Deck 17: Chi-Square Tess

Question 1

Legal researchers were interested in whether a rape defendant's perceived guilt depends on the type of argument introduced by his defense attorney.Researchers classified each defense by whether it introduced the idea that the female victim was asking for the encounter-the "she shouldn't have worn that short skirt" defense.In a simulated juror situation, the researchers counted the number of guilty verdicts "returned" by the 20 participating mock jurors.This number is a _____ dependent variable, which means that the data can be analyzed using a _____ test.

A) nominal; nonparametric
B) scale; nonparametric
C) nominal; parametric
D) scale; parametric

Answer 1

nominal; nonparametric

Question 2

Chi-square tests assess whether:

A) two interval variables are associated.
B) the means of two samples differ significantly from each other.
C) the observed data differ from the pattern expected given the null hypothesis.
D) the distributions of two variables are normally distributed.

Answer 2

the observed data differ from the pattern expected given the null hypothesis.

Question 3

Degrees of freedom for the chi-square test of goodness-of-fit is determined by:

A) subtracting 2 from the number of cells that there are (k - 2).
B) subtracting 1 from the number of cells that there are (k - 1).
C) multiplying the number of rows minus 1, times the number of columns minus 1.
D) subtracting 1 from the number of participants that there are (N - 1).

Answer 3

subtracting 1 from the number of cells that there are (k - 1).

Question 4

For two nominal variables, the appropriate statistical test to use is a chi-square test for:

A) goodness-of-fit.
B) independence.
C) association.
D) dissociation.

Question 5

A chi-square test for goodness-of-fit has a sample size of 25.What are the degrees of freedom for this chi square?

A) k - 1
B) 24
C) 26
D) k - 2

Question 6

A college instructor is interested in whether assigning mandatory homework affects students' grades.Since she is currently teaching two sections of the same class, the instructor assigns homework to one of the classes and just recommends homework problems to the second class.Then she determines the number of people in each class that will receive an A, B, C, D, or F on the final exam.What type of statistical test should the instructor use to analyze her data?

A) chi-square test for goodness-of-fit
B) chi-square test for independence
C) analysis of variance
D) independent-samples t test

Question 7

In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summation of a case involving a battered woman who had shot and killed her husband.Of the 80 participants, 27 assigned a verdict of guilty, 40 a verdict of not guilty by reason of self-defense, and 13 a verdict of not guilty by reason of insanity.What type of chi-square test should the researchers use to determine whether the distribution of verdicts differed from what could have been expected by chance?

A) association
B) dissociation
C) goodness-of-fit
D) independence

Question 8

When using a nonparametric test (in contrast to a parametric hypothesis test) the:

A) range of variables available for statistical analysis is reduced.
B) sample size should be very large.
C) testing power is increased.
D) risk of committing a Type II error is increased.

Question 9

A researcher wonders whether talking to plants improves their health and conducts a study in which he talks to one group of five plants and does not talk to a second group of five plants.As an indicator of health, the researcher counts the number of blooms that each plant produces after two weeks of talk treatment (or no-talk treatment).This number is a _____ dependent variable, which means that the data can be analyzed using a _____ test.

A) nominal; nonparametric
B) scale; nonparametric
C) nominal; parametric
D) scale; parametric

Question 10

A college instructor is interested in whether assigning mandatory homework affects students' grades.Since she is currently teaching two sections of the same class, the instructor assigns homework to one of the classes and just recommends homework problems to the second class.Then she determines the number of people in each class who will receive an A, B, C, D, or F on the final exam.What would be the appropriate statistical advice to give to the instructor?

A) It is unnecessary for her to analyze her data because you know from experience that mandatory homework improves your own grades in college classes.
B) Unfortunately, she will not be able to analyze the data because the dependent variable is not a scale variable.If, however, she were to use percentages or points on the final exam as the dependent variable, she would be able to analyze the data.
C) It is unnecessary for her to analyze the data because you know from experience that mandatory homework does not improve your own grades in college classes.
D) She would have more power to detect an effect of homework if she used the students' percentage grades (or point grades) and performed a parametric statistical test.

