UX Research by the Numbers: Task Success, Time on Task, Questionnaire Scores, Problems Found per Tester, Card Sorting and Target-Size Laws

What you'll build

What you will be able to do

Each lab ends with a concrete result that you build yourself and the checker confirms, cell by cell.

• Turn twelve test sessions into a success rate per task, add an adjusted 95 % range with the fixed constant z = 1.96 and give every task a verdict against the 70 % target: Pass, Fail or Unclear.
• Summarise twelve booking times: you keep only the completed attempts, compare the median with a geometric mean built from an LN helper column and flag the sessions that were unusually slow.
• Measure how directly ten testers changed their delivery address: path efficiency against the shortest path, a lostness score from pages visited, a lost flag and the errors per tester for new and experienced users.
• Score a fictional ten-item usability questionnaire for eight testers: convert positive and negative items into contributions, turn each tester's total into a 0–100 score, flag straight-lined answers and compare new with experienced users.
• Estimate from six sessions how likely one tester is to hit a problem, draw the share of problems that n testers find, work out how many testers reach a 90 % target and chart the discovery curve.
• Turn eight usability problems into a priority score, rank them, sort them into effort-vs-impact quadrants and name the top issue for the next sprint.
• Count how often eight participants put two menu cards into the same group, turn the counts into agreement percentages, and score a tree test for success and directness.
• Predict how long it takes to hit five buttons before and after a redesign with a target-size law, estimate the menu choice time for 11 vs 5 items with a choice-time law and present the before/after result in a chart.