Goes Goes in false false neg negative ative box in 2X2 tabl tablee Goes Goes in fal false se positi positive ve box box in 2X2 2X2 tabl tableeīet a error (T y pe pe II) → false Negative error → The N in negative has two ( II) vertical lines, so corresponds to type II Power Power is when when the ex experime perimental ntal hyp othesis othesis is is true and and the study study also also shows that that it is trueĪ lp lp ha ha error (T (T y pe pe I) → false Positive error → The P in Positive has one ( I) vertical line, so corresponds to type I Make a 2X2 2X2 tabl table e agai again, n, use use the ABC rule rule to place place R eality eality on top and S tudy results results on the side S tatistical hypotheses hypotheses and error error Pygmalion effect → Researcher’s belief in efficacy of Tx changes the outcome of the treatmentĬhanges its behavior behavior ow ing to knowledge of being studied Hawthorne effect → Group changes
rela relative tive risk: risk: crea create te 2X2 table using the alphebetic alphebetical al order order rule rule Disease comes before R isk isk factor formulas formulas are p retty logica logicall Positi Positive ve and neg negative tive skew skew ■Įdian is always in the middle when mean, median and mode are ranked mode is logically placed Median (point (p oint at peak), median median is middle, mean mean is on ot her end Vertica Verticall → Sensitiv Sensitivity/sp ity/sp ecifi ecifici city ty formula formulas s vagina vagina is is VERTICAL VERTICAL slit the vag vagina is is SENSITIVE
51 note that disea disease se which which come comess first alphabetically alphabetically is on t op ), can use the fo llowing: llowing: If can can create create 2X2 disea disease/ se/test test table (FA 2011 P. Case Case contro controll (OR) (OR) is in in past, cohort cohort study is in future future (re (rela lative tive risk) risk) OR comes comes before RR alphabeticallyĮvalu valuati ation on of of dia diagnosti nosticc tests tests ■Įcificty rules IN SNOUT = S eNsitivity rules OUT SPIN = S Pecificty
0 kommentar(er)