Question 11

A chi-square test for goodness-of-fit is an appropriate statistical test to use when the study has:

A) one nominal dependent variable.
B) one scale independent variable.
C) two nominal dependent variables.
D) two nominal independent variables.

Question 12

A researcher wonders whether talking to plants keeps them from dying, and conducts a study in which he talks to one group of five plants and do not talk to a second group of five plants.After two weeks the researcher records how many plants have died in each group.This number is a _____ dependent variable, which means that the data can be analyzed using a _____ test.

A) nominal; nonparametric
B) scale; nonparametric
C) nominal; parametric
D) scale; parametric

Question 13

Legal researchers were interested in whether a rape defendant's perceived guilt depends on the type of argument introduced by his defense attorney.Researchers classified each defense by whether or not it introduced the idea that the female victim was asking for the encounter-the "she shouldn't have worn that short skirt" defense.In a simulated juror situation, the researchers asked each of the 20 participating mock jurors to rate the guilt of the defendant on a scale from 1 to 7.This score is a _____ dependent variable, which means that the data can be analyzed using a _____ test.

A) nominal; nonparametric
B) scale; nonparametric
C) nominal; parametric
D) scale; parametric

Question 14

Which statement does NOT state an assumption of the chi-square test?

A) Participants are randomly selected from the population.
B) The dependent variable is nominal.
C) Each observation is independent of all other observations.
D) The dependent variable is scale.

Question 15

One drawback of many nonparametric tests is that:

A) there is no way of calculating effect sizes.
B) the APA does not allow data analyzed with nonparametric tests to be published in APA journals.
C) it is very difficult to meet the assumptions necessary to use the tests.
D) they have more power than their parametric equivalent.

Question 16

In which situation would it be inappropriate to use a nonparametric test?

A) The sample size is small and the population distribution might be skewed.
B) The dependent variable is scale.
C) The dependent variable is nominal.
D) The dependent variable is ordinal.

Question 17

A chi-square test for independence is an appropriate statistical test to use when the study has:

A) one nominal dependent variable.
B) one scale independent variable.
C) two nominal dependent variables.
D) two nominal independent variables.

Question 18

Corsi et al.(2007) conducted a study of the factors mediating whether injection drug users enter a drug treatment program.In their sample of 491 drug users, 38.1 percent were single, 27.5 percent were married or had a partner, and 34.4 percent were divorced, separated, or widowed.When considering the drug users who actually entered a treatment program, 67 were single, 66 were married or had a partner, and 61 were divorced, widowed, or separated.What kind of chi-square test should Corsi use to analyze the data?

A) goodness-of-fit
B) independence
C) association
D) dissociation

Question 19

The appropriate statistic for analyzing data in a situation in which both the independent and dependent variables are nominal is a(n):

A) z test.
B) analysis of variance.
C) chi square.
D) independent-samples t test.

Question 20

When there is one nominal variable, the appropriate statistical test to use is a chi-square test for:

A) goodness-of-fit.
B) independence.
C) empowerment.
D) association.

Question 21

With a p level of 0.05, a chi-square test for goodness-of-fit for which there are three categories would have _____ degrees of freedom, and the critical value for the chi square would be _____.

A) 2; 4.605
B) 2; 5.992
C) 3; 9.49
D) 4; 11.14

Question 22

In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays.One of the questions the researchers were interested in was whether a person's "outness" depended on his or her class in school (freshman, sophomore, junior, or senior).All participants were classified as closeted or as out.What are the degrees of freedom associated with the chi-square analysis of these data?

A) 8
B) 4
C) 3
D) 2

Question 23

In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband.Of the 80 participants, 27 assigned a verdict of guilty, 49 a verdict of not guilty by reason of self-defense, and 4 a verdict of not guilty by reason of insanity.What are the degrees of freedom associated with this chi square?

A) 2
B) 3
C) 77
D) 79

Question 24

In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband.In particular, the researchers manipulated the race of the defendant (African American or Caucasian) and the type of abuse suffered (emotional or physical).The dependent variable was the number of guilty verdicts.What is the critical cutoff for this test, given a p level of 0.05?

