#separator:Tab
#html:false
#tags column:5
IMF-101-02-0001-en	What does a hypothesis test assess, and what can it not directly decide?	It asks whether observed data are sufficiently incompatible with a specified statistical model to challenge it; it does not decide whether a theory is true, an effect is clinically important, or bias is absent.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0002-en	Why is a p value not the probability that the null hypothesis is true?	It is calculated by assuming the null model and all test assumptions: it is the probability of a result at least as incompatible with that model as the one observed.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0003-en	Why is a nonsignificant result not automatically a type two error?	Type one and type two error rates and power are properties of a procedure under assumptions over repeated use, not retrospective labels for one result.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0004-en	Why does post hoc power calculated from the observed effect add little?	It merely restates the p value; once data exist, the estimated effect and its uncertainty are more informative.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0005-en	When is a one-sided test appropriate, and why is that condition rarely met?	Only when effects in the opposite direction would be treated as equivalent to no effect for the decision; harm in the unexpected direction is rarely irrelevant.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0006-en	Why can an odds ratio mislead as a measure of relative risk when events are common?	Odds ratios approximate risk ratios only when events are rare; when events are common they can look more extreme.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0007-en	Why is comparing a confidence interval with zones of meaningful benefit, trivial effect and harm richer than asking whether it excludes zero?	An estimate can be statistically significant yet too small to matter, or nonsignificant while still compatible with important benefit and harm.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0008-en	What goes wrong when an independent-group test is applied to paired data, and a paired test to unrelated data?	The independent test wastes information because it cannot remove stable between-unit variation; the paired test invents dependence that does not exist.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0009-en	Why do many tests at a five per cent threshold require multiplicity control, and what do family-wise and false-discovery methods limit?	With many true nulls the chance of at least one false rejection rises; family-wise control limits the probability of any false rejection, false-discovery-rate methods the expected proportion among rejections.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0010-en	How can convincing p values arise without deliberate fabrication, and what distinguishes planned from exploratory work?	Choosing outcomes, time points, exclusions, transformations or models after seeing results exploits researcher degrees of freedom; prespecification, registration, time-stamped protocols and code separate planned work.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0011-en	Why is a significant result in one subgroup and a nonsignificant result in another not evidence that the effect differs?	Evidence that effects differ across patient characteristics requires a direct interaction test, not a comparison of separate significance labels.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0012-en	Why can bias toward similarity falsely support non-inferiority, and how is this usually checked?	Non-inferiority only requires excluding an unacceptable loss, so poor adherence or outcome misclassification that blurs differences favours the claim; intention-to-treat and per-protocol views are often both examined.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0013-en	Why does pooling systematically biased studies in a meta-analysis not yield a trustworthy estimate, however precise?	Meta-analysis is only as meaningful as the studies' estimands, bias and heterogeneity permit, so a precise pooled estimate from biased studies remains precisely wrong.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending
IMF-101-02-0014-en	Why does repeatedly checking accumulating data and stopping when significance appears distort the evidence?	It inflates false-positive probability unless the design accounts for it, and stopping early for benefit when few events have accrued can exaggerate the effect estimate.	Integrated Medical Foundations; chapter 101, TTS module 2; https://theisegoria.github.io/medical-textbook/integrated-medical-foundations-expanded-english.md; module SHA256 at export: de1b4d957e07fc042e8104d4c0e36bda254ec2bb87dba76e2d87c7fe256e1e19; editorial review pending	imf lang::en chapter::101 module::101-02 review::pending