Safety Does Not Set Teacher Pay in El Paso County

Two El Paso County districts report behavior incidents at rates that nearly overlap — Lewis-Palmer D38 at 31 to 35 per thousand students, Cheyenne Mountain D12 at 36 to 45 — and in both, dangerous-weapons incidents are too few for the state to publish. Cheyenne Mountain pays its teachers $8,484 more. What separates the two is not what the state records about behavior. It is that Cheyenne Mountain’s voters approved 18.698 override mills and D38’s approved 4.294.

There is a tidy explanation for why teachers in Lewis-Palmer School District 38 are paid less than teachers a few miles south: they have an easier job. Economists call it a compensating differential. Harder conditions command a premium; safer, calmer schools do not have to pay one. If that were what the numbers showed, the argument of the third article in this series — that D38’s pay gap is a revenue problem rather than a spending one — would weaken considerably.

It is a testable claim, and Colorado publishes the data to test it. This series has run the same exercise before, on the theory that new apartments were dragging down test scores and on the question of what actually cost D38 its “Distinction” rating. The state’s workbooks record behavior incidents, discipline actions, truancy and law-enforcement referrals for every district, alongside average teacher salaries. Set them side by side across El Paso County, and the theory does not survive.

What the State’s Data Shows

D38 does have the lowest rate of reported behavior incidents of the four. That much of the premise holds, and it is worth saying plainly. What fails is the mechanism. Cheyenne Mountain School District 12 sits alongside it — 36.5 to 44.9 incidents per thousand students against D38’s 31.3 to 35.3, ranges that nearly overlap — and pays the highest teacher salary in the region. Two districts in one county, effectively the same incident profile, $8,484 apart on pay.

Neither district is uniformly safer than the other, which is what makes the pairing useful. D38 records a lower rate of behavior incidents and a lower rate of discipline actions — 32.6 per thousand students against 44.1 — and a law-enforcement referral count too small for the state to publish. In raw counts, which flatter the smaller district, D38 records more discipline actions than Cheyenne Mountain, not fewer. Cheyenne Mountain records markedly less truancy — 1.6% of its pupils habitually truant against D38’s 4.8% — and fewer robberies. On the measures the state publishes, they are closely matched. On pay they are not close at all.

An Asterisk Is Not a Zero

The state suppresses small counts to protect student privacy, marking them with an asterisk. In the workbook’s district incident-count table, 2,179 cells are suppressed, and the lowest figure the state publishes anywhere in it is 4 — so a suppressed cell is somewhere between zero and three. It is not nothing, and it should not be reported as nothing.

The honest wrinkle sits in that table. D38 reported five robberies; Cheyenne Mountain’s count was suppressed. D38 is not lower than its better-paid neighbor in every category, and a fair reading has to say so.

The Hardest District Is Not the Best Paid

If difficulty set the price of a teacher, the four districts above would be led by Colorado Springs School District 11, which reports roughly fourteen to sixteen times D38’s incident rate, 17.7% of its pupils habitually truant, and 16.32 law-enforcement referrals per thousand students, against a D38 count the state suppressed as too small to report. D11 pays $6,807 less than Cheyenne Mountain. Of these four, the district reporting the highest rates of behavior incidents, truancy and law-enforcement referrals is not the best paid. It is third among them, and roughly seventh among the county’s seventeen districts.

One comparison does run the other way, and it belongs here. Academy District 20 pays about $6,000 more than D38 and does have modestly worse truancy — 5.8% against 4.8% — and more recorded discipline actions. It is a single comparison, the truancy gap is about one percentage point, and D20 is four times D38’s size with a far more varied enrollment. It is real. It is not enough to carry a theory.

What the “#1 in Education” Ranking Measures

Colorado’s first-place finish in U.S. News & World Report’s 2026 Best States education category is sometimes read as a statement about how safe or well-run its schools are. It is not one.

The half that measures schools is the one Colorado does worse on. Beneath the Pre-K-12 ranking sits a high school graduation rate of 82.7%, against a national average of 85.8%. U.S. News attributes the state’s rise chiefly to associate-degree outcomes and preschool enrollment rather than to K-12 results.

Two Further Tests

The compensating-differential theory makes two predictions that can be checked against D38’s own records, and it fails both.

Departing teachers went to other districts. Of the 86 licensed staff who left D38 in 2022-23, 34 gave better pay as a reason and 32 had taken a position in another district. They did not leave teaching; they left D38 for districts the theory holds to be harder. That is hard to square with the idea that D38’s conditions were worth accepting a lower salary for, though separation forms record what people chose, not what they would have done otherwise.

The district ran the experiment. D38 raised pay 18.7% in a single year. Teacher turnover fell from 27.6% to 24.4% to 18.8%, and its teaching headcount grew for the first time in the series. The pay moved and the retention moved. This is a before-and-after comparison with no control, and CDE changed its discipline collection in 2023-24, so the safety data cannot be tracked cleanly across the same interval.

What D38’s Own Finance Office Says

The clearest answer is not in the state’s files but in the district’s. D38’s trial balance of 30 September 2025 carries a page headed “‘Actionable’ Revenue Comparisons with Competitors and Comparators,” on which the district ranks itself against sixteen districts of its own choosing.

Colorado Springs 11 — the district reporting the highest rates on every measure above — collects nearly six times D38’s override revenue per student. Its teachers are not paid more for hazard. Its voters approved more money — 19.778 override mills, the most of the four. That does not make locally approved revenue the only thing setting pay: D11 collects the most override money per pupil and still pays less than Cheyenne Mountain, because it also carries far more need. What it does rule out is difficulty as the thing being compensated. Cheyenne Mountain’s approved more than four times as much as D38’s did.

