Lecture 3 - Threats to Validity

Question 1

External Validity

Answer 1

The extent to which research can be applied to situations outside of the experimental setting.

Question 2

Internal Validity

Answer 2

The extent to which effects can be reliably connected with the manipulation of the independent variable.

Question 3

Construct Validity

Answer 3

The extent to which selected measures map onto the theoretical constructs they are purported to be measuring.

Question 4

Treatment Variance

Answer 4

Variance among scores traceable to our own methodology.

Question 5

Confound Variance

Answer 5

Variance among scores caused by uncontrolled variables.

Question 6

Error Variance

Answer 6

Variance among scores caused by extraneous variables and measurement errors.

Question 7

Ceiling Effect

Answer 7

An error of measurement sensitivity in which an artificial upper limit on scores skews the distribution of results; in certain tests, if the task is too easy for participants, an artificial upper limit is created, and many achieve perfect scores.

Question 8

Floor Effect

Answer 8

An error of measurement sensitivity in which an artificial lower limit on scores skews the distribution of results; in certain tests, if the task is too difficult for participants, an artificial lower limit is created, and many achieve scores near zero.

Question 9

Order Effects

Answer 9

A potential confound that results from the order in which an experimental procedure is carried out.

Question 10

Practice Effects

Answer 10

The changes in performance resulting from familiarity with the task being completed in which participants are seen to improve over time.

Question 11

Fatigue Effects

Answer 11

The changes in performance resulting from overexposure to dull or tedious tasks in which participants become bored.

Question 12

Counterbalancing

Answer 12

A method designed to reduce order effects; one group experiences the conditions in one particular order, while another experiences them in the opposite order.

Question 13

Latin-Square Design

Answer 13

A technique used to reduce order effects by partially counterbalancing the conditions of an experiment; each condition is assigned a position in a column and row where each only occurs once in their respective column and row. Individual participants are assigned their own row and experience that particular sequence of conditions.

Question 14

Randomization

Answer 14

A technique to limit the likelihood of order effects emerging; arbitrarily organizes stimuli and trials.

Question 15

Demand Characteristics

Answer 15

Cues within a study that are intended to help participants fulfill their role in the experiment; may potentially alter the behavior of participants as a result of perceived expectations.

Question 16

Pre-Test, Experimental Treatment, and Post-Test

Answer 16

A pre-test consists of measuring a baseline for the variable of interest before an experiment begins.

An experimental treatment refers to the condition of experiencing the independent variable.

A post-test consists of a re-assessment following the experimental treatment.

Question 17

Maturation Effects

Answer 17

Refers to any of the natural biological or intrapsychic processes that change over time and pose a threat to the internal validity of an experiment.

Question 18

History Effects

Answer 18

Refers to any changes in the circumstances surrounding the lives of individual participants outside of an experiment that may influence its outcome.

Question 19

Testing Effects

Answer 19

Refers to any effect of taking a test on respondents. It could refer to the sophistication of a test; it could also refer to the influence of having taken a pre-test on re-assessment during the post-test.

Question 20

Passive Control Group

Answer 20

A control group involved in the completion of a meaningless alternative task.

Question 21

Active Control Group

Answer 21

A control group involved in something that could be reasonably assumed to have an effect.

Question 22

Wait List

Answer 22

A control group that believes they will receive the treatment at some point later.

Question 23

Attrition

Answer 23

Refers to when participants start but do not complete a study. It is wise to be particularly mindful of differential attrition in which participants withdraw from one condition or treatment more than any other.

Question 24

Matched Design

Answer 24

Design in which participants are matched as closely as possible on some characteristics and then assigned to separate groups.

Question 25

Random Group Assignment

Answer 25

Participants are allocated to groups by some method that involves chance.