Prompt hierarchy examples for IEP data

A prompt hierarchy turns "needed help" into data. Instead of marking a response right or wrong, you mark the level of help it took. Over weeks, the codes should drift from full physical toward independent. That drift is progress, even before the student meets the goal.

Each example below names the skill, the hierarchy, the wait time and how it shows on the data sheet, with a note on what makes it strong. All names and numbers are fictional. For the wider data plan, see the IEP data collection guide.

The standard prompt levels and codes

Use one set of codes for every sheet on your caseload, so any adult reads them the same way. The prompt hierarchy glossary entry has the short definition.

Some teams add a visual prompt (a picture card or written cue) between gestural and verbal. Add it if you use it, and keep the order fixed.

Least to most vs most to least

Least to most starts with the natural cue, waits, then adds help one level at a time until the student responds. It fits skills the student partly knows, because it gives every chance to respond alone.

Most to least starts with the level that guarantees success, then fades help over sessions. It fits new skills and routines where errors are costly or tend to repeat. Most-to-least plans need a written fading rule, or the full prompt never goes away. The prompt fading glossary entry covers the term.

Communication and requesting (examples 1 to 5)

1. Requesting a preferred item with a device. Least to most: I, G, V, M, PP. Wait 5 seconds at each level. Code each request on the opportunity data sheet. Strong because the wait time is fixed, so no adult jumps in early.
2. Asking for a break with a card. Most to least: PP for the first week, then G, then I. Fade when 4 of 5 requests in a session happen at the current level. Strong because the fading rule is written down.
3. Answering "what's your name?" Least to most: I, V (repeat the question), M (say the answer). Score 10 trials across the day. Strong because it skips physical prompts that do not fit a verbal skill.
4. Greeting a peer on arrival. Least to most: I, G (point to peer), V ("say hi"). Only I counts toward the goal. Strong because the goal criterion and the code agree.
5. Using a 2-word phrase to request. Least to most: I, V (expectant pause plus "tell me more"), M (say the 2-word phrase). Strong because the model level gives the exact target phrase.

Weak: "Will request items with prompts as needed." Strong: "Given a desired item in view but out of reach, will request it with a 2-word phrase on the device independently in 4 of 5 opportunities across 3 sessions, scored with a least-to-most hierarchy (I, G, V, M) and a 5-second wait." The strong version tells the collector what counts and how to code the rest. Templates are in the AAC goals.

Daily living and self-care (examples 6 to 10)

1. Washing hands. Most to least across a 7-step task analysis. FP on all steps in week 1, PP in week 2, G by week 4. Code each step. Strong because each step has its own code, so you see which steps still need help.
2. Zipping a coat. Least to most: I, G, V, M, PP, FP, wait 5 seconds. Strong because it uses all six levels on a motor skill where physical help is meaningful.
3. Putting lunch items away. Least to most with a picture sequence as the visual level. Strong because the visual cue is listed as a level, not left out of the data.
4. Toileting routine. Most to least, with prompts faded from V to G on each step after 3 consecutive days at the current level. Strong because the fading criterion is a number of days.
5. Brushing teeth. Graduated guidance: hand over hand, shadowing, then no touch. Code FP, PP, I. Strong because it records the shadow stage as partial physical instead of calling it independent.

Weak: "Will wash hands with minimal assistance." Strong: "Will complete 7 of 7 handwashing steps at the gestural level or less for 5 consecutive days, as recorded on a task analysis sheet." Use the task analysis data sheet and see the task analysis examples for step lists. For goal wording, see the daily living goals.

Academic tasks (examples 11 to 15)

1. Reading sight words. Least to most: I, V (first sound), M (say the word). Wait 3 seconds. Strong because the first-sound prompt is a weaker cue than the full model, so the levels are truly ordered.
2. Solving a single-digit addition fact. Least to most: I, G (point to counters), M (show with counters). Strong because the prompts point to a strategy the student can later use alone.
3. Writing a name. Most to least: FP, PP (hand at wrist), then tracing as the visual level, then I. Strong because tracing is coded as its own level.
4. Matching pictures to words. Least to most: I, G (tap correct card), V ("which one says dog?"). Strong because it caps prompting at verbal so the student is never moved through the answer.
5. Starting a writing task. Least to most: I, G (tap the paper), V ("write your first sentence"). Measure latency to the first word as well. Strong because the code shows how much help the start took, and the latency shows how fast.

