INTRO TO EPIDEMIOLOGY- HLTH 298 CAL POLY SLO
QUESTIONS AND CORRECT DETAILED ANSWERS
(VERIFIED ANSWERS) ALREADY GRADED A+
Ecologic Ans✓✓✓Use it early in a line of investigation (for hypothesis
generation or development)
Requires limited resources
ONLY ONE WHERE UNIT OF OBSEVATION IS A POPULATION
Problems: We don't know if people
cross-sectional study Ans✓✓✓A study in which a representative cross
section of the population is tested or surveyed at one specific time.
Exposed or unexposed?
Disease free or has disease?
Case-control study Ans✓✓✓A type of epidemiologic study where a
group of individuals with the diseases, referred to as cases, are compared
to individuals without the disease, referred to as controls
Cohort (Prospective longitudinal) study Ans✓✓✓two samples, one
exposed one unexposed, followed for long time
quasi-experimental design Ans✓✓✓An experiment that does not require
random assignment to conditions. Used to prep for true experimental, to
extend true exper. to real world.
, Limits: Non-equal comparison grps
true experimental design Ans✓✓✓research using an experimental group
and a control group, to which test units are randomly assigned,
best to minimize confounding variables and is strongest, used to
establish temporality
-no recall bias
Biases: - attrition bias (minimized, but still there)
- observer bias, not real world conditions,
causal inference Ans✓✓✓exposure causes disease
ecologic fallacy Ans✓✓✓observations made at the group level may not
represent the exposure-disease relationship at the individual level
confounding variable Ans✓✓✓- There is another factor that could be
explaining the variation.
(- cannot be a mediator (cannot be in a causal pathway)
(- could be + (inflates) or - (attenuates/masks) confounder)
(measures to address it: randomization for experimental,matching esp
for case-control, restriction, stat adjustment)
QUESTIONS AND CORRECT DETAILED ANSWERS
(VERIFIED ANSWERS) ALREADY GRADED A+
Ecologic Ans✓✓✓Use it early in a line of investigation (for hypothesis
generation or development)
Requires limited resources
ONLY ONE WHERE UNIT OF OBSEVATION IS A POPULATION
Problems: We don't know if people
cross-sectional study Ans✓✓✓A study in which a representative cross
section of the population is tested or surveyed at one specific time.
Exposed or unexposed?
Disease free or has disease?
Case-control study Ans✓✓✓A type of epidemiologic study where a
group of individuals with the diseases, referred to as cases, are compared
to individuals without the disease, referred to as controls
Cohort (Prospective longitudinal) study Ans✓✓✓two samples, one
exposed one unexposed, followed for long time
quasi-experimental design Ans✓✓✓An experiment that does not require
random assignment to conditions. Used to prep for true experimental, to
extend true exper. to real world.
, Limits: Non-equal comparison grps
true experimental design Ans✓✓✓research using an experimental group
and a control group, to which test units are randomly assigned,
best to minimize confounding variables and is strongest, used to
establish temporality
-no recall bias
Biases: - attrition bias (minimized, but still there)
- observer bias, not real world conditions,
causal inference Ans✓✓✓exposure causes disease
ecologic fallacy Ans✓✓✓observations made at the group level may not
represent the exposure-disease relationship at the individual level
confounding variable Ans✓✓✓- There is another factor that could be
explaining the variation.
(- cannot be a mediator (cannot be in a causal pathway)
(- could be + (inflates) or - (attenuates/masks) confounder)
(measures to address it: randomization for experimental,matching esp
for case-control, restriction, stat adjustment)
