Exam 2

Question 1

What is a hypothesis?

Answer 1

A testable prediction

Question 2

What is hypothesis testing?

Question 3

What are the four steps of hypothesis testing?

Answer 3

1. state the hypothesis in terms of a population
2. Set the criteria for a decision
3. Collect data and compute sample stats
4. Make a decision based on the obtained sample data with the prediction that was made earlier.

Note two of these (1, 2) occur before collecting data. This is a criticism of hypothesis testing.

Question 4

State the hypothesis in terms of _________________.

Answer 4

Population parameters

Question 5

What is a null hypothesis?

Answer 5

states that there is no effect, no change, no difference (nothing happened)

Question 6

What is the notation for a null hypothesis?

Answer 6

H0

Question 7

What is an alternative hypothesis?

Answer 7

states that there is a change, difference, or relationship for the general population; predicts that the independent variable will have an effect on the dependent variable

Question 8

What is the notation for alternative hypotheses?

Answer 8

H1

Question 9

Why do we avoid “prove” language in scientific writing?

Answer 9

Because we don’t have definitive “proof” of anything; only have support for hypothesis

Question 10

In hypothesis testing, we test the ____________________ not the ______________.

Answer 10

Null hypothesis (not the alternative)

Question 11

Distribution of sample means is divided into:

Answer 11

1) sample means that are likely if H0 is true
2) sample means that are very unlikely if H0 is true

Question 12

What is an Alpha level (a)

Answer 12

A probablility that is used to identify the “very unlikely” values

Question 13

What is the most common alpha value?

Answer 13

a=.05

Question 14

What is the critical region?

Answer 14

it’s composed of the extreme sample values that are very unlikely

Question 15

Boundaries of the critical region are dtermined by the ___________.

Answer 15

Alpha value

Question 16

The critical region is determined by

Question 17

What would be the critical value at a=.01?

Question 18

What would the critical value be at a=.001?

Question 19

How do you determine the critical region? and what is this value called?

Question 20

What does a z-score of 4 mean?

Answer 20

It is very extreme, on the furthest end of the distribution

Question 21

What do you do if the sample data falls within the critical region?

Answer 21

Reject the null

Question 22

The critical region is where on the distribution?

Answer 22

Not the middle, on the edges

Question 23

What do you do if the sample dataa does not fall in the critical region?

Answer 23

fail to reject the null

Question 24

A result is ______ or ______ when the result is sufficient to reject the null hypothesis. This is when the _______ is in the ____________.

Answer 24

significant; statistically significant

z-score, critical region

Question 25

Remember that you reject the null hypothesis with _______________, ____________ located in the _______________ (or __________ of the distribution).

Answer 25

extreme, low probability values critical region; or tails of the distribution

Question 26

A significant result that _________ the null hypothesis corresponds to p<.05.

Answer 26

rejects

Question 27

DO THIS EXAMPLE: Let’s say a researcher wants to examine the effect of prenatal alcohol on birth weight.  Pregnant rats are given daily doses of alcohol and at birth one pup is selected from each litter to produce a sample of  n = 16 newborn rats. The average weight for this sample is M = 15 grams. The researcher wants to compare the sample mean to the mean weight of the general population of baby rats. It is known that regular newborn rats (not exposed to alcohol) have an average weight of μ = 18 grams.  The distribution of weights is normal with σ = 4 grams. 

Question 28

Explain the difference between a one-tailed and a two-tailed test.

Question 29

A one-tailed hypothesis test is also known as a ____________ hypothesis.

Answer 29

directional

Question 30

When are one-tailed tests used?

Question 31

In a _____________ test, H0 and H1 specify either an increase or decrease in the population mean score.

Answer 31

one-tailed

Question 32

In one-tailed tests, what are the two changes from the two-tailed testing procedure?

Answer 32

1. directional prediction is incorporated into the statement of the hypothesis 2. the critical region is located entirely in one tail of the distribution (after this, the rest of the one-tailed test proceeds exactly as the two-tailed test --- calculate z-score stat then make a decision about H0, depending on whether or not the z-score is in the critical region).

