Deck 9: The Basics of Experimentation and Testing for a Difference Between Means

Question 1

To decide that the independent variable is responsible for the difference between the means of two groups in an experiment, it is necessary to determine the.

A)difference expected from sampling error alone
B)effect expected from the dependent variable
C)variation due to the level of alpha
D)difference due to the independent variable

Answer 1

A

Question 2

Sampling error implies that.

A)mistakes always will be made in selecting participants for an experiment
B)the means for samples selected from the same population will differ by some chance amount
C)the means for samples selected from the same population will always be equal
D)we can decide if an independent variable has an effect by visually inspecting the means of the different groups in an experiment

Answer 2

B

Question 3

Equivalent groups typically are formed by.

A)selecting two or more samples randomly from the population
B)ensuring that the proportions of males and females are the same in each group
C)matching participants for extraneous characteristics
D)randomly assigning participants to groups

Answer 3

D

Question 4

Extraneous variables in an experiment can affect the.

A)size of the population
B)size of the needed sample
C)dependent variable
D)independent variable

Question 5

The dilemma facing the experimenter is determining how large a difference between the sample means is enough to decide that something other than is at work in the experiment.

A)the independent variable
B)a treatment effect
C)sampling error
D)systematic variation

Question 6

A simulated treatment condition in an experiment is called a(n) control.

A)simulated
B)experimenter
C)confounded
D)placebo

Question 7

An independent variable must have at least levels.

A)one or more
B)two or more
C)three or more
D)four or more

Question 8

Equivalent groups are expected to.

A)be exactly the same in every way
B)not differ in any systematic way before being given the independent variable
C)represent different populations
D)have bimodal distributions

Question 9

A placebo control group in an experiment would.

A)not be administered the dependent variable
B)receive a simulated treatment condition
C)be randomly selected but not randomly assigned
D)not be considered an equivalent group

Question 10

If two groups are expected to differ only by chance before the independent variable is given, then the groups are considered to be.

A)separate but equal
B)identical
C)equivalent
D)matched

Question 11

When two or more groups are used in an experiment, the number of subjects in each group is indicated by.

A)N
B)#S
C)nA
D)NS

Question 12

When two or more groups are used in an experiment, the total number of subjects in the experiment is indicated by.

A)N
B)#S
C)nA
D)NS

Question 13

In actual experimentation, random sampling rarely is used. Rather, sampling typically is used.

A)stratified
B)clustered
C)occasional
D)convenience

Question 14

Obtaining participants for an experiment by asking people accessible to the experimenter to participate is called sampling

A)simple random
B)valid
C)convenience
D)experimental

Question 15

Which of the following occurs first when performing an experiment?

A)Predicting a relationship between an independent variable and a dependent variable.
B)Selecting the level of alpha.
C)Identifying extraneous variables.
D)Forming equivalent groups.

Question 16

Random assignment of participants requires that.

A)every person selected for the experiment has an equal chance of being assigned to any one of the treatment groups
B)the selection of one person to participate in the experiment is independent of the selection of any other person
C)all members of a population have an equal chance of being selected to participate in the experiment
D)all the above

Question 17

Variables other than the independent variable that can affect the dependent variable are called variables.

A)distracter
B)interrupter
C)intervening
D)extraneous

Question 18

An experiment in which one independent variable is manipulated and two equivalent groups of subjects are formed is using a design.

A)one-factor within-subjects
B)one-factor between-subjects
C)two-factor within-subjects
D)two-factor between-subjects

Question 19

In an experiment designed to measure the effect of a drug on behavior, one group may be given a medication with no active ingredient; this group is identified as a(n) control group.

A)dependent variable
B)confounded
C)experimental treatment
D)placebo

Question 20

An experimenter manipulates the independent variable of the difficulty of a list of words a person is asked to memorize. One group of participants is given an easy list to memorize, the second group of participants is given a difficult list of words to memorize. The levels of this independent variable are.

A)list difficulty
B)the number of participants in each group
C)easy list and difficult list
D)the number of words memorized

Question 21

In a directional tind test, a rejection region is located in of the sampling distribution of t.

A)the left tail
B)the right tail
C)either the left tail or the right tail, but not both tails
D)both tails

Question 22

The standard error of the difference between means is given by

Question 23

The numerator of tind is.

A)a measure of error variation
B)a measure of how much two sample means differ from each other
C)the standard error of the difference between means
D)the sampling distribution of the difference between means

Question 24

The distribution of the difference between two sample means when all possible pairs of samples of size n are selected from a population is called a.

A)standard error of the sampling distribution
B)sampling error distribution
C)t distribution
D)sampling distribution of the difference between means

Question 25

The null hypothesis for tind is H0:.

Question 26

The standard error of a theoretical sampling distribution of differences between means is .
A
B)
C)the standard deviation of the distribution of differences between sample means
D)the same as s

Question 27

In a nondirectional tind test, a rejection region is located in of the sampling distribution of t.

A)the center
B)the left tail
C)the right tail
D)both tails

Question 28

The degrees of freedom for tind are found by.

