Controlled usability testing borrows its logic from experimental science. Before you can say that one interface is "better" than another, you need a plan that specifies what you are changing, what you are measuring, and what you are holding constant. That plan is your experimental design, and it determines whether the numbers and observations you collect actually mean anything.
This topic introduces the vocabulary and decisions behind that plan: independent variables (what you manipulate, such as two versions of a navigation menu), dependent variables (what you measure, such as time-on-task, error counts, or completion rates), and the control of everything else so that differences in results can be traced back to the design and not to noise. You will also look at internal validity (did my manipulation really cause the observed effect?) versus external validity (will this hold up outside my lab?), the difference between formative studies that shape a design in progress and summative studies that evaluate a finished one, and the ethical obligations you owe participants: informed consent, the right to stop, confidentiality, and never making a participant feel that they are the thing being tested.
As you watch the videos below, keep a running list of three things: what was manipulated, what was measured, and what could have gone wrong (a confound, a biased task order, a leading question). Notice also where the videos emphasize structured quantitative comparison versus open-ended qualitative insight. Both are legitimate; the choice depends on the question you are asking and on where you are in the design cycle.
A focused tutorial on the distinction between independent and dependent variables, the foundational vocabulary of any experiment. Expect clear definitions, worked examples of identifying which variable is manipulated and which is measured, and likely discussion of how to phrase a hypothesis around them. The test-prep framing keeps the explanation concrete and example-driven.
This maps directly onto the Independent and Dependent Variables subtopic and gives you the precise language you will need when you write up a usability study plan. Mometrix Academy produces instructional material for exam preparation, so the treatment is deliberately unambiguous and repeatable.
A short roadmap from Nielsen Norman Group laying out six steps for planning a systematic program of A/B tests rather than one-off comparisons. The description points to sequencing decisions, defining what you will vary, and optimizing a design over the long term. A/B testing is the most common applied form of controlled comparison in UX work.
It shows experimental design in its real industry form: one variable changed at a time, an outcome metric chosen in advance, and results used to decide. NNgroup is a leading authority in usability research, and the video fits the summative, measurement-oriented side of the formative-versus-summative distinction.
A longer introduction to qualitative user research, covering how it differs from quantitative approaches and how to get started conducting it. Based on the title and description, it is likely to address planning research questions, recruiting and talking with participants, and gathering the kind of observational and interview data that explains why users behave as they do. It is framed for people new to UX research.
Qualitative methods supply the formative, exploratory half of the study-type decision and the think-aloud style observations that complement your numbers. CareerFoundry produces structured introductory UX training, so this gives useful context on when a controlled experiment is the wrong tool and an open inquiry is the right one.
Together these videos prepare you to run and interpret a controlled session rather than just attend one. The variables tutorial and the A/B testing roadmap give you the design discipline behind MO3: if you know which variable you manipulated and which you measured, you can recognize when an uncontrolled factor such as task order, facilitator prompting, or a participant's prior experience has confounded an observation. The A/B roadmap also reinforces MO2 by treating metrics as things you decide on before testing, which is how completion rates, error counts, and time-on-task become comparable across participants. The qualitative research video supports MO1 and the qualitative half of MO2, modeling the neutral, non-leading way you should guide participants through task scenarios and capture think-aloud commentary without steering their behavior.
These videos were chosen from YouTube on October 3, 2026. Each has captions; the player turns them on. Videos on YouTube can be changed or removed — if one no longer plays, tell your instructor. If a player does not appear, use its Watch on YouTube link.