Weak: "Will read sight words with 80% accuracy with prompting." Strong: "Given 20 pre-primer sight words, will read 18 of 20 independently within 3 seconds each, on 3 consecutive weekly probes." Prompted correct answers are recorded but do not count. Score these on the accuracy data sheet.

Behavior and classroom routines (examples 16 to 20)

1. Lining up at the door. Least to most: I (bell), G, V. Strong because the natural cue, the bell, is named as the start of the hierarchy.
2. Following a one-step direction. Least to most: I, V (repeat once), G, PP. Wait 5 seconds. Strong because the verbal prompt is a single repeat, so it does not turn into nagging.
3. Transitioning from preferred activity. Least to most with a visual timer as the first level, then V, then G. Strong because the timer is the cue to fade to, not the adult.
4. Using a calm-down strategy. Least to most: I, G (point to the strategy card), V ("pick one"). Strong because it leaves out physical prompts, which can escalate a student who is already upset.
5. Unpacking at arrival. Most to least across a 6-step routine, faded one level after 3 consecutive days at 6 of 6 steps. Strong because a full-routine criterion stops early fading.

Weak: "Will follow directions with reminders." Strong: "Given a one-step direction to the class, will begin following it within 5 seconds, independently or with one verbal repeat, in 8 of 10 directions across 2 observed days." Templates are in following adult directions goals.

Vocational and community (examples 21 to 24)

1. Sorting mail at a school job. Most to least: M, G, I across 3 weeks. Strong because model prompts fit a task the student must see done before trying.
2. Paying at a cashier. Least to most: I, V ("how much?"), G (point to the total), M. Strong because the prompts can fade into the natural cue of the register display.
3. Crossing a parking lot. Most to least: PP (hand on shoulder), G, I, with a safety rule that any error goes back one level. Strong because it treats safety as a reason to keep more support.
4. Clocking in at a work site. Least to most: I, visual (picture card on the time clock), V. Strong because the visual card can stay as a permanent support after adult prompts fade.

How to record prompt data on the sheet

Write the code of the prompt that produced the correct response. If the student responds independently, write I. If a gesture was needed, write G. If the student made an error at one level and then responded after a higher prompt, write the higher code. Never write I for a response that came after any prompt.

Summarize each session two ways: the number of independent responses (for the goal) and the most common prompt level (for fading). A session with 2 independent, 6 gestural and 2 verbal shows real gain over one with 2 independent and 8 full physical, even though both have 20% independent.

The goal must state how progress is measured,^[1] so the goal and the sheet should agree on what counts. For staff training on evidence-based instruction, the IRIS Center offers free modules.^[2] The AFIRM module on prompting covers how to plan prompts, give them in a set order and fade them.^[3]

Printable prompt code card

Print this section and tape it to the data clipboard, or copy it onto the back of each sheet.

Wait time: 3 to 5 seconds at each level unless the plan says otherwise. Fade one level after the written criterion is met.

Frequently asked questions

Use it in IEP Atlas

• Prompt hierarchy in the IEP glossary
• Prompt fading in the IEP glossary
• Measurement methods
• Opportunity data sheet
• Task analysis data sheet

Related guides

• IEP data collection: methods, sheets and schedules that hold up
• Task analysis examples for life skills goals
• How to choose a data collection method for an IEP goal
• Percent accuracy vs percent of opportunities: what the denominator means
• Setting up an IEP data binder
• Interobserver agreement for IEP data

Sources

1. 34 CFR §300.320: Definition of individualized education program, Legal Information Institute, Cornell Law School
2. IRIS Center, Vanderbilt University
3. Prompting (AFIRM Module), AFIRM, Frank Porter Graham Child Development Institute, UNC Chapel Hill

Written and checked by the IEP Atlas editorial team under our editorial policy. This guide describes general practice and federal rules. It is not legal advice; your state may add requirements, and the IEP team makes the decisions for each student.