Item Weighting

Item weighting in categories

For tests which contain different categories, the weighting system can also be applied at the level of the category.

The image below shows the results (the test variables only) for a test containing the same items with the same weights as above, but this time the items have been assigned to two different categories. (This simple test containing just three items has been constructed for the purpose of this course, so one of the categories contains only one item. Normally, there would be more than one item in a category.)

Scores from a weighted test with categories


The test-taker gave the same answers in the original test which had no categories. As you can see, the value of SCORE_TOTAL_MAX is 3 and SCORE_TOTAL_MAX_WEIGHTED is 6, as before. The SCORE_RATIO still shows that the test-taker got 2/3 questions right, and the SCORE_RATIO_WEIGHTED shows that when the weights were applied, he/she got 50%.

Now, however, there is also a set of scores relating to each category. Let's have a look at the first category in more detail. These values tell you that there were two items in this category (SCORE_CATEGORY_MAX_CAT1 = 2) and that the test-taker answered them both correctly (SCORE_CATEGORY_CAT1 = 2). There were three marks available for this category (SCORE_CATEGORY_WEIGHTED_MAX_CAT1 = 3), and the test-taker was awarded all of them (SCORE_CATEGORY_WEIGHTED_CAT1 = 3).

By contrast, he/she got none of the marks (SCORE_CATEGORY_WEIGHTED_CAT2 = 0) of the three available (SCORE_CATEGORY_WEIGHTED_MAX_CAT2 = 3) for  the second category.

Note: It is possible to assign items to multiple categories, and then choose between them in different test scenarios. (This may be useful, if, for example, the emphasis of a test is category-based, and an item falls into more than one category.)

Details on how to configure weighting are given in the chapter 'Summary and configuration of scoring methods' below.

Generating a PASS or FAIL

You may want to define a minimum score which the test-taker needs to achieve in order to pass the test. In other words, if the test-taker gets this score or above, they will pass. If not, they will fail. This is set as a ratio of the total score, so if the test-taker needs to get half the questions right (or half the marks available), then this is defined as: 0.5.

Following the example used before, if you set the pass mark as 20, and the test-taker got 24 points out of a possible 40, they will be awarded a PASS. This is shown as:

PASS_ALL: true

If you have different categories in your test (you might have sections on Theory, or Critical Thinking, for example), you can also set a pass mark for each category. Let's say, for example, that the marks were divided equally over the two categories, and you define the pass mark as 10 for each category.

If the test-taker got 15/20 in Theory and 9/20 in Critical Thinking, this will appear in the results table as follows:



Note: Details on how to configure this option are given in the chapter 'Summary and configuration of scoring methods' below.

Customizing scoring methods

If you wish to define a scoring method other than those available in TAO, you can do this. 

In this scenario, you can edit the scoring method using an interface which displays editable XML code called Edit Processing Rule. By way of example, for a group of three questions you might want to specify that the marks from the third one should only be awarded if the first two were both answered correctly.

XML code for customizing scoring method

Note: This option is only available for Ignite/Premium users.