13.3 Assessing Language Development
Key Takeaways
- PECT 0006.8 applies screening, diagnostic, benchmark, formative, and summative assessment specifically to language development — naming, sequencing, describing, inferring, retelling — not to a generic worksheet grade.
- A brief universal oral-language screener flags who needs a closer look; it does not diagnose a language disability or become an IEP by itself.
- Language samples and oral-language rubrics are core classroom tools: collect talk, interpret it, and change instruction. Do not invent commercial cut scores Pearson and PDE do not publish.
- Results exist to plan, modify, and differentiate language instruction and to make accommodations — extra wait time, visuals, home-language evidence, AAC — not to file and wait for Grade 3 PSSA.
- Home-language competence is language evidence. Scoring a dual language learner only in English and calling the child 'low language' misreads 0006.8.
13.3 Assessing Language Development
Quick answer: PECT 0006.8 asks you to select appropriate assessments of language development for different purposes (screening, diagnosis, benchmark, formative, summative), interpret the results, and use them to plan, modify, and differentiate instruction and to make accommodations. The evidence is oral language — conversation, play talk, retell, description, inference — gathered with tools such as language samples and oral-language rubrics. Do not invent commercial cut scores. Chapter 4 (0002.1) taught the SAS types; this descriptor applies those types to language, not to math packets or PSSA prep in PreK.
Language development here means the 0006.6–0006.7 skills: naming, sequences, associations, ideas, inferences, descriptions, retell, story sequence, personal connections, tone, and prediction talk. It is not yet a phoneme-segmentation score (0007.8) and not yet a fluency rate (0008.8). If the stem’s decision is “who cannot blend CVC,” you have left 0006.8.
Match the type to the language decision
Use the same SAS logic as 0002.1, with talk as the evidence.
| Type | Language decision | PreK–4 snapshot | Misuse |
|---|---|---|---|
| Screening | Who might need a closer look at language? | Brief, universal check: every kindergartner retells a very short familiar story or names a set of objects/pictures | Treating a 4-minute task as a speech-language diagnosis |
| Diagnostic | Which language skills are thin, and in which contexts? | After a flag: longer language sample, targeted probes (sequence vs. inference vs. vocabulary), SLP collaboration when indicated | Giving a deep probe to the whole class every Monday |
| Formative | What should I teach in tomorrow’s talk? | Anecdotal notes during dialogic reading; a science-talk checklist | Filing notes with no change in prompts |
| Benchmark | Is the class on track for age/grade oral-language expectations at this checkpoint? | Periodic common retell or description task aligned to ELS / PA Core Speaking and Listening | Using a winter oral-language checkpoint as tomorrow’s only lesson plan for one child |
| Summative | After teaching, can they retell, describe, or explain at the standard? | End-of-unit oral rubric; documented conference | Calling PSSA the PreK language test (PreK–2 do not take PSSA) |
| Authentic (method) | Can they use language in a real task? | Explaining a block structure to a peer; giving the lunch count; retelling a class trip | A multiple-choice “vocabulary test” with no talking |
Screening ≠ diagnosis. A child who flags on a universal oral-language screener needs more information: a language sample, classroom observation, family input on home language, hearing check if indicated, and, when concerns persist, referral pathways the district uses. One quiet day is not a disability. Skipping screening until May is the opposite error.
Formative versus summative. If you change your CROWD prompts tomorrow because three children could not sequence the story, the retell was formative. If you score an end-of-unit rubric after you have taught sequence language, it is summative. The same retell protocol can serve both; use decides the type.
Benchmark is a periodic pulse against speaking-and-listening / ELS language expectations — often fall, winter, spring — so the team can see whether core talk-rich instruction is working. It is not a daily sticky note (formative) and not a full diagnostic inventory.
Language samples — the classroom workhorse
A language sample is a documented stretch of the child’s real talk: play, a conversation, a story retell, or an explanation of a content task. You are collecting what the child can do with language, not how neatly they sit.
Practical PreK–4 method (no invented norms):
- Choose a context that invites language (familiar book, blocks, science table), not a yes/no quiz.
- Record or write verbatim as much as you can — phone audio with family permission per district policy, or live notes. Capture child utterances, not only your impression (“good talker”).
- Look for the 0006 skills: Does the child name? Sequence? Associate? Convey an idea? Infer? Describe? For a book context: retell key events, order, connect, notice tone, predict?
- Note support: How many prompts did you need? Gestures? Home language? Pictures?
- Compare to more than one sample before a high-stakes decision. Tired, shy, or new-to-English children look different on Tuesday.
What you may describe without fake numbers: mean length of utterance as a qualitative impression (“mostly two-word combinations” versus “multiword explanations”), vocabulary diversity, use of first/then, presence of story elements, and whether inferences appear. What you may not do on this exam: announce “the child scored 82 on Publisher X, which is 1.5 standard deviations below the mean,” unless a stem actually hands you that publisher’s score. Pearson does not publish commercial language-test cut scores for PECT. PDE does not ask PreK–4 teachers to memorize CELF or PLS scaled-score tables. If a district uses a norm-referenced speech-language test, that is typically specialist territory; your job is to interpret classroom language evidence and know when to collaborate, not to invent a cutoff.
