A gentle onramp into statistics on the way to more advanced quant training.
A foundational, UX-grounded introduction to the statistical concepts under quantitative UX research, built as warm-up prep for courses like Statistical Methods and Survey Methodology (not required for either, just helps you absorb faster).
Course launches soon. Get notified when it does.
Lifetime access to lessons and materials
Revisit every concept, example, and resource whenever you need a refresher.
First access when enrollment opens
Waitlist members get notified, and get first pick of seats, before this goes public.
Certificate of completion
A shareable credential when you finish.
Build the foundation you need to gain real quantitative skills.
It starts with a simple question. A stakeholder challenges the validity of an insight, a recommendation, a design decision, and goes straight for the numbers: does this observation hold for all our users? If you're doing product work, you need to be able to speak this language. Quantitative fluency used to be a specialist track. It isn't anymore, and the moment you start down that path, you run into statistics; there's no version of "doing quant work" that skips it.
Learning basic stats on your own can be slow and painful. There's a small set of concepts, like a normal curve, standard error, and p-value, that aren't exactly intuitive. Most quant training assumes you already have these, and rarely explains them in plain terms. But once those are in place, everything that comes after gets easier to follow.
This course provides UX and product professionals with the foundational concepts needed in order to build the kind of quantitative skills taught in our Statistical Methods, Survey Methodology, and Conjoint courses. It works as a more gentle onramp to quantitative thinking that steers clear of the dull stats-textbook approach and teaches through real, relevant UX examples. You'll learn essential concepts like what a sample can and can't tell you about a population, the logic of distributions, and which data type calls for which kind of stat.
Taught by HarmoniJoie Noel, PhD, a senior survey methodologist and current UXR Institute instructor, the approach is deliberately non-scary and concept-first: building working vocabulary and intuition, not deriving formulas, so you can walk into Statistical Methods, Survey Methodology, or whatever quant course comes next already warmed up.
-
Explain what a p-value, a confidence interval, and an effect size actually tell you, and how they work together.
-
Distinguish standard deviation from standard error.
-
Match a data type, categorical (binary, nominal, ordinal) or continuous, to the kind of statistical analysis to run (chi-square, t-test, ANOVA, correlation).
-
Recognize when a research question calls for counts instead of averages, and explain why.
-
Explain what a normal curve distribution is and how it relates to p-values and statistical significance testing.
A solid foundation for further quantitative study, even (and especially) starting from zero.
"I want to learn to analyze analytics data, but haven't done any stats or math in a long time."
Whether it's been years or you're not sure you ever really had it, this course doesn't assume a math background you're supposed to already have. It starts at zero on purpose, so picking up analytics work is a matter of building the foundation now, not feeling behind for not already having it.
"I want to get into more quantitative UX research, but I come from a 100% qualitative background."
You don't need any prior quant background to make that move; that's the whole point of a warm-up. This course is the foundation under quantitative UX research, taught from scratch, so the switch is possible without pretending you already have skills you don't.
"I'm being asked to look at quant data more often, and I'm not confident about the claims I can make about it."
This is the exact gap the course targets, whether you're a researcher, a PM, or a designer: not turning you into a statistician, but giving you enough of a real foundation that you're not guessing at what a p-value or a confidence interval even means when the data lands in front of you.
Is this course for you?
Good fit if…
- You want to take Statistical Methods or Survey Methodology but keep bouncing off unfamiliar terms before you get to the part you actually want to learn.
- You're a PM, designer, or researcher who's planning to move into quantitative UX research and wants the basic vocabulary in place before the first real quant course, not during it.
- You learned stats once (in school, or a past job) and it didn't stick, and you want the fundamentals rebuilt before you rely on them again.
Not a fit if…
- You already know what a p-value, a confidence interval, and an effect size each measure, and can match a data type to the right statistical analysis on sight. Skip ahead to Statistical Methods.
- You're looking to build statistical models or run analyses in R, Python, or SPSS. This course stays at the concept level; it doesn't teach a tool.
- You're looking for study design or sampling strategy (how to write a survey, how to recruit a sample). That's Survey Methodology; this course is the vocabulary underneath it, not the method itself.
You'll need
- Comfortable with basic arithmetic and reading a simple chart or table. No formal math or statistics background required.
- Some exposure to research findings that include numbers (usage metrics, survey results, A/B test readouts) you didn't fully follow at the time.
- Genuine curiosity about what's underneath a quant finding, not just how to report one.
Helpful, not required
- You're currently planning to take, or already enrolled in, a follow-up quant course.
- You've encountered terms like "p-value" or "confidence interval" in a past readout and wanted to actually understand them instead of taking them on faith.
Five modules. Every basic concept you need before the next course.
Why we're doing this at all
Explain why a product decision calls for a quantitative question in the first place, and recognize the basic data type behind it.
What this covers
- Why we ask quant questions in a product world instead of relying on qualitative judgment alone
- Data types: categorical (binary, nominal, ordinal) versus continuous
- Measures of central tendency: mean, median, and mode, and when each one is the right one to report
Counts vs. averages
Recognize when a question is really asking for counts and proportions instead of an average, and explain why.
What this covers
- Why some questions are answered by calculating counts or proportions rather than averages
- Connecting the data type from Module 01 to the right summary statistic
Sampling and probability
Reason about what a sample can and can't tell you about the population behind it.
