From Data to Diagnosis to Action: A Repeatable Math Growth Cycle
Used state, i-Ready, item-level, domain, proficiency, progress-monitoring, student-work, and pre/post evidence to identify performance patterns, diagnose root causes, target support, and verify whether changes worked—contributing to a pattern in which my classes consistently led their schools in annual mathematics growth across eight tested-grade years.
Verified Professional Experience
Public K–12 education across multiple schools and districts
2017–2026 | Eight tested-grade years
Skills Demonstrated
Data Analysis, Root-Cause Analysis, Learning Analytics, Assessment Analysis, Continuous Improvement, i-Ready, Adaptive Learning, Progress Monitoring, Intervention Design, Differentiation, Evidence-Based Decision Making, Data Synthesis, Outcomes Measurement
Challenge
A single assessment score rarely explains why performance is strong or weak. Across multiple schools, districts, grade levels, and assessment systems, I needed to determine not only where students were struggling, but why—and whether the problem came from content knowledge, prerequisite skills, question interpretation, scoring, timing, technology, instruction, or the assessment itself. The challenge was turning large amounts of performance evidence into precise decisions rather than reacting to one overall score.
Role & Scope
Across eight tested-grade years of mathematics instruction, I regularly analyzed state-assessment results, i-Ready data, item-level performance, standards and domains, proficiency patterns, progress-monitoring evidence, student work, and pre/post assessments. I used that evidence to identify individual and class-level patterns, prioritize needs, form differentiated and guided groups, target intervention, communicate next steps, reassess learning, and adjust support. I also used assessment evidence to question the measure itself when results suggested an item, scoring method, technology issue, or test condition might be obscuring what students actually knew.
Approach & Execution
I followed a recurring evidence cycle: identify the pattern, investigate the likely cause, select the response, and then verify the result. I looked beyond aggregate scores to individual students, commonly missed items, standards and domains, proficiency trends, and changes over time. When the evidence pointed to a genuine learning gap, I adjusted instruction, intervention, grouping, practice, or resources. When the evidence suggested the assessment itself was contributing to the problem, I reconsidered item wording, scoring, timing, technology, or measurement design. I then reassessed and compared the new evidence rather than assuming the first intervention had worked.
Key Decisions & Rationale
One of the most important decisions was not treating every incorrect response as the same kind of problem. A low result could reflect missing content knowledge, an unfinished prerequisite skill, difficulty interpreting the question, a scoring issue, time constraints, technology friction, or a poorly designed assessment task. Distinguishing among those causes mattered because each required a different response. I also resisted treating data as a final judgment; I used it as a signal for where to investigate next, then looked for additional evidence before deciding what to change.
Outcome & Impact
Across eight tested-grade years, my classes consistently led their schools in annual mathematics growth and produced substantial growth across state and i-Ready measures. More importantly, that result came from a repeatable improvement process rather than one isolated cohort or intervention: analyze evidence, diagnose the underlying issue, target the response, reassess, and adjust. Over time, that cycle became a consistent operating method for using data to improve learning outcomes without reducing students to a score.
What This Demonstrates
This work demonstrates my ability to move from raw performance evidence to useful insight and action. It reflects strengths in quantitative and qualitative data analysis, root-cause diagnosis, learning analytics, assessment interpretation, continuous improvement, intervention design, adaptive-learning data use, progress monitoring, evidence synthesis, outcomes measurement, and decision-making under uncertainty. It also shows that I can distinguish between a performance problem and a measurement problem—and verify whether a change actually improved the outcome rather than assuming it did.