ConductScienceConductScience
ToolsConductScience tool
Turn-TakingFree in-browser calculator

Conversation Turn Analyzer.

Paste a child-partner dialogue transcript with speaker tags (e.g. C: and P:) and the analyzer returns turns per speaker, average turn length, total speaker-to-speaker turn switches, the child topic-maintenance ratio, a four-tier topic-maintenance classification (poor, emerging, adequate, strong), and a three-tier turn-balance classification (partner-dominant, balanced, child-dominant) in under five minutes. Tier thresholds are derived from Fey (1986), Brinton & Fujiki (1989), Mentis & Prutting (1991), and Timler (2008). Built for school-based SLPs, clinic SLPs, autism-assessment teams, graduate SLP students, and paediatric language researchers screening pragmatic-discourse in children with DLD, ASD, ADHD, and TBI.

Run calculator Request a Quote
PrivateData stays in your browser
LiveNo sign-up required
Validated2026-04-06
CitableMethods and citation included

Calculator

Results update in place

Paste a child-partner dialogue transcript

One turn per line. Start each line with a speaker tag (e.g. C: for child, P: for partner) followed by the utterance. Append [on] or [off] at the end of each child turn to mark topic maintenance.

Paste a dialogue transcript above to compute turn-taking metrics, topic maintenance, and a pragmatic-discourse classification.

Get the full analysis

Unlock normative comparison and error analysis

This free tool covers the basic case. ConductSpeech adds normative comparison, error categorisation, and a parent-ready report.

Get the full analysis with ConductSpeech

When to use

  • Initial screening of a new school-based SLP caseload for pragmatic-discourse concerns — collect one 20-turn dialogue per child and flag children with topic maintenance below 70 %
  • IEP goal writing — use the topic-maintenance ratio and the turn-balance tier to draft a SMART pragmatic-discourse goal that the classroom teacher can report on
  • Weekly progress monitoring during active social-communication intervention — collect a fresh dialogue, re-score, and compare to baseline
  • Autism-assessment team screening — collect a conversational sample as part of a multidisciplinary autism evaluation and report the topic-maintenance ratio alongside the other pragmatic measures
  • ADHD caseload triage — screen children referred for "interrupts a lot" and "goes off on tangents" with a 20-turn dialogue and report the turn-balance and topic-maintenance tiers
  • Traumatic brain injury (TBI) re-evaluation — track conversational-discourse recovery across sessions after paediatric TBI
  • Graduate SLP training — teach pragmatic-discourse analysis with a concrete, reproducible rubric before moving to the full Brinton & Fujiki (1989) or Mentis & Prutting (1991) coding schemes
  • Discharge / dismissal screening — collect two fresh dialogues at the end of an intervention block and confirm that topic maintenance and turn balance have moved into the adequate or strong tier

Do not use for

  • As a stand-alone diagnostic instrument for pragmatic-language disorder, autism spectrum disorder, or social (pragmatic) communication disorder — always pair with a standardised pragmatic measure (CCC-2, CASL-2 Pragmatic Judgment) and the full case history
  • For samples shorter than 20 turns — the topic-maintenance ratio is unstable with fewer than 10 child turns
  • For monologue samples (narrative retells, fiction generation, expository samples) — the analyzer assumes two-party dialogue and the topic-maintenance ratio is not meaningful for monologue
  • For dual-language learners in the second language without considering language exposure — a partner-dominant or poor-topic pattern may reflect L2 attrition rather than a pragmatic deficit
  • For children producing single-word or two-word utterances — turn-taking and topic maintenance assume the child is producing connected discourse
  • As the only pragmatic measure in a formal evaluation report — supplement with a published standardised pragmatic assessment
  • To re-score the same transcript after intervention — collect a fresh sample at every progress-monitoring point to avoid overstating the gain

Watch for partner-driven topic shifts

Topic maintenance is only meaningful when the partner is maintaining topic too. If the partner shifts topic every two turns, the child has no opportunity to maintain topic and the resulting score will artificially flag the child as "off-topic." Re-transcribe with partner topic-shifts marked explicitly and exclude those child turns from the denominator before scoring.

