Government Employee Engagement Survey Questions

Federal agencies participating in the FEVS are not free to design their annual engagement survey from scratch. Under 5 CFR Part 250, Subpart C, agencies conducting an annual employee survey through OPM must include the 16 prescribed Annual Employee Survey questions. The statutory basis is Pub. L. 108-136 §1128, which made annual federal employee surveys a legal requirement. OPM administers the FEVS as the mechanism for satisfying this obligation across most of the civilian executive branch, and agencies must post survey dates and results — or a statement declining to post — no later than 120 days after completing administration.

Those 16 questions cover the core dimensions of federal work experience: whether employees understand agency priorities, whether their supervisors listen and communicate, whether performance differences are recognized, and whether the agency is managed effectively. They are not optional starting points — they are the floor. Agencies may add supplemental items, and many do, but the 16 statutory questions produce the governmentwide benchmarks that make cross-agency comparison and year-over-year FEVS trend analysis possible. Understanding which items feed which EEI subscale, and which score lowest, is more useful than treating the survey as a generic satisfaction instrument.

For HR leads designing supplemental or follow-up pulse surveys, the statutory structure has a practical implication: you are adding depth to a known instrument with established benchmarks, not inventing one from scratch. The smartest starting question is not "what should we ask?" but "which of the 16 statutory items scored lowest in our agency or work unit, and how do we write a pulse question that tracks movement on that specific dimension?"

For state and local governments that do not participate in the FEVS, the lesson transfers in structure if not in specific obligation: adopt a validated, consistent instrument — whether MissionSquare's state/local survey, Gallup Q12, or a university-partnered tool — rather than assembling ad-hoc questions each cycle. Consistency over time is what makes trend data interpretable and comparable. A jurisdiction that changes its survey instrument every two years cannot tell whether score changes reflect real workforce shifts or question changes.

The FEVS Employee Engagement Index (EEI) is built from three subscales that each measure a distinct driver of engagement. In 2024, those subscale scores were: Leaders Lead at 63% positive, Supervisors at 81% positive, and Intrinsic Work Experience at 75% positive (OPM FEVS, 2024). These three figures are what should drive HR dashboard conversations — they are more diagnostic than the composite EEI alone because each subscale points to a different set of owners and a different intervention pathway. Leaders Lead is about senior executive behavior and organizational trust; Supervisors is about direct manager behavior; Intrinsic Work Experience reflects how employees feel about the nature and value of the work itself.

Two individual items anchor the extremes of the 2024 FEVS and are worth quoting verbatim, because the exact wording determines how to align supplemental and pulse questions. The lowest-scoring item on the entire survey was Q.17: "In my work unit, differences in performance are recognized in a meaningful way" — 47% positive (OPM FEVS, 2024). The highest-scoring item was Q.90: "It is important to me that my work contribute to the common good" — 92% positive (OPM FEVS, 2024). Together these two numbers define the public-sector engagement equation: federal employees are nearly universally mission-motivated, but fewer than half feel that differentiated performance is seen and acknowledged.

Federal employees are nearly universally mission-motivated — 92% say it matters that their work contributes to the common good — yet fewer than half feel performance differences are recognized meaningfully. That gap between Q.90 and Q.17 is the highest-leverage intervention point in federal engagement.

A third item that sits between those extremes but carries direct practical weight for recognition program design is Q.69: "How satisfied are you with the recognition you receive for doing a good job?" — 57% positive in 2024, among the lowest-scoring items on the full survey (OPM FEVS, 2024). The gap between 92% mission motivation and 57% recognition satisfaction is the leverage point most agencies can influence without legislation, budget supplementals, or pay-scale changes.

One explicit rule for downstream survey designers: the verbatim item wording for Q.17, Q.69, and Q.90 quoted above comes from OPM's published FEVS reports and may be cited as such. Do not reproduce or paraphrase additional FEVS item text as if it were validated wording — use official OPM documentation as your source, and describe non-quoted items by theme or subscale rather than fabricated item text.

