Reference no: EM131548852

DIRECTIONS: READ THE FOLLOWING DISCUSSION AND RESPOND TO STUDENT. CITE REFERENCES PLEASE.

Discussion 1

Before looking at the meaning of the scores and interpreting these results, lets define reliability. According to our textbook, Standards for Educational and Psychological Testing, reliability is defined as " the precision or consistency of measurements" (American Psychology Association, 2014).

This means that our data is to be accurate and must have the able to consistently obtain the same results no matter who is giving the test. Without a test being reliable, we can not conclude that the test is valid. "There are 4 methods used when assessing reliability. These 4 methods include: test retest, alternate forms, split-half, and internal consistency" (Cohen, Swerdlik, Sturman, 2013).

The case study provided, used alternate test forms; these were titled thing 1 and thing 2. The coefficient values used with this test are internal consistency reliability coefficient=0.92, alternate forms reliability=0.82, and test-retest reliability coeffient=0.50. The higher the coefficient, the higher the reliability of that test would be.

These numbers indicate that the test would be pretty reliable. The test-retest is still a high number but only at .50, so the test-retest could still be considered reliable, but would not be nearly as reliable as the other two. The internal consistency reliability would be the most reliable out of these three test. I also believe that these data are acceptable, because they are consistent and will provide the same results over time.

When looking just at the alternate forms reliability coefficient, for the test THING, we see the coefficient= 0.83. This is a high reliability score for this test. we can conclude that over time we will see that both test, even though different will provide similar scores. " Certain traits are presumed to be relatively stable in people ocer time, we would expect test measuring those traits to reflect that stability" (Cohen, Swerdlik, Sturman, 2013).

Overall, all three test are reliable and can be considered acceptable. We can show a little hesitation toward the third test (test-retest) since even though 0.50 is considered a high number it is right at the line of what is reliable, but for this test, THING, I would consider it to still be reliable and acceptable.

References:

Cohen, R. J., Swerdlik, M., &Sturman, E. (2013). Reliability. In Psychological Testing and Assessment (8th ed., pp. 77-117). New York, NY: McGraw- Hill Companies. doi:978-0-07-803530-2
APA. (2014). Reliability/Precision and Errors of Measurement. In Standards for Educational and Psychological Testing (pp. 33-47). Washington , DC: American Educational Research Association. doi:978-0-935302-35-6