A)N
B)N - 1
C)N + 2
D)N - 2

Question 29

The denominator of tind is.

A)the standard error of the difference between means
B)the difference between the population means
C)a measure of how much two sample means differ from each other
D)the sampling distribution of the difference between means

Question 30

In practice, the standard error of the difference between means is estimated by.

Question 31

For tind, H0.

Question 32

When rejection regions are located at both ends of the sampling distribution for t, a test is used.

A)two-tailed
B)parametric
C)directional
D)one-tailed

Question 33

The formula for tind is.

Question 34

If a control group has 20 participants and an experimental group also has 20 participants, then the degrees of freedom for tind on the two groups are.

A)40
B)38
C)20
D)19

Question 35

The alternative hypothesis for tind is H1:.

Question 36

If group A1 has 14 participants and group A2 has 12 participants, then tind on the two groups would have degrees of freedom.

A)12
B)14
C)24
D)26

Question 37

If H0 is rejected when tind falls into either the extreme left or extreme right region of the sampling distribution, then it is called a test.

A)one-tailed
B)two-tailed
C)directional
D)bi-directional

Question 38

The sampling distribution of the difference between means.

A)is positively skewed and unimodal
B)approaches a normal distribution as sample size increases
C)is bimodal with the standard error dependent upon sample size
D)is negatively skewed and unimodal

Question 39

The tind is appropriate for analyzing data from a one-factor design using levels of an independent variable.

A)within-subjects; two
B)within-subjects; three or more
C)between-subjects; two
D)between-subjects; three or more

Question 40

The rejection region for tind represents the values of t that have a probability alpha if .

A)greater than; H0 is true
B)greater than; H1 is false
C)less than or equal to; H0 is true
D)less than or equal to; H1 false

Question 41

The power of tind can be increased by.

A)maximizing the effect of the independent variable
B)minimizing the effect of the independent variable
C)increasing the value of s
X1 − X 2
D)decreasing the difference between

Question 42

If tobs falls into a rejection region, then the decision is to.

A)reject the null hypothesis
B)fail to reject the null hypothesis
C)do not accept the alternative hypothesis
D)increase the value of alpha

Question 43

If tobs does not fall into a rejection region, then the decision is to.

A)reject the null hypothesis
B)fail to reject the null hypothesis
C)accept the alternative hypothesis
D)decrease the value of alpha

Question 44

If H0 is not rejected in a tind test, then.

A)H1 is accepted
B)the alpha level should be increased
C)the observed difference between the sample means is attributed to sampling error
D)the distribution of scores in the sample is normal

Question 45

If tobs falls into a rejection region, then which of the following is true?

A)The probability of obtaining tobs if H0 is true is greater than the alpha level.
B)The alternative hypothesis is not accepted.
C)The null hypothesis is rejected.
D)The null hypothesis is not rejected.

Question 46

If the results of an experiment using tind were summarized as t(48) = 1.632, p > .05, then you would know that the total number of participants used (N) was and the null hypothesis was.

A)48; rejected
B)48; not rejected
C)50; rejected
D)50; not rejected

Question 47

Increasing the size of the sample in an experiment will the value of.

A)increase; the standard error of the difference between means
B)decrease; the standard error of the difference between means
C)increase; alpha
D)decrease; alpha

Question 48

If tobs does not fall into a rejection region, then the.

A)value of alpha should be changed from .05 to .01
B)mean of the sampling distribution of the t statistic is not equal to zero
C)standard error is smaller than the hypothesized difference in population means
D)null hypothesis is not rejected

Question 49

The use of tind rests on the assumption that the.

A)variances of the samples are equal
B)variances of the populations are equal
C)standard errors of the samples are equal
D)standard errors of the populations are equal

Question 50

A "robust" test implies that violations of the assumptions have little effect on the.

A)distribution of the standard error
B)distribution of the degrees of freedom
C)probability of a Type I error
D)probability of a Type II error

Question 51

Saying that a statistical test is "robust" means that violating an assumption of the test has little effect on the.

A)probability of a Type I error
B)probability of a Type II error
C)obtained value of the standard error of the difference between means
D)sampling distribution of differences between sample means

Question 52

The power of tind can be increased by.

A)minimizing the effect of the independent variable
B)increasing the value of s
C)increasing the difference between
D)decreasing the difference between

Question 53

The use of tind rests on the assumption that.

A)each subject is randomly selected
B)each subject is randomly assigned to a group
C)the population sizes are equal
D)the population means are equal

Question 54

If tobs does not fall into the rejection region, then which of the following is true?

A)The probability of obtaining tobs if H0 is true is less than .05.
B)The alternative hypothesis is accepted.
C)The null hypothesis is rejected.
D)The null hypothesis is not rejected.

Question 55

The power of a statistical test is the probability of rejecting H0 when H0 is and H1 is .

A)true; true
B)true; false
C)false; false
D)false; true

Question 56

If the sample means in an experiment differ significantly using tind, then the.