Effective supplemental and pulse questions start with the domains the FEVS identifies as highest-leverage. Two themes stand out from the 2024 data — recognition and mission connection — with supervisor quality as the third because it carries the most within-agency variance and the most manageable intervention pathway (PLAY-029, PLAY-001). Designing questions by theme rather than by asking "what else do we want to know?" keeps the instrument short, targeted, and usable at the work-unit level.

Recognition. The FEVS anchor for this domain is Q.17: "In my work unit, differences in performance are recognized in a meaningful way" (47% positive, OPM FEVS, 2024). Pulse questions in this domain should stay close to that wording and ask whether specific recognition occurred recently — whether a supervisor acknowledged a specific contribution by name, whether a peer's effort was publicly noted, or whether the employee felt their work was visible to leadership in the last month. The MSPB's research on job characteristics confirms that recognition — as a concrete reward dimension supervisors directly control — is something work-unit managers can change without budget authority or agency-wide initiative (MSPB, Federal Employee Engagement). Non-monetary peer recognition under Title 5 §4503 makes this actionable immediately.

Mission connection. The FEVS anchor is Q.90: "It is important to me that my work contribute to the common good" (92% positive, OPM FEVS, 2024). This domain starts at a high floor — the challenge is maintaining visible connection when priorities shift, processes slow, or leadership communication breaks down. Pulse questions in this domain ask whether the employee can currently draw a clear line between their daily tasks and the agency's stated priorities, or whether recent changes made the mission feel more or less tangible. Partnership for Public Service analysis consistently identifies mission commitment as the top driver of federal engagement, with mission connection scoring highest among all engagement driver categories (PLAY-001).

Supervisor quality. The Supervisors subscale at 81% positive (OPM FEVS, 2024) is the highest of the three EEI components, but it is also the dimension with the most within-agency variance — meaning the same agency-level score can mask high-performing and struggling work units just floors apart. Pulse questions in this domain should focus on observable, recent supervisor behaviors: whether the supervisor communicated clear priorities in the last month, whether they held a development conversation, whether they took a visible action on a problem the team raised. Behavioral questions — anchored in specific actions rather than general impressions — are more useful because they diagnose where specific behaviors are missing rather than asking employees to assess personality.

A pulse survey answers one question: did conditions change in the specific area we said we would address? It is not a replacement for the annual survey; it is a between-cycles check-in that confirms whether an action actually moved something. The governing discipline for government HR leads: keep the annual census — FEVS or its state/local equivalent — as the system of record, treat pulse results as directional indicators useful for in-cycle course correction, and never position a pulse as statistically comparable to the validated annual instrument (PLAY-009).

Wording alignment matters more than question count. When pulse items are worded differently from FEVS items, score changes become uninterpretable. If the annual survey asks whether performance is recognized "in a meaningful way" and the pulse asks whether the employee "feels appreciated," you cannot tell whether a change in scores reflects a real shift in the work environment or a wording effect. The practical rule: use language as close as possible to the FEVS item anchors — even when adapting for brevity — so that movement on the pulse points toward a specific FEVS domain rather than an ambiguous sentiment.

The 2025 survey cycle has created a real planning gap for federal HR leads. OPM cancelled the 2025 FEVS in August 2025 — the first cancellation since the survey became annual in 2010 — and OPM indicated the FEVS would return in 2026. In the gap, the Partnership for Public Service fielded its 2025 Public Service Viewpoint Survey (PSVS): 23 questions administered November through December 2025, with 11,083 respondents across 30 reported agencies (PLAY-008). However — and this is a mandatory guardrail for every federal HR lead — the Partnership and OPM both state the PSVS is NOT directly comparable to FEVS: different methodology, different sample size, different administration windows. Never present PSVS scores as year-over-year FEVS movement (PLAY-008).

For the 2025–2026 gap, the right posture is to run a lightweight pulse aligned to FEVS item wording, track directional trends only, and preserve the 2024 FEVS as your performance baseline when the official survey resumes. Do not design an in-house instrument to replace FEVS — design a check-in to complement it and then step aside when the official cycle resumes.

