Exam 2

Question 1

Sampling Bias

Answer 1

Occurs when a sample was not randomly selected from the population

Question 2

Selection Bias

Answer 2

Participants were not selected at random

Question 3

Nonresponse Bias

Answer 3

Occurs when respondents differ in meaningful ways from nonrespondents

Question 4

Quota Sampling

Answer 4

By time zone and region

Question 5

Sampling Error

Answer 5

The sample collected is not representative of the population due to chance alone. Outliers get absorbed, so the spread gets much tighter in a sampling distribution. Always tighter and smaller range

Question 6

Standard Error

Answer 6

Measures the variability. Can never be larger than the SD. Formula is SE= SD/SR of n (square root)

Question 7

95% Margin of Error (MOE)

Answer 7

If we sampled repeatedly, 95% of these intervals would contain the true population mean. Formula is mean +/- 2 SE

Question 8

Null Hypothesis

Answer 8

Nothing happened (H0)

Question 9

Alternative Hypothesis

Answer 9

Something happened (Ha)

Question 10

Conformation Bias

Answer 10

A tendency for people to favor information that confirms their hypothesis

Question 11

Type 1 Error

Answer 11

Null is true, reject the null (false alarm)

Question 12

Type 2 Error

Answer 12

Null is false, refuse to reject the null (miss and/or false negative)

Question 13

P-Value

Answer 13

Probability of a false alarm. If the p-value is low, we should reject the null

Question 14

P < .05 means what?

Answer 14

Reject null (statistically significant)

Question 15

P > .05 means what?

Answer 15

Failed to reject null (nonsignificant)

Question 16

Cofounding Variables

Answer 16

Has to change with the IV to affect the DV. It is an extraneous variable.

Question 17

We can get really small p-values by…

Answer 17

Having large sample sizes

Question 18

Replication Crisis

Answer 18

If you found a significant effect, someone else should be able to replicate it; however, many studies can’t be replicated

Question 19

Reasons why replication crisis exists:

Answer 19

Publication bias: journals only want to publish significant results.
“Publish or perish” mentality: researchers wouldn’t have a job if they didn’t publish results

Question 20

P-Hacking

Answer 20

Making the p-value smaller than it actually is. May not be intentional, and is due to common practices in psychology.

Question 21

Effect sizes quantify the strength of a relationship between variables because…

Answer 21

It tells us how meaningful the effect is, is unaffected by sample size, and can help mitigate the replication crisis

Question 22

Between-Group Comparisons

Answer 22

Different people in the same group, and group size is independent of each other

Question 23

Within-Group Comparisons

Answer 23

Same person throughout time, and groups are dependent on each other

Question 24

Independent Samples T-Test

Answer 24

Used for between-group comparisons
Formula is t= m1+m2/SE

Question 25

Quality Control Testing

Answer 25

Allowed only small samples to work with, used to calculate p-values, and uses degrees of freedom. Formula for df is df=n1+n2-2

Question 26

Trick question on quiz:

Answer 26

If all you see is a figure, do a statistical test! If only figure, not enough info

Question 27

Significant Differences

Answer 27

Differences that exist are greater than what we would expect from chance alone

Question 28

APA Format for statistical results with t-tests:

Answer 28

t (test value) =df(degrees of freedom), p (p-value), d (effect size)

Question 29

APA format for statistical results with ANOVA:

Answer 29

f (test value) = df (degrees of freedom for both IVs), p (p-value), n2p (effect size)

Question 30

Repeating t-tests on the same data set increases the risk of false alarm