cumulative

C.7 Measuring efficiency: trials to criterion, cost, and training time

What the task asks: Measure how efficiently a procedure produces its outcome, using trials to criterion, training time and cost-benefit comparisons.

Measurement, Data Display, and Interpretation4 min read4 practice questionsOfficial outline (opens in a new tab)

The big idea

Two teaching procedures can both work. The better choice is often the one that gets there with fewer trials, fewer errors, less instructional time or lower cost. Efficiency measures ask what it took to produce an outcome, not just whether the outcome happened.

That question matters because learners’ time is limited and someone pays for every service hour. A procedure that needs 400 trials where another needs 150 uses instruction time the learner could have spent on other skills.

Breaking it down

Trials to criterion

Trials to criterion is the number of response opportunities a learner needs to reach a preset mastery criterion, such as 90% correct across two consecutive sessions. Fewer trials mean faster acquisition. Close relatives include sessions to criterion, errors to criterion and blocks to criterion.

Trials to criterion is used to:

  • compare teaching procedures, such as two prompting methods (D.8);
  • track a learner’s progress across sets of targets. If each new set is mastered in fewer trials than the last, the learner is acquiring new material faster, sometimes called learning to learn.
04080120160140Set 1115Set 290Set 370Set 455Set 5Target sets, in the order taughtTrials to criterion
Trials to criterion for one learner across five sets of targets, all held to the same mastery criterion. Each set takes fewer trials than the last: the learner is acquiring new material faster.

The comparison is fair only if the criterion is the same. A stricter criterion takes more trials to meet, whatever the procedure.

Training duration and time to criterion

Training duration is the total instructional time needed to reach criterion, in minutes, hours, sessions or days. It catches what trial counts miss: trials differ in length. A procedure that plays a video model before every trial may need fewer trials yet take more minutes.

The same measures apply to staff. The hours of behavioral skills training a new technician needs before implementing a protocol at 90% accuracy is a training duration, and it is a common way to compare training packages.

Cost-benefit analysis

Cost-benefit analysis weighs the resources a procedure uses against the outcomes it produces.

  • Costs: staff hours and wages, materials, training and supervision time, the learner’s time, and the effort or discomfort the procedure involves.
  • Benefits: skills gained, problem behavior reduced, greater independence, less need for later services, better quality of life.

A formal cost-benefit analysis puts both sides in money. A simpler comparison, often called cost-effectiveness, divides cost by a unit of outcome. Program A costs $2,400 and produces 12 mastered skills: $2,400 ÷ 12 = $200 per skill. Program B costs $1,800 and produces 6: $1,800 ÷ 6 = $300 per skill. B is cheaper overall, but A delivers each skill for less.

Program A $2,400 total 12 skills mastered $200 per skill Program B $1,800 total 6 skills mastered $300 per skill
Each bar's length is a program's total cost, split into one block per mastered skill. Program B costs less overall, but Program A's blocks are narrower: $200 per skill against $300.

Efficiency is not the whole picture

A fast procedure that does not maintain or generalize, or that the learner and family dislike, is not efficient in the long run. Weigh efficiency alongside maintenance, generalization and social validity when recommending a procedure (H.2).

Seeing it in practice

A team compares two prompting procedures for teaching Asha to read community signs, using 15-minute sessions for both.

MeasureConstant time delayLeast-to-most prompting
Trials per session2010
Trials to criterion12080
Sessions to criterion120 ÷ 20 = 680 ÷ 10 = 8
Instructional time6 × 15 = 90 minutes8 × 15 = 120 minutes
Errors before criterion631

Least-to-most prompting wins on trials, but constant time delay reached criterion 30 minutes sooner and with far fewer errors. Each least-to-most trial took longer because it could pass through several prompt levels. Which procedure is “more efficient” depends on the measure, so report more than one.

Common mix-ups

  • Effective is not efficient. Both procedures may reach criterion; efficiency asks at what cost.
  • Fewer trials does not always mean less time. Check trial length and the number of sessions.
  • Different criteria make comparisons meaningless. Hold the criterion constant.
  • Cheapest overall is not most cost-effective. Compare cost per unit of outcome.
  • Trials to criterion is not percentage correct. It is a single count of opportunities until a criterion is met.
  • Speed without durability. An efficient procedure must still produce behavior that lasts.

Exam tip: When a stem describes two procedures that both worked, the question is about efficiency. Work out each measure given (trials, minutes, dollars per outcome) before you choose, and check whether the measures point the same way.

Quick recap

  • Efficiency asks how much it took to reach an outcome.
  • Trials to criterion counts opportunities until a mastery criterion is met; the criterion must match across comparisons.
  • Training duration counts instructional time and catches differences in trial length.
  • Cost-benefit analysis weighs resources against outcomes; cost per unit of outcome is a practical shortcut.
  • Efficiency matters alongside maintenance, generalization and social validity.

Further reading

  • Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied behavior analysis (3rd ed.), chapter 4.

Key terms

Trials to criterion
The number of response opportunities a learner needs to reach a set mastery criterion. Fewer trials suggest faster learning or a more efficient procedure, but comparisons are fair only when the criterion is the same.
Training duration (Time to criterion)
The total instructional time, in minutes, hours, sessions or days, that a learner or trainee needs to reach a criterion. It captures differences in efficiency that trial counts miss when trials differ in length.
Mastery criterion
A predetermined level of performance that shows a skill has been learned, usually stated as an accuracy level sustained across a set number of sessions, opportunities or settings.
Cost-benefit analysis
A comparison of the resources a procedure uses, such as staff time, money, materials and client effort, with the outcomes it produces. A related comparison, cost-effectiveness, expresses cost per unit of outcome, such as dollars per skill mastered.

Check yourself

4 questions on C.7. Choose an answer to see whether you're right and why.

  1. Question 1Program A costs $2,400 for a term and produces 12 mastered skills. Program B costs $1,800 and produces 6. Which is more efficient in cost per skill?
    Show the answer

    Answer: C. Divide cost by outcome: $2,400 ÷ 12 = $200 per skill for Program A and $1,800 ÷ 6 = $300 per skill for Program B. B is cheaper in total, but A produces each skill for less. The $150 figure wrongly divides B's cost by A's number of skills.

  2. Question 2Method X reached criterion in 60 trials, at 20 trials per 10-minute session. Method Y reached criterion in 45 trials, at 9 trials per 10-minute session. Which used less instructional time?
    Show the answer

    Answer: D. Method X needed 60 ÷ 20 = 3 sessions, or 30 minutes. Method Y needed 45 ÷ 9 = 5 sessions, or 50 minutes. Y used fewer trials, but each trial took longer, which is why training time should be reported alongside trials to criterion.

  3. Question 3Rafael is taught a series of matching sets with the same procedure and criterion. The first set takes 150 trials to reach criterion, the second 90 and the third 50. What do these data most likely show?
    Show the answer

    Answer: B. Falling trials to criterion across new sets, with the procedure and criterion held constant, suggest that Rafael is acquiring new material faster, sometimes called learning to learn. Reaching criterion in fewer trials is the opposite of less accurate responding, and the stem rules out changes to the criterion or procedure.

  4. Question 4Why must two teaching procedures be held to the same mastery criterion when trials to criterion is the measure?
    Show the answer

    Answer: C. Trials to criterion counts opportunities until a target is met, so a tougher target (for example, 100% across three sessions instead of 80% across two) inflates the count whatever the procedure. The comparison would then reflect the criteria, not the procedures. Trial length and graphing are unaffected.

Adds a step to your study record. Saved in this browser only.

Type to search all 104 guides and the glossary.