chapter 9 Flashcards
(20 cards)
quasi-experiment
- a “sort of” experiment
- a research design that investigates causal relationships, but is not a true experiment, it lacks random assignment of participants to groups, often due to ethical or practical constraints
quasi experiments have:
- a control group
- a treatment (or program) group
- variables
quasi experiments do not have:
- random assignment to groups because if a variable is a participant variable that is not malleable there can’t be random assignment
- e.g. studying gender and voting, you must categorize based on gender, assignment can’t be random to each gender category
quasi experiment designs
- one-group posttest only design
- one-group pretest-postrest design
- non-equivalent control group design
one-group posttest only design
- single experimental group, where the dependent variable is tested once, post manipulation/treatment
- weakness: no comparison; we only know the results after and cant compare to what the data would have been before
- e.g. testing smoking habits, but only collecting data at the end means you can’t compare to see if they smoked more before the experiment only how much they smoked after
one-group pretest-posttest design
- add a baseline measure to provide a basis for comparison
- within subjects design
- the dependent variable is measured once before the treatment
is implemented and once after it is implemented - weakness: outside influences; single groups threats to internal validity
nonequivalent control group design
- one of the most frequently used quasi-experimental designs but looks just like a pretest-posttest design
- lacks random assignment to groups; not a true experiment b/c there is nonequivalence at the start
- as a result of having no random assignment, the treatment and control groups may be different at the study’s start
- raises a selection threat to internal validity
postest only non-equivalent group design
participants in one group are exposed to a treatment, a nonequivalent group is not exposed to
the treatment, and then the two groups are compared
selection threat to internal validity
- the people we selected for the groups resulted in groups that are different
- the possibility that differences between groups in a study, observed after the treatment, are due to pre-existing differences between the groups before the treatment, rather than the treatment itself
non-equivalent control group pretest-posttest design
- an improvement on the nonequivalent control group design
- a treatment group that is given a pretest,
receives a treatment, and then is given a posttest - the additional of a pre-test give it the strongest internal validity you can have in a quasi-experimental design because having the pretest and two groups we can assess group equivalence at the beginning
- can show cause and effect
- weakness/threat: effect of the pretest; in this case treat it as another independent variable to counter the weakness
interrupted time series design
- examines the dependent variable over an extended period of time, both before and after the independent variable is implemented
- lacks a control group
- can use archival research to go back further
control series design
- improves interrupted time series design by finding an appropriate “control group”
- involves finding a similar population that did not receive a particular manipulation
- e.g. data from ontario and made a control group by combining data from manitoba and new brunswick
single subject design
- can be one person or a small group
- each individual is still treated as a group unto itself; seen as a single case
- looking at the effects of the treatment individually
- e.g. one rat in a cage, undergoing an experiment, testing results
single case experimental design
- measured from baseline period to treatment periods
- test dependent variable until we get a good baseline measure; dependent variable = before treatment
- there are external threats to validity such as a history threat
- with causation you can argue for a limited cause in a single case design
- generalizability with a sample size of one, or a small group is not easy
reversal (ABA) or withdrawal design
- baseline → treatment → baseline
- stronger version of the single case experimental design
- withdraw the treatment and return to baseline to see what happens
If the behaviour returns the its baseline value we are more confident that it is the treatment having the effect and vice versa - can be improved by extending to ABAB or ABABAB designs
benefits of an ABA
- increased power: reversal designs are more likely to detect an effect if there is one
- more ethical to land on “B” because it is the treatment
multiple baseline design
- multiple measures taken before and after the manipulation and more then one baseline is measured
- increase the power
- treatment is effective if change in behaviour occurs following manipulation, under multiple circumstances
- can be measured across subjects, situations and behaviours
interrupted time-series design with non-equivalent groups
involves taking a set of measurements at
intervals over a period of time both before and after an intervention of interest in two or more nonequivalent
groups
pretest-posttest design with switching replication
nonequivalent groups
are administered a pretest of the dependent variable, then one group receives a treatment while a
nonequivalent control group does not receive a treatment, the dependent variable is assessed again, and
then the treatment is added to the control group, and finally the dependent variable is assessed one last
time
switching replication with treatment removal design
the treatment is
removed from the first group when it is added to the second group
