ch 10 Flashcards
Non-experimental approaches are used when
the researcher is unable to control the variables or the nature of the question is not casual.
In general a non-experimental design includes research where:
a. The researcher does not manipulate an IV
b. Has limited or no control over the nature or timing of the treatment
c. When casual relationships are not the primary focus of the researcher
Quasi-experiment
a. Design that look like true experience and that are statistically analyzed in similar ways
b. The researcher can compare groups but does not control the nature and/or timing of the treatment may be a participant variable.
c. Casual interpretation cannot be made
Participant variable:
Variables associated with the participant themselves (eg. Gender, a treatment the participant chose.)
Time series design
a. A quasi-experiment design where participants who have been exposed to a treatment are tested both before and after the intro of treatment.
b. The researcher does not control the nature of the treatment or the time that it was introduced.
c. Usually involves pre- posttest measures and may include a comparison group: both are methods to try to control extraneous variables.
Mult. Time series design
a. Like an interrupted time series design but include a control group that was not exposed to the treatment
b. Even with a control group be cautious about interpretations. An alternative explanation may explain the outcome (ie. The outcome may not have resulted because of the treatment.)
Non-equivalent groups design
a. Used to compare groups that they know or suspect, are different at the outset of the study.
b. Compares changes in behavior between the groups.
c. By comparing changes you control for initial group difference.
Longitudinal research:
a. Involves studying a group of individual over a long period of time to determine how characteristics measured earlier in life related to behavior later in life
b. Difficult and expensive to conduct.
c. Attrition (loss of participants) can be a major problem and may affect the internal validity of the study.
Cross sectional
a. Used to study groups of people who are different ages.
b. Cohort effect: variables that are confounded with age
Sequential research:
a. Combined cross-sectional and longitudinal research by selecting a cross section of ages over a number of years. Cohort effects are controlled by following a number of age cohorts (cross-sectional) over time (longitudinal)
Microgenetic method
i. Involves carefully observing behavior during periods when rapid change is occurring and collecting both quantitative and qualitative information.
ii. Used to study the process of change
iii. Example: piaget’s stage model-observation.
Case studies
i. In depth studies of a single individual
ii. Used when a researcher is interested in studying a single individual on many variables and is not assessing a treatment
iii. The objective is to describe the characteristic of the individual case, not to generalize to a population of similar cases.
iv. Often qualitative, but can be more quantitative.
Correlational research
i. Used to study relationship between variables.
ii. Variables that are systematic related are said to be correlational.
iii. Often used to research topics that would otherwise be unethical to study.
iv. Correlation does not infer causation.
v. Confounding variables may affect the results.
Multiple regression
i. A powerful technique that allows us to look at the relationship between a number of predictor variables and a single criterion variable.
ii. Results tell us if the predictor variables are positively or negatively correlated with the criterion variable. Also show the relative importance of each variable in determining behavior.