Audio-record the sample when possible

Real-time topic-maintenance coding is biased toward the turns the SLP remembers. Audio-record the dialogue and score from the transcript — even a quick gist transcript will produce more accurate counts than memory. Reserve real-time scoring for triage and use transcript-based scoring for IEP progress notes and evaluation reports.

Use open-ended prompts and 3-5 second wait-time

A closed-ended prompt ("Did you have fun?") pulls the sample toward a partner-dominant pattern and limits the child's opportunity to maintain topic. Use open-ended prompts ("Tell me about your favourite thing to do at recess") and wait 3-5 seconds after each child turn to give the child a chance to initiate.

Re-sample with a non-preferred topic

A child-dominant pattern in a preferred-topic sample may reflect age-appropriate excitement rather than a turn-yielding deficit. Re-sample with a non-preferred topic (e.g. tell me about your chores) and compare — a persistent child-dominant pattern across topics is the clinically meaningful signal.

Rule out receptive-language contributors

Off-topic turns can be driven by a receptive-comprehension deficit (the child did not understand the partner turn) rather than a pragmatic deficit. Rule out receptive language with a standardised receptive measure (CELF-5 Receptive, CASL-2 Receptive) before committing to a pragmatic-discourse intervention plan.

Use 3-5 second wait-time in the elicitation

A short wait-time pushes the partner into the next turn before the child has a chance to initiate, which looks like a partner-dominant pattern in the analyzer. Explicitly wait 3-5 seconds after every child turn before the partner takes the next turn — this reflects the clinical target for conversational-assertiveness elicitation (Fey 1986; Timler 2008).

1

Method

The Conversation Turn Analyzer parses a line-per-turn dialogue transcript using a simple regular-expression grammar. Each non-empty line must begin with a speaker tag (e.g. "C:" or "Parent -") followed by the utterance text. Recognised child tags (C, CH, Child, Kid, Target, T, Student) are normalised to "child"; recognised partner tags (P, Pa, Partner, Parent, Adult, A, E, Examiner, Clinician, SLP, Teacher, Mom, Mother, Dad, Father) are normalised to "partner"; unrecognised tags default to "partner" (the conservative default — the child metric is the one of clinical interest). An optional "[on]" or "[off]" flag at the end of each child turn sets the topic-maintenance flag; child turns without a flag default to on-topic. Per-speaker metrics (turn count, total words, average turn length, longest turn, shortest turn) are computed from the parsed turns. The number of speaker-to-speaker turn switches is counted by iterating through the turn list and incrementing a counter each time the speaker of turn n differs from the speaker of turn n-1. The child topic-maintenance ratio is on-topic child turns / total child turns. The ratio is classified into one of four tiers (poor ≤ 0.50, emerging 0.51-0.70, adequate 0.71-0.89, strong ≥ 0.90) based on Brinton & Fujiki (1989) and Mentis & Prutting (1991) typical-developer expectations. The child share of total turns is classified into one of three tiers (partner-dominant < 0.40, balanced 0.40-0.60, child-dominant > 0.60) based on Fey (1986) and Brinton & Fujiki (1989) conversational-assertiveness / responsiveness targets. The analyzer then surfaces 3-6 clinical caveats driven by the specific pattern (short-sample rule-out, partner-driven topic-shift rule-out, preferred-topic rule-out, receptive-language rule-out, and dual-language rule-out) and a per-speaker and per-turn breakdown. The output is a planning aid for conversational-discourse screening — it is not a stand-alone standardised test and must be paired with a published pragmatic measure (CCC-2, CASL-2 Pragmatic Judgment, CELF-5 Pragmatics Profile), a microstructure measure (MLU, NDW, IPSyn, DSS, PCC, PGU), and a narrative macrostructure measure (NSS, story grammar checklist) for a complete language-sample profile.

2

Validated

Last validated 2026-04-06. Calculations are designed for planning and documentation support; verify procurement decisions against manufacturer specifications or institutional SOPs.

3

How to cite

How to Cite

ConductScience Conversation Turn Analyzer (v1.0). ConductScience, Inc. 2026. Available at: https://conductscience.com/tools/conversation-turn-analyzer

Fey ME. Language intervention with young children. San Diego, CA: College-Hill Press; 1986.

Brinton B, Fujiki M. Conversational management with language-impaired children: Pragmatic assessment and intervention. Rockville, MD: Aspen; 1989.

