Blackboard Item Analysis

Master

Media Component

Media

Content

Question analysis provides statistics on overall performance, assessment quality, and individual questions. This data helps you recognize questions that might be poor discriminators of student performance. Question analysis is for assessments with questions. You can run a report before all submissions are in if you want to check the quality of your questions and make changes.

Uses for question analysis:

Improve questions for future assessments or to adjust credit on current attempts
Discuss assessment results with your class
Provide a basis for remedial work
Improve classroom instruction

Example:

After the question analysis, you notice that the majority of students answer one question incorrectly. Why the low success rate?

Is the wording of the question confusing?
Are the answer options unclear?
Were students given the right content to learn to successfully answer this question?
Was the content to learn easily accessible and clear?

Based on what you discover, you can improve the question to truly assess what students know or don't know.

After you run a report, you can view overall summary information and details about each question. Select the assessment on the Question Analysis page to view the summary.

Only submitted attempts are used in calculations. When attempts are in progress, those attempts are ignored until they're submitted and you run the analysis report again. Automatic zeros assigned for late work aren't included in calculations. [view screenshot]

Summary of statistics for the individual assessment:
1. Average score: The score shown is the average score reported for the assessment in the gradebook. The average score can change if more attempts are submitted and graded.
2. Possible questions: The total number of questions in the assessment.
3. Completed attempts: The number of submitted assessments.
4. Average time spent: The average completion time for all submitted attempts.
Rerun a report or edit the assessment to make a change to questions.
Use the graphs to filter the table of questions. Make selections in both graphs to refine your search. If you make no selections, all the questions appear in the table at the bottom of the page.
1. Discrimination: Indicates how well questions differentiate between students who know the subject matter and those who don’t.
  1. Shows the number of questions that fall into these categories:
    1. Good (greater than 0.3)
    2. Fair (between 0.1 and 0.3)
    3. Poor (less than 0.1) categories
    4. Can't calculate: A question's difficulty is 100% or all students received the same score on a question.
  2. Questions with discrimination values in the Good and Fair categories differentiate between students with higher and lower levels of knowledge.
  3. Questions in the Poor category are recommended for review.
2. Difficulty: Percentage of students who answered the questions correctly
  1. Shows the number of questions that fall into these categories:
    1. Easy (greater than 80%)
    2. Medium (between 30% and 80%)
    3. Hard (less than 30%)
  2. Questions in the Easy or Hard categories are recommended for review.
Select a heading to sort the questions. For example, sort the Review column so questions that need review appear first.
Clear Filters: Clear the filters you selected in the graphs and display all questions in the table
Download the question analysis report

Note: The question statistics table provides item analysis statistics for each question in the test. Questions that are recommended for your review are indicated with red circles so that you can quickly scan for questions that might need revision.

The questions table provides analysis statistics for each question in the assessment. After you use the graphs to filter the questions table, you can view and sort the results.

In general, good questions fall in these categories:

Medium (30% to 80%) difficulty
Good or Fair (greater than 0.1) discrimination values

In general, questions recommended for review fall in these categories. They may be of low quality or scored incorrectly.

Easy ( > 80%) or Hard ( < 30%) difficulty
Poor ( < 0.1) discrimination values

Reminder: If you make no selections, all the questions appear in the table at the bottom of the page. [view screenshot]

To investigate a specific question, select the title and review the question details.

Information for each question appears in the table:

Needs review: Triggered when discrimination values are less than 0.1. Also, when difficulty values are either greater than 80% (the question was too easy) or less than 30% (the question was too hard). Review the question to determine if it needs revision.
Question Modified: Displays Yes if you run a report, then change part of a question, and rerun the report. Yes also appears if you copied the question from another assessment when you created the assessment.

If Yes appears in the Question Modified column for a question, the Yes doesn't carry over when you archive and restore the course.

Discrimination: Indicates how well a question differentiates between students who know the subject matter and those who don't. A question is a good discriminator when students who answer the question correctly also do well on the assessment. Values can range from -1.0 to +1.0. A question is flagged for review if its discrimination value is less than 0.1 or negative. Discrimination values can't be calculated when the question's difficulty score is 100% or when all students receive the same score on a question.
- Discrimination values are calculated with the Pearson correlation coefficient. X represents the scores of each student on a question and Y represents the scores of each student on the assessment. [view screenshot]
- These variables are the standard score, sample mean, and sample standard deviation, respectively: [view screenshot]
Difficulty: The percentage of students who answered the question correctly. The difficulty percentage is listed along with its category: Easy (greater than 80%), Medium (30% to 80%), and Hard (less than 30%). Difficulty values can range from 0% to 100%. A high percentage indicates the question was easy. Questions in the easy or hard categories are flagged for review.
- Difficulty levels that are slightly higher than midway between chance and perfect scores do a better job differentiating students who know the tested material from those who don't. High difficulty values don't assure high levels of discrimination.
Graded Attempts: Number of question attempts where grading is complete. Higher numbers of graded attempt produce more reliable calculated statistics.
Average Score: The score that appears is the average score reported for the assessment in the gradebook. The average score might change after all attempts are graded.

You can investigate questions flagged for your review and view student performance. From the Question Analysis questions table, select a linked question title to access the question's summary. [view screenshot]

After you access a question, use the question title's menu to access any question in the assessment. You can navigate to other questions sequentially on either side of the page.
The summary table displays statistics for the question.
Select Edit Assessment to access the assessment and make changes.
The question text and answer choices appear. You can see how many students chose each answer choice or the percentage answered correctly. For example, for a Matching question, you see what percentage of students matched the pairs correctly. Only the question text appears for Essay questions.

About multiple attempts, question overrides, and question edits

The analysis handles some common scenarios in these ways:

When students take an assessment multiple times, the last submitted attempt is used as the input for the analysis. As soon as a student submits another attempt, subsequent analyses will include that newest attempt.
Gradebook overrides don't impact the analysis data because the analysis generates statistical data for questions based on completed student attempts.
When you make changes to a question or manually grade questions, you must run the analysis again to see if the changes affect the data.

Examples

Question analysis can help you improve questions for future assessment administrations. You can also fix misleading or ambiguous questions in a current assessment.

In a Multiple Choice question, an equal number of students chose A, B, and C. Examine the answer choices to determine if they're ambiguous, if the question is too difficult, or if the material wasn't covered.
A question is recommended for review because it falls into the hard difficulty category. You determine the question is hard, but you keep it to adequately evaluate your course objectives.
A Multiple Choice question is flagged for your review. More Top 25% students chose answer B, but A is the correct answer. You realize you didn't choose the correct answer when you created the question. You edit the assessment question and it's automatically regraded.

General Inquiries

Education

Blackboard Item Analysis