Module 7: Quantitative Research Designs Flashcards
What are the 3 broad design categories used in quantitative research?
- Nonexperimental/observational
- Quasi-experimental
- Experimental
Nonexperimental/Observational studies can be either ______ or ______.
descriptive (correlational or univariate)
correlational/relationship (retrospective or prospective)
What do descriptive correlational studies show?
A relationship between two variables (whether or not they covary) without looking for explanation or causality
What do univariate descriptive studies do and what are the two types?
They describe the occurrence of something
- Prevalence (cross-sectional)
- Incidence (longitudinal)
How are prevalence (cross-sectional) and incidence (longitudinal) univariate descriptive studies different?
Prevalence studies that are cross-sectional explore the occurrence of something at one point in time. Incidence or longitudinal studies estimate the development of new events or new cases over time.
What is a retrospective study?
A type of correlational/relationship nonexperimental study that links present phenomena to phenomena that has occurred in the past. The researcher begins with the dependent variable and examines whether or not it is correlated with one or more previously occurring independent variables
What is a case control design?
A type of retrospective study (correlational/relationship nonexperimental study) which looks at similar groups of people, one which have a specific outcome and one which does not
(ex- similar people, one group with lung cancer and one without)
What is a prospective or cohort study design?
A type of correlational/relationship nonexperimental study that starts with a presumed cause and then goes forward in time to the presumed effect
(strongest design for prognosis and etiology question when randomization is impossible)
Which type of research study can be described as controlled trials without randomization?
Quasi-experimental studies
Quasi-experimental Designs include: (4)
- Pretest-posttest nonequivalent control group
- After-only nonequivalent control group
- One group pretest-posttest
- Time series
How is a time series experiment conducted?
(quasi-experimental)
Data about the dependent variable is collected over an extended period during which an intervention is introduced and following the intervention. Repeated measures
In experimental and quasi-experimental designs researchers use an ___________ to bring about a desired effect.
Intervention
How is systematic bias reduced?
Randomization
Experimental Designs include what 5 types?
- True experimental (RCT)
- Pretest-Posttest
- Posttest Only
- Factorial
- Crossover
What are some design features of True experiments (RCTs)?
Manipulation
Control
Randomization
Blinding
Which variable is manipulated in an RCT?
Independent variable (researchers used manipulation to do something to study participants)
What are examples of control groups used in RCTs?
Patients receiving a placebo
Standard of care practices
Blinding in an RCT is used to decrease bias that comes from ________ .
Awareness
In randomization participants have an ________ chance of being placed in groups randomly.
equal
How is the pretest-posttest design of an experiment conducted?
Outcome variable data is measure on both groups before the intervention and again after the intervention
What is a downside to posttest only experimental designs?
Not testing the participants prior to the intervention doesn’t allow the researchers to know if the groups were the same prior to the intervention
In a pre-test post test design where participants have data collected from them at multiple post intervention points this is called________ _____ _____.
repeated measures design
What is factorial experimental design?
A design that allows for manipulation of two or more variables (factors) simultaneously
In addition to ‘main effects’ (those from experimentally manipulated variables) factorial designs also allow researchers to test _________ _______ .
Interaction effects
How is a crossover experimental study conducted?
The same participants are exposed to more than one condition and serve as their own controls
When can a crossover experimental design not be used?
When there is expected carry over effect from one condition to another.
To avoid carryover effect in a crossover experimental study what do researchers sometimes allow for?
Washout period in between treatments
What aspect of correlational research. related to sampling, is a weakness?
Self-selection of groups