Term
| three major pairs of components in a classic experiment |
|
Definition
(1) Independent and dependent variables
(2) Random assignment to experimental and control groups
(3) Pretesting and posttesting |
|
|
Term
|
Definition
| Participants are measured on the DV (pretested), exposed to a stimulus that represents the IV, and then re-measured on the DV (posttested). |
|
|
Term
|
Definition
| Produces experimental and control groups that are statistically equivalent. |
|
|
Term
|
Definition
Both researchers and subjects are “blind” with respect to who is receiving the experimental treatment and who is not.
Another researcher knows which subjects are in which group, but that person is not responsible for administering the experiment. |
|
|
Term
|
Definition
| ensures that the cause precedes the effect in time by taking post-test measurements of the DV after introducing the experimental stimulus (IV). |
|
|
Term
|
Definition
determined by comparing the pretest to the posttest for the experimental group.
A change from pretest to posttest measures demonstrates correlation. |
|
|
Term
|
Definition
observed correlation not being due to the influence of some third variable because:
(1) random assignment produces experimental and control groups that are equivalent and will not differ on some other variable.
(2) the posttest for the experimental group is compared with the posttest for the control group. |
|
|
Term
| History (Internal Validity Threat) |
|
Definition
| historical events may occur during the course of the experiment that confound the experimental results |
|
|
Term
| Maturation (Internal Validity Threat) |
|
Definition
people are continually growing and changing and those changes affect the results of the experiment
In a long-term experiment, subjects grow older and wiser
In shorter experiments, they may become tired, bored, hungry, or change in other ways that affect their behavior |
|
|
Term
| Testing (Internal Validity Threat) |
|
Definition
| the process of testing and retesting influences people’s behavior and confounds (confuses; fails to distinguish) the experimental results |
|
|
Term
| Instrumentation (Internal Validity Threat) |
|
Definition
| if we use different measures of the DV they may not be comparable |
|
|
Term
| Statistical Regression (Internal Validity Threat) |
|
Definition
| includes experiments on subjects who start out with extreme scores on the DV. |
|
|
Term
| Selection Biases (Internal Validity Threat) |
|
Definition
random assignment eliminates the potential for bias in selecting subjects.
Volunteers are often solicited for experiments conducted on college campuses.
Students who volunteer for an experiment may not be typical of students as a whole. |
|
|
Term
| Experimental Mortality (Internal Validity Threat) |
|
Definition
| subjects often drop out of an experiment before it is completed, and that can affect statistical comparisons and conclusions. |
|
|
Term
| Diffusion of Treatments (Internal Validity Threat) |
|
Definition
| if in communication with one other, experimental subjects could pass on some elements of the treatment to the control group. |
|
|
Term
| Compensatory Treatment (Internal Validity Threat) |
|
Definition
in some experiments subjects in the control group may be deprived of something of value to them.
Compensatory Rivalry = subjects deprived of the treatment may try to compensate by working harder.
Demoralization = feelings of deprivation among the control group may result in subjects giving up. |
|
|
Term
|
Definition
A question of whether results from experiments in one setting (time and place) will be obtained in other settings.
Whether a treatment found to be effective for one population will have similar effects on a different group |
|
|
Term
|
Definition
A posttest only design reduces testing as a threat to validity by eliminating the pretest.
Random assignment is the key to the posttest only design. |
|
|
Term
|
Definition
Involves examining a series of observations on some variable over time.
Instrumentation can be a threat to validity in time-series designs for two reasons:
1. Observations are usually made over a relatively long time period, which increases the likelihood of changes in measurement instruments.
2. Time-series designs often use measures that are produced by an organization, such as a police department, criminal court, probation office, or corrections department. |
|
|
Term
|
Definition
| special type of design in which a series of observations is compared before and after some intervention/treatment is introduced. |
|
|
