Sw Research Ch 10-11

Front 1

case-control design

Back 1

A design for evaluating interventions that compares groups of cases that have had contrasting outcomes and then collects retrospective data about past differences that might explain the difference in outcomes. It relies on multivariate statistical procedures.

Front 2

causal inference

Back 2

An inference derived from a research design and findings that logically imply that the independent variable really has a causal impact on the dependent variable.

Front 3

control variable

Back 3

A variable that is held constant in an attempt to further clarify the relationship between two other variables.

Front 4

cross-sectional study

Back 4

A study based on observations that represent a single point in time.

Front 5

distorter variable

Back 5

A variable that, when controlled in a multivariate analysis, shows that the direction of a bivariate relationship reverses.

Front 6

elaboration model

Back 6

A way to better understand the meaning of a relationship (or lack of a relationship) between two variables; done by examining multivariate frequency tables to study the effects on the original bivariate relationship that are produced by introducing additional variables into the tables.

Front 7

explanation (in the elaboration model)

Back 7

One possible result in the elaboration model that occurs when an original bivariate relationship is explained away as spurious.

Front 8

external validity

Back 8

Refers to the extent to which we can generalize the findings of a study to settings and populations beyond the study conditions.

Front 9

history

Back 9

A threat to internal validity referring to extraneous events that coincide in time with the manipulation of the independent variable.

Front 10

inference

Back 10

A conclusion that can be logically drawn in light of our research design and our findings.

Front 11

internal invalidity

Back 11

Refers to the possibility that the conclusions drawn from experimental results may not accurately reflect what went on in the experiment itself.

Front 12

internal validity

Back 12

The degree to which an effect observed in an experiment was actually produced by the experimental stimulus and not the result of other factors.

Front 13

interpretation

Back 13

A technical term used in connection with the elaboration model. It represents the research outcome in which a control variable is discovered to be the mediating factor through which an independent variable affects a dependent variable.

Front 14

maturation

Back 14

A threat to internal validity referring to aging effects or developmental changes that influence the dependent variable.

Front 15

mulitvariate analysis

Back 15

The analysis of the simultaneous relationships among several variables.

Front 16

passage of time

Back 16

A threat to internal validity referring to changes in a dependent variable that occur naturally as time passes and not because of the independent variable.

Front 17

recall bias

Back 17

A common limitation in case-control designs that occurs when a person’s current recollections of the quality and value of past experiences are tainted by knowing that things didn’t work out for them later in life.

Front 18

research design

Back 18

A term often used in connection with whether logical arrangements permit causal inferences; also refers to all the decisions made in planning and conducting research.

Front 19

selection bias

Back 19

A threat to internal validity referring to the assignment of research participants to groups in a way that does not maximize their comparability regarding the dependent variable.

Front 20

specification

Back 20

A technical term used in connection with the elaboration model, representing the elaboration outcome in which an initially observed relationship between two variables is replicated among some subgroups created by the control variable and not among others.

Front 21

spurious relationship

Back 21

A relationship between two variables that are no longer related when a third variable is controlled; the third variable explains away the original relationship.

Front 22

statistical regression

Back 22

A threat to internal validity referring to the tendency for extreme scores at pretest to become less extreme at posttest.

Front 23

suppressor variable

Back 23

A variable that, when controlled in a multivariate analysis, shows that two variables that appear unrelated or weakly related in the bivariate analysis are actually more strongly related than they appeared.

Front 24

alternative treatment design with pretest

Back 24

An experiment that compares the effectiveness of two alternative treatments. Participants are assigned randomly to two experimental groups, each of which receives a different intervention being evaluated, and to a control group that does not receive any intervention. Each group is tested on the dependent variable before and after the experimental groups receive the intervention.

Front 25

attrition

Back 25

A threat to the validity of an experiment that occurs when participants drop out of an experiment before it is completed.

