A5

A school board was receiving complaints it could not act on. Roughly 150 students and about 300 stakeholders across staff, parents, and students, and classes of eight or nine; small enough that every dissatisfaction arrived as a named individual account. Leadership effectiveness, program quality, and family churn were each understood to be problems but none had a measurement attached to it. Parent sentiment, the signal closest to churn, was not being collected at all.

Staff were being asked to evaluate the administrator who set their schedule. Students were being asked to evaluate teachers who would grade them the following week. Therefore, anonymity became the instrument’s operating condition.

Discovery

The first discovery was designed to locate the problem. Twenty-five rating items on a 1–4 scale across five categories fielded to staff returned 28 responses and an overall average. Looking at the categories separated, a quarter of a workforce reporting that candor carries a penalty prompted me to look further.

Framing

In addition to aggregates, open-ended responses named specific mechanisms like oversight density and after-hours expectation. Discovery 1 pointed at a particular office, so discovery 2 was built as separate targeted instruments rather than one universal form.

Tradeoffs

• Separate instruments per administrator over one comparable template
• Full anonymity over follow-up capability
• Frequency scales over agreement scales for conduct items

Data access was held to me

I held sole access to raw responses for the duration of the contract. The head of school was excluded from the dashboard intentionally and the board only received synthesized findings. I distributed the staff and student surveys personally rather than through administration. Each of those is a small operational cost but an instrument that measures fear of retaliation cannot be delivered through the channel the fear is about.

The student instrument was worded to its condition

Discovery 1 and subsequent conversations established a culture of silence among students. Standard evaluative phrasing assumes a respondent who believes the answer is consequence-free. The grades 6–12 instrument was rewritten around that, with all questions across culture, curriculum, and staff phrased so that a student describing a condition was not required to accuse a person.

Two administrator-specific staff surveys at 22 and 24 questions, a student voice survey for grades 6–12 at 15 questions, a parent feedback survey covering academics, culture, staff attitude, facility, a parent exit survey, and a staff exit survey added in the third wave at seven questions across three sections, built to capture why a departure happened and what would have changed the outcome.

From that, a live, auto-syncing dashboard across separate channels reading published endpoints on a 30-second refresh, carrying cross-survey score comparison, section-level breakdowns, flagged high and low questions, multiple-choice response distributions, and a churn-risk signal on the parent tab.

Reporting the average suppressed distribution.

The institutional composite read as adequate, which is the number a board would file rather than act on. Underneath it, the items touching conduct carried a tail of strong disagreement that the category averages concealed as thoroughly as the composite did.