Practice Quiz Chapter 09

* not completed

. Which of the following is not a reason why researchers should be critical of measuring race and racialization through statistical analysis?

a) The failure to clearly distinguish between "race," ethnicity, and nationality on surveys correct incorrect

b) Aboriginal people are excluded from survey analysis correct incorrect

c) The problematically-designed "visible minority" label correct incorrect

d) Canada's official crime statistics do not include information about the race of the victim correct incorrect

e) Canada's official crime statistics include information about the race of the victim correct incorrect

* not completed

. Lambda is a PRE measure that measures what?

a) The magnitude of an association between two ordinal-level variables correct incorrect

b) The magnitude of an association between two nominal-level variables correct incorrect

c) The magnitude of an association between two ratio-level variables correct incorrect

d) The magnitude of an association between a nominal-level variable and a ratio-level variable correct incorrect

e) The magnitude of an association between an ordinal-level variable and a ratio-level variable correct incorrect

* not completed

. Which two components does the calculation of lambda require?

a) G1 and G2 correct incorrect

b) A nominal-level variable and a ratio-level variable correct incorrect

c) E1 and E2 correct incorrect

d) Gamma and PRE correct incorrect

e) 1E and 2E correct incorrect

* not completed

. There are 200 people overall to consider for the following question: 80 people in Group 1 say they have no contact with police and 30 people in Group 2 say they have no contact with police. What does E1 equal?

a) 900 correct incorrect

b) 9.9 correct incorrect

c) 0.90 correct incorrect

d) 90 correct incorrect

e) 9 correct incorrect

* not completed

. In the same group of 200 people from the question above, 20 people in Group 1 had contact with the police and 70 people in Group 2 had contact with the police. What does E2 equal?

a) 1.5 correct incorrect

b) 0.5 correct incorrect

c) 0.05 correct incorrect

d) 5.0 correct incorrect

e) 50 correct incorrect

* not completed

. After calculating E1 and E2 in the previous two questions, now calculate the lambda. What does it equal?

a) 0.44 correct incorrect

b) 4.4 correct incorrect

c) 44 correct incorrect

d) 0.044 correct incorrect

e) More information is necessary correct incorrect

* not completed

. Which of the following is not true about the lambda?

a) A lambda of 0 indicates that knowing group membership improves predictions correct incorrect

b) Lambda ranges from 0 to 1 correct incorrect

c) A lambda of 0 indicates that knowing group membership doesn't mean improved predictions correct incorrect

d) A lambda of 1 indicates that knowing about group membership results in perfectly predicting the dependent variable correct incorrect

e) Calculating lambda relies on calculating E1 and E2 first correct incorrect

* not completed

. Which of the following can't be said about the relationship between the gamma and the lambda?

a) The gamma uses more information than the lambda correct incorrect

b) The gamma accounts for the order or ranking of attributes whereas the lambda does not correct incorrect

c) the gamma is a PRE measure whereas the lambda is not correct incorrect

d) The gamma is a symmetric measure of association whereas the lambda is asymmetric correct incorrect

e) The gamma provides information about the direction of the relationship and the lambda does not correct incorrect

* not completed

. If there are 144 cases in a sample, what is the number of pairs?

a) 10,692 correct incorrect

b) 103 correct incorrect

c) 100,296 correct incorrect

d) 1,296 correct incorrect

e) 10,296 correct incorrect

* not completed

. Which of the following is not a component of the formula for gamma?

a) Concordant pairs times discordant pairs correct incorrect

b) Concordant pairs correct incorrect

c) Discordant pairs correct incorrect

d) Concordant pairs minus discordant pairs correct incorrect

e) Concordant pairs plus discordant pairs correct incorrect

* not completed

. Which of the following is not true about the gamma calculation?

a) The gamma ranges from -1 to +1 correct incorrect

b) A positive gamma indicates that people with higher attributes of the independent variable have higher attributes of the dependent one correct incorrect

c) A negative gamma indicates that people with higher attributes of the independent variable have lower attributes of the dependent variable correct incorrect

d) The gamma relies on the lambda for calculation purposes correct incorrect

e) The size of the gamma provides information about the magnitude of the relationship between two variables correct incorrect

* not completed

. In a hypothetical example, the number of concordant pairs was calculated to be 56,680,861 and the number of discordant pairs was calculated to be 41,269,757. If this is the case, then what is the gamma?

a) 174 correct incorrect

b) 74 correct incorrect

c) 0.0174 correct incorrect

d) 17.4 correct incorrect

e) 0.174 correct incorrect

* not completed

. In a hypothetical example using similar data, the number of concordant pairs was calculated to be 56,680,861 and the number of discordant pairs was calculated to be 41,269,757. If the number of cases under consideration is 25,146, then how many tied cases are there in the data?

a) 219,555,111 correct incorrect

b) 261,432,444 correct incorrect

c) 216,197,467 correct incorrect

d) 266,147,667 correct incorrect

e) 296,333,777 correct incorrect

* not completed

. Which of the following is not true of non-parametric tests?

a) Chi-square tests of independence are considered non-parametric tests correct incorrect

b) They assume an underlying normal distribution in the population correct incorrect

c) They do not rely on any assumption about the underlying distribution correct incorrect

d) They do not involve estimating population parameters correct incorrect

e) They rely on comparing the frequencies that are observed in the sample, with the frequencies that are expected if the null hypothesis is true correct incorrect

* not completed

. Which of the following is not part of the process for calculating the expected frequencies?

