{"id":"res-col-discussion-post-cited-claims","audience":"college","category":"research","technique":"grounding","title":"Pull author-page cited claims from two readings for a discussion post","free":false,"task":"For your Sociology 150 discussion post on urban gardens, you want claims from the two assigned readings with citations. You want exactly 3 numbered lines, each a claim under 15 words ending with an author-page citation like \"(Ramos, p. 22)\". Use only the two readings, not the blog quote in your notes, and output nothing else in plain text.","materialLabel":"Excerpts from your two assigned readings and your notes (attached to your prompt automatically)","material":"READING 1 - Ramos, L. (2023). \"Growing Together: Community Gardens in Midwest Cities.\"\np. 21: Of 48 gardens surveyed in four cities, 36 were on formerly vacant lots.\np. 22: Gardeners reported knowing an average of 6 more neighbors by name after one season.\np. 24: Most gardens depended on city water access agreements that had to be renewed every year.\n\nREADING 2 - Okafor, T. & Bell, S. (2024). \"Who Benefits? Garden Access and Neighborhood Change.\"\np. 9: In two of the four neighborhoods studied, home prices rose faster near new gardens than elsewhere in the city.\np. 11: Long-time residents made up a smaller share of garden members than newer residents.\np. 13: The authors note their data cannot show whether gardens caused price changes.\n\n-----\nMY NOTES\n- Saw a blog that said \"gardens cut crime by 40%\" - from GreenCityLife blog, no study linked\n- Post due Thursday 11:59 p.m.; Prof. Hale wants at least one claim from each reading","rules":[{"type":"numberedItems","count":3},{"type":"plainText"}],"mistakes":[{"title":"Using the blog","why":"The crime stat is the most striking line in your notes, and the AI may include it.","fix":"Say: \"Only claims from the two readings.\"","checks":["r0"]},{"title":"Sloppy citations","why":"The AI cites 'Ramos 2023' without pages, or swaps pages between claims.","fix":"Show it: \"End each with (Author, p. N).\"","checks":["r1"]},{"title":"Overstating a finding","why":"The AI turns 'prices rose near gardens' into 'gardens raised prices'.","fix":"Add: \"Don't claim more than the reading says.\"","checks":["r2"]},{"title":"Not saying how many","why":"The AI decides the length itself and usually returns more claims than you can use.","fix":"Be exact: \"exactly 3 numbered claims.\"","checks":["numberedItems"]},{"title":"Getting markdown back","why":"AI defaults to bold labels and headings, which show up as stray symbols in the discussion board.","fix":"Ask for \"plain text, no bold, no headings.\"","checks":["plainText"]}],"promptHints":[{"label":"Asks for page citations","patterns":["p\\.\\s*\\d","page","author.{0,10}page","cite"]},{"label":"Rules out the blog","patterns":["blog","only.{0,30}readings","two readings"]},{"label":"Asks for both readings","patterns":["each reading","both readings","one from each"]},{"label":"Says exactly how many","patterns":["\\b(3|three)\\b.{0,25}(claim|line)","exactly \\d+"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]}],"hint":"A discussion post is only as good as its citations. Tell the AI which sources count, show the citation shape, and keep claims within what the readings say.","rubricLabels":["Every claim comes from Reading 1 or Reading 2, includes at least one claim from each, and none is the blog's 40% crime claim","Each citation names the correct author and page for its claim (for example, vacant lots = Ramos, p. 21; home prices = Okafor & Bell, p. 9)","If the home price finding appears, it does not say gardens caused prices to rise; and nothing besides the 3 numbered lines is output"]}