The process of creating tests is riddled with difficulties that compound the complexity of the process. One needs to ensure that the test is reliable and possesses practical value. Discrimination and difficulty are among the test measures that are often applied to ensure that tests serve their purpose. The analysis of the THING test items allows one to appreciate the difficulty of creating useful tests. For this test to work better, some items should be removed.
Difficulty is a measure used to determine the ease of answering questions on a test. Optimal difficulty should be such that not all takers of the test answer all the questions correctly and that not all of the takers fail to provide the correct answers (“Chapter 8”, n.d). Given this guideline, it is determined that item 1 in the THING test should be removed. With a difficulty level of 1, this means that all the takers answered this item correctly. With difficulty levels of 0.60 each, items 2, 3 and 4 can be retained for further analysis. The further analysis should focus on the discrimination indices of these items. According to the text, 0.75 marks the optimal discrimination index (“Chapter 8”, n.d). With a discrimination index of 0.2, item 2 deviates wildly off the optimal value and should therefore be removed. For optimal effectiveness, a test should have a positive discrimination index. A positive index shows that the test takers who answered a particular item correctly also performed well in the overall test (“Item Analysis”, n.d). Since it has a discrimination index of -1.00, item 4 is unideal and should therefore be eliminated. Item 3 seems the most ideal as its discrimination and difficulty indices are closest to the optimal values. This item’s discrimination and difficulty indices are provided as 0.6 and 0.6 respectively. These figures suggest that there is a strong positive association between this item and the overall THING test and that its difficulty level is such that most takers stand a fairly good chance of providing the correct responses.
Delegate your assignment to our experts and they will do the rest.
References
Chapter 8. Test development. (n.d).
Item analysis. (n.d). Pennsylvania State University. Retrieved February 7, 2019 from http://sites.psu.edu/itemanalysis/discrimination/
