Question 1

Which of the following has a correlation matrix as its output?&#10;A)Item-total analysis&#10;B)Cronbach's alpha&#10;C)Test-retest reliability&#10;D)Criterion-related validity&#10;E)None of the above

Accepted Answer

Item-total analysis often involves calculating correlations between each item and the total score, excluding that item, which can be represented in a correlation matrix.

Question 2

When conducting an item-total analysis, the items in the scale are considered internally consistent when all of the item-total correlations are:&#10;A)Greater than 1.0&#10;B)Less than 1.0&#10;C)Greater than 0.7&#10;D)Less than 0.7&#10;E)Greater than 0.3

Accepted Answer

Item-total correlations indicate how well each item in a scale correlates with the total score of the scale. It is ideal for these correlations to be positive and significant, but not too high (as this may indicate redundancy among items). A commonly accepted threshold for item-total correlations is 0.3 or higher. Therefore, option C is the best choice as 0.7 is a high enough threshold to ensure good internal consistency among the items.

Question 3

Which of the following is NOT true for criterion-related validity?&#10;A)Determines the extent to which the scale produces similar results over time.&#10;B)Determines the extent to which the scale you are testing correlates with a criterion.&#10;C)Determines whether the measure is valid for its intended purpose.&#10;D)Determines whether a scale is measuring what it says it is measuring.&#10;E)Requires that all items be measured on an interval or ratio scale.

Accepted Answer

Criterion-related validity does not involve producing similar results over time; that is related to reliability. Criterion-related validity focuses on the correlation with a criterion and the validity of the measure for its intended purpose.

Question 4

Which of the following is FALSE about Cronbach's alpha?&#10;A)All items must be measured on an interval or ratio scale.&#10;B)It is a way to determine if you are measuring the correct construct.&#10;C)It is a test of the internal consistency of a test.&#10;D)It is a way to assess the reliability of a test.&#10;E)It determines the degree to which all the items measure the same construct.

Accepted Answer

Cronbach's alpha is primarily a measure of internal consistency or reliability of a test, not a direct measure of whether the correct construct is being measured. Validity, not reliability, concerns whether the correct construct is being measured.

Question 5

Test-retest reliability is:&#10;A)a measure of validity.&#10;B)a measure of internal consistency.&#10;C)a measure of consistency over time.&#10;D)a measure of the same kind of reliability as Cronbach's alpha.

Accepted Answer

Test-retest reliability measures how consistent results are over time, therefore it is a measure of consistency over time, which is different from internal consistency and validity. Cronbach's alpha is also a measure of internal consistency, but it is not the same as test-retest reliability.

Question 6

When assessing the value of your Cronbach's alpha,&#10;A)Numbers close to 1.00 indicate poor internal consistency.&#10;B)Numbers close to 0.00 indicate poor internal consistency.&#10;C)Numbers close to 1.00 indicate poor reliability.&#10;D)Numbers close to 0.00 indicate excellent reliability.&#10;E)You hope the number is less than 0.00.

Accepted Answer

Cronbach's alpha ranges from 0 to 1, and values closer to 0 indicate poor internal consistency, while values closer to 1 indicate better internal consistency. Therefore, numbers close to 0.00 indicate poor internal consistency.

Question 7

Which of the following is NOT true for the test-retest reliability coefficient?&#10;A)It is a correlation coefficient for the relationship between total scores on two administrations of the same measure.&#10;B)It is calculated using the Pearson r correlation coefficient with the two administrations of the measure as the two variables.&#10;C)It should be positive.&#10;D)It is considered strong if it is close to 1.00.&#10;E)It is considered strong if it is close to 0.00.

Accepted Answer

Test-retest reliability coefficient is considered strong if it is close to 1.00, not 0.00. Therefore, E is the correct option.

Question 8

Which of the following is NOT an aspect of Item-total analysis?&#10;A)All items must be measured on an interval or ratio scale.&#10;B)It is a test of the internal consistency of a test.&#10;C)It is a way to assess the reliability of a test.&#10;D)It is a way to assess the validity of the construct in the test.&#10;E)It determines the degree to which all the items measure the same construct.

Accepted Answer

Item-total analysis does not assess the validity of the construct in the test. It is a measure of internal consistency, which is a type of reliability.

Question 9

What are the assumptions for item-total analysis?
I)All items should be measured on an interval or ratio scale.
II)Each item should be normally distributed.
III)The analysis is conducted using Pearson r but can be conducted with Spearman rho if the data are ordinal in nature.

A)I only
B)II only
C)III only
D)I & II only
E)I, II, & III

Accepted Answer

Item-total analysis assumptions include that items should be measured on an interval or ratio scale, each item should be normally distributed, and the analysis is typically conducted using Pearson r but can use Spearman rho for ordinal data.

Question 10

When conducting an item-total analysis, if an item-total correlation is negative, you should:&#10;A)Report the correlation as negative.&#10;B)Average the correlation with the other correlations.&#10;C)Remove the question and recalculate the analysis.&#10;D)Remove the question only if the item-total correlation is below 0.3.&#10;E)Add another question and recalculate the analysis.

Accepted Answer

A negative item-total correlation suggests that the item is not measuring the same construct as the rest of the items, so it is best to remove the question and recalculate the analysis.

Deck 9: Test Construction