Front 26

compensatory equalization

Back 26

A threat to the validity of an evaluation of an intervention’s effectiveness that occurs when practitioners in the comparison routine-treatment condition compensate for the differences in treatment between their group and the experimental group by providing enhanced services that go beyond the routine-treatment regimen for their clients, thus potentially blurring the true effects of the tested intervention.

Front 27

compensatory rivalry

Back 27

A threat to the validity of an evaluation of an intervention’s effectiveness that occurs when practitioners in the comparison routine-treatment condition decide to compete with the therapists in the other unit. They may start reading more, attending more continuing education workshops, and increasing their therapeutic contact with clients. Their extra efforts might improve their effectiveness and thus blur the true effects of the tested intervention.

Front 28

control group

Back 28

In experimentation, a group of participants who do not receive the intervention being evaluated and who should resemble the experimental group in all other respects.

Front 29

diffusion or imitation of treatments

Back 29

A threat to the validity of an evaluation of an intervention’s effectiveness that occurs when practitioners who are supposed to provide routine services to a comparison group implement aspects of the experimental group’s intervention in ways that tend to diminish the planned differences in the interventions received by the groups being compared.

Front 30

dismantling studies

Back 30

Experiments designed to test not only whether an intervention is effective, but also which components of the intervention may or may not be necessary to achieve its effects. Participants are assigned randomly to groups that either receive the entire intervention package, separate components of it, or a control condition, and are tested on a dependent variable before and after the intervention components are provided.

Front 31

experimental demand characteristics

Back 31

research participants learn what experimenters want them to say or do, and then they cooperate with those expectations.

Front 32

experimental design

Back 32

A research method that attempts to provide maximum control for threats to internal validity by: (1) randomly assigning individuals to experimental and control groups, (2) introducing the independent variable (which typically is a program or intervention method) to the experimental group while withholding it from the control group, and (3) comparing the amount of experimental and control group change on the dependent variable.

Front 33

experimental group

Back 33

In experiments, a group of participants who receive the intervention being evaluated and who should resemble the control group in all other respects.

Front 34

experimental mortality

Back 34

A threat to the validity of an experiment that occurs when participants drop out of an experiment before it is completed.

Front 35

experimenter expectancies

Back 35

Research participants learn what experimenters want them to say or do, and then they cooperate with those “demands” or expectations. !!

Front 36

interrupted time-series with a nonequivalent comparison group time-series design

Back 36

The most common form of multiple time-series design, in which an experimental group and a control group are measured at multiple points in time before and after an intervention is introduced to the control group.

Front 37

intervention fidelity

Back 37

The degree to which an intervention being evaluated is actually delivered to clients as intended.

Front 38

matching

Back 38

In connection with experiments, the procedure whereby pairs of subjects are matched on the basis of their similarities on one or more variables, and one member of the pair is assigned to the experimental group and the other to the control group.

Front 39

multiple time-series designs

Back 39

A form of time-series analysis in which both an experimental group and a nonequivalent comparison group are measured at multiple points in time before and after an intervention is introduced to the experimental group.

Front 40

nonequivalent comparison groups design

Back 40

A quasi-experimental design in which the researcher finds two existing groups that appear to be similar and measures change on a dependent variable before and after an intervention is introduced to one of the groups.

Front 41

novelty and disruption effects

Back 41

A form of research reactivity in experiments in which the sense of excitement, energy, and enthusiasm among recipients of an evaluated intervention—and not the intervention itself—causes the desired change in their behavior.

Front 42

obtrusive observation

Back 42

This occurs when the participant is keenly aware of being observed and thus may be predisposed to behave in socially desirable ways and in ways that meet experimenter expectancies.

Front 43

one-group pretest/-posttest design

Back 43

A pre-experimental design, with low internal validity, that assesses a dependent variable before and after a stimulus is introduced but does not attempt to control for alternative explanations of any changes in scores that are observed.

Front 44

one-shot case study

Back 44

A pre-experimental research design, with low internal validity, that simply measures a single group of subjects on a dependent variable at one point in time after they have been exposed to a stimulus.

