Question 1

Somers' d is actually created from another statistic; that statistic is:&#10;A)Pearson's r&#10;B)Gamma&#10;C)Uncertainty Coefficient (U)&#10;D)Phi

Accepted Answer

Somers' d is derived from the Gamma statistic, which measures the strength and direction of association between two ordinal variables.

Question 2

In general,ordinal measures of association tend to provide an inaccurate estimate of the real relationship when:&#10;A)the sample size is very large&#10;B)the data are measured at the ordinal level&#10;C)there are many ties in the data&#10;D)all of the above

Accepted Answer

Ordinal measures of association are less accurate when there are many ties in the data. This is because ties reduce the amount of variation in the data, which can make it difficult to assess the true relationship between the variables. The sample size and level of measurement may also affect the accuracy of the measure, but these factors do not have as strong an impact as the presence of ties in the data.

Question 3

While the Gamma statistic has other problems,which of the answers below is an advantage of Gamma:&#10;A)expressly designed for 2 X 2 tables&#10;B)it uses interval level data&#10;C)it calculates asymmetric relationships&#10;D)it has a PRE interpretation

Accepted Answer

Gamma has a Proportional Reduction in Error (PRE) interpretation, which is an advantage as it provides a measure of the strength of association between variables.

Question 4

Which of the following is the best interpretation of a significant Somers' d with a value of .30?&#10;A)9% of the variance in the dependent variable is explained&#10;B)9% of the variance in the independent variable is explained&#10;C)15% of the variance in the independent variable is explained&#10;D)30% of the variance in the dependent variable is explained

Accepted Answer

Somers' d measures the strength and direction of the relationship between two ordinal variables. A value of .30 indicates a moderate to strong relationship, meaning that 30% of the variance in the dependent variable can be explained by the independent variable. Therefore, the correct interpretation is that 30% of the variance in the dependent variable is explained. It is not appropriate to interpret Somers' d in terms of the variance in the independent variable.

Deck 23: Ordinal Level Tests of Association