The same page puts D38’s total per-pupil revenue last of the sixteen, a gap of $1,612 per pupil which the district’s own finance office converts to 10.9 mills. It ends: “The only district (of all 178 in the entire state) that we are funded better than: Pueblo County 70.” The same constraint runs through the district’s 2026-27 budget, built around a smaller district and a staffing reduction of about 35 positions.

Across these four districts, the reported-incident data provides no evidence that safer conditions explain D38’s pay gap. Neither did the explanations this series tested before it. New apartments, opt-out rates, per-pupil spending, class size, the size of the district’s administration, superintendent turnover, working conditions and now reported school safety have each been checked against the state’s own records, and D38 comes out of every one of them the same way: the lowest-paid of the districts it is measured against.

What differs, consistently and by a wide margin, is how much operating revenue its voters have authorized: 4.294 override mills against Cheyenne Mountain’s 18.698, and $633 per pupil against $2,750. That is a documented gap rather than an inference. No single cross-section proves causation, and pay also reflects experience, staffing mix and the shape of a salary schedule. But after this many alternatives have been tested and set aside, it is the account left standing.

Sources & further reading

The Independent’s own coverage: Blame the Buildings? What D38’s Test Scores Actually Show — part one, the first alternative explanation this series tested · A Quiet Downgrade: How D38 Lost Its “Distinction” — part two, which established D38 as the lowest-paid and highest-turnover district of its three peers · Four Million Dollars, Frozen — part three, on the 1999 override that never grew and the turnover that followed · Right-Sizing the Rolls — the 2026-27 budget built around the revenue constraint described here · School Spotlight: Monument Academy — the charter whose staff sit inside D38’s state-reported turnover figures.

State data: CDE FY2024-25 mill levy table — the voter-approved override column, alongside total mills and each district’s net assessed valuation · Colorado School Finance Project — the same override mills ranked statewide, used as an independent check · CDE Discipline Statistics — the 2024-25 Behavior Data Workbook (incident counts by behavior) and the separate Discipline Data Workbook (actions and law-enforcement referrals) · CDE Attendance Statistics — 2024-25 Attendance and Truancy Rates by LEA · CDE School/District Staff Statistics — 2025-26 Average Salary Reports and Personnel Turnover Rates.

District record: Lewis-Palmer D38 Financial Transparency — where the monthly trial balance and management reporting series is posted · Colorado Springs D11 Mill Levy Override — documents the 2000 and 2017 voter-approved overrides behind D11’s per-pupil figure · Cheyenne Mountain School District 12 · Academy District 20.

Rankings: U.S. News & World Report, 2026 Best States — Colorado — the state profile carrying the education, Pre-K-12, higher education and crime & corrections ranks cited here, and the graduation-rate figure beneath them.

Methodology

El Paso County contains seventeen school districts. This article compares four of them — the three the series has used throughout, plus Colorado Springs 11 — and every ranking or superlative here should be read as applying to that group, not to the county. Smaller districts were not compiled, and suppression would in any case leave several of them with ranges too wide to place. Behavior incidents are drawn from the Colorado Department of Education’s 2024-25 Behavior Data Workbook and expressed per 1,000 students using each district’s PK-12 count from the 2024-25 attendance file. They are reported as ranges because the state suppresses counts below four: every suppressed cell is treated as spanning 0 to 3, which sets the low and high ends. Discipline actions and law-enforcement referrals come from the separate 2024-25 Discipline Data Workbook, where D38’s referral count is itself suppressed — it is reported here as suppressed rather than as zero, which is how an earlier version of this data had it; truancy and attendance from the 2024-25 Attendance and Truancy Rates by LEA.

Salaries are CDE’s 2025-26 average teacher salary, non-charter column. The distinction matters: charter teachers are paid substantially less and the four districts carry very different charter loads, so an all-schools figure is not comparable across them. Average salary is total salary divided by total FTE.

El Paso County District 49 is included in the serious-incident and override-revenue tables, where the underlying records are complete, and left out of the first table, where its salary and mill levy figures are not yet compiled. Override mill levies are from CDE’s 2024-25 mill levy table and count voter-approved override mills only. That is a narrower figure than a district’s total levy: Colorado Springs 11 levies 19.778 override mills within a total of 40.605, and using the larger number would overstate what its voters approved for operating revenue against the other districts shown. The levy figures were checked against the revenue ones: multiplying each district’s funded pupil count by its override revenue per pupil and dividing by its override mills reproduces the net assessed valuation CDE publishes for that district — $4.121bn for Colorado Springs 11, $2.780bn for Academy 20, $931.5m for D38 and $535.0m for Cheyenne Mountain — in every case to within a thousandth of a percent. Override revenue per pupil and the sixteen-district comparison are taken from D38’s own trial balance dated 30 September 2025, not recalculated by this newspaper. As an internal check, 6,314.9 funded pupils multiplied by $633.42 returns $3,999,984 — the $4 million authorized by voters in 1999, which confirms the district’s figures are being read as intended.

Three limits are worth stating. Behavior incidents and discipline actions are reported events: a district with a stricter code of conduct or a lower reporting threshold records more of them, so these figures reflect policy and practice as well as underlying behavior. The comparison between D38 and Cheyenne Mountain is the most like-for-like available — both are small, affluent, suburban districts in one county — which is why it carries the argument. And this is a single cross-section: behavior data for 2024-25, pay for 2025-26. It tests whether pay tracks difficulty across districts today; it does not test how either has moved over time.

Turnover figures are LEA-level and cannot be split by charter, so D38’s rate blends its own schools with Monument Academy. The reasons departing staff gave are taken from separation forms submitted to the district’s human resources office, not from exit interviews, and the form allows more than one reason, so the percentages exceed 100.