Click to upload or drag and drop
PDF, DOCX, PPTX, TXT, JPG, JPEