{"id":"study-regression-office-hours-questions","audience":"college","category":"studying","technique":"context","track":"study","title":"Turn your confused notes into office hours questions","free":false,"task":"You have 10 minutes at your stats professor's office hours, so you want to arrive with sharp questions. Get exactly 3 numbered questions, each starting with the slide number in parentheses like '(Slide 7)' and under 20 words, plain text, with no answers. Ask only about the parts you marked ??? and haven't figured out yet.","materialLabel":"Your lecture notes (attached to your prompt automatically)","material":"STAT 2100 - Lecture 11 - Simple linear regression (Prof. Nakamura)\n\nSlide 4: Regression line: predicted y = b0 + b1*x. b1 = slope.\nSlide 6: Slope interpretation: \"for each 1-unit increase in x, predicted y changes by b1.\"\n  ??? In the hours-studied example, slope = 4.2. Does that mean every student gains exactly 4.2 points per hour?\nSlide 9: Residual = actual y - predicted y.\n  ??? Why is a negative residual not \"bad\"? Prof said it's just below the line.\nSlide 11: R-squared = proportion of variation in y explained by x.\n  ??? What does 0.64 mean in words?\n  (nvm, figured it out from the textbook: 64% of the variation in exam scores is explained by hours studied. Crossed off.)\nSlide 13: Extrapolation = predicting outside the range of the data. \"Dangerous.\"\n  ??? Our data goes from 0 to 12 hours. Is predicting for 13 hours extrapolation, or is that close enough?\n\nThings I'm fine on: scatterplots, correlation r, the residual plot shape stuff.\n\nHW 6 due Monday. Office hours Wed 1-3, room 214 Hartley Hall. She said \"come with specific questions, not 'I don't get it.'\"","rules":[{"type":"numberedItems","count":3},{"type":"maxWords","max":75},{"type":"plainText"}],"mistakes":[{"title":"Asking what to ask","why":"The AI suggests generic questions about regression, not the exact spots you're stuck on.","fix":"Say: \"Only about the parts I marked ???.\"","checks":["r0"]},{"title":"Missing what you resolved","why":"The AI sees four ??? marks and asks about all of them, including the one you crossed off.","fix":"Add: \"Skip the R-squared one; I figured it out.\"","checks":["r1"]},{"title":"Getting answers instead","why":"The AI's instinct is to answer your questions, which turns into a long lecture.","fix":"Add: \"Just the questions, no answers.\"","checks":["r2"]},{"title":"Not saying how many questions","why":"The AI picks its own number, and it usually picks more questions than you need.","fix":"Be exact: \"Exactly 3 numbered questions.\"","checks":["numberedItems"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting. Ask for a total and it often lands 20% over.","fix":"Cap each question: \"each under 18 words.\"","checks":["maxWords"]},{"title":"Forgetting where it will be read","why":"AI formats for a screen, with bold and headings that show up as stray symbols in your notes.","fix":"Say: \"Plain text, no bold, no headings.\"","checks":["plainText"]}],"promptHints":[{"label":"Points to your ??? marks","patterns":["\\?\\?\\?","marked","confus"]},{"label":"Skips what you solved","patterns":["skip","figured","r-?squared","crossed"]},{"label":"Asks for slide numbers","patterns":["slide"]},{"label":"Says no answers","patterns":["no answers?","(don't|do not).{0,15}answer","just.{0,10}questions","only.{0,10}questions"]},{"label":"Says how many questions","patterns":["\\b(3|three)\\b.{0,25}question"]}],"hint":"The AI doesn't know which confusions are still open. Point it to your ??? marks, say which one you already solved, and ask for questions only, tagged by slide.","rubricLabels":["Each question starts with its slide number in parentheses and is about one of the 3 unresolved ??? marks (Slides 6, 9 and 13)","No question asks about R-squared or Slide 11, which the student already figured out, or about topics marked as fine","Gives only the questions, with no answers, explanations or extra tips before or after them"]}