OPM protects FEVS respondent confidentiality through what practitioners call the Rule of Ten: results are not reported for a work unit unless at least ten employees responded, and cells below ten are masked in published data files (OPM FEVS Technical Report, 2019). This rule serves a dual purpose — protecting individual employees from identification and ensuring that reported scores reflect a meaningful sample at the unit level rather than the response of one or two people during a bad week (PLAY-010).

For HR leads designing supplemental or pulse surveys, the Rule of Ten creates a practical floor with real operational consequences. Small field offices, single-shift crews, and specialized technical units may never reach ten respondents in a unit-level analysis, particularly if survey administration windows are short. The options are limited: aggregate units for reporting (losing granularity), raise the threshold above ten for additional conservatism, or provide qualitative summaries only for small units — with an explicit note that quantitative reporting was withheld to protect anonymity. The Partnership for Public Service applied a thirty-respondent floor for agency-level PSVS reporting, demonstrating that even well-resourced survey programs use conservative thresholds when respondent pools are small (PLAY-010).

Commercial pulse tools commonly advertise thresholds as low as three to five respondents, reported as vendor defaults (PLAY-010). Public-sector buyers should approach low thresholds with skepticism. In a small district office or a tight-knit shift crew, a five-respondent minimum can make individual respondents identifiable to supervisors — especially on emotionally charged items about recognition or leadership. The question to ask vendors before signing: what is the minimum threshold, what is the default, and can it be raised by the agency administrator? Then disclose the threshold to employees before the survey opens.

Anonymity is ultimately a trust decision, not just a statistical parameter. Agencies that underinvest in anonymity protection often discover the cost in the next cycle's response rate: employees who suspect their responses could be traced respond with surface-level positivity, which makes the data feel good but provides no diagnostic value. The more you can credibly commit to protection — including small-unit masking — the more honest the data you collect.

The most common government survey-design failure is asking about things no one in the building has authority to change. Items about GS-scale pay, political leadership, agency reorganizations, and mandated return-to-office policies may be legitimate for research but are frustrating in a pulse context because the results cannot drive work-unit-level action. OPM's Simple Approach to Action is the right model to design backward from: pick one focus area — any item where the unfavorable response rate crosses OPM's defined threshold for needing improvement — take two or three concrete actions, name owners, and discuss progress four times a year (OPM, "A Simple Approach to Action," PLAY-004). Questions should map cleanly to that model before they are asked.

Practical wording rules for government survey designers:

• Be specific, not atmospheric. "I received recognition for a specific contribution in the last 30 days" generates more actionable data than "I feel valued at work." The FEVS uses precise wording because specific items diagnose specific gaps.
• Match the FEVS scale format. The FEVS uses an agree/disagree response format; using the same format in supplemental items allows direct comparison with FEVS subscale benchmarks rather than introducing a scale-translation problem.
• Ask about observable behaviors. "My supervisor provided specific feedback on my work in the last month" is more actionable than "My supervisor respects me." Behavioral items locate the gap in a concrete, manageable action rather than a personality impression.
• Limit items to what the work-unit manager can act on. Questions about agency-wide IT infrastructure, facilities, or executive leadership belong in the annual survey where results route to the owners who can act. At the work-unit level, focus on recognition, communication, development conversations, and goal clarity.
• Avoid double-barreled items. "My supervisor listens and acts on my feedback" can be simultaneously true and false, and a single response is uninterpretable. Split every compound claim into two separate items.

The overarching discipline: before adding any item to a pulse survey, answer the question "what action would a low score require from this work unit, and who owns it?" If you cannot answer that question, the item should not be on the survey.

The 2024 FEVS achieved a 41% response rate — its highest since a pandemic-era dip — up from 39% in 2023 (OPM FEVS, 2024). The trend since 2021 is encouraging: 2021 (34%), 2022 (35%), 2023 (39%), 2024 (41%). OPM notes that the large volume of respondents creates the statistical foundation for agency-level and work-unit-level reporting down to the Rule of Ten threshold, making the FEVS one of the largest recurring workforce surveys conducted anywhere in the world.