Question 33

In a one-tailed distribution the critical region is ___________.

Answer 33

On only one end of the distribution (the one specified in H1). The whole 5% is on that side (not split between sides like in 2-tailed). If the finding is on the extreme end of the opposite side, you will still reject the null.

Question 34

When the variability for a population is not known, we use ---------- in its place. ## Footnote 11_T-Statistic and Type I and Type II Error

Answer 34

sample variability

Question 35

Degrees of Freedom | Definition

Answer 35

describe the number of scores in a sample that are independent and free to vary

Question 36

What is the shortcoming of the Z statistic? ## Footnote 11_T-Statistic and Type I and Type II Error

Answer 36

A z-score requires that we know the value of the population standard deviation, which is usually not known. Therefore, we can’t compute the standard error and the z-score.​

Question 37

What typically happens if we don't make the degrees of freedom correction?

Answer 37

sample variability tends to underestimate the population variability

Question 38

Dividing by a smaller number (n-1) produces a --------- result and makes sample variability a ------------------------- estimator of population variability.

Answer 38

larger; more accurate/unbiased

Question 39

Used as an estimate of the real standard error when the value of standard deviaion is unknown. ???????

Question 40

Estimated standard error is computed from the ----------------- or -----------------.

Answer 40

sample variance; standard deviation

Question 41

What is estimated standard error?

Answer 41

provides an estimate of the standard distance between a sample mean and the population mean

Question 42

LEFT OFF ON SLIDE 7 IN 11_T-STATISTIC IN TYPE I AND II ERROR

Question 43

---------- is used as an estimate of the real standard error when the value of σ is unknown.

Answer 43

estimated standard error

Question 44

Estimated standard error is computed from the ------- or -------.

Answer 44

sample variance; standard deviation

Question 45

When we have a z-score, we have --- data.

Answer 45

population

Question 46

In principle, a t-score is ------ to the z-score.

Answer 46

parallel

Question 47

Substitute the estimated standard error in the denominator of the z-score formula. The result is a new test statistic called the -------.

Answer 47

t-statistic

Question 48

When t approaches infinity, it looks just like ------------.

Answer 48

normal distribution (or distribution of the z-score)

Question 49

When df gets larger, the wings will ---------- and the t-value (proportion in the tails) will -------.

Answer 49

shrink down; decrease

Question 50

How is the sample standard deviation different than the population standard deviation calculated?

Answer 50

The sample standard deviation uses degrees of freedom (n-1) instead of n

Question 51

What are two assumptions of the t test?

Answer 51

1. the values in the sample must conssit of **independent observations** 2. The population sampled must be **normal** ## Footnote slide 27

Question 52

Two observations are independent if the occurrence of the first event ---------------- probability of the second event. ​

Answer 52

has no effect on the

Question 53

The assumption that the ---------------- is a necessary part of the mathematics underlying the development of the t statistic and the t distribution table.

Answer 53

population sample must be normal

Question 54

When we use a sample to draw inferences about a larger population, there is ----

Answer 54

always the possibility that an incorrect conclusion will be made (e.g. there’s a chance that the sample was not representative of the population (even with random sampling); the sample size may not have been large enough) ## Footnote Slide 28 - Type 1 and II errors

Question 55

Type I error

Answer 55

When you reject the null hypothesis (N0), but the null hypothesis is true | You're saying that you found an effect but there is actually not one

Question 56

Type II Error

Answer 56

Fail to reject the null hypothesis (H0) but the null hypothesis is false

Question 57

When the data lead you to reject the null hypothesis when the treatment actually has no effect

Answer 57

Type I error

Question 58

when a treatment effect really exists but the hypothesis test fails to detect it

Answer 58

Type II error

Question 59

The consequences of a ------- error can be serious. ## Footnote Slide 31

Answer 59

Type I

Question 60

----------- minimizes the risk of a Type I error. ## Footnote Slide 31

Answer 60

hypothesis testing