4.3 Using Assessment to Drive Instruction
Key Takeaways
- PECT 0002.3 scores the full loop: design an assessment that matches the standard and purpose, implement it fairly, interpret the evidence, then change instruction — not merely file a score.
- Use results to name strengths, needs, and preferred modes of learning so you can differentiate or modify how children access the same standard, not so you can lower the standard for children who struggle.
- If most of the class misses the target, treat that as evidence that instruction was not effective and reteach; do not treat a whole-class miss as a row of individual child deficits.
- PECT 0002.4 requires multiple indicators — including standardized assessments and progress monitoring — before you identify individual need; one winter percentile is not a triangulation.
- Progress monitoring asks whether this instruction is working for this child over time; do not invent RTI/MTSS cut scores the exam does not publish — defend the logic of converging evidence.
4.3 Using Assessment to Drive Instruction
PECT lens: Objective 0002.3 asks you to design, implement, and interpret assessments to identify strengths, needs, and preferred modes of learning; to differentiate and modify instruction; to track progress; and to monitor whether instruction is working. Objective 0002.4 adds that you must use multiple indicators, including standardized assessments and progress monitoring, to identify individual need and support every student. Together they are one professional habit: evidence in, teaching changed, evidence checked again.
A score that never changes a lesson is decoration. Module 1 items in this cluster almost always include a next instructional move. If you stop at "the child scored low," you have not finished the objective.
Design, implement, interpret — then act
Design. Start from the Pennsylvania standard and the decision (Section 4.1). A PreK design might be a 10-minute observation protocol for counting in the snack routine. A Grade 3 design might be a five-item CRT plus one authentic explanation item on equivalent fractions, with a rubric. Design also means planning access: large print, extra time specified in an IEP, oral directions for a child who cannot yet decode the math word problem (ethics in 4.4; SWD/ELL detail in the next chapter). If the tool cannot show the skill you named, no amount of later "data talk" will save it.
Implement. Give the assessment the way you designed it. For a running record, sit beside the child, code live, and avoid teaching through the assessment until you have a sample. For a whole-class exit ticket, protect enough quiet that the product is the child's. For a standardized or benchmark measure, follow administration rules so scores remain comparable. Implementation errors — coaching answers, skipping items, testing a hungry child at dismissal — are your validity problem, not the child's deficit.
Interpret. Ask three questions of every set of results:
- What can this child (or this class) already do? Strengths are instructional starting points, not consolation prizes.
- What is the next need relative to the standard? Name a skill, not a label ("needs short-vowel blending," not "low kid").
- How does this child best show learning? That is preferred mode of learning — not a learning-styles myth that you should only teach "kinesthetic children" with beanbags. It means the child currently demonstrates mathematical thinking with cubes more clearly than with oral-only number talks, or retells a story with drawings better than with a blank page. You keep the standard and vary the access and expression, then you teach toward more independent modes.
Act. Differentiation and modification are the payoff:
- Differentiate when children need different paths to the same standard: small-group reteach, extra guided practice, enrichment that deepens the same idea, language supports, manipulatives, or a different text at the instructional level.
- Modify when the team has changed the expectation itself (often via IEP). Do not casually "modify" a Grade 2 child out of the PA Core standard because a single quiz went poorly; that is lowering the bar, not using assessment.
Track progress with repeated measures of the same skill (CBM probes, dated writing samples, a phonics inventory given every few weeks). One Friday quiz cannot show a trend.
When the data indict the teaching
0002.3 explicitly includes monitoring the effectiveness of instruction. If 22 of 24 first graders miss silent-e words after a week of worksheets, the professional interpretation is not "this generation cannot read." It is "this instruction did not work." Reteach with blending in connected text, more modeling, and a different practice structure; then assess again. PECT will offer a distractor that blames children or moves on to stay "on pace." The correct option changes teaching.
The same logic applies to a PreK unit: if no child used the target spatial words after a week of unplanned block play, the environment and the adult language were the intervention that failed. Add modeled language, a simple observation checklist, and a reason to use under and beside; then watch again.