Front 45

placebo control group design

Back 45

An experimental design that controls for placebo effects by randomly assigning subjects to an experimental group and two control groups and exposing one of the control groups to a stimulus that is designed to resemble the special attention received by subjects in the experimental group.

Front 46

placebo effects

Back 46

Changes in a dependent variable that are caused by the power of suggestion among participants in an experimental group that they are receiving something special that is expected to help them. These changes would not occur if they received the experimental intervention without that awareness.

Front 47

posttest-only control group design

Back 47

A variation of the classical experimental design that avoids the possible testing effects associated with pretesting by testing only after the experimental group receives the intervention, based on the assumption that the process of random assignment provides for equivalence between the experimental and control groups on the dependent variable before the exposure to the intervention.

Front 48

posttest-only design with nonequivalent groups

Back 48

A pre-experimental design that involves two groups that may not be comparable, in which the dependent variable is assessed after the independent variable is introduced for one of the groups.

Front 49

pre-experimental designs

Back 49

Pilot study designs for evaluating the effectiveness of interventions; they do not control for threats to internal validity.

Front 50

pretest\-posttest control group design

Back 50

The classical experimental design in which subjects are assigned randomly to an experimental group that receives an intervention being evaluated and to a control group that does not receive it. Each group is tested on the dependent variable before and after the experimental group receives the intervention.

Front 51

quasi-experimental design

Back 51

Design that attempts to control for threats to internal validity and thus permits causal inferences but is distinguished from true experiments primarily by the lack of random assignment of subjects.

Front 52

randomization

Back 52

A technique for assigning experimental participants to experimental groups and control groups at random.

Front 53

randomized clinical trials (RCTs)

Back 53

Experiments that use random means (such as a coin toss) to assign clients who share similar problems or diagnoses into groups that receive different interventions. If the predicted difference in outcome is found between the groups, it is not plausible to attribute the difference to a priori differences between two incomparable groups.

Front 54

reactivity

Back 54

A process in which change in a dependent variable is induced by research procedures.

Front 55

replication

Back 55

(1) Generally, the duplication of a study to expose or reduce error or the reintroduction or withdrawal of an intervention to increase the internal validity of a quasi-experiment or single-case design evaluation. (2) One possible result in the elaboration model elaboration model that occurs when an original bivariate relationship appears to be essentially the same in the multivariate analysis as it was in the bivariate analysis.

Front 56

research reactivity

Back 56

A process in which change in a dependent variable is induced by research procedures.

Front 57

resentful demoralization

Back 57

A threat to the validity of an evaluation of an intervention’s effectiveness that occurs when practitioners or clients in the comparison routine-treatment condition become resentful and demoralized because they did not receive the special training or the special treatment. Consequently, their confidence or motivation may decline and may explain their inferior performance on outcome measures.

Front 58

selection bias

Back 58

A threat to internal validity referring to the assignment of research participants to groups in a way that does not maximize their comparability regarding the dependent variable.

Front 59

simple interrupted time-series design

Back 59

A quasi-experimental design in which no comparison group is utilized and that attempts to develop causal inferences based on a comparison of trends over multiple measurements before and after an intervention is introduced.

Front 60

Solomon four-group design

Back 60

An experimental design that assesses testing effects by randomly assigning subjects to four groups, introducing the intervention being evaluated to two of them, conducting both pretesting and post testing on one group that receives the intervention and one group that does not, and conducting post testing only on the other two groups.

Front 61

static-group comparison design

Back 61

A cross-sectional design for comparing different groups on a dependent variable at one point in time. The validity of this design will be influenced by the extent to which it contains multivariate controls for alternative explanations for differences among the groups.

Front 62

time-series designs

Back 62

A set of quasi-experimental designs in which multiple observations of a dependent variable are conducted before and after an intervention is introduced.

Front 63

unobtrusive observation

Back 63

Unlike in obtrusive observation, the participant does not notice the observation and is therefore less influenced to behave in socially desirable ways and ways that meet experimenter expectancies.