inferential Statistics Review

Question 1

what is a type 1 error?

Answer 1

when the null hypothesis is true but it is rejected
(false positive)

Our decision is that the population means are not equal when they actually are equal.

Question 2

what is a type 2 error?

Answer 2

when the null hypothesis is false but it is accepted
(false negative)

The population means are not equal, but the results of the experiment do not lead to a decision to reject the null hypothesis.

Question 3

what are the types of t-test and when do you use them?

Answer 3

• use t-test when comparing means between 2 groups

types:
- independent groups (T-test compare means of 2 separate groups)

• repeated measures/ paired samples t-test (Compare means of same group 2 times)

Question 4

what are the types of ANOVA and when to use them?

Answer 4

• use when comparing means for more than 2 groups

types:
- independent -groups ANOVA (comparing 3 or more separate groups)

• repeated measures/ paired samples ANOVA (comparing 3 or more repeated measures of the same group)

Question 5

when to use chi-square test?

Answer 5

use when you are comparing counts/frequencies in categories (e.g., # of ppl who prefer cats v. dogs)

Question 6

when to use correlation?

Answer 6

use when you want to see how 2 continuous variables move together (e.g., hours studied and exam score)

Question 7

how do you calculate DF?

Answer 7

for independent samples: df = N1 + N2 – 2 f

for repeated measures/ paired samples: df = N – 1

Question 8

what is Power of statistical test?

Answer 8

Probability of correctly
detecting an effect

Question 9

how to calculate power?

Answer 9

Power = 1 - β
(β = probability of a Type II error)

Question 10

what is power analysis?

Answer 10

Given an effect size (for a
particular IV in a particular situation) and level of
significance, determine N needed to detect effect

Question 11

what is the primary way researchers control power?

Answer 11

researchers control power by changing the sample size

• To increase power, increase the sample size
• To detect smaller effects, increase sample size

Question 12

what are reasons results may be non-significant even
though H0 is false (Type II error)?

Answer 12

Level of significance (α, probability of a Type I
error) is very low…Increases probability of a

Type II error
Sample size is too small for effect size

Question 13

A statistically significant result has little practical
significance when:

Answer 13

• the study has poor external validity
• the effect size is very small
• the treatment is too costly to implement
• the effect size is comparable to that for an
  existing treatment