A zero denominator ABC conditional probability is undefined because the conditioning event never occurred in the valid sample. Record 0 eligible exposures and label the probability N/A or undefined. A value of 0 would mean exposures occurred and none met the outcome. Preserve observation coverage, investigate whether the sample could contain the event, and gather suitable data when the question remains important.

Separate undefined from zero

Undefined means there were no eligible denominator units. Zero means one or more eligible units occurred and the numerator was zero.

Reject automatic imputation

Adding an invented exposure, substituting the background rate, or entering 0 changes the evidence. Preserve the empty cell and the reason it is empty.

Check sampling opportunity

Review whether the observation included settings and times where the conditioning event could reasonably occur. Keep client safety and ordinary access ahead of attempts to manufacture exposure.

Set a re-observation rule

If the question matters clinically, name the settings, observation window, stop condition, owner, and minimum descriptive information needed for review. Do not arrange a risky event solely to populate a cell.

Build the probability table from raw cells

List every eligible exposure, whether the outcome occurred, and whether the full sequence window was observed. Keep missed, truncated, invalid, and outside-scope records separate. The conditional denominator is the count of valid exposures to the conditioning event. The numerator is the subset followed by the defined outcome within the stated window.

If the exposure count is zero, division cannot be performed. Write “0 eligible exposures; conditional probability undefined” and retain the background count as a separate descriptive result. A software field that accepts only numbers needs an explicit missing-value code and display label so downstream reports do not turn the empty cell into zero.

Identify why the cell is empty

Review the observation plan, actual coverage, response and event definitions, setting mix, and time of day. The conditioning event may have been absent, difficult to observe, coded under another label, or excluded by an overly narrow window. Each explanation leads to different follow-up work.

A true absence in a representative sample is useful information about that sample. A coverage failure calls for a measurement repair. An event that would require creating distress or removing access should remain unprovoked while the qualified clinician considers safer assessment options.

Report the result without inventing precision

Use a table that shows exposure count, outcome count, conditional result, background probability, valid observation, and missingness. A zero denominator ABC conditional probability should appear as undefined, with the zero exposure count beside it. The narrative can state: “No denied-request exposures occurred in 30 valid intervals, so the requested conditional probability is undefined. The response occurred in 4 of 30 valid intervals overall.”

Keep the observation scope visible. An undefined cell supplies no evidence that the conditional relation is present or absent outside the sample. Record the next evidence step, owner, stop rule, and review date only when the question remains important to the assessment.

Frame one clear review question

The zero-denominator exception log for Sami names the observable events, sequence direction, time or interval window, planned exposure, valid denominator, missing-state rule, raw cells, calculation owner, and review date before percentages are interpreted.

Worked example: no eligible exposures

Sami has 30 valid intervals and no denied-request exposures. P(response | denied request) is undefined because its denominator is 0. The response occurs in 4 of 30 intervals, giving a background interval probability of 0.133. That value cannot fill the missing conditional cell.

Audit the exposure denominator

Sami's log distinguishes three states: no denied-request exposure, denied-request exposure without response, and a missed observation. Only the second state could contribute a valid zero outcome after an actual exposure.

Interpret the empty cell cautiously

The empty cell may reflect a rare event, a poorly selected routine, a definition that observers could not apply, or a genuinely absent condition. The reason guides the next observation plan.

Protect access during observation

During Sami's undefined conditional probabilities review, keep AAC, interpreters, mobility, food, water, bathroom use, prescribed care, health support, rest, relationships, and emergency help available. Record access or health changes and use accessible communication to ask about relevant events, comfort, assent, dissent, and priorities.

Use evidence within its stated scope

The BACB ethics hub and CASP public summary provide professional context. The BCBA Test Content Outline includes direct, indirect, and product measures; validity and reliability; representative measurement; and indirect, descriptive, and experimental functional assessment as examination content.

For Sami's undefined conditional probabilities question, primary descriptive-assessment research compares conditional with background probabilities and shows why event direction, sequence windows, cell counts, and context matter. Antecedent versus consequent predictors, contingency space analysis, and comparisons of descriptive methods document limited correspondence between descriptive and experimental findings. Research on reducing functional-assessment ambiguity distinguishes observed association from experimental demonstration. ASHA says AAC users should always have access to their tools or devices.

Close the descriptive-data review

Review the zero-denominator exception log with Sami and the responsible clinician. Record raw observations, definitions, windows, coverage, calculations, limitations, the selected next assessment step, its owner, and a review date. Reopen the method when the response, context, access, health, observer, software, or decision changes.

Clinician review checklist

  • Does zero mean an observed exposure with no outcome, or were there zero exposures?
  • Are truncated, missed, invalid, and outside-scope records excluded under written rules?
  • Does the software preserve undefined as a distinct state?
  • Are background and conditional probabilities labeled separately?
  • Was the observation sample capable of containing the conditioning event?
  • Does follow-up preserve safety, AAC, assent, and ordinary access?
  • Is the next step owned by the qualified clinician responsible for the assessment?

Related resources

Sources