Multiple indicators and triangulation (0002.4)
Multiple indicators means you do not identify a child's "need" from one number. Standardized assessments (a district benchmark, a published screener, the PSSA in tested grades) are one source. They are efficient and comparable, and they miss context: home language, a bad day, a skill the child shows in play but not on a timer. Progress monitoring is a second source: brief, repeated probes that answer "is this instruction working for this child?" Classroom evidence is a third: observation, running records, work samples, conferences, family information.
Triangulation is the logic of laying those sources next to each other:
| Question | Weak (single indicator) | Stronger (converging evidence) |
|---|---|---|
| Does this kindergartner need more phonological-awareness instruction? | One 4-minute screener on a noisy Monday | Screener plus two weeks of anecdotal notes at circle plus a diagnostic sound inventory plus family report of home language |
| Is Grade 1 core reading instruction working? | One child's running record | Grade-level benchmark, several CBM probes, and classroom decoding work across the class |
| Should we suspect a need for further evaluation? | One low NRT percentile | Persistent need across classroom work, progress-monitoring slopes that stay flat after targeted teaching, and standardized flags — then the team process, not a solo teacher label |
Do not invent RTI/MTSS cut scores, percentile gates, or "after 12 weeks in Tier 2" numbers the PECT outline does not publish. Districts set procedures; Pennsylvania does not ask you to memorize a national triangle's unpublished numbers. What the exam does reward is the logic: screen → teach → progress-monitor → if the line does not move, change instruction or seek a deeper look → never leap from one score to a disability label.
Progress monitoring is not the same as a once-per-term benchmark. If you teach a six-week decoding cycle, you need frequent probes along the way so you can abandon a failing approach in week two instead of discovering failure at the next seasonal benchmark. Conversely, a flat CBM line is a prompt to triangulate, not an automatic special-education referral.
Preferred modes of learning fit here as another indicator, not as a separate science. A child who builds a solid argument with cubes but freezes at the rug is showing a strength you can use and an oral-language need you still teach. Removing all talk so the child "stays comfortable" abandons the speaking standard. Ignoring the cubes because "real math is mental" abandons the child's current access point. Design the next week with both: cubes as a scaffold, structured language as the stretch.
Scenario: Grade 1 — strengths, needs, mode, then grouping
After a phonics CRT, running records, and two CBM probes, Ms. Nguyen sees that Priya blends CVC words in writing and with magnetic letters but stalls on timed oral lists. Strength: encoding and hands-on blending. Need: automaticity in connected text. Preferred mode right now: letters she can move. Instructional move: keep magnetic letters as a scaffold, add daily blending in short decodable sentences, progress-monitor oral reading weekly, and do not assign a "visual learner only" packet that never requires sounding out.
Scenario: Grade 3 — whole-class miss
Most of the class fails a CRT on area. Mr. Ortiz does not send a pile of worksheets home as punishment. He interprets the assessment as instructional evidence: the lesson stayed on formula without tiling. He reteaches with arrays, connects tiling to the formula, uses a short formative check the next day, and only then re-administers a brief CRT. He monitored instruction, not just children.
Scenario: PreK — triangulation before "delay"
A screening flag, one quiet morning, and a missed art project are not enough to tell a family their child is delayed. The teacher adds anecdotal notes across two weeks, a second structured observation, input from the family about home language and sleep, and a look at health/hearing concerns. Multiple indicators either dissolve the worry or justify a request for further evaluation through the proper team. One indicator does not.
Exam traps for 0002.3 and 0002.4
- Assess, then keep the same lesson. Interpretation without action fails 0002.3.
- One score = identification of need. Fails 0002.4.
- Invented RTI cut scores or week-counts as if they were PECT facts.
- Preferred mode = drop the standard. Use the mode as access; still teach the missing skill.
- Whole-class failure blamed on children instead of instruction.
- Standardized tests versus classroom evidence as enemies. 0002.4 wants both, plus progress monitoring.
When you read a stem, find the standard, the evidence you already have, the evidence you still need, and the teaching move. If an option skips straight from one percentile to retention, referral, or giving up on the standard, it is almost certainly wrong.
A Grade 1 teacher is concerned about a child's reading. Which approach best matches PECT's emphasis on multiple indicators?
After three days of a science investigation, most of the class cannot explain their evidence. The teacher's best use of that assessment information is to:
Kindergarten screening and two observations show that a child explains math thinking clearly with cubes but freezes during oral-only number talks. How should the teacher use that information?