{"args":{},"checks":[{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":null,"kind":"tool_called","min_lre":null,"note":"","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":"infer.density_test","value":null,"weight":1.0,"where":null},{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":"running","kind":"arg_equals","min_lre":null,"note":"the vote margin is the running variable","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":"infer.density_test","value":"margem","weight":1.0,"where":null},{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":"cutoff","kind":"arg_equals","min_lre":null,"note":"the cutoff of a vote margin is zero","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":"infer.density_test","value":0.0,"weight":1.0,"where":null},{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":null,"kind":"any_of","min_lre":null,"note":"headline figure against the truth","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":null,"value":[{"abs_tol":0.05,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":"p_value","kind":"figure_close","min_lre":null,"note":"","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":"infer.density_test","value":0.0,"weight":1.0,"where":null},{"abs_tol":0.05,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":"p_value","kind":"diagnostic","min_lre":null,"note":"","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":"infer.density_test","value":0.0,"weight":1.0,"where":null}],"weight":1.0,"where":null},{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":null,"kind":"answer_contains","min_lre":null,"note":"the answer must tell the reader the margin was sorted around zero (keywords; negation is not detected)","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":null,"value":["manipul","aglomer","acumul","ac\u00famul","amontoa","bunching","sorting","autossele","auto-sele","salto na densidade","descontinuidade na densidade"],"weight":1.0,"where":null},{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":null,"kind":"answer_backed","min_lre":null,"note":"every number traced to a tool call or the question","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":null,"value":null,"weight":1.0,"where":null},{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":null,"kind":"any_of","min_lre":null,"note":"the answer cites the expected tool's figures","open_interval":false,"rel_tol":null,"required":false,"source_id":null,"table":null,"tool":null,"value":[{"abs_tol":null,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":null,"kind":"answer_cites","min_lre":null,"note":"","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":"infer.density_test","value":null,"weight":1.0,"where":null}],"weight":1.0,"where":null}],"datasets":{"dados":{"columns":null,"description":"Uma linha por partido numa elei\u00e7\u00e3o municipal. margem: margem de votos do partido (positiva quando venceu, negativa quando perdeu; zero \u00e9 o empate); vencedor: 1 se o partido venceu a elei\u00e7\u00e3o; reeleito: \u00edndice de desempenho do partido na elei\u00e7\u00e3o seguinte (maior \u00e9 melhor).","file_sha256":null,"format":null,"frame_sha256":null,"generator":"magi_bench.generators:rdd_sharp","kind":"generator","licence":null,"member":null,"object":null,"params":{"effect":0.23,"manipulation":0.78,"n":3000,"seed":851057454,"slope_left":0.75,"slope_right":0.78},"path":null,"read_kw":{},"records":null,"role":"agent","sha256":null,"transforms":[],"url":null}},"difficulty":"H","id":"tool_selection.rdd_manipulated.test.s1003","level":"T2","meta":{"acceptable_tools":[],"expected_tool":"infer.density_test","family":"rdd_manipulated","features":["cutoff","running_variable","manipulation"],"headline_keys":{"infer.density_test":["p_value"]},"lesson":"When units may have sorted around the cutoff, run infer.density_test first; a density jump means the comparison at the cutoff is not credible and the answer must say so.","params":{"effect":0.23,"manipulation":0.78,"n":3000,"seed":851057454,"slope_left":0.75,"slope_right":0.78},"reference":{"cite":1,"note":"O teste de densidade indica manipula\u00e7\u00e3o da margem de votos perto de zero: a compara\u00e7\u00e3o entre quem venceu e quem perdeu por pouco n\u00e3o \u00e9 confi\u00e1vel.","steps":[{"args":{"cutoff":0.0,"dataset":"dados","running":"margem"},"tool":"infer.density_test","why":"Suspeita de manipula\u00e7\u00e3o na apura\u00e7\u00e3o: teste de densidade da margem em zero."}]},"seed":1003,"tolerance":0.05,"trap_tools":[],"truth_value":0.0,"variant":3},"question":"H\u00e1 suspeita de que partidos que perderiam por pouco conseguiram virar o resultado na apura\u00e7\u00e3o. A compara\u00e7\u00e3o entre quem venceu e quem perdeu por pouco ainda \u00e9 confi\u00e1vel?","requires":[],"runner":null,"sources":[],"split":"test","suite":"tool_selection","tags":["tool_selection","rdd_manipulated","test"],"target":"agent","tool":null,"truth":"DGP generators.rdd_sharp(manipulation=0.78): that share of units just below zero moved just above it; truth = manipulation present, so the density test's p-value must be at most 0.05 (a binomial count test at \u00b10.1 gives p = 3.1e-55)"}