Chapter 13

Question 1

significant result

Answer 1

low probability of occurring by chance (doesn’t automatically mean it’s important)

Question 2

hypothesis testing

Answer 2

can be used to reveal differences between µhyp may and µtrue that are large enough that we care about them

Question 3

effect size

Answer 3

The magnitude of the difference between µhyp may and µtrue expressed in standard deviation unit

Question 4

cohen’s d

Answer 4

measure of effect size using X bar instead of the mean (because we don’t know what mean is)

Question 5

hedge’s g

Answer 5

measure of effect size used when we don’t know what the stdv is

Question 6

small/med/lrg effect sizes (d and g)

Answer 6

Small: .2
Medium: .5
Large: .8

Question 7

alpha

Answer 7

probability of making a type 1 error

Question 8

beta

Answer 8

• probability of making a type 2 error

- Increased chance of doing this when your alpha level is very low (ie. 0.001)

Question 9

power

Answer 9

• probability of correctly rejecting a false null hypothesis
• Any condition that increases beta increases power

Question 10

conditions that affect power

Answer 10

• Greater the effect size (different between utrue and uhyp), greater power
• Larger the sample size, smaller standard error of mean, greater power (simplest method for increasing power)
• The less variability, the smaller stdv, greater power
• The greater the alpha level, greater power (increased risk of making type 1 error decreases the risk of a type 2 error)
• One-tailed test, greater power

Question 11

statistical conclusion vs. research conclusion

Answer 11

• Statistical conclusion: says something about the parameter of population/data
• Research conclusion: says something about the subject matter - the meaning of the study