{"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.sample_size","value":null,"weight":1.0,"where":null},{"abs_tol":42.02969224576248,"bind":"any","column":null,"failure_class":null,"field":"value","interval":false,"key":"n_required","kind":"figure_close","min_lre":null,"note":"headline figure against the truth","open_interval":false,"rel_tol":null,"required":true,"source_id":null,"table":null,"tool":"infer.sample_size","value":4202.969224576248,"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.sample_size","value":null,"weight":1.0,"where":null}],"weight":1.0,"where":null}],"datasets":{},"difficulty":"E","id":"tool_selection.power.test.s1000","level":"T2","meta":{"acceptable_tools":[],"expected_tool":"infer.sample_size","family":"power","features":["design","power_analysis"],"headline_keys":{"infer.sample_size":["n_required"]},"lesson":"Planning questions go to the design tools with the question's inputs: infer.sample_size for how many units, infer.mde for the detectable effect, with icc and cluster_size when whole clusters are assigned.","params":{"effect":40,"effect_sd":0.2,"icc":0.2,"k":40,"m":25,"n":600,"power":90,"sd":400},"reference":{"cite":1,"note":"","steps":[{"args":{"alpha":0.05,"effect":40.0,"power":0.9,"sd":400.0},"tool":"infer.sample_size","why":"Pergunta de desenho: os n\u00fameros v\u00eam da pergunta."}]},"seed":1000,"tolerance":42.02969224576248,"trap_tools":[],"truth_value":4202.969224576248,"variant":2},"question":"Para detectar um aumento de 40 reais na renda mensal das fam\u00edlias, cujo desvio-padr\u00e3o \u00e9 de 400 reais, com poder de 90% e signific\u00e2ncia de 5%, quantas fam\u00edlias precisamos sortear no total, metade para o programa?","requires":[],"runner":null,"sources":[],"split":"test","suite":"tool_selection","tags":["tool_selection","power","test"],"target":"agent","tool":null,"truth":"closed form (z\u2081\u208b\u03b1/\u2082 + z_power)\u00b2 with design effect 1 + (m \u2212 1)\u00b7icc, half treated; truth = 4202.9692; tolerance 1% or rounding up to whole clusters in each arm = 42.0297"}