A) 3.841
B) 7.815
C) 9.488
D) 18.307

Question 25

A chi-square test for goodness-of-fit for which there are four categories and 20 participants would have _____ degrees of freedom.

A) 57
B) 19
C) 16
D) 3

Question 26

A chi-square test for goodness-of-fit is conducted on a nominal variable with four levels.What is the critical cutoff for a test with a p level of 0.05?

A) 5.992
B) 6.252
C) 7.815
D) 9.488

Question 27

In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays.One of the questions the researchers were interested in was whether a person's "outness" depended on his or her gender.All participants were classified as closeted or as out.What type of chi-square test should the researchers use to answer the question of whether outness varies with gender?

A) association
B) dissociation
C) goodness-of-fit
D) independence

Question 28

The same formula is used to calculate the chi-square statistic in the chi-square test for goodness-of-fit and the chi-square test of independence.Which calculation differs along the way for these two tests?

A) observed frequencies
B) expected frequencies
C) difference between observed and expected frequencies
D) sample size

Question 29

In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband.In particular, the researchers manipulated the race of the defendant (African American or Caucasian) and the type of abuse suffered (emotional or physical).The dependent variable was the number of guilty verdicts.What are the degrees of freedom associated with the chi-square analysis of these data?

A) 4
B) 3
C) 2
D) 1

Question 30

The chi-square test to use for two nominal variables is called the chi-square test for independence because one is assessing:

A) an assumption of the chi square that the two variables are independent of one another.
B) whether the dependent variable is related to the independent variable.
C) whether the effect of each variable is independent of the other variable.
D) whether the distribution in the sample is independent of the distribution in the population.

Question 31

The table constructed for a chi-square test of independence is a _____ table.

A) contingency
B) source
C) frequency
D) periodic

Question 32

In a 2006 study, Gortmake and colleagues surveyed 80 lesbian and gay students at a large Midwestern state university to assess their experiences and their perceptions of the campus climate regarding lesbians and gays.One of the questions the researchers were interested in was whether a person's "outness" depended on his or her class in school (freshman, sophomore, junior, or senior).All participants were classified as closeted or as out.What is the critical cutoff for this test, given a p level of 0.05?

A) 3.841
B) 7.815
C) 9.488
D) 18.307

Question 33

Which of these correctly states the results of a chi-square test reported in APA format?

A) $\chi$ ²(N = 1042) = 12.85, df = 1, p < 0.05
B) $\chi$ ²(1, N = 1042) = 12.85, p < 0.05
C) $\chi$ ²(1) = 12.85, p < 0.05
D) $\chi$ ²(1) = 12.85, N = 1042, p < 0.05

Question 34

In a study of simulated juror decision making, Braden-Maguire, Sigal, and Perrino (2005) investigated the type of verdict assigned by study participants after they read a 12-page summary of a case involving a battered woman who had shot and killed her husband.In particular, the researchers manipulated the race of the defendant (African American or Caucasian) and the type of abuse suffered (emotional or physical).The dependent variable was the number of guilty verdicts.What type of chi-square test should the researchers use to analyze these data?

A) association
B) dissociation
C) goodness-of-fit
D) independence

Question 35

Chi-square tests for independence assess whether:

A) two scale variables are associated.
B) the means of two samples differ significantly from each other.
C) two nominal variables are independent of each other.
D) the distributions of two variables are normally distributed.

Question 36

What is the correct formula for the chi-square statistic?

A) $x ^ { 2 } = \sum \left[ \frac { ( O - E ) } { E } \right]$
B) $x ^ { 2 } = \sum \left[ \frac { ( O - E ) ^ { 2 } } { E } \right]$
C) $\chi ^ { 2 } = \sum ( O - E ) ^ { 2 }$
D) $x ^ { 2 } = \sum \left[ \frac { ( O - E ) } { E } \right] ^ { 2 }$

Question 37

The expected frequencies across all cells of the chi square should:

A) always add up to the size of the sample.
B) always add up to the size of the population.
C) be five times the number of participants.
D) be three times the number of participants.