a) Square the total number of cases in the same row as the cell correct incorrect

b) Calculate the percentage of people overall in some group being considered correct incorrect

c) For any cell in the cross-tabulation, find the total number of cases in the same row as the cell correct incorrect

d) Multiply the total number of cases in the same row as the cell by the total number of cases in the same column as the cell correct incorrect

e) Divide the result of multiplying the total number of cases in the same row as the cell by the total number of cases in the same column as the cell by the total number of cases in the cross-tab correct incorrect

* not completed

. Which of the following is not part of the formula for calculating the chi-square statistic?

a) Subtract the observed frequencies from the expected frequencies correct incorrect

b) Square the result of subtracting the expected frequencies from the observed frequencies correct incorrect

c) Subtract the expected frequencies from the observed frequencies correct incorrect

d) Divide the result of squaring the difference between the observed frequencies and the expected frequencies from the observed frequencies correct incorrect

e) Sum the results of squaring the difference between observed and expected frequencies and dividing these results by expected frequencies correct incorrect

* not completed

. Which of the following is not true of degrees of freedom for the chi-square distribution?

a) Like the F-distribution, the chi-square distribution is a probability distribution that changes shape depending on the number of degrees of freedom correct incorrect

b) The degrees of freedom of a chi-square distribution account for the number of cells in the cross-tab correct incorrect

c) It is not easier to understand the idea of degrees of freedom with a cross-tab correct incorrect

d) It is easier to understand the idea of degrees of freedom with a cross-tab correct incorrect

e) For chi-square tests of independence, the number of degrees of freedom is the number of rows in the cross-tab minus 1, multiplied by the number of columns in the cross-tab minus 1 correct incorrect

* not completed

. Which of the following is not a measure of the magnitude of a relationship between two categorical variables?

a) V correct incorrect

b) ? correct incorrect

c) Chi-Square correct incorrect

d) Lambda correct incorrect

e) ? correct incorrect

* not completed

. Which of the following is not true about phi and Cramér's V?

a) An absolute interpretation of phi and Cramérs V is possible correct incorrect

b) Both range from 0 to 1 correct incorrect

c) There is no absolute interpretation of phi and Cramér's V correct incorrect

d) Researchers use these two measures to classify the effect of the independent variable on the dependent variable correct incorrect

e) The effect of the independent variable on the dependent one can be classified as small, medium, or large correct incorrect

* not completed

. Which of the following is not a step necessary to calculating phi?

a) Count the total number of cases (n) but do not include cases that are missing correct incorrect

b) Find the chi-square statistic for the relationship between two variables correct incorrect

c) Divide the chi-square statistic by the total number of cases correct incorrect

d) Find the square root of the result from dividing the chi-square statistic by the total number of cases correct incorrect

e) Multiply the result of dividing the chi-square statistic by the total number of cases by the Cramér's V correct incorrect

* not completed

. It is difficult to measure "race" or racialization with statistics since"race" is a social construct that varies over time and in different contexts.

a) True correct incorrect

b) False correct incorrect

* not completed

. There is very limited data available about "race" and racism in the Canadian criminal justice system.

a) True correct incorrect

b) False correct incorrect

* not completed

. A strategy that researchers should not use to describe the size of a relationship between two categorical variables is to show how the distribution of the dependent variable changes depending on group membership.

a) True correct incorrect

b) False correct incorrect

* not completed

. PRE measures show how much better predictions about the dependent variable are if group membership is taken into account.

a) True correct incorrect

b) False correct incorrect

* not completed

. Lambda relies on the mode to predict the attribute of a dependent variable for every case.

a) True correct incorrect

b) False correct incorrect

* not completed

. E1 is the number of prediction errors made when group membership is considered.

a) True correct incorrect

b) False correct incorrect

* not completed

. E2 shows the number of incorrect guesses if the most common attribute in each group is predicted for every case in the group.

a) True correct incorrect

b) False correct incorrect

* not completed

. A lambda of 0.44 indicates that the error in predicting something can be reduced by 44 per cent if you know which group they belong to.

a) True correct incorrect

b) False correct incorrect

* not completed

. Gamma is used to measure the magnitude of a relationship between two nominal-level variables.

a) True correct incorrect

b) False correct incorrect

* not completed

. Gamma relies on counting pairs of cases so that every case in the sample is compared to every other case in the sample.

a) True correct incorrect

b) False correct incorrect

* not completed

. The word "concordant" refers to something that is not consistent or in agreement.

a) True correct incorrect

b) False correct incorrect

* not completed

. The word "discordant" refers to something that is consistent or in agreement.

a) True correct incorrect

b) False correct incorrect

* not completed

. "Tied" cases occur in pairs where a case is ranked the same as another on either the independent or dependent variable.

a) True correct incorrect

b) False correct incorrect

* not completed

. Once you have tallied the number of concordant and discordant pairs of cases it is easy to calculate gamma.

a) True correct incorrect

b) False correct incorrect

* not completed

. If there are more discordant pairs than concordant pairs, gamma will be positive.

a) True correct incorrect

b) False correct incorrect

* not completed

. The gamma calculation uses the number of cases in each cell but not the percentages.

a) True correct incorrect

b) False correct incorrect

* not completed

. Although gamma is easy to understand, it has been criticized by some researchers in the past because it ignores tied pairs.

a) True correct incorrect

b) False correct incorrect

* not completed

. The most common way that researchers assess the reliability of a relationship between two categorical variables is by using a chi-square test of independence.

a) True correct incorrect

b) False correct incorrect

* not completed

. Non-parametric tests assume that the variables being tested have an underlying normal distribution in the population.

a) True correct incorrect

b) False correct incorrect

* not completed

. The size of the difference between the expected frequencies and the observed frequencies is used to determine the likelihood of randomly selecting a sample with the observed relationship from a population in which group membership is not related to the dependent variable.

a) True correct incorrect

b) False correct incorrect