon training covered sergeants and second lieutenants leading about 20; squad training covered corporals leading 7 or 8. Effect on the probability of holding a manager occupation, in percentage points. The authors de-emphasize the high end of the range; robustness checks run from about 3 to 9 points. Men only; managers defined broadly by occupation code.
Show the numbers
Training tierEffect on becoming a manager
Platoon (about 20 led)+4.0 to +7.4 points across the five main specifications; about +5 on a 6.7% baseline
Squad (7 or 8 led)−0.4 to +0.1 points; never significant

Enlisted veterans start civilian work less satisfied than officers, and the gap holds

17 pointsgap in work satisfaction between officers and enlisted veterans at the first survey

Both groups slipped over the next six months. In a related study of the same cohort, about three years after service, enlisted veterans, NCOs among them, reported more underemployment than officers; that study counts a gap in leadership experience as underemployment (Davenport et al. 2022, abstract only).

First survey, within about 3 months of leaving serviceAbout 6 months later

Share of employed post-9/11 veterans satisfied with their work, officers and enlisted, at two points in the first year after serviceOfficers: 82.41% at the first survey, 77.63% about six months later. Enlisted: 65.22% and 61.87%. Vogt et al. 2020, weighted national cohort of 9,566.0%25%50%75%100%Officers82%78%Enlisted65%62%
Vogt et al., "U.S. Military Veterans' Health and Well-Being in the First Year After Service," American Journal of Preventive Medicine 58(3), 2020. Prospective national cohort of post-9/11 veterans (The Veterans Metrics Initiative), N = 9,566 at the first survey, weighted. Share of employed veterans satisfied with their work on the Well-Being Inventory (average response of somewhat or very satisfied). Differences between officers and enlisted significant at p < .01. Gap: 82.41% minus 65.22%.
Show the numbers
First surveyAbout 6 months later
Officers82.41%77.63%
Enlisted65.22%61.87%

What the evidence does and doesn't say

Grades use a simple scale. Strong: a randomized trial, natural experiment, meta-analysis or large weighted cohort. Moderate: peer-reviewed work with a design limit. Weak: convenience samples or grey literature. Most sources were read at abstract level, so every grade is provisional.

It does say

  • StrongBeing trained to lead a platoon of about 20 made Swedish conscripts about 75% more likely to become civilian managers later. Training to lead a squad of 7 or 8 made no detectable difference.Grönqvist & Lindqvist 2016. It measures getting the manager job, not doing it well, and found no significant wage gain.
  • StrongIn their first months out, enlisted veterans are markedly less satisfied with civilian work than officers: about 65% against 82%.Vogt et al. 2020, weighted national cohort of 9,566 post-9/11 veterans.
  • StrongThe only randomized trial of veteran-supportive supervisor training we found showed no evidence of an overall effect. It helped only where supervisor and coworker support was already higher.Hammer et al. 2019, 35 organizations. That is why the Check-Up diagnoses before anyone trains.
  • StrongStructured debriefs like the after-action review improve team and individual performance by about 20 to 25% on average.Tannenbaum & Cerasoli 2013, 46 samples. A result about the practice and performance, not retention.
  • ModerateVeterans who felt underemployed were 90% more likely to leave their jobs; those who also felt underpaid were about four times more likely.TVMI national cohort, VETERANetwork 2026. Associational and self-reported.
  • ModerateInside the military, platoon sergeants' leadership ratings, alongside platoon leaders', predicted unit performance, and NCO behavior predicted soldiers' stigma about seeking care over and above officer behavior.Bass et al. 2003; Britt et al. 2012. Military settings only; both read at abstract level.
  • ModerateIn one hiring audit, employers did not favor veterans across the board: Black veterans with transferable, non-combat skills got preferential treatment over Black non-veterans, and combat-arms veterans got none.Kleykamp 2009, résumé audit of entry-level jobs in one metropolitan area; read at abstract level.

