{"id":"study-which-stats-test-classify","audience":"college","category":"studying","technique":"examples","track":"study","title":"Practice choosing the right statistical test","free":false,"task":"Your Research Methods final asks which test fits each study, and you freeze on that. Get exactly 5 numbered study descriptions, then an 'Answers' line labeling each exactly t-test, chi-square or correlation, under 160 words total, plain text. Use each label at least once and follow your instructor's decision rule, including her correction about the chi-square.","materialLabel":"Your Research Methods notes (attached to your prompt automatically)","material":"PSYC 2010 - Research Methods - Dr. Okafor - Choosing a test\n\nDECISION RULE (Dr. Okafor):\n1. Comparing the MEANS of two groups on a numeric outcome? -> t-test\n   (ex: do students who sleep 8+ hours score higher on a quiz than those who sleep less?)\n2. Two CATEGORICAL variables, counting how many fall in each category? -> chi-square\n   I wrote: \"chi-square compares two averages.\"\n   ** Dr. O correction: NO. Chi-square is for COUNTS in categories (like yes/no, major, class year). If you're comparing averages, it's a t-test. **\n3. Two NUMERIC variables, asking if they move together? -> correlation\n   (ex: hours of TV per week and GPA)\n\nOnly these 3 tests are on the final. ANOVA and regression are in the next course.\n\nFinal format: \"Read the study. Name the test.\" 10 items, no calculations.\n\nMy problem: I always pick t-test when it should be chi-square.\n\nFinal exam: Dec 10. Review sheet posted Monday.","rules":[{"type":"questionCount","count":5},{"type":"maxWords","max":160},{"type":"plainText"}],"mistakes":[{"title":"Not saying how many studies","why":"The AI picks its own number, and it usually picks more studies than you need.","fix":"Be exact: \"Exactly 5 numbered studies.\"","checks":["questionCount"]},{"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 item: \"each study under 25 words.\"","checks":["maxWords"]},{"title":"Not limiting the tests","why":"The AI includes ANOVA and regression, which aren't on your final.","fix":"Say: \"Only t-test, chi-square or correlation.\"","checks":["r0"]},{"title":"Missing the correction","why":"Your wrong chi-square note is still in the material.","fix":"Add: \"Chi-square is for counts in categories, per Dr. Okafor.\"","checks":["r1"]},{"title":"Answers under each item","why":"Seeing the answer right away means you never test yourself.","fix":"Add: \"Put all answers at the end under 'Answers'.\"","checks":["r2","plainText"]}],"promptHints":[{"label":"Says how many studies","patterns":["\\b(5|five)\\b.{0,25}studie"]},{"label":"Gives exact labels","patterns":["t-?test.{0,20}chi","label"]},{"label":"Uses the correction","patterns":["correct","counts","categor"]},{"label":"Puts answers at the end","patterns":["answers?.{0,30}(end|bottom|after|separate)","(end|bottom).{0,20}answers?"]},{"label":"Sets a length limit","patterns":["\\d+\\s*words","word (limit|count)","under \\d+"]}],"hint":"To stop freezing, you need practice that looks like the final. Give the AI the three labels, the decision rule with its correction, and where answers go.","rubricLabels":["Each answer uses exactly one label, t-test, chi-square or correlation, each label is used at least once, and no other tests (ANOVA, regression) appear","Every answer is correct by the decision rule: two-group mean comparisons are t-tests, counts across categories are chi-square, two numeric variables moving together are correlation","All answers appear together after the last study under an 'Answers' line, with no explanations or calculations"]}