Test 1 Flashcards
The application of statistical principles in medicine, public health, or biology.
biostatistics
Virtually all statistics are an ().
estimate or guess
(prevalence or incidence) exposed/unexposed
relative risk
A field of study focused on the study of health and illness in human populations, patterns of health or disease, and the factors that influence these patterns.
epidemiology
More effective from a statistical viewpoint; Intervening and measuring a response happens here; participants are, with no selectiveness, put into one of several comparison treatments or groups.
randomized controlled trial or experimental design
What is an example of a randomized controlled trial?
clinical trial
The methodology that is used to collect the information to address the research question.
study design
In what type of study do we observe a phenomenon?
observational studies
A study conducted at a single point in time; appropriate design when research question is focused on prevalence of disease, a present practice, or an opinion between participant groups; nonrandomized.
cross-sectional study
A study conducted to study the exposure or risk factor status of participants looking back in time.
retrospective cohort study
A distortion of the effect of an exposure or risk factor on the outcome by other characteristics.
confounding
Study often used in epidemiological research when seeing if there is an association between a particular risk factor or exposure and an outcome, particularly for a rare outcome.
case-control study
Clinical trials which include multiple study centers
Multicenter trials
number of persons with disease / the number of persons examined at baseline
point prevalence
number of persons who develop a disease during a specified period / number of persons at risk at baseline
cumulative incidence
possibility of going from disease free to diseased
hazard
number of persons who develop disease during a specified period / sum of lengths of time during which persons are disease-free
incidence rate
(PP/CI/IR) exposed - unexposed
risk difference
(PP/CI/IR) (overall-unexposed)/overall
population attributable risk
(PP/CI) exposed/unexposed
relative risk or risk ratio
IRexposed/IRunexposed
rate ratio
(CI/PP) (exposed/1-exposed)/(unexposed/1-unexposed)
odds ratio
Subset of individuals from the population
sample
Another name for characteristics in a study
variables, outcomes, endpoints
() variables only have two responses
dichotomous
() variables have more than two possible responses and are ordered
ordinal
() variables have more than two responses and are unordered
categorical or nominal
() variables take on an unlimited number of responses between defined minimum and maximum values
continuous or quantitative or measurement
numerical summary measures computed on samples
statistics
summary measures computed on populations
parameters
What is the appropriate graphical display for dichotomous variables?
bar charts
What is an ideal way to display descriptive statistics for ordinal and categorical variables?
frequency distribution table
What is the appropriate graphical display for ordinal variables?
histograms
What is the appropriate graphical display for categorical variables?
bar charts
sum of all the values / sample size (most appropriate measure of typical value if no outliers)
sample mean
A second measure of the average value (ideal if there are outliers)
sample median
Most frequent value
mode
Important measure of variability in a sample
sample range
Most widely used measure of variability for a continuous variable
standard deviation
the difference between the first and third quartiles (ideal if there are outliers)
IQR
The value in the dataset that holds 25% of the values below it
first quartile
The value in the dataset that holds 25% of the values above it
third quartile
What is the appropriate graphical display for the distribution of a continuous variable?
box-whisker plots
Numbers that reflect the likelihood that a particular event occurs
probabiltiies
Making generalizations about unknown population parameters based on sample statistics
inferential statistics
When each member of population has a known probability of being selected
probability sampling
Each member of the population is selected without use of probability
nonprobability sampling
Number that reflects the likelihood that a particular event occurs focusing on a subset of the population
conditional probability
P(+|D)
sensitivity
P(-|DF)
specificity
P(+|DF)
false positive fraction
P(-|D)
false negative fraction
P(D|+)
positive predictive value
P(DF|-)
negative predictive value
The probability of one event is not affected by the occurrence or nonoccurrence of the other ; P(A|B) =P(A)
independence
Probability rule that can be used to compute conditional probability based on specific available info ; P(A|B) = (P(A|B)P(A))/ P(B)
Bayes’ Thereom
What are examples of observational studies?
cross-sectional and cohort
() and incidence have large effect on prevalence
duration
high incidence and low duration = () prevalence
low
low incidence and high duration = () prevalence
high
probability of what you are trying to get /probability of anything else
odds
() + 1.5 * IQR for larger value outlier(s)
Q3
() - 1.5 * IQR for smaller value outlier(s)
Q1
If the observed values vary widely around the sample mean, the standard deviation is ()
large
If all observed values are close to the sample mean, the standard deviation is ()
small
If all values in sample are identical, the standard deviation is ()
zero
Describes info collected in a study sample
descriptive statistics