Question 38

A researcher decides to examine whether people prefer milk chocolate or dark chocolate.A chi-square test for goodness-of-fit is conducted.What is the critical cutoff for a test with a p level of 0.05?

A) 3.841
B) 5.992
C) 6.252
D) 7.815

Question 39

The values in the cells of a chi-square test for goodness-of-fit are:

A) proportions.
B) medians.
C) means.
D) counts or frequencies.

Question 40

Degrees of freedom for the chi-square test of independence is determined by:

A) subtracting 2 from the number of cells there are (k - 2).
B) subtracting 1 from the number of cells there are (k - 1).
C) multiplying the number of rows minus 1 times the number of columns minus 1.
D) subtracting 1 from the number of participants that there are (N - 1).

Question 41

Even with scale data, one may choose to run a nonparametric test because there is a small sample size and the data are skewed.

Question 42

When we calculate relative risk, we are essentially determining the ratio of two:

A) observed frequencies.
B) conditional proportions.
C) relative likelihoods.
D) expected frequencies.

Question 43

When graphing the results of a chi-square test it is more useful to graph _____ rather than _____.

A) frequencies; conditional proportions
B) proportions or percentages; frequencies
C) means; frequencies
D) frequencies; means

Question 44

Corsi et al.(2007) conducted a study of the factors mediating whether injection drug users enter a drug treatment program.One of the factors they explored was marital status.The researchers found that 66 of 135 married (or partnered) drug users entered a treatment program and 67 of 185 single drug users entered a treatment program.They also found that the relative likelihood of entering a treatment program if a user is married is 1.35.This means that:

A) married drug users are 1.37 times more likely to enter a treatment program than are single drug users.
B) single drug users are 1.37 times more likely to enter a treatment program than are married drug users.
C) for every married drug user that enters a treatment program, only 0.37 single drug users enter a treatment program.
D) for every single drug user that enters a treatment program, only 0.37 married drug users enter a treatment program.

Question 45

The adjusted standardized residual tells which of these?

A) whether there is an interaction between two nominal variables
B) the expected value of a cell when the null hypothesis is false
C) the expected value of a cell when the null hypothesis is true
D) the number of standard errors an observed frequency is from its expected frequency

Question 46

What is the formula for calculating expected frequencies in the chi-square test for independence?

A) $\left( \frac { \text { Total } _ { \text {row } } } { N } \right) ^ { 2 }$
B) $\frac { \text { Total } _ { \text {rw } } } { N }$
C) $\frac { \text { Total } _ { \text {conwx } } } { N }$
D) $\frac { \mathrm { Total } _ { c a \mathrm { own } } } { \mathrm { N } } \left( \mathrm { Total } _ { \mathrm { ran } } \right)$

Question 47

Post hoc tests are carried out for chi square by:

A) calculating correlations between observed and expected frequencies in each cell.
B) performing all possible two-group chi-square tests.
C) calculating the adjusted standardized residual for each cell.
D) using the Bonferroni correction for the p level.

Question 48

For a study, a chi-square statistic of 3.69 was calculated.There were 65 participants in total, and 3 degrees of freedom for the first variable and 2 degree of freedom for the second variable.What is the effect size?

A) 0.337
B) 0.168
C) 0.138
D) 0.028

Question 49

The adjusted standardized residual is calculated as the difference between the observed and expected frequency in a given cell:

A) divided by the degrees of freedom for the row.
B) divided by the standard error.
C) minus the total sample size.
D) minus the total for the row and minus the total for the column.

Question 50

Using Cohen's conventions for effect size, how would one label a Cramer's V = 0.30, if the row df = 3 and the column df = 3?

A) small
B) medium
C) large
D) 0

Question 51

A chi-square test for independence is calculated on two variables, each with 2 degrees of freedom.An effect size of 0.23 is calculated.What size effect is this?

A) small
B) medium
C) medium-to-large
D) large

Question 52

Another name for relative risk is:

A) observed frequency.
B) chi-square statistic.
C) relative likelihood.
D) Cramer's V.

