Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
28 Cards in this Set
- Front
- Back
- 3rd side (hint)
what is stratified?
what is double blind ? |
stratified is broken down into groups by age, sex, age, ethnicity. we stratify what we find important.
double blind is a procedure in which the experimenter and the subjects dont know the controls, to protect from bias |
|
|
necessary and sufficient
|
one to one relationship
|
|
|
necessary but not sufficient
|
one needs the other to complete
A + B + C = Factor |
|
|
sufficient but not necessary
|
one or the other can complete the factor
A or B Or C |
|
|
neither sufficient nor necessary
|
A + B
C + D E + F need two factors to complete |
|
|
Differential
|
bias is focused on one group. usually stratified across groups.
|
|
|
non differential
|
bias is in the same amount for all four age groups
|
|
|
random error
|
error by chance alone
|
|
|
systematic error
|
error in the study design
|
|
|
temporal
|
a relationship between exposure and disease that had occured before
|
|
|
strength
|
association measured by RR
|
strong = causal
|
|
dose response
|
a dose of risk increases, does its dose of disease increase ?
|
dose to dose relationship
|
|
replication
|
consistent with other findings
|
to replicate this must be adjusted or be
|
|
biologic plausability
|
coherence/ must work with modern biologic knowledge
|
in your biology book that you cant refute
|
|
consideration
|
ruling out all/other explanations
|
to consider
|
|
cessation of exposure
|
exposure goes down as factor disappears
|
like the mcrib except exposure
|
|
consistency
|
consistent with other findings to be able to study.
|
|
|
specificity
|
is a disease specific with one type of disease
|
|
|
selection bias
|
an error in the design
|
i go to the beach and study the relationship between people and lung cancer. i only study when their are events, rather than normal days
|
|
** information bias
|
subjects know or provide too much information
|
|
|
** surveillance bias
|
subjects know their being watched and may react differently
|
like psychology's observation bias
|
|
recall bias
|
a remember effect. asked to recall but provides other than at present information
|
cant remember the shyt
|
|
detection bias
|
a bias that changes because the follow up was more intense
|
|
|
measurement bias
|
data collection different from true value
|
|
|
confounding bias
|
one factor missing not include that effects the other two factors to become biased
|
|
|
interaction bias
|
how the study is approached when and how? to create bias
|
|
|
precision
|
reliability of experiment
|
|
|
accuracy
|
how close the results are to the actual true value
|
|