SMRM Flashcards
1/8
0.125
sqrt(1)
1
sqrt(2)
1.41
sqrt(3)
1.73
sqrt(4)
2
sqrt(5)
2.24
sqrt(6)
2.45
sqrt(7)
2.67
sqrt(8)
2.83
sqrt(9)
3
sqrt(10)
3.16
sqrt(11)
3.32
sqrt(12)
3.46
sqrt(13)
3.61
sqrt(14)
3.74
sqrt(15)
3.87
deci
10^-1
centi
10^-2
milli
10^-3
micro
10^-6
nano
10^-9
pico
10^-12
femto
10^-15
deca
10^1
hecto
10^2
kilo
10^3
mega
10^6
giga
10^9
tetra
four measures
sin(0)
0
sin(30)
1/2
sin(45)
sqrt(2)/2
sin(60)
sqrt(3)/2
sin(90)
1
cos(0)
1
cos(30)
sqrt(3)/2
cos(45)
sqrt(2)/2
cos(60)
1/2
cos(90)
0
cosecant
1/sin
secant
1/cosine
cotan
1/tan
tangent
sin/cos
pi radians
180 degrees
distance
x=(1/2)at^2
Force
F=Gmm/r^2
F=ma
velocity
v=sqrt(2gh)
ideal gas
PV=nRT
selection bias
method is not random, results are not representative
types of selection bias
specific real area self-selection pre-screening exclusion healthy user Berkson's fallacy overmatching survivorship
specific real area
neglecting a subset of the population that is not present
self-selection
decision whether to participate
pre-screening
choosing a subset to prove a hypothesis
exclusion
keeping a subset out of the population
healthy user
subset better off than the general population
Berkson’s fallacy
subset less healthy than general population (hospital)
overmatching
correlation of data with a confounder as a result of exposure
survivorship
only the remaining subjects are selected, neglecting those not left anymore
observer bias
subject knows the goals and/or hypotheses of the study
demand characteristics
formed interpretation of the purpose of the study
information bias (two types)
wrong or inexact recording of variables
measurement error: involves a continuous variable
misclassification: categorical variable
confounding variable
extraneous variable that influences the study variable
placebo effect: pt with a blank treatment experiences health benefits
detection bias
systematic differences between groups in terms of treatment (doctor cares better for a test subject)
experimenter bias (two types)
errors based on expectations
confirmation bias: more likely to favor information that confirms the hypothesis and to ignore information that denies the hypothesis
reporting bias: systematic difference in what information is reported
accuracy
the degree to which a measurement conforms to the standard “correct” value
systematic error
precision
relates to the reproducibility or repeatability of the results
random error
reliability (two types)
results that are consistent and repeatable
test-retest: test results are consistent over time
inter-rater: degree of consensus among judges
validity (two types)
test what it is meant to measure and is done under scientific standards
external: generalizable, extrapolation
internal: justifiable
correlation
r^2
amount of variance in y that is accounted for by x
causation
one thing is found to be both necessary and sufficient to cause another
VERY DIFFICULT TO PROVE
Hill’s criteria (9)
temporality strength consistency testable specificity plausibility dose-response relationship coherence analogy
normal distribution
1SD: 68%
2SD: 95%
3SD: 99%
probability
assumes outcomes are independent and mutually exclusive
AND: multiply
OR: add
directional hypothesis testing
p>alpha, cannot reject Ho
non-directional hypothesis testing
p>alpha/2, cannot reject Ho
standard error of mean (SEM)
how precisely the mean represents the true population mean
decreases with increasing population size
type I error
claimed different when none existed
type II error
no claimed difference when one existed
how do you deal with percentages in proportionality
convert to fractions
distance
d=rt
how is the density of air related to elevation?
density decreases with height
is a bird a mammal
no
how does a linear relationship change on a semilog plot
it becomes non-linear