Statistics Flashcards
What test determines significant different b/w means of 2 normally distributed groups
T-test
What test compares means among 3+ normally distributed groups
ANOVA
What test is used when population parameters are definitely known
Z-test
What test is used to determine if there is significant different b/w expected frequencies and observed frequencies in 1+ categories
Chi-square test
compares differences of categorical (ordinal) data. eg: comparing presence of lung ca in smokers vs nonsmokers
used in case-control studies
Test that estimates the relationships among variables. Estimates the conditional expectation of the dependent variable given the independent variables
Regression analysis
QI approach to improve efficiency
Small scale: Rapid cycle improvement (RCI)
Large scale: Lean (Value stream mapping)
QI approach to ID & correct root causes of variation & enhance efficiency (roadmap for problem-solving and process improvement)
Six sigma
Lean Six Sigma
Improve quality of communication
SBAR
TeamSTEPPS
Type II error
Statistically significant difference does exist but it is not detected and you accept the null inappropriately
Probability of committing type II error is estimated by beta and is set to 0.20. Common reason is due to low power
Cross-sectional Study
Purely observational. Looks at single point in time.
Used for determining disease prevalence.
No causal relationship is inferred.
Randomized controlled-trial (RTC)
prospective
subjects randomly assigned to groups and separated based on exposure, and followed to see outcomes
Gold standard for determining causality
Cohort study
prospective
Subjects grouped according to exposure status (present or absent) for a risk factor, and then followed to see outcomes
Case-control study
Retrospective
Subjects are grouped based on outcomes and then are researched to see whether there was any exposure