Mentis M, Prutting CT. Analysis of topic as illustrated in a head-injured and a normal adult. Journal of Speech and Hearing Research. 1991;34(3):583-595.

Timler GR. Social communication: A framework for assessment and intervention. The ASHA Leader. 2008;13(15):10-13.

Adams C. Practitioner review: The assessment of language pragmatics. Journal of Child Psychology and Psychiatry. 2002;43(8):973-987.

Bishop DVM. Children's Communication Checklist, Second Edition (CCC-2). London: Pearson; 2003.

Bishop DVM, Baird G. Parent and teacher report of pragmatic aspects of communication: Use of the Children's Communication Checklist in a clinical setting. Developmental Medicine & Child Neurology. 2001;43(12):809-818.

Tager-Flusberg H, Paul R, Lord C. Language and communication in autism. In: Volkmar FR, Paul R, Klin A, Cohen D, editors. Handbook of Autism and Pervasive Developmental Disorders, Volume 1: Diagnosis, Development, Neurobiology, and Behavior. 3rd ed. Hoboken, NJ: Wiley; 2005. p. 335-364.

Tomblin JB, Records NL, Buckwalter P, Zhang X, Smith E, O'Brien M. Prevalence of specific language impairment in kindergarten children. Journal of Speech, Language, and Hearing Research. 1997;40(6):1245-1260.

American Speech-Language-Hearing Association. Social Communication Disorder — Practice Portal. Rockville, MD: ASHA; 2024. Available at: https://www.asha.org/practice-portal/clinical-topics/social-communication-disorder/

Why Turn-Taking and Topic Maintenance Matter

Conversational turn-taking and topic maintenance are the two most direct measures of a child's ability to participate in connected, reciprocal discourse with a partner. A child who cannot yield a turn when the partner is speaking, or who cannot stay on topic across three or four exchanges, cannot participate in classroom discussion, peer play, or a structured interview — even when every other language-sample metric is intact. For this reason, turn-taking and topic maintenance are routinely the first pragmatic-discourse measures on the caseload-triage list for school-based SLPs assessing children with developmental language disorder (DLD), autism spectrum disorder (ASD), attention-deficit / hyperactivity disorder (ADHD), and traumatic brain injury (TBI).

Turn-taking and topic maintenance are the pragmatic dimensions the IEP team cares about. Classroom teachers, special-education teachers, and parents describe pragmatic-discourse deficits in the vocabulary of turn-taking ("he interrupts all the time") and topic maintenance ("she goes off on tangents and I can't follow her"). The IEP meeting is held in this vocabulary, and the IEP goal that results is written in this vocabulary. A pragmatic-discourse measure that reports turn-taking and topic-maintenance numbers directly is the measure the IEP team can act on.
The measures are fast to collect and fast to score. A 20-turn child-partner dialogue can be collected in 5-8 minutes during a session and scored in 2-3 minutes with this analyzer. That is fast enough to use on every child on the caseload at triage, fast enough to use as a weekly progress-monitoring metric during active social-communication intervention, and fast enough to use at discharge / dismissal screening. Contrast this with the published standardised pragmatic tests (CCC-2, CELF-5 Pragmatics Profile, CASL-2 Pragmatic Judgment), which each take 20-40 minutes to administer and require a licensed test kit — those are the right tools for a formal evaluation, but they are too slow for weekly progress monitoring.

How to Collect a Conversational Sample

A good conversational sample has three properties: (1) the topic is rich enough that the child has something to say, (2) the partner asks open-ended rather than closed-ended questions, and (3) the partner waits 3-5 seconds after each child turn to give the child a chance to initiate. The combination produces a balanced dialogue in which the child has a genuine opportunity to maintain topic, yield turns, and repair breakdowns.

Recommended prompts. Use open-ended prompts keyed to the child's interests: "Tell me about your favourite thing to do at recess" for a school-age child, "What did you do at the park yesterday?" for a preschooler, or "Tell me about your favourite game" for an older child. Avoid yes / no questions, which pull the sample toward a partner-dominant pattern, and avoid topic shifts by the adult, which confound the child topic-maintenance count.
Transcription. Audio-record the sample and transcribe verbatim. Use "C:" for the child and "P:" for the partner (or other recognised tags — see the FAQ). After transcription, go back through the child turns and mark each one as "[on]" or "[off]" for topic maintenance, using the previous turn as the anchor. The Mentis & Prutting (1991) topic-maintenance definition is: the child turn is on-topic if it maintains, shades, or extends the topic set by the previous turn; it is off-topic if it introduces a new topic, re-introduces a previous topic, or does not relate to any prior topic.