Question 53

For a study, a chi-square statistic of 1.25 was calculated.There were 40 participants in total, and 2 degrees of freedom for the first variable and 1 degree of freedom for the second variable.What is the effect size?

A) 0.566
B) 0.177
C) 0.125
D) 0.031

Question 54

Corsi et al.(2007) conducted a study of the factors mediating whether injection drug users enter a drug treatment program.One of the factors explored was marital status.The researchers found that 66 of 135 married (or partnered) drug users entered a treatment program and 67 of 185 single drug users entered a treatment program.What is the relative likelihood of entering a treatment program if a user is married compared to if he or she is single?

A) 0.36
B) 0.50
C) 1.35
D) 1.55

Question 55

For a study, a chi-square statistic of 4.42 was calculated.There were 87 participants in total, and 2 degrees of freedom for the first variable and 1 degree of freedom for the second variable.What is the effect size?

A) 0.502
B) 0.225
C) 0.172
D) 0.051

Question 56

If an adjusted standardized residual is larger than _____, the observed frequency in that cell differs significantly from the expected frequency.

A) 0.05
B) 0.5
C) 1.0
D) 2.0

Question 57

The statistic for reporting the effect size of a chi square is:

A) Cramer's V, or $\varphi$ .
B) Cohen's d.
C) R².
D) proportionate reduction in error.

Question 58

When graphing the results of a chi-square study, it is important for the y-axis to depict:

A) only the range of values observed in the data.
B) values of the dependent variable.
C) the scale of proportions from 0 to 1.0.
D) values of the independent variable.

Question 59

For a study, a chi-square statistic of 1.82 was calculated.There were 45 participants in total, and 2 degrees of freedom for the first variable and 3 degrees of freedom for the second variable.What is the effect size?

A) 0.954
B) 0.201
C) 0.142
D) 0.116

Question 60

One danger of reporting relative risk or relative likelihood information rather than frequencies is that:

A) people may not understand what the relative risk information means.
B) the base rate occurrence of an event is not reported, and the data can be misleading.
C) people may be frightened on hearing the relative risk for a positive event.
D) the proportion of events is not reported.

Question 61

When working with one nominal variable with three categories, a chi-square test for goodness-of-fit would be used.

Question 62

When parametric assumptions are met, nonparametric tests have less statistical power than parametric tests.

Question 63

When deciding on the critical value for the chi-square test, one must first decide whether to use a one-tailed or two-tailed hypothesis test.

Question 64

When working with one nominal variable, a chi-square test for goodness-of-fit would be used.

Question 65

When parametric assumptions are met, researchers are in greater danger of making a Type II error with a nonparametric test than with a parametric test.

Question 66

The suggested minimum for expected frequencies in each cell of the chi square is 10.

Question 67

When parametric assumptions are met, nonparametric tests have more statistical power than parametric tests.

Question 68

When working with two nominal variables, a chi-square test for goodness-of-fit would be used.

Question 69

When parametric assumptions are met, researchers are in greater danger of making a Type I error with a nonparametric test than with a parametric test.

Question 70

When working with two nominal variables, each with two levels, a chi-square test for goodness-of-fit would be used.

Question 71

When working with one nominal variable, a chi-square test for independence would be used.

Question 72

When working with one nominal variable with three categories, a chi-square test for independence would be used.

Question 73

The null hypothesis for the chi-square test for independence is that the two variables have a dependent relationship.

Question 74

Delucchi suggested there should be at least ten times as many participants as cells in a chi-square study.

Question 75

Even with scale data, one may choose to run a nonparametric test because there is a large sample size and the data are normal.

Question 76

When working with two nominal variables, a chi-square test for independence would be used.

Question 77

The suggested minimum for expected frequencies in each cell of the chi square is 5.

Question 78

Delucchi suggested there should be at least five times as many participants as cells in a chi-square study.

Question 79

When working with two nominal variables, each with two levels, a chi-square test for independence would be used.

Question 80

The null hypothesis for the chi-square test for independence is that the two variables have an independent relationship.