It doesn't say

  • How former NCOs perform as civilian managers.We found no study that measures it. In one Estonian study, subordinates rated none of their junior commanders' relational competencies as sufficient (Meerits & Kivipõld 2020). This is the gap, not a finding.
  • That supervisor training on its own improves veteran retention.The one trial we found showed no evidence of a main effect; its specific work outcomes still need checking against the full text.
  • That mentoring, ERGs or onboarding programs improve veteran retention.No outcome evaluation was found. Absence of evidence is not evidence of no effect, but it can't be claimed either.
  • That after-action reviews reduce turnover, or that NCOs run them better than civilian managers.The debrief evidence is about performance and the practice, not retention or the facilitator.
  • That half of all veterans leave their first civilian job within a year.That figure comes from a 2014 convenience survey of about 1,250 veterans with an average age of 48.6, and many left for a better opportunity.
  • That studies of military-experienced CEOs tell you anything about NCOs.Those studies cover older, draft-era executives of public companies, many of them former officers. Applying them to NCOs is a category error.
  • That the Rank-to-Role model below is validated.Each factor has published evidence. How they combine has never been tested; that is what my research plans to test.

The Rank-to-Role model

Working model: each factor has published evidence; the combination is untested

Veteran retention as a conversion problem: how well a company turns the experience someone brought from service into a role that fits it. In this model, four drivers an employer can change are proposed to shape job satisfaction, which is linked to intention to leave and then leaving. Two proposed moderators may change how strongly the drivers work. How the parts combine, moderators included, is an untested hypothesis.

Proposed moderators (untested)

Prior rank groupJunior enlisted, NCO, senior NCO, officer. Enlisted veterans are less satisfied (Strong, descriptive) and report more underemployment (Moderate); in Sweden only platoon-level training led to management (Strong). We found no test of whether rank changes how the drivers work.Moderate
Supervisor readinessSupport from the supervisor and team, and the supervisor's attitude toward veterans. In the only trial we found, training helped only where these were already higher.Moderate

Untested hypothesis: both may change how strongly the drivers below act on satisfaction

Drivers an employer can change

Level fitThe gap between the skills and leadership gained in service and what the job asks. For NCOs, possibly span of control: leading people, then placed with no one to lead. The Check-Up measures it.Moderate
PaySatisfaction with pay and benefits for the role and level.Moderate
Advancement and meaningful workA clear path to the next level, especially first-line manager, and work that doesn't feel tedious.Weak to Moderate
Supervisor supportHow much the veteran's own supervisor and coworkers back them.Moderate

Outcomes, in order

Job satisfactionThe focal outcome.
Intention to leaveIn one survey of 450 serving military personnel, support was linked to intention to leave through job satisfaction (Dupré & Day 2007).
LeavingSplit into voluntary, involuntary and leaving for a better opportunity.

Every box is also measured for non-veteran colleagues in the same company, so a veteran problem can be told apart from a company-wide one. Grades are for each factor's evidence. No study has estimated how much the four drivers explain together, or whether they behave differently for former NCOs.

One thing to try this weekIf you hire former NCOs, pick one veteran on your team and compare how many people they led in service with how many they lead now. If the gap is big, ask them about it.

A mini Veteran Atlas

The 18 most relevant veteran and NCO studies from my evidence review, each summarized in one plain sentence and graded by how strong its evidence is.

4 Strong13 Moderate1 WeakEach card says how much of the source was read.

NCO leadership inside the military

What first-line enlisted leadership does in service, and its limits. 3 studies.

Moderate2020Leadership & Organization Development Journal 41(8)

Meerits & Kivipõld 2020

Among 89 first-level commanders in the Estonian Defence Forces, rated by 1,655 subordinates, only two task and two change competencies, and none of the relational ones, were rated at a sufficient level. The study rated competency levels; it did not measure effectiveness.

Survey with cluster analysis, 89 commanders rated by 1,655 subordinatesAbstract onlyRead the original paper
Moderate2012Psychological Services 9(1)

Britt, Wright & Moore 2012

After a 15-month deployment to Afghanistan, soldiers' ratings of NCO and officer behavior both predicted stigma about seeking care, but when the two were modeled together only NCO behavior predicted it uniquely.

Multilevel survey, three waves after deploymentAbstract onlyRead the original paper
Moderate2003Journal of Applied Psychology 88(2)

Bass, Avolio, Jung & Berson 2003

In 72 US Army light infantry platoons, leadership ratings of both platoon leaders and platoon sergeants, taken under stable conditions, predicted how the unit later performed in combat simulations, partly through cohesion and confidence.

Prospective field study, 72 platoonsAbstract onlyRead the original paper

From military leadership to civilian management

Whether leading in service leads to managing later. In one Swedish study, the level someone was trained to lead at mattered. 3 studies.