What this covers
- The central limit theorem
- Repeated sampling and the basics of probability theory
- Randomization: what it protects against and why it matters
- Distributions and the normal curve
Spread and uncertainty
Read a spread-of-data statistic and a confidence interval for what they actually claim, not more than that.
What this covers
- Standard deviation versus standard error: what each one measures and when you'd report one over the other
- Confidence intervals, at a high level: what "95% confident" actually means
- Absolute value, and where it shows up in these calculations
Basic vocabulary for testing a claim
Recognize what a p-value, an effect size, and a Type I or Type II error are pointing at, at a basic conceptual level, without yet running a single test yourself.
What this covers
- P-values and what they do (and don't) tell you
- One-tailed versus two-tailed tests
- Effect size, at a high level: why "significant" and "meaningful" aren't the same thing
- Type I and Type II errors: the two ways a test can mislead you
- Parametric versus non-parametric statistics, and how to recognize which family a method belongs to
- Bringing it together: what data you'd need to answer different kinds of quant questions, and which statistics fit which question, the bridge into Statistical Methods and Survey Methodology
HarmoniJoie Noel, PhD
Instructor, UXR Institute · Senior Survey Methodologist
HarmoniJoie has spent fifteen years making sure survey data means what people think it means, work that starts with exactly the fundamentals this course covers: sampling, distributions, standard error, and what a result can and can't claim to prove. As a senior survey methodologist, she's built and tested surveys for some of the country's highest-stakes health research, including the CMS Health Insurance Marketplace survey and NHANES for the CDC, after senior methodology roles at RTI International, Booz Allen Hamilton, and the American Institutes for Research, and three years as a behavioral scientist at the CDC's National Center for Health Statistics. She holds a PhD in sociology and survey research methodology from the University of Nebraska-Lincoln, and also teaches UXR Institute's Survey Methodology for Product Impact course, one of the courses this warm-up prepares you for.
What is this course, and what will I learn?
This is a basic statistics primer built specifically for UX researchers, PMs, and designers, not a general stats 101 class and not a substitute for Statistical Methods. You'll get the vocabulary and intuition behind sampling and probability, distributions, standard deviation versus standard error, confidence intervals, p-values, and the difference between parametric and non-parametric approaches, taught with UX and product examples instead of textbook ones. The goal is recognizing and understanding these terms at a basic level, not running the tests yourself; that's what Statistical Methods teaches next.
Do I need to know how to code or use statistical software?
No. This course stays at the concept level: what each idea means, when it applies, and how to read it in someone else's findings. There's no R, Python, SPSS, or formula-heavy computation here. If you eventually want to run your own analyses, that's what Statistical Methods and Survey Methodology build toward.
Is this too basic for me if I already do some quantitative work?
If you can already explain what a p-value and a confidence interval measure, and you know on sight whether a method is parametric or non-parametric, this course won't teach you much new; go straight to Statistical Methods. But a lot of researchers who "do some quant work" have gaps in exactly these fundamentals; if you've ever reported a p-value without being fully sure what it claims, this is squarely for you.
How is this different from your Statistical Methods course?
Statistical Methods teaches you to actually run and interpret tests of statistical significance and correlation. This course stays one level below that: the basic vocabulary and concepts those tests are built on, what a distribution is, what standard error means, why a test might be one-tailed instead of two-tailed, at a purely conceptual level. You won't run a single test in this course. Statistical Methods moves much faster once those fundamentals aren't new, which is the whole point of taking this one first.
Should I take this before Survey Methodology?
It's not required, but it helps. Survey Methodology assumes some comfort with data types, sampling logic, and basic statistical reasoning as it teaches you to design and analyze surveys. If those ideas are shaky, this warm-up closes that gap first so you can focus on survey design itself rather than relearning statistics mid-course.
I've tried to learn statistics before and it never stuck. Will this be different?
Probably, if what didn't stick before was formula-first instruction disconnected from your actual work. A lot of people freeze up around statistics because they expect it to produce one clean right answer, and get discouraged when it doesn't work that way. This course is built the opposite direction: concept first, UX example immediately after, so the ideas have something real to attach to instead of sitting in the abstract.
Is this useful for PMs and designers, or only researchers?
All three, as long as you're headed toward the same next step: an eventual quant course like Statistical Methods, Survey Methodology, or Conjoint. This course isn't building a workplace skill on its own; it's building the vocabulary those courses assume you already have, whether you'll eventually be the one running the analysis or the one making sense of it.
What will I be able to do after completing this course?
Nothing you can apply on its own; that's not what this course is for. What you'll have is the vocabulary and basic conceptual grounding, what a p-value, a confidence interval, and a standard error each mean, which data type calls for which kind of stat, that Statistical Methods, Survey Methodology, and future quant courses assume you already have. The value of this course is entirely in what it prepares you for next.
When does this course launch, and how do I find out?
This course doesn't have confirmed dates yet. Join the waitlist above and you'll be the first to know when enrollment opens, with first pick of seats before it's announced publicly.
Does this course offer a certificate?
Yes, a certificate of completion when you finish.
What's the real time commitment?
The exact format and schedule aren't finalized yet. Join the waitlist above and we'll share the time commitment details as soon as they're set, before this goes live for everyone else.
Am I ever going to have to pick the "right" statistical test on my own after this?
Not in this course. You'll leave able to recognize what family a test belongs to (parametric vs. non-parametric) and what kind of question it's suited for, which is the exact uncertainty that trips people up most. Actually selecting and running specific tests is what Statistical Methods teaches next.
Loading the question form…