Interpreting the Metrics

Average turn length (in words). Typically-developing school-age children produce 4-10 word turns in structured dialogue. Shorter average turns are common in DLD and in dual-language learners in L2; longer average turns are common in ADHD and pragmatic-language disorder (rapid, tangential turns). Compare child average turn length to partner average turn length — a large gap in either direction is clinically meaningful.
Turn switches. In a perfectly alternating dialogue, turn switches = total turns − 1. Fewer switches than that indicate consecutive same-speaker turns, which are common when one speaker produces multiple short utterances in a row. Turn switches are less diagnostic than topic maintenance and turn balance but are useful as a reproducibility check.
Child topic maintenance. The most diagnostic metric in the analyzer. A ratio at or below 50 % is a conservative clinical red flag — fewer than half of child turns stay on topic. A ratio of 51-70 % is emerging — the child can stay on topic some of the time but not reliably. A ratio of 71-89 % is adequate — the child can carry a structured dialogue across several turns. A ratio of 90 % or higher is strong — topic maintenance is a relative strength. The tier classification is an interpretation framework, not a cutoff — always pair the ratio with the full case history and clinical judgement.
Turn balance. A child share between 40 % and 60 % is balanced — the clinical target. Below 40 % is partner-dominant — the adult is carrying the conversation, which is common in DLD and ASD. Above 60 % is child-dominant — the child monopolises the floor and does not yield turns, which is common in ADHD and pragmatic-language disorder.

IEP Goal Examples from Analyzer Output

The analyzer is designed to feed directly into the IEP goal-writing process. Here are three example IEP goals keyed to the three most common patterns the analyzer identifies.

Pattern 1: Poor topic maintenance (ratio ≤\leq 50 %). "Given a 15-turn child-partner dialogue on a preferred topic with open-ended prompts, [child] will maintain topic across at least 4 consecutive turns on 4 of 5 progress-monitoring probes across 8 weeks, as measured by the ConductScience Conversation Turn Analyzer." Pair with explicit topic-maintenance instruction using scripted dialogue, visual topic anchors, and partner-implemented cuing (Timler 2008).
Pattern 2: Partner-dominant balance (child share < 40 %). "Given a 20-turn child-partner dialogue with 3-5 second wait-time and comment-based stimuli, [child] will produce at least 40 % of total turns on 4 of 5 progress-monitoring probes across 8 weeks, as measured by the ConductScience Conversation Turn Analyzer." Pair with conversational-assertiveness intervention using comment-based stimuli, wait-time scaffolds, and modelled initiation (Fey 1986).
Pattern 3: Child-dominant balance (child share > 60 %). "Given a 20-turn child-partner dialogue on a non-preferred topic, [child] will yield the conversational floor after each turn (producing between 40 % and 60 % of total turns) on 4 of 5 progress-monitoring probes across 8 weeks, as measured by the ConductScience Conversation Turn Analyzer." Pair with turn-yielding intervention using visual turn-tokens, stop-signals, and partner-implemented cues (Brinton & Fujiki 1989).

Frequently asked

325
Free tools
1,200+
Institutions
100%
Client-side
0
Uploads required

Next steps

Protocol

Narrative Scoring Scheme (NSS) Calculator — score a narrative sample on the 0-5 Heilmann 2010 rubric for each macrostructure dimension

Reference protocol for the next planning step.

Read guide Protocol

Story Grammar Scorer — tick the six Stein & Glenn (1979) story grammar elements and get a completeness tier in under five minutes

Reference protocol for the next planning step.

Read guide Protocol

MLU Calculator — compute baseline mean length of utterance from the same dialogue sample

Reference protocol for the next planning step.

Read guide Protocol

Lexical Diversity Calculator — compute NDW, TTR, and tokens from the dialogue sample

Reference protocol for the next planning step.

Read guide