Should I average the session percentages or pool the totals?

Whichever your approved measurement protocol specifies. Pooling weights every opportunity equally and is the usual reading of "percent of opportunities". Averaging the session percentages weights every session equally, which suits a goal written in terms of sessions or days. They give different answers whenever sessions have different numbers of opportunities, so the important part is saying which one you used.

Does a session where the student got nothing right count as zero or as missing data?

It counts as zero. Opportunities were administered and the student met criterion on none of them, which is a real observation. Dropping it inflates the result. Missing data means no opportunities were administered at all, and that session is left out of the percentage entirely rather than scored.

What happens if I rescore a session?

The corrected record replaces the original instead of joining it. Counting both is a common way a pooled percentage drifts, because the same opportunities end up in the denominator twice.

Is anything I type here uploaded?

No. The calculation runs in the browser tab and the page makes no network requests for the values you enter. If you allow analytics, Google Analytics counts the visit and button use, never the values. No advertising scripts. Downloading or printing the result sends it wherever you choose to send it.

IEP trial percentage calculator

Two sessions of 4 out of 5 and one of 9 out of 10 do not average to what most people expect. This works out both summaries, shows the totals behind each, and refuses to produce a number when the data will not support one.

The arithmetic runs in this browser tab. What you type is never sent anywhere. If you allow analytics, Google Analytics counts the visit and which buttons were used (for example “Print”), never the values you enter. No advertising scripts. Printing, downloading or sharing a result sends it wherever you point it, so follow your district’s approved process for anything that identifies a student.

Sessions

One row per session, sitting or sheet.

Are these sessions comparable?

Goal or skill

Optional, and it stays in this tab. Use a generic description rather than the wording from a student’s IEP.

Protocol version

Prompt condition

If the sessions below were collected under different conditions, they are measuring different things and should be summarised separately.

Session data

Result

Updates as you type.

Take it with you

The download is built in this tab from what you entered.

The part that trips people up

Four sessions: 2 of 5, 9 of 10, 4 of 5, and 0 of 5. Same goal, same conditions, four different days.

Why 60% and 52.5% are both correct

Pooled, you have 15 opportunities met out of 25 administered, which is 60%. That answer weights every opportunity the same, so the ten-opportunity session pulls harder than the five-opportunity ones. It is the usual reading of a goal written as “percent of opportunities”.

Averaged across sessions, you have 40%, 90%, 80% and 0%, which is 52.5%. That answer weights every session the same regardless of how many opportunities it contained. It suits a goal written in terms of days or sessions.

Neither is a mistake. Reporting one while your goal is written for the other is. The fix is to say which you used, every time, in the same sentence as the number.

Why the 0 of 5 session has to stay in

Drop it and you get 15 of 20, which is 75%. That is fifteen percentage points of difference produced by removing a session where the student was present, five opportunities were administered, and criterion was met on none of them.

An observed zero is data. A missed session is not. The two look similar on a sheet and mean opposite things: one says the skill was not demonstrated, the other says nothing at all happened to observe. A percentage that quietly treats the first as the second is reporting a result that never occurred.

Why a corrected session replaces the original

You score a session from memory at the end of the day, then find the tally sheet and revise it from 3 of 5 to 4 of 5. If both rows reach the total, those five opportunities are counted twice, and the pooled denominator is ten opportunities that were never administered.

The original stays in your records with a reason attached. It just does not reach the sum. In the calculator, mark the revised row as superseding the earlier one and it is handled for you.

Why it sometimes refuses to give an answer

A session with zero administered opportunities has no denominator, so there is no percentage to report. The tool says “not calculable” instead of showing 0%, because those are different claims about the same child.

The same applies when a session has more met than administered, or a negative count, or a fractional trial. Those are input errors, and repairing them silently means the number you quote in a meeting came from a guess this page made on your behalf.

Record the observations once, not twice

This calculator works on totals you have already transcribed from somewhere. The transcription is the part that eats the afternoon. In the app you score the trial at the table by voice, tap or watch, and the summary is already built when you need it.

Free for up to three students. No card to start.

Get the app See how capture works

Questions

Can I use this with a real student’s data?

The arithmetic never leaves this tab, so the numbers are as safe here as they are in a calculator app on the same device. What we would avoid is typing a student’s name or their IEP wording into the goal field, printing the result to a shared network printer, or downloading it somewhere synced. Removing a name does not make a record non-identifying, and your district’s process governs the rest.

Does it handle interval or duration data?

Not in this tool, on purpose. Partial interval, whole interval, momentary sampling and duration each need their own denominator and their own account of what a missed observation means. Running them through a trial calculator produces a number with no defensible meaning. The worked examples page shows how each is handled.

Why is there no email gate on the result?

Because the result is the useful thing, and putting it behind an address would make the tool a lead form wearing a calculator costume. If it helps, use it. If the app is not right for you, that is a legitimate outcome.