Strong2016Journal of Labor Economics 34(4)

Grönqvist & Lindqvist 2016The Making of a Manager: Evidence from Military Officer Training

In Sweden, conscripts trained to lead a platoon of about 20 were about 75% more likely to become civilian managers later, while those trained to lead a squad of 7 or 8 showed no detectable difference; it measures getting the manager job, not doing it well.

Natural experiment (fuzzy regression discontinuity), about 302,000 menFull text (working paper)Read the original paper
Moderate2022Strategic Management Journal 43(4)

Gagliardi & Mariani 2022Trained to lead: Evidence from industrial research

After the US ended the draft in 1973, men affected by the change became less likely to hold team-leadership positions in industrial research. The authors say this is most likely explained by education (fewer men stayed in school to defer the draft), not by military training itself, and the effect faded over time.

Quasi-experiment (end of the US draft)Abstract onlyRead the original paper
Weak2020Forces in Mind Trust report (UK)

Poole, Barnes, Owen & Lyonette 2020Senior Non-Commissioned Officers and employment after leaving the UK Armed Forces

The only NCO-specific employment study we found, a UK report on senior NCOs leaving all three services, cites earlier evidence of low confidence in translating military experience for civilian employers and lower levels of formal education.

Mixed methods, grey literatureSummary onlyRead the summary

Satisfaction, fit and leaving

The drivers behind the Rank-to-Role model, measured in veterans after service. 4 studies.

Strong2020American Journal of Preventive Medicine 58(3)

Vogt et al. 2020U.S. Military Veterans' Health and Well-Being in the First Year After Service

In the months after leaving service, 65% of employed enlisted veterans were satisfied with their work against 82% of officers, and satisfaction slipped for both over the next six months.

Prospective national cohort, N = 9,566, weightedFull textRead the original paper
Moderate2026Penn State Clearinghouse research brief

VETERANetwork 2026 (TVMI cohort)Underemployment, compensation dissatisfaction, job turnover and salary change

About 1 in 3 veterans in the cohort felt underemployed; those who did were 90% more likely to leave their jobs, and those who also felt underpaid were about four times more likely.

Prospective national cohort, post-9/11 veterans who left service in 2016Research briefRead the brief (PDF)
Moderate2022Work 72(4)

Davenport, Morgan et al. 2022The subjective underemployment experience of post-9/11 veterans after transition to civilian work

About three years after leaving service, enlisted veterans reported more pervasive underemployment than officers, with underemployment defined to include a gap in leadership experience, not just skills or education.

Longitudinal cohort (TVMI), about 3 years after serviceAbstract onlyRead the original paper
Moderate2007Human Resource Management 46(2)

Dupré & Day 2007

In 450 military personnel, support from supervisors and the organization, work-life balance and job quality were tied to intentions to leave indirectly, through job satisfaction.

Survey, 450 military personnelAbstract onlyRead the original paper

What employers see and do

Hiring signals, stereotypes, climate, and the one randomized trial of an employer intervention we found. 6 studies.

Strong2019Journal of Applied Psychology 104(1)

Hammer, Wan, Brockwood, Bodner & Mohr 2019Supervisor support training effects on veteran health and work outcomes in the civilian workplace

A one-hour computerized training for supervisors on supporting veteran employees, plus behavior tracking, showed no evidence of direct effects on veterans' health and work outcomes; it helped only where supervisor and coworker support was already higher.

Cluster randomized trial, 35 organizationsAbstract onlyRead the original paper
Strong2013Human Factors 55(1)

Tannenbaum & Cerasoli 2013

Structured debriefs, the practice behind the military's after-action review, improved team and individual performance by about 25% on average; the evidence is about the practice, not who facilitates it, and not about retention.

Meta-analysis, 46 samples (N = 2,136)Abstract onlyRead the original paper
Moderate2020Journal of Occupational and Organizational Psychology 93(2)

Hammer, Brady & Perry 2020Training supervisors to support veterans at work: Effects on supervisor attitudes and employee sleep and stress

In the same trial, the training improved supervisors' attitudes toward veteran employees, and its effects on employees' sleep and stress depended on how positive those attitudes were to begin with.

Same randomized trial, 189 supervisor–employee pairsAbstract onlyRead the original paper
Moderate2020Human Relations 73(10)

Gonzalez, Tillman & Holmes 2020Coming home: Why veterans with disabilities withhold workplace accommodation requests

Among veterans with disabilities, a clash between military identity and civilian work was linked to lower psychological safety and, through it, to withholding accommodation requests; a strongly inclusive climate weakened the first link.