Oral-language rubrics
A rubric turns a sample into a decision. Keep criteria observable and tied to 0006.6–0006.7 and PA Core Speaking and Listening / ELS language, not to personality.
| Criterion (example) | Emerging | Developing | Secure (classroom language) |
|---|---|---|---|
| Naming / vocabulary in context | Relies on pointing or a general word (that, stuff) | Names some objects with support | Names objects and uses topic words in the activity |
| Sequence | Events in random order | Two events in order with prompts | First–next–last with a connector word |
| Description / ideas | Labels only | Adds one attribute or reason | Describes with attributes and conveys an idea |
| Retell / story | Isolated labels from pictures | Some key events, order shaky | Recognizable retell of a simple story |
| Inference / prediction talk | Does not go beyond what is literally named | Infers or predicts with heavy prompting | Uses a clue (picture, event) plus knowledge |
Do not turn the rubric into a public shame chart (0002.5 ethics; 0005.1 community). Use it to group for language, to choose dialogic prompts, and to document growth for families.
Interpret, then change teaching
0006.8’s second half is use. A sample that sits in a folder is not assessment.
| Finding | Instructional response | Accommodation / access | |---|---|---|---| | Names objects but cannot sequence | Teach first–next–last with photos in science and stories; fewer events | Photo sequence strip; extra wait time | | Strong home-language retell, thin English | Build English with L1; dialogic reading with visuals | Accept L1 retell as evidence of story knowledge; bilingual peer or family story | | Describes in play, silent at rug | Small-group dialogic reading; reduce audience size | Alternative seating; talking piece; AAC / core board if the child uses one | | Cannot infer even with pictures | Model “I think… because I see…” on one clue at a time | Highlight a single picture clue; think-aloud | | Flags on universal screener | Diagnostic sample + family conversation; do not label | More information before any disability claim | | Meets oral rubric, cannot decode | Language is a relative strength; word-reading instruction is the other Simple View factor | Do not cancel oral language; add phonics (later chapters) |
Differentiate the language demand, not the right to hear books: more modeling, more turns, simpler sequence length, richer vocabulary for children who are ready. Modify a language task only when an IEP requires a changed standard; most 0006.8 items want accommodations (time, visuals, response mode, home language, assistive communication) while keeping the language target.
ELLs. Chapter 6 (ELPS, BICS/CALP) still applies. On 0006.8, the professional error is an English-only rubric that scores a fluent Spanish storyteller as “no language.” Assess what the child can do in each language you can access, and use ELPS-informed supports so content talk is comprehensible. Do not delay oral-language instruction until English is “done.”
Students with disabilities. Follow the IEP for accommodations during language assessment (extra time, scribe, AAC, quiet setting). Universal Design still applies to the task design (0001.7, 0003.5). A child who uses a communication device is producing language; count those utterances.
Scenario: kindergarten September
Ms. Brennan gives every child a two-minute object-naming and familiar-rhyme retell as a screener. Four children flag. She does not email “language disability.” She collects a diagnostic play sample, asks families what languages are used at home, and watches the same children at the snack table (authentic). Two need small-group naming and sequence intervention; one is a dual language learner with rich L1; one was new and silent only on day one. Screening found them; interpretation sorted them; instruction changed.
Scenario: Grade 2 formative retell
During dialogic reading, three children cannot order events. Mr. Okonkwo rebuilds tomorrow’s lesson with four picture cards and first/then/last stems. That retell was formative because teaching changed. He does not invent a publisher scaled score to justify the regrouping.
Scenario: Grade 4 “we’ll wait for PSSA”
A child reads grade-level words but cannot explain a science demonstration or infer a character’s feeling. Waiting for the Grade 4 ELA PSSA is the wrong clock: PSSA is a statewide summative of a broad ELA construct in tested grades, not a PreK–4 oral-language diagnostic, and it is too late to teach this unit’s descriptions. 0006.8 wants a language sample/rubric now and a change in talk opportunities.
Exam traps for 0006.8
- Invented cut scores. No fake CELF, PLS, or “language IQ” numbers.
- Screener as diagnosis. Brief and universal versus deep and targeted.
- PSSA as PreK language assessment. Wrong grade and wrong construct for 0006.8.
- English-only verdict on a dual language learner.
- File and ignore. Interpretation without planning, modifying, differentiating, or accommodating.
- Wrong construct. Phoneme deletion and words-correct-per-minute are other objectives.
- Authentic = ungraded chaos. Language samples still map to named skills.
Ask: What language decision is this teacher making? Who needs a closer look, which skill, what to teach tomorrow, whether the class is on track this term, or whether the taught language target was met? Choose the type, read the talk honestly, and change the next lesson.
A kindergarten teacher collects a five-minute play language sample from every child in September. Several children produce mostly pointing and one-word labels. What is the most appropriate 0006.8 next step?
A PreK teacher needs evidence of whether children can retell a simple familiar story after two weeks of dialogic reading. Which plan best matches assessment of language development?
Which statement correctly describes selecting and using language-development assessments on 0006.8?