Chapter 9 - Final

Question 1

Experimental Designs

Answer 1

• Experiments
• Quasi-experiments
• Non-experiments

Question 2

Types of Variables

Answer 2

• Independent (IV)
• Dependent (DV)
• Random - likely with random techniques, we hope they will even out
• Control - might affect our results, however,
  awareness of them can mitigate their effects.
• Extraneous (Disturbance) -
  a. Confounding or intervening variables
  i. not found until the study is in process and unable to control
  ii. usually difference between the levels of IV other than DV
• Moderator or Mediator -
  a. Moderators affect the strength of the effect
  b. Mediators (intervening variables) get in the way of effects

Question 3

Types of Variation

Answer 3

• Non-systematic - Error (Noise)
• Systematic - Treatment (variation due to IV)
• Confounds & Artifacts - variation due to variables other than IV
  Examples: Demand Characteristics, Response
  sets, Bias, Threats to Internal Validity

Question 4

Null Hypothesis

Answer 4

The alternate hypothesis (either reject or fail to reject)
H1(Hypothesis)
H0(Null Hypothesis)

Question 5

Reject H0 Null Hypothesis

Answer 5

The significant effect found, therefore we reject the null hypothesis

Question 6

Fail to reject H0 Null Hypothesis

Answer 6

No effect found (Not significant), therefore we fail to reject the null hypothesis

Question 7

Type I Error

Answer 7

False Positive [found something that is not there]
specify how vulnerable you will be choosing your significance level
(p < .05 means, p < .01 means, p < 0?)

Question 8

Type II Error

Answer 8

False Negative [failed to find something that is there]
To reduce the likelihood of Type II error,
a. reduce random error
i. use reliable measures and standardized procedures
ii. carefully code data
iii. use a homogenous group of participants

Question 9

Examples of changing your alpha

Answer 9

• lowering for risk of drugs that work
• raising for safety claims (anything that is an potential risk)

Question 10

Power

Answer 10

Probability of rejecting the null hypothesis
When it is false, increase by lowering the Beta and increasing the probability that you are not rejecting the null hypothesis.
a. If B is too high, power is low
b. if B is too low, power is high

   Power = 1-Beta

Question 11

Trade-offs between Type I and II errors

Answer 11

• Overlooked effects, “Mirages,” Making false claims
• Risks of convicting an innocent person vs. letting off a guilty person
• Burglar Alarms