Item Scoring Rules

Naturally, the first question that a Test-taker has after completing a Test is, "What score did I get?" Scores are generated through performance in the various interactions that make up the Items, that themselves compose the Test. Interactions will generate score numbers, which will determine what the Item provides, and each item will be tallied to produce the final Test result. Knowing how to set up scoring within an interaction is therefore useful for determining the grade achieved by the Test-taker from the Test.

Step 1: While selecting responses (answers) for an interaction, look within the Response Properties panel for a pull-down choice box offering the two options of Response Processing: Match Correct and Map Response.

Match Correct: This refers to the processing option where either the question is answered right or wrong. There is no partial credit.

Map Response: This refers to the processing option where getting part of the response correct. This is useful for questions where the answer is given in multiple parts. A singular part can be weighed more heavily than other parts in this option.

Step 2: Select the Response Processing that is appropriate for the question.

This is going to depend in large part on the preference of the test author, particularly with respect to partial credit.

Step 3: If Match Correct is selected, stop here - scoring for this interaction is complete. If Map Response is selected, review the responses in the interaction and determine the scores to be assigned to potential responses.

This is where partial credit is given. Find within the interaction displayed in the Item Canvas the boxes where scores are assigned, and assign these according to relative importance.

Step 4: Within the Response Properties panel, review the Score Range fields.

It is here that minimum and maximum scores can be entered, setting these as the bounds by which the Test-taker can do no better or no worse in this interaction. Its use is optional. If the correct response involves selecting more than one response, the maximum should be set as the sum of anticipated scores. If only one response is expected, the maximum should equal that of the highest score. It should be noted that these can be set higher or lower to any arbitrary level, if required by the interaction author.

Other values in the Score Range include that for Mapping Default, which is described as the default value given if no specific scores are assigned to a response, and the check box used to Define Correct Response. The latter should be checked if there are correct answers, and left unchecked if the correctness of the answer is dependent on the sum of scores accumulated by the Test-taker in answering the interaction.

Lastly, it should be noted that TAO assigns a Response Identifier to each interaction response. It is recommended that the interaction author not change this.