chapter 13 Flashcards
(32 cards)
properties of a well-designed experiment
manipulation
random assignment
control
manipulation
vary at least one IV to assess its effects on participants’ responses
random assignment
have the power to assign participants to the various experimental conditions in a way that assures their initial equivalence
control
ability to control all other potential variables (i.e., extraneous variables) that may influence participants’ responses
quasi-experiments
aim to demonstrate causality between an intervention and an outcome without randomization or IV manipulation
They use:
1. Comparisons are made between groups that already exist or
2. Within a group before and after a quasi-experimental intervention has occurred
independent variables
In an experiment, the researcher manipulates one or more independent variables
An independent variable must have two or more levels (different values of the independent variable)
These levels can be:
Quantitative → coffee dosages in the forms of 100mg, 200mg, 300mg, 400mg
Quantitative → coffee vs decaf
True independent variable
one you have full control over; you can manipulate these variables
Quasi-independent variable
not a true independent variable that is manipulated by the researcher but rather is an event that occurred for other reasons
Ex: langer and rodin used a field experiment to investigate the effects of personal choice among nursing home residents. They found that residents in the responsibility-induced group were 48% happier than the control group and more alert at the end of the experiment.
Criteria for Inferring Causality
covariation
directionality
extraneous variables
Correlational research satisfies the first )and sometimes the second) criterion, but never the third
covariation
changes in one variable are associated with changes in the other variable; same as correlation (i.e., high school GPA → SAT score)
directionality
the presumed causal variable preceded the presumed effect in time (i.e., smoking → lung cancer)
extraneous variables
all other variables that may affect the relation between the two target variables are controlled or eliminated (think of discrimination and depression)
Evaluating Quasi-Experimental Designs
Quasi-experimental designs can show:
Covariation – the cause and effect covary
Directionality – the presumed causal variable preceded the effect in time
Quasi-experimental designs do not:
Control for extraneous variables – eliminate all other alternative explanations of the results through randomization and experimental control
Questionable Internal Validity
The internal validity of quasi-experiments is always questionable
No control over the independent variable and/or the assignment of participants to conditions
However, some quasi-experimental designs are more internally valid than others. The extent to which a quasi-experimental design can eliminate possible threats to internal validity determines its usefulness.
Common Threats to Internal Validity in Quasi-Experimental Designs
history
maturation
regression to the mean
pretest sensitization
history
something other than the quasi-independent variable that occurred between the pretest and posttest caused the observed change
maturation
normal changes that occur over time, such as those associated with development, may be mistakenly attributed to the quasi-independent variable
regression to the mean
when participants were selected because they had extreme scores, their scores may change in the direction of the mean between pretest and posttest even if the quasi-independent variable had no effect
pretest sensitization
taking the pretest changes participants’ reactions to the posttest
Pre-experimental design: one-group pretest-posttest design
O1 —————> X ————–> O2
Pretest intervention post-test
Measures are collected before and after an intervention
–> Assess DV (focus), introduce IV (coffee), assess DV again
This is an experimental design (rather than a quasi-experimental design) because it lacks control and has no internal validity
- Fails to eliminate most threats to internal validity
- Can be influenced by regression to the mean
- This design should NEVER be used
Quasi-Experimental designs: nonequivalent control group designs
measure both groups after one receives the quasi-independent variable
Quasi-experimental group: X O
Nonequivalent control group: – O
Potential threat to internal validity:
- Pre-exist group differences
- Selection bias is a threat - can’t be sure the groups were the same before the treatment
- Check Rodin & Langer (1977) study in the book
Quasi Experimental Designs: Nonequivalent groups pretest-posttest
both groups are measured before and after the quasi-independent variable
Quasi-experimental group: O1 X O2
Nonequivalent control group: O1 – O2
Potential threat to internal validity:
- Pre-existent group differences (selection bias!)
- Local history effect – something else may happen to one group that does not happen to the other group (also called a selection-by-history interesting). Plus, you will always have all the other threats related to pre existing group differences
Ensuring similarity in nonequivalent control designs
There are two ways to ensure similarity across groups in quasi-experimental designs:
1. Search for nonequivalent control groups that are similar to quasi-experimental groups
2. Collect additional information about the participants in both groups for additional statistical controls
Simple interrupted time series design
O1 O2 O3 O4 X O5 O6 O7 O8
Should be able to tell whether or not an effect is due to the quasi-experimental variables as opposite to aging or maturation
Possible threat to internal validity → contemporary history – possibility that the observed effects are due to some other event that occurred at the same time as the quasi-independent variable
