Exam One - research designs Flashcards
What are the 3 characteristics of quantitative research
- manipulation
- randomization
- control
manipulation
imposing a set of predetermined experimental conditions
- having an IV
IV
independent variable
what you manipulate
DV
dependent variable
what you measure
each IV is called a…
factor
a 3 x 2 design would mean?
you have three different conditions and you measure the DV twice
random assignment
subjects unsystematically appointed to groups
affect Internal validity
does random assignment mean the same thing as random selection?
NONONONONONONO
randomized matched assignment
subjects are paired based on some criteria to artificially control the influence of that variable on DV
subjects within each matched pair randomly assigned separate conditions
control
most effective design element to eliminate confounding variables
expect to see no change in control group
improves internal validity
interaction effect
demonstrate a covariation of cause and effect
“the treatment effect depends on group assignment”
what are the 3 main categories of research design?
1 - experimental
2 - quasiexperimental
3 - nonexperimental
experimental design is best for?
showing cause and effect
quasiexp is best for?
if you lack random assignment or conrol group
nonexperimental is good for?
collecting data as it naturally exists but worst for cause and effect
What are examples of non-experimental designs?
-epidemiological studies: ID distribution and determinants of disease
- correlational studies: test if 2 variables are linked in some way
- cross sectional: snapshot of the characteristics of a defined pop at a single point in time
- survey research: questionnaires, interviews
- case report: systematic, detailed description of a clinical practice or case
- methodological research: reliability or validity of clinical measures
_______- study designs often form the basis for new theories, new hypothesis, and future intervention studies
non experimental
prevalence
% of pop with a certain condition at a given point in time
incidence
rate that new cases are developing during a time period
______________ designs are also used to analyze determinants of injury or disease
epidemiological
______ studies are used to describe the distribution of health and illness in human pop (prevalence and incidence)
epidemiological
Two types of cohort studies for epidem purposes
- prospective
- retrospective
prospective cohort study
data collection before development of condition or injury
*only way to determine etiology of injury
restrospective cohort study
data collection after development of condition or injury
cohort study (epidem)
observational study of naturally occuring group over time
What is the first step in every cohort study?
establish 2 groups: those exposed and those not exposed to risk factor
case control studies
compare exposure information among people with and without an outcome or disease
what is the goal of all analytical epidem study designs?
identify the possible determinate of disease
Is there an association?
case control
exposure <-> outcome
Can we infer causality?
cohort design
exposure->outcome