Research Test 1 Shitt Chapter 8

Front 1

What is validity?

Back 1

The extent to which the conclusions of that research are believeable and useful.

Front 2

What is internal validity? How do you control threats?

Back 2

The results of a study demonstrate that a causal relationship exists between the independent and dependent variables. Randomized controlled trial to control threats

Front 3

What are the 11 threats to internal validity?

Back 3

Listed below

Front 4

1) History :

Back 4

events unrelated to the treatment of interest occur during the course of the study and may be change the dependent variable (ex. Handwriting)
3 strategies to minimize effects of history: planning, use a randomly selected control group, and description of unavoidable historical events.

Front 5

2) Maturation

Back 5

changes within a participant caused by the passage of time
To minimize: use of a control group, take multiple baseline measurements before treatment

Front 6

3) Testing:

Back 6

does repeated testing lead to changes in DV?
To minimize: use randomly selected experiemental and control groups so that the effects of testing in the control group can be removed by comparison with the effects of the testing and treatment in the experiemnental group, eliminate multiple testing through use of a posttest only design, conduct familiarization sessions with the testing personnel so the effects of learning (pilot study) are accounted for before the IV is manipulated.

Front 7

4) Instrumentation

Back 7

Calibration, researcher as the instrument Changes in measuring tools themselves are responsible for observed changes in the DV.

Front 8

5) Statistical regression to the mean:

Back 8

a threat when participants are selected based on extreme scores on a single administration of a test. Your GPA, grouping participants based on one particular item
To minimize: Select participants for groups based on reliable, stable measures. If measure used to form groups are inherently variable, then participants are best assigned to groups based on a distribution of scores collected over time rather than a single score (GPA again)

Front 9

6) Assignment (subject selection):

Back 9

When groups of participants are different from one another on some variable that is related to the DV of interest.

Front 10

7) Subject attrition (mortality):

Back 10

when participants are lost from the different study groups at different rates or for different reasons.
To minimize: Collect data on folks who dropped out; try to document characteristics of folks who dropped out; if these folks have the same characteristics as those who remained in the study, then differential mortality has not occurred

Front 11

9) Diffusion or Imitation of treatments:

Back 11

some participants give themselves extra treatment, they share information

Front 12

10) Compensatory equalization of Treatments:

Back 12

Researcher has preconceived idea of the experimental treatment and gives more attention to those participants who are receiving the ‘less desirable’ treatment (oh, those poor pitiful people)
Remind the researchers of the protocol; blind the researchers

Front 13

11) Compensatory Rivalry or resentful demoralization:

Back 13

Members of the group who feel they are getting the less desirable treatment
• Compensatory rivalry (we’ll show ‘em)
o May mask true differences between groups (there is no differences because the control group did more than they were supposed to)
• Resentful demoralization (why bother)
o May mask true differences between groups (greater difference than what one expected because the control group gave up and didn’t even do the minimum they were supposed to do)
• Mask participants, mask researchers, treat every one the same

Front 14

What is construct validity?

Back 14

Construct as labeled
Construct as implemented
Threats to construct validity are realized when the IV or DV within a study are not well developed or are incorrectly labeled.

Front 15

What is construct underrepresentation?

Back 15

constructs are not fully developed within a study.

Front 16

Define Experimenter Expectancies:

Back 16

A Threat to construct validity when the participants are able to guess the ways in which the experimenter wishes them to respond.

Front 17

What does interaction between different treatments mean?

Back 17

when treatments other than the one of interest are administered to participants.

Front 18

What does interaction between testing and treatment mean?

Back 18

when a test itself can be considered an additional treatment.

Front 19

Define External validity:

Back 19

Concerned with to whom, in what settings and at what times the results of research can be generalized.

Front 20

What are the treats of external validity?

Back 20

1) Selection: biased to yield participants who are in some manner different from the population to which the researcher or readers hope to generalize the results.
2) Setting: a threat when peculiarities of the setting in which the research was conducted make it difficult to generalize the results to other settings.
3) Time: a threat when the results of a study are applicable to limited time frames.