TEST ed 360

TEST ed 360 - Notes on Evaluating a Test 1. Begin by...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
Notes on Evaluating a Test 1. Begin by looking at the handouts at the end of the packet to see what is there. 2. Don’t forget to think about the type of reliability being addressed when responding to questions in part B. Accuracy does not equal stability. What is meant by consistency? 3. When computing the standard deviation by hand (if you don’t have a calculator to do it) be sure to subtract each score from the mean then square it then add them up. Don’t add them up first and then square the whole thing. 4. On page 7, part E. You haven’t read about grading in criterion-referenced testing, but you can still get the idea of the item (establishing the criteria for cut-off scores is arbitrary, but every teacher needs to be able to justify why they choose the score they do for their students, parents etc.) Invent a rationale.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Evaluating a Test Mr. G. has been working with his 7th grade class on a unit about weather. (The class consists of 20 students.) Because Mr. G. will be using students’ performance on a summative measure written for this unit (in combination with two other summative measures from units on electricity and mechanics) to assign grades, he has written a 50 item norm-referenced test for them to take. Below are the students’ test results. Use these results to answer the following questions. Student ID 1 2 3 4 5 6 7 8 9 10 Test Score 39 49 48 30 50 29 28 35 44 32 Student ID 11 12 13 14 15 16 17 18 19 20 Test Score 33 16 40 42 35 36 49 35 20 25 A. 1. What is the mean? 2. What is the median? 3. What is the mode? 4. Is this a good norm-referenced distribution? Why/why not? 2
Background image of page 2
B. 1. Compute the reliability of this test using the K-R21 formula on your hand-out of computational formulas 2. Is this a reliable test? Why/why not? 3. Mr. G. gave this test twice last year to the same group of 30 students. He computed a Pearson r and got a coefficient of .66. Is it statistically significant? What does that tell you in this case? Explain. 4. According to the information on the hand-out labeled “Interpretation of Correlation Coefficients” how consistent is this test? 5. How much better is this test than a test with a reliability coefficient of .29? Explain. (Hint: use the interpretation handout.) Why is this knowledge helpful? C. Mr. G.’s students wanted to know what their grades on this test were. To give this information to a few of his students, complete the following: 3
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Uncorrected Distribution Raw Score Distribution T Score Distribution Distribution Corrected for Error Raw Scores T Scores 1. Fill in the mean of the raw scores on this test. Then determine the raw scores at each of the 4 deviation points on the graph. (Hint: 2 standard deviations above the mean can’t be greater than 50.) 2. Fill in the mean T score on this test. Then determine the T scores for each of the 4 deviation points on the graph. 3. Compute the standard error of measurement for this test.
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 14

TEST ed 360 - Notes on Evaluating a Test 1. Begin by...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online