Chapter 6:Reliability and Validity

Category 1

100

Which term best describes the consistency of an assessment measure? 

Reliability

Validity

Correlation

Variance

RELIABILITY

200

r= +.91 is best described as:

_{A strong positive relationship between variables} 

_{A weak positive relationship between variables} 

_{A weak negative relationship between variables}

_{A strong negative relationship between variables}

A strong positive relationship between variables

300

r= -.21 is best described as: 

A weak positive relationship between variables 

A weak negative relationship between variables 

A strong positive relationship between variables 

A strong negative relationship between variable

A weak negative relationship between variables

400

Smoking leads to an increased risk of lung cancer is what type of correlation? 

Negative correlation 

Weak correlation 

Positive correlation 

Strong correlation

Positive correlation

500

This is a graphic representation of a correlation. 

Scattergram 

Frequency polygon 

Standard deviation 

Frequency distribution

Scattergram

Category 2

100

Increased exercise is likely to lead to a loss of weight is what type of correlation? 

Weak correlation 

Positive correlation 

Negative correlation 

Strong correlation

Negative correlation

200

r= 00 is best described as A weak correlation A perfect correlation A positive correlation No correlation

No correlation

300

Pearson r is best used with what types of scales? Nominal and ordinal scales Ratio and interval scales Interval and nominal scales Ratio and ordinal scales

Ratio and interval scales

400

Reliability most commonly measured using which of the following metrics? Pearson's Product Moment Standard Error of Measurement Kuder-Richardson 20 Coefficient Alpha

Pearson's Product Moment

500

Research has proven that the Best Achivement Test in America had poor internal consistency. What does that mean? The test items do not repeatedly measure what they say they measure The test is not produce the same results across time The difference versions of the test do not produce the same responses The test, when split and half and compared, does not produce the same results

The test items do not repeatedly measure what they say they measure

Category 3

100

Mr. Jones, the classroom teacher, administers a test to Joey in September and then again in October. Joey's scores are one point off. This test could be said to have Test-retest reliability Alternate forms reliability Interrater reliability Split Half reliability

Test-retest reliability

200

Mrs. Rodriguez administers version A of and achievement test in September and then version B in May to determine progress for IEP reporting. Research must determine________ reliability in order to ensure that version A is the same as version B Alternate forms reliability Split half-reliability Test-retes reliability Interrater reliability

Alternate forms reliability

300

Mark and Eve collect data on the same student using the same assessment and find their data is almost exactly the same. It could be said that Mark and Eve have Test-retest reliability Interrater reliability Split half-reliability Alternate forms reliability

Interrater reliability

400

This formula is used to check consistency across items of an instrument with right/wrong responses Standard Error of Measurement Kuder-Richardson 20 Pearson's Coefficient Alpha Product Moment

Kuder-Richardson 20 Pearson's

500

This formula is to check consistency across terms of an instrument with responses with varying credit. Kuder-Richardson 20 Product Moment Coefficient Alpha Standard Error of Measurement

Coefficient Alpha

Category 4

100

Mrs. Morgan reviews an assessment that has a correlation coefficient of .45. Should Mrs. Morgan use this test? Why? No; the correlation coefficient is too low. No; a correlation coefficient must be at .80 to be considered reliable. Yes; if the test was published, it must be reliable to use. Yes; the correlation coefficient is considered adequate.

No; the correlation coefficient is too low.

200

Test A has a standard error of measurement of 4.3 and Test B has a standard error of measurement of 1.3. Which test should be administered to ensure the best obtained score as possible? Test A Either test should be reasonable to give Test B Neither

Either test should be reasonable to give

300

Given the following information, determine the standard error of measurement. Standard deviation = 10.83, r = +.83. 17 Unable to be determined .41 4.47

4.47

400

Given an obtained score of 87, a standard error of measurement of 3.5 and a correlation coefficient of .83, which of the following would be the correct confidence interval? 86-88 83.5 - 90.5 84 – 90 86.17 – 87.83

83.5 - 90.5

500

Given the following information, estimate the true score. Mean of the group = 76, correlation coefficient = .83, obtained score = 81, standard deviation = 2.23 -64.85 71.85 11.15 80.15

80.15

Category 5

100

What happens to the range of scores when comparing obtained and true scores? The true score provides a wider confidence interval. The true score provides a narrower confidence interval. There is no difference. The obtained score provides a narrower confidence interval.

The true score provides a narrower confidence interval.

200

When a test actually measures what it purports to measure, we say that it has good Correlation Variance Validity Reliability

Validity

300

Suppose you created a new measure of reading comprehension and you wanted to test its validity. Therefore you compare the results of your new measure to existing validated measures of reading comprehension and find that your measure compares well with these other measures. We could then say that your new measure has good____. reliability content validity concurrent validity criterion-related validity

criterion-related validity

400

A test is said to accurately assess specific psychological and personality traits, psychological concepts, attributes, and theoretical characteristics. This assessment is likely to have good predictive validity. concurrent validity. content validity. construct validity.

construct validity.

500

Mrs. Beth noticed that many of the students in her class who are Latino answered the same questions wrong when administering a standardized assessment. Mrs. Beth may be noticing _____ in the test creation. racism item bias profiling sexism

item bias