A)null hypothesis was accepted
B)difference is attributed to chance
C)difference is attributed to the dependent variable
D)difference is attributed to the independent variable

Question 57

The denominator of tind is.

A)the difference between the population means
B)the difference between the sample means
C)a measure of the variation in differences between population means
D)a measure of the error variation in an experiment

Question 58

If tobs falls into a rejection region, then.

A)the null hypothesis is rejected and the alternative hypothesis accepted
B)the null hypothesis is not rejected and the alternative hypothesis is not accepted
C)a Type II error may have occurred
D)the population means probably do not differ

Question 59

If the sample means in an experiment differ significantly using tind, then.

A)the null hypothesis was rejected
B)the experiment could not have been confounded
C)the alternative hypothesis was not accepted
D)all the above

Question 60

The power of tind can be increased by.

A)minimizing the effect of the independent variable
B)increasing the value of s
C)decreasing the difference between
D)decreasing the value of

Question 61

If knowing the level of an independent variable received by a participant does not at all improve prediction of that participant 's score in an experiment, then will be.

A)-1.0
B)100 percent
C)1.0
D)0.0

Question 62

If the result of a tind on an experiment is reported as t(28) = 2.633, p < .05, then you know that the.

A)critical value of t was 2.633
B)alpha level was less than .05
C)total number of subjects was 28
D)total number of subjects was 30

Question 63

If the results of an experiment using tind were summarized as t(24) = 1.974, p > .05, then you would know that the probability of the difference between the means occurring by chance is and that the difference is.

A)less than or equal to.05; nonsignificant
B)less than or equal to.05; statistically significant
C)greater than .05; nonsignificant
D)greater than .05; statistically significant

Question 64

In a within-subjects design,.

A)all groups of subjects receive different levels of each independent variable
B)a single group of subjects is given each level of the independent variable
C)all groups of subjects receive the same levels of each independent variable
D)all subjects are matched for attributes on extraneous variables

Question 65

If knowing the level of an independent variable received by a participant allows perfect prediction of that participant 's score in an experiment, then will be.

A)100
B)100 percent
C)1.0
D)either -1.0 or +1.0

Question 66

If the results of an experiment using tind were summarized as t(20) = 2.146, p < .05, then you would know that the probability of the difference between the means occurring by chance is and that the difference is.

A)less than or equal to.05; nonsignificant
B)less than or equal to.05; statistically significant
C)greater than .05; nonsignificant
D)greater than .05; statistically significant

Question 67

A journal report of the results of a tind on two groups in an experiment should include which of the following?

A)The sample size.
B)The mean and a measure of variability for each group.
C) ∗
D)All the above.

Question 68

Increasing the sample size in an experiment will.

A)decrease the standard error of the difference between means
B)increase the standard error of the difference between means
C)increase the difference between
D)decrease the difference between

Question 69

Eta squared represents the proportion of total variation in scores that is accounted for by .

A)the independent variable
B)the dependent variable
C)error variation
D)variation within treatment groups

Question 70

For tind, is obtained by.

Question 71

If the results of an experiment using tind were summarized as t(30) = 2.817, p < .05, then you would know that the total number of participants used (N) was and the null hypothesis was.

A)32; rejected
B)32; not rejected
C)30; rejected
D)30; not rejected

Question 72

Eta squared is a measure of the.

A)power of a statistical test
B)strength of association
C)amount of error variation within groups
D)probability of making correct decisions about the statistical null and alternative hypotheses

Question 73

Extensive variation in the measure of the dependent variable in an experiment will.

A)increase the probability of a Type I error
B)increase the power of a statistical test
C)decrease the power of a statistical test
D)enable the use of a smaller sample

Question 74

The value of may range from.

A)0.0 to 100.0
B)0.0 to 1.0
C)-1.0 to +1.0
D)-10.0 to +10.0

Question 75

Extensive error variation within the groups of an experiment will.

A)increase the probability of a Type I error
B)result in a more detectable treatment effect
C)increase the power of a statistical test
D)decrease the power of a statistical test

Question 76

Eta squared provides a measure of.

A)error variation
B)the value of the difference between population means
C)the minimum sample size required to achieve a specific level of power
D)the strength of effect of an independent variable

Question 77

Rejecting H0 with a tind and a two-tailed test.

A)does not necessarily provide support for a research hypothesis in an experiment
B)provides direct support for a research hypothesis in an experiment
C)can only occur if α equals .01
D)is always a Type II error

Question 78

Which of the following is a criticism of statistical hypothesis testing?

A)The probability of a Type I error cannot be precisely specified.
B)The probability of a Type II error cannot be precisely specified.
C)Null hypotheses are almost always true.
D)Humans are better intuitive judges of statistical significance than are statistical tests.

Question 79

The 95 percent confidence interval for the difference between two independent population means is given by.

Question 80

If the result of a tind on an experiment is reported as t(28) = 2.633, p < .05, then you know that the.

A)critical value of t was 2.633
B)alpha level was less than .05
C)observed value of t was 2.633
D)total number of subjects was 28