Supporting students as they moved from memorizing formulas to reasoning with uncertainty, distributions, estimators, confidence intervals, and hypothesis tests.
Overview
As a Teaching Assistant for Probability and Statistics at Stony Brook University, I helped students connect lecture theory with the practical judgment needed to solve unfamiliar problems. The work combined technical accuracy, patient communication, and consistent classroom operations.
Many students could identify a formula after seeing a worked example, but struggled to choose a method when the wording changed. My goal was to make the decision process visible: what is random, what is conditioned on, which assumptions matter, and what the final number means.
I prepared recitation examples, answered conceptual and computational questions, supported office hours, graded assignments with transparent rubrics, clarified common mistakes, and helped students build repeatable approaches for probability distributions, sampling, estimation, and hypothesis testing.
Responsibilities
Beyond answering questions, the role required diagnosing where confusion started and designing explanations that made the next step feel reachable.
Built step-by-step problem walkthroughs for conditional probability, Bayes’ theorem, random variables, expectation, variance, normal approximation, confidence intervals, and hypothesis tests.
Worked with students individually and in small groups, often translating broad questions like “I do not know where to start” into a concrete plan: identify givens, choose a model, compute, then interpret.
Graded assignments and exams with attention to partial reasoning, notation, assumptions, and final interpretation. Feedback focused on the exact point where a solution diverged, not only the final answer.
Helped maintain a steady feedback loop between students and course staff by surfacing recurring misconceptions, clarifying expectations, and keeping explanations consistent across sections.
Teaching approach
Before calculating, I encouraged students to classify the problem: counting, conditional probability, distribution behavior, estimator behavior, interval estimation, or a test of a claim. This reduced formula hunting and built decision-making confidence.
Symbols were always paired with plain-language interpretation: sample mean as a statistic, population mean as a parameter, p-value as a conditional probability under the null model.
Common mistakes became mini-lessons: confusing independence with mutual exclusivity, mixing standard deviation and standard error, or treating a confidence interval as a probability statement about a fixed parameter.
Every solution ended by returning to the original context. A calculation was not complete until the student could say what the result implies and what assumptions made it valid.
Project showcase
Reusable structures helped students move from problem text to model selection. Each frame highlighted givens, assumptions, target probability, computation method, and interpretation.
Feedback separated setup, method, algebra, computation, notation, and conclusion so students could see which part of the reasoning needed repair.
Questions were grouped into conceptual, notation, computation, and exam-preparation needs, making limited support time more useful for each student.
Visual maps connected probability topics to inference topics, helping students understand how early course material reappeared in later units.
Skills developed
Strengthened fluency with probability models, sampling distributions, estimation, confidence intervals, and hypothesis testing through repeated explanation and assessment.
Learned to adapt the same concept for different levels of preparation, from intuitive examples to formal notation.
Practiced writing feedback that is direct, fair, and actionable, especially when students made partially correct attempts.
Built confidence facilitating review sessions, managing uncertainty, and helping students stay engaged with challenging quantitative material.
Developed habits around preparation, deadline awareness, consistency, and coordination with instructors and course staff.
Reflection
Teaching Probability and Statistics made me more precise as a communicator and more patient as a problem solver. It reinforced that technical expertise is not only knowing the answer; it is being able to guide someone else through the reasoning that makes the answer trustworthy.