SMRM research methods Flashcards
What is a testable hypothesis?
A scientific question that is specific and answerable
A scientific experiment must have ___ ___ and verification
peer review
verification- aka replication
The two types of human subjects research is __ and __
experimental
observational
What is the term that describes the end of a study because of positive results?
beneficence
What is the term that describes the end of a study because of negative results?
nonmaleficence
___ - patient autonomy and informed consent
autonomy
__ - equal treatment of all people
justice
What is a cohort study?
longitudinal study observing characteristics of members of a cohort across time (ex. smokers are more likely to get lung cancer before 50)
What is a cross-sectional study?
analysis of data collected from a population at one specific time (ex. survey of population of us to see prevalence of covid)
What is a case-control study?
observational study of individuals with condition compared to control of the population
What is selection bias?
method to select participants is not completely random
What is specific real area bias?
conducting the study in a specific area that does not include a representative sampling of the population being studied
What is self-selection bias?
participants in the study have the ability to choose to participate or not to, or determine their level of involvement (surveys)
What is pre-screening or advertising bias?
advertising process itself results in an unrepresentative sample
What is exclusion bias?
exclusion of an entire group from the population (ex. excluding homeschooled kids in a study ab education)
What is healthy user bias?
when persons included in the study are healthier than normal people
What is Berkson’s Fallacy?
selection of participants from hospitals, where they are more likely to be unhealthy
What is Overmatching?
negative outcome resulting from what is normally a good practice- matching for potentially confounding variables
What is observer bias?
observers know the goals of the study or the hypothesis and allow the knowledge to influence observations
What is information bias?
wrong or inexact recording of variables or data
What are confounding variables?
affects both independent and dependent variables and may skew an experiment
What is detection bias?
systematic differences between groups caused by inconsistency in the method of detection or diagnosis
What is performance bias?
systematic differences between groups in terms of the actual care or treatment provided
reporting bias vs confirmation bias
reporting bias- not reporting all the findings
confirmation bias- favor info that supports hypothesis
Accuracy vs precision
accuracy - how correct you are
precise - how repeated trials are close together in value
validity vs reliability
reliability - results are consistent and repeatable
validity - the test measures what it purports to measure and uses methods that meet scientific standards
What is test-retest reliability?
measure of the degree of consistency between administration of a test and a subsequent administration of that same test
What is inter-rater reliability?
measure of consistency between multiple raters assigning the same values or making the same observations
External vs internal validity
external - the degree to which findings can be applied to general population
internal - findings of truth or causation are justifiable
What is Hill’s criteria?
set of guidelines that evaluates whether there is a causal relationship
Define the terms of Hill’s criteria
- temporality
- strength
- consistency
- specificity
- plausibility
- dose -response relationship
- testable by experiment
- coherence
- analogy
temporality - exposure of treatment MUST precede outcome
strength- statistical significance
plausibility - fits logically within understanding of process
dose-response- effects proportional to dosage of treatment
coherence- association is compatible with pre-existing science
analogy - similar associations are known to exist
necessary vs sufficient
necessary- condition that must be satisfied for event to occur
sufficient - if satisfied that event is guaranteed to occur
double blind vs single blind experiments
double - both experimenter and subject is unknowing
single - experimenter or subject is unknowing
In normal distribution, there is a bell shape curve, what percentages of the population are at each SD
1 SD- 68
2 SD - 95
3 SD - 99
null vs alternative hypothesis
null- there is no relationship
alternative - there is a significant difference
The presence of an asterisk generally indicates ___
data point is statistically significant
SEM in stats is__
standard error of the mean - how precise the mean represents the true mean of population SEM decreases as sample size increases