Validity & Control Unit 3

Question 1

Internal validity and what’s its biggest threat?

Answer 1

(FUNDAMENTAL) study provides evidence (logic) of a cause-effect relationship between the independent and dependent variable. (Independent variable changes dependent variable)

You strive to rule out alternative variables as potential causes of the behavior

Biggest threat - confounding validity (THIMMS)

Question 2

What are the 4 types of validity (designing & evaluating) ICES

Answer 2

1. Internal validity
2. Construct validity
3. External validity
4. Statistical conclusion validity

Question 3

Confounding validity and what are the different types of threats?

THRRMMS

Answer 3

Biggest threat to (internal) validity - error that occurs when the effects of 2 variables in experiment cannot be separated, resulting in confused interpretation

1. History - specific events that occur outside of the experiment that could influence results (time of day/week, weather)
2. Maturation - changes in an individual’s internal conditions (biological & psychological) that occurs over time and causes error in experiment
3. Testing (repeated test) - performance on a second test is influenced by simply having taken a first test (become less anxious)
4. Random error - the value of a variable that can be attributed to chance
5. Selection bias - different selection procedures are used for placing subjects in the various groups
6. Mortality - subjects dropping out at different rates
7. Ambiguous temporal precedence (which comes before the other?)
8. Regression effect - tendency of subjects with extreme scores on a first measure to score close to the mean on a second testing

Question 4

Construct validity and what’s its biggest threat?

How to determine if a test has CV?

Answer 4

A test actually measures what it is designed to measure

You must rule out other possible theoretical explanations of the results

Biggest threat -

1. loose connection between theory and method (using nail biting as sign of anxiety)
2. ambiguous effect of independent variable (some may see the situation the same as experimenter and their expectations may change the results)

How to determine for construct validity?

• Manipulation check, to see if its working the way you think it is
• Use various measurements of the construct to determine if it is being altered

Question 5

Extraneous validity vs. Confounding validity

Answer 5

(both LOWERS internal validity)

(extraneous) any factors that are IN the experiment, but are not being studied
VS.
(confounding) –> related to independent variables AND effect the dependent variable

Question 6

External validity & biggest threats??

Answer 6

Extent in which results can be generalized to another study using different subject, setting, time, etc.

Biggest threats -

1. other subjects - findings may not apply to everyone (students + rats)
2. other times - society changes over time, thus people’s reactions to things change too(conduct similar results?)
3. other settings - lab work may not be the same as natural settings

Question 7

Statistical conclusion validity

Answer 7

(similar to internal) did the IV cause change in the DV OR was the result accidental?

Question 8

What is the most important control strategy and name other strategies

Answer 8

Randomization - ensure that extraneous variables will not systematically bias the results.

Other:

• Matching: matching subjects in the various experimental groups **can increase the sensitivity of an experiment
• Building: building the extraneous variable into the research design to allow any effects of that variable to be examined

Question 9

2 broad categories of bias?

Answer 9

1. Role demands - participant’s expectations of what an experiment requires them to do

To overcome: keep the real aim of the experiment a secret

2. Experimenter bias - when an experimenter’s wishes or expectations influence the results of a study

To overcome: keep experimenters blind to the conditions

Question 10

Control

Answer 10

Any means to rule out possible threats to the validity of a piece of research

Question 11

What are the 2 meanings of control?

Answer 11

1. Control experiment - allows one to conclude that a dependent variable is associated with an independent variable, and no other
2. Experimental control - facilitates this conclusion (above) by limiting the number of variables operating in the situation

Question 12

Control group –> between-subjects

Answer 12

subjects in a between-subjects design experiment who are like the experimental group in every aspect except they do NOT receive treatment

Question 13

Between-subjects

Answer 13

each subject experiences only one of the conditions in the experiment

Question 14

Control condition –> within-subjects

Answer 14

condition in a within-subjects design that does not contain experimental manipulation

Question 15

Within-subjects

Answer 15

each subject experiences every condition in experiment

Question 16

Replication

Answer 16

Repeating of an experiment to see if results are the same

Question 17

How does one (internal/external) validity increase or decrease the other?

Answer 17

When you INCREASE internal validity you control more variables making it specific to the situation.

To REDUCE the chance of confounding variables, you must control more variables making it less generalizable.

Therefore, the more you INCREASE internal validity, you decrease external validity.

Question 18

What are the different types validity in relation to tests and measurement

Answer 18

Construct validity - measures what its designed to measure
Face validity - test should APPEAR to test what its supposed to test
Content validity - test should sample the RANGE of behavior that is represented
Criterion validity - should relate closely to other measures of the same construct