Data-Based Decision Making in Psychological Practice
This lesson introduces the conceptual framework used throughout the course: student achievement data become useful when they are interpreted in relation to an instructional question, translated into an action, and followed by new evidence about whether the action worked.
Open the Reading ↗
What the Practice Guide Is Actually Asking Educators to Do
The guide begins from a problem that remains highly relevant: schools often have abundant achievement data but considerably less clarity about how those data should influence instruction. The availability of scores is not itself evidence of effective data use. The work begins when educators organize information around a meaningful question, interpret patterns, identify a plausible instructional response, and then gather new evidence to evaluate that response.
The panel frames data use as a way to support decisions such as changing the emphasis of a lesson, modifying assignments, reallocating instructional time, regrouping students, targeting additional instruction, refining methods, or examining broader curricular strengths and weaknesses. In other words, the goal is not simply to describe performance. The goal is to use evidence to improve the next instructional decision.
A psychologically informed interpretation
For psychologists, this distinction is especially important. A low score can identify concern, but it does not automatically identify the cause of the concern, the intervention that should be selected, or whether a disability is present. Interpretation should consider the nature of the measure, the skill domain sampled, timing, instructional exposure, consistency across sources, and the question the team is trying to answer.
How the Guide Treats Evidence
The practice guide does not present every recommendation as if it were supported by equally strong causal research. Instead, it explicitly separates the importance of a recommendation from the strength of evidence establishing that the recommendation causes improved student outcomes.
Requires research with strong causal validity and sufficient generalizability, such as multiple well-designed experiments or a strong multisite trial with no comparable contradictory evidence.
High confidence in causal inferenceReflects evidence that is strong in either internal validity or generalizability but has meaningful limitations in the other dimension.
Useful but qualified causal supportMay reflect descriptive studies, case studies, expert judgment, related theory, or direct evidence that does not meet the threshold for moderate or strong evidence.
Insufficient causal evidenceA low rating does not mean the panel judged the practice unimportant or ineffective. It means the available research did not permit a strong causal claim about effects on achievement.
Importance ≠ evidence ratingWhat Counts as Useful Achievement Data?
The guide focuses primarily on common assessments administered routinely and consistently by a state, district, or school. Examples include statewide accountability tests, benchmark and interim assessments, early-grade reading measures, end-of-course assessments, and common school- or district-developed prompts.
The guide also recognizes classroom-specific evidence such as tests, student work, and projects. The central principle is not that one assessment type is always superior. Different sources answer different questions, and every assessment has limitations related to timing, detail, comparability, technical quality, or instructional proximity.
| Data source | What it may help answer | Important limitation |
|---|---|---|
| Annual statewide assessment | Broad areas of relative strength/weakness; larger-group or schoolwide patterns | May be old by the time instruction is planned and may lack enough detail for immediate instructional decisions |
| Interim / benchmark assessment | Patterns during the year; performance that can often be compared across classrooms | Still may not isolate the exact prerequisite skill or explain why a student is struggling |
| Curriculum-based or progress-monitoring data | Rate of improvement, response to intervention, and change over time | Interpretation depends on measure quality, administration, goal selection, and instructional alignment |
| Classroom work / projects / unit assessments | Error patterns, strategy use, task-specific strengths and weaknesses | May be less standardized or less comparable across settings |
Scope: What the Guide Includes—and What It Does Not
K–12 instructional decision making using student achievement data to improve teaching and learning.
Classroom teachers, school administrators, and district leaders, with responsibilities distributed across classroom, school, and system levels.
Using common assessment and classroom evidence to understand learning, set goals, modify instruction, support student self-evaluation, and build organizational structures for sustained data use.
Using data primarily for management efficiency, teacher rewards, scheduling, staffing, or other administrative operations.
The Five Recommendations as One System
The recommendations are designed to operate together. Recommendations 1 and 2 focus most directly on classroom practice. Recommendations 3 and 4 address the school structures and culture that make effective data use more likely. Recommendation 5 addresses the district-level technological infrastructure needed for sustainable access to useful information.
Collect and prepare data, interpret patterns and develop hypotheses, modify instruction, and evaluate new evidence.
Classroom level · Low evidenceClarify expectations, provide useful feedback, give students tools for learning from feedback, and use students’ analyses to inform instruction.
Classroom level · Low evidenceCreate leadership structures, define important concepts, develop a written plan, and sustain data leadership.
School level · Low evidenceUse facilitation, structured collaboration time, and recurring professional development to strengthen shared data practices.
School level · Low evidenceInclude stakeholders, define user requirements, make intentional build-or-buy decisions, and stage implementation.
District level · Low evidenceThe Psychologist’s Role in Data-Based Decision Making
Psychologists often enter the data-use process when the information is incomplete, conflicting, or being interpreted too quickly. Their contribution is not simply to produce more scores. It is to improve the quality of the inference the team makes from the available evidence.
Questions that strengthen a team’s reasoning
- What is the decision? Clarify what the team is actually trying to decide before collecting more information.
- What does each measure contribute? Identify the construct, timing, and level of detail provided by each source.
- Do the sources converge? Look for consistent patterns across measures, settings, tasks, and time.
- What remains uncertain? Separate what the data support from what the team is merely assuming.
- What instructional hypothesis follows? Translate the pattern into a testable explanation of what may improve learning.
- What change will test that hypothesis? Select a specific instructional modification rather than a vague recommendation.
- What evidence will count as improvement? Define the outcome data and decision rule before evaluating the change.
Evidence Calibration: What Can You Actually Conclude?
Select every statement that correctly evaluates this claim.
Match the Data Source to the Decision
For each decision, select the best starting source. Several sources may eventually be useful, but choose the one that most directly addresses the immediate decision.
A. What specific error pattern is interfering with one student’s multi-step computation?
Choose the source that gives the most instructionally specific information.
B. Which grade-level mathematics domains appear weak across several classrooms midway through the year?
Choose the source that provides comparable, current information across classrooms.
C. Is a student responding adequately to a new intervention over the next six weeks?
Choose the source best suited to evaluating change over time.
Build the Instructional Inquiry Cycle
Click the steps in the most defensible order. The task is deliberately difficult: each step is plausible, but the cycle requires a coherent chain from evidence to hypothesis to action to evaluation.
Branching Consultation Case: Avoid the Premature Conclusion
Stage 1: What should you do first?
Stage 2: New information
Stage 3: Make the plan testable
Analyze the Instructional Improvement Cycle
Apply the concepts from the reading to a complex case. Your response will receive individualized feedback based on the quality of your reasoning, attention to baseline data, intervention fit, measurable goals, decision rules, and ongoing monitoring.
Key Principles of Data-Based Instructional Decision Making
The practice guide treats data-based decision making as a structured process rather than a score-review exercise. Educators should use evidence from multiple sources, interpret patterns in relation to a specific instructional question, develop a plausible hypothesis, modify instruction, and then collect new evidence to determine whether the modification improved learning.
The guide also models an important professional habit: being precise about evidence. All five recommendations are presented as useful priorities, but all five were rated as supported by a low level of evidence. That distinction matters because good professional judgment requires both acting on the best available information and accurately representing what the research can—and cannot—establish.