What drives those rates up? OPM's own documentation and Partnership for Public Service case studies identify a consistent pattern: visible action from the prior cycle predicts participation in the next one. Employees who saw survey results produce real changes — and who heard about those changes directly from their supervisor — are more likely to respond again. Agencies that fail to close the loop — presenting results at an SES briefing while never communicating outcomes to the work unit — watch response rates trend down over successive cycles and scores become less interpretable as the responding population self-selects toward those who already expect nothing to change (PLAY-005). Participation is a lagging indicator of institutional trust, not just a survey-administration metric.

For state and local government, the response-rate picture is different and more honest: there is no credible nationwide engagement-survey participation benchmark for the state/local workforce. No sector-wide employee-level engagement survey response-rate data exists for cities, counties, and special districts (STAT-031-MISSING). MissionSquare's annual surveys sample HR administrators and budget officials — not the broad employee population — and do not report an employee-level engagement-survey response rate. HR leads at municipalities and counties should track their own response-rate trends over time rather than benchmarking against FEVS figures. The instruments, populations, and administration contexts differ too much for a direct comparison to be meaningful or fair to either number.

Survey data is only worth collecting if someone is accountable for acting on it. The most reliable failure mode in government — documented in OPM guidance and confirmed by Partnership for Public Service postmortems — is the action plan that lives in a slide deck, gets presented to senior leadership, and never reaches the work-unit level where the engagement variance actually lives (PLAY-003). When that pattern repeats, response rates drop in the next cycle and trust in the instrument erodes — a self-reinforcing loop that takes multiple cycles to reverse.

OPM's Simple Approach to Action keeps the method achievable: pick one focus area, take two or three concrete actions including at least one quick win, assign a named owner for each step, and communicate progress four times throughout the year (PLAY-004). The close-the-loop commitment — "you said, we did" — is the visible mechanism that rebuilds belief that the survey has real consequences. High-performing agencies like HHS have built this discipline into their annual cycle: a written work-unit summary communicated directly by the team supervisor, within weeks of results posting, naming what changed and what is still in progress (PLAY-005). The accountability pattern matters because employees who believe the survey will be acted upon respond more candidly, which produces more diagnostic data, which enables better action — and the cycle runs forward instead of backward.

Where Actify fits — and where it does not. Actify is the post-survey action layer: a platform for running the recognition cadences, activity-first engagement rhythms, and participation programs that respond to gaps your annual survey or pulse identified. If FEVS surfaces a recognition gap — 47% positive on Q.17 is the record (OPM FEVS, 2024) — Actify is where you run peer recognition under Title 5 §4503 and build the daily habit of specific, non-monetary acknowledgment that the MSPB identifies as a direct supervisor lever (PLAY-029). Actify is not a survey engine, not a FEVS replacement, and not capable of producing FEVS-grade statistical comparability. The lightweight automatic monthly pulse Actify provides is a directional temperature check on activity participation — it is not a validated engagement instrument and should not be presented as one.

For state/local/municipal HR leads, Actify's clearest value comes after a listening exercise surfaces the action items: mobile delivery with no .gov email required reaches deskless and field staff who never sit at a government workstation, non-monetary peer recognition operates within Title 5 constraints, and flat pricing fits cash-constrained jurisdictions. Federal agencies face FedRAMP and Authority to Operate procurement requirements that gate most software adoption; the fit is strongest at state, local, and municipal jurisdictions where those procurement gates are lighter (PLAY-023, PLAY-028). The honest framing for the HR/engagement lead persona (PLAY-028): Actify does not fix pay, staffing shortages, or poor leadership — those require structural decisions that software cannot make. It is the execution layer for the recognition and engagement commitments that come after you have listened, analyzed the data, decided what to change, and found the owners who will do it.