Researchers believe that Drug A will have fewer negative side effects than Drug B. State the null hypothesis the investigators would use to determine if their results are statistically significant.
Source: PH 250A Problem Set 12
Wednesday, July 18, 2018
Tags
- study designs (46)
- bias and confounding (31)
- measures of disease (31)
- measures of association (23)
- biostatistics review (20)
- confounding (20)
- bias (19)
- experimental studies (19)
- prevalence (19)
- trials (17)
- cohort studies (16)
- effect modification (14)
- person-time (14)
- incidence (13)
- case-control studies (12)
- incidence density (12)
- stratification (12)
- cumulative incidence (11)
- loss to followup (11)
- odds ratio (11)
- selection bias (11)
- Causal inference (10)
- causality (10)
- prospective/retrospective (8)
- randomization (8)
- rate ratio (8)
- risk ratio (8)
- screening (8)
- recall bias (7)
- power (6)
- sufficient-component cause model (6)
- Hill's criteria (5)
- information bias (5)
- 2x2 table (4)
- differential misclassification (4)
- hypothesis testing (4)
- masking (4)
- matching (4)
- misclassification bias (4)
- necessary cause (4)
- non-differential misclassification (4)
- sufficient cause (4)
- attributable risk/rate (3)
- blinding (3)
- chi-squared test (3)
- counterfactual (3)
- effectiveness (3)
- interviewer bias (3)
- mantel-haenszel measure of effect (3)
- population attributable risk (3)
- population attributable risk/rate percent (3)
- preventive trials (3)
- risk (3)
- sample size (3)
- sampling (3)
- standardization (3)
- statistical significance (3)
- study population (3)
- therapeutic trials (3)
- attributable rate (2)
- attributable risk (2)
- causal complement (2)
- cluster randomization (2)
- compliance (2)
- component cause (2)
- efficacy (2)
- equipoise (2)
- fixed/dynamic populations (2)
- mantel-haenszel odds ratio (2)
- minimum detectable effect (2)
- multivariable modeling (2)
- placebo (2)
- population controls (2)
- precision (2)
- primary/secondary study base (2)
- respondent bias (2)
- risk difference (2)
- sample size and power (2)
- steady state (2)
- study base principle (2)
- validity (2)
- DAG (1)
- IRB (1)
- additive interaction (1)
- additive/multiplicative scale (1)
- adherence (1)
- attributable risk % (1)
- backdoor pathway (1)
- berkson's bias (1)
- block randomization (1)
- case cohort sampling (1)
- causal-effect (1)
- community trials (1)
- concurrent/non-concurrent (1)
- confidence intervals (1)
- crossover trials (1)
- crude rate (1)
- double blind (1)
- eligibility (1)
- exchangeability (1)
- factorial study design (1)
- healthy worker bias (1)
- individual trials (1)
- informed consent (1)
- institutional review board (1)
- intention to treat (1)
- internal validity (1)
- interview bias (1)
- multiplicative interaction (1)
- multivariate modeling (1)
- nested case-control (1)
- non-collapsibility (1)
- non-missclassification bias (1)
- observer bias (1)
- odds (1)
- open/closed cohort (1)
- parallel trials (1)
- phases (1)
- population attributable risk/rate (1)
- prevention trials (1)
- primary outcomes (1)
- random error (1)
- rare disease assumption (1)
- rate (1)
- reference population (1)
- relative risk (1)
- residual confounding (1)
- results (1)
- risk set (1)
- run in period (1)
- secondary outcomes (1)
- sensitivity (1)
- sham (1)
- social-desirability bias (1)
- specific rate (1)
- specificity (1)
- stratified randomization (1)
- systematic error (1)
- target population (1)
- test of heterogeneity (1)
- treatment (1)
- type 1/ type 2 errors (1)
- validation (1)