Peer-reviewed studyAbstract onlyRead the original paper
Moderate2019Organizational Behavior and Human Decision Processes 153

Shepherd, Kay & Gray 2019Military veterans are morally typecast as agentic but unfeeling: Implications for veteran employment

Across laypeople, managers and employees, veterans were seen as high in agency but low in feeling, and judged less suited to careers that call for emotion and empathy.

Multi-study experiments and surveys, including managersAbstract onlyRead the original paper
Moderate2009Armed Forces & Society 35(2)

Kleykamp 2009A great place to start? The effect of prior military service on hiring

In a résumé audit of entry-level jobs in one metropolitan area, employers favored Black veterans with transferable skills over Black non-veterans, while veterans from the combat arms got no preferential treatment, whatever their race.

Field audit with matched résumésAbstract onlyRead the original paper

The field's own maps

Two recent reviews of veterans and military-connected people in civilian work. Neither names rank as an organizing variable at abstract level. 2 studies.

Moderate2026Journal of Management

Jasionowicz, Triana, Vogus & Delgado 2026Bridging the Gap: An Integrative Review of Research on Veterans' Transition to the Civilian Workforce

This review covers when veterans' moves into civilian work succeed, how their skills transfer and how work affects their wellbeing, and calls for including veterans in the study of careers.

Integrative reviewAbstract onlyRead the original paper
Moderate2026Organization Management Journal 23(1)

Peat, Perrmann-Graham & Stone 2026Veterans and military-connected individuals in the civilian workforce: An integrative review and research agenda

Across 78 studies of military-connected workers (veterans, reservists, spouses and dependents), this review finds distinctive human capital held back by identity strain, organizational misalignment and inconsistent institutional support, and employers underusing it through poor skill translation and stereotypes.

Integrative review, 78 studies from 36 journalsSummary onlyRead the original paper

Summaries are written in plain language by Unfrozen from the evidence review's reading notes, not copied from publishers; most rest on abstracts, and each card says so. Studies are graded by evidence strength, and coding is provisional until an independent human coder checks it. Spot an error? Email info@unfreeze.us and it gets fixed. For the 173 studies on middle managers, 105 of them about change (as of 30 October 2026), see the Atlas of the Middle.

For employers: the Veteran Retention Check-Up

A diagnostic for companies that hire veterans. It measures the model's drivers in your company, compares veterans with the non-veterans who work alongside them, and recommends interventions only where the data, and the climate, can support them.

What it measures

  • Level fit: perceived underemployment, and the gap between the span of control someone had in service and their role now.
  • Pay: satisfaction with pay and benefits, and pay position within band.
  • Advancement and meaningful work: career-path clarity and actual time to first-line manager.
  • Supervisor readiness: support from supervisors and coworkers, and supervisors' attitudes toward veterans. This is the gate for any supervisor-facing work.
  • Job satisfaction: the focal outcome.
  • Embeddedness and intention to leave.
  • Your HR records, in aggregate: retention at one and two years, voluntary and involuntary exits, exits for a better opportunity, and promotion to first-line manager.

Identity strain, stereotypes and mission commitment can be added as exploratory items, labeled as hypotheses in every report.

How it works

Everyone in scope takes the same survey, so non-veterans provide the benchmark. Veteran status and rank group are self-reported and voluntary, never pulled from HR files. No result is shown for any group under 10 people. You get an executive readout, a driver map with each factor's evidence grade next to it, a supervisor readiness index by unit with a go or not-yet call, a level-fit snapshot, a retention and promotion baseline, a 90-day action plan, and a plan to re-measure at 6 and 12 months.

  1. Week 1Kickoff and survey build
  2. Weeks 2–3Survey in the field; HR extract
  3. Week 4Analysis
  4. Week 5Executive readout
  5. Week 6Action plan
Why diagnose firstThe only randomized trial of veteran-supportive supervisor training we found showed no evidence of an overall effect; it helped only where support was already higher. So training is recommended only where the readiness index says it can land. Elsewhere, placement and climate come first.
Not booking yetThe Check-Up isn't open for booking. Join the interest list and I'll email you when it is.
Join the Check-Up interest listFor HR, talent and veteran ERG leads.