Simple Experiments

Question 1

internal validity

Answer 1

Making sure the IV is the only systematic source of group differences on the DV

Question 2

external validity

Answer 2

how well its findings generalize to other samples and procedures & potentially even to “real-world” settings

Question 3

Is internal or external validity more important?

Answer 3

Internal validity
Researchers tend to care more about internal validity because while it is important for a researcher to be able to generalize their results (external validity) if internal validity is weak it is hard to draw conclusions

Question 4

**Two crucial features of an experiment:

Answer 4

Random assignment
Manipulation of the Independent variable while controlling for all other variables

Question 5

What are the two types of non-random assignments?

Answer 5

Arbitrary assignment (using pre-existing grouping)
Self assignment (letting participants choose their condition)

Question 6

What is a problem related to arbitrary assignment and/or self assignment?

Answer 6

A problem occurs if the participants drop out of experiments in non-random ways
e.g., if participants with one level of IV are more likely to drop out than those with another; this might inflate or cancel out the group differences in the DV

Question 7

experimental expectancies:

Answer 7

researchers behave differently towards participants or assess them differently depending on which condition they are in

Question 8

Demand characteristics (participant expectancies):

Answer 8

participants might perceive some cues as to how they’re “supposed” to behave, and these cues can differ between conditions

Question 9

why are “double blind” experiments used?

Answer 9

they are set up so neither the researcher nor participants know which condition a given participant is in.
ensure experimenter expectancies and demand characteristics, and placebo effects wont influence the groups differently

Question 10

confounds due to improper control:

Answer 10

If groups are systematically treated differently in some way other than level of IV

Question 10

Examples of confound variables:

Answer 10

Placebos
demand characteristics
experiment expectancies
groups differ in experimenter
groups differ in setting
groups differ in day or time
groups differ in treatment prior to experiment

Question 11

Placebos:

Answer 11

participants who think they’re in a treatment condition may be affected by it simply because they believe it will

Question 12

If conditions differ not just in _________ but also in ________ for _________ then any effects on DV might be due to that _________

Answer 12

If conditions differ not just in treatment but also in vehicle for treatment then any effects on DV might be due to that vehicle
** this is why you need a condition in which participants get only the placcebo**

Question 13

By avoiding what problems should the DV be due to the IV itself (internal validity)

Answer 13

• non-random assignment
• confounds due to improper control
• non-independent observations (participants influencing each other’s behaviour

Question 14

If a sample is not ________ of the intended population, then results might not ________ to the population

Answer 14

If a sample is not representative of the intended population, then results might not generalize to the population

Question 15

“ecological validity”:

Answer 15

highly artificial experimental settings can make effects seem stronger than in the real world
(experiment can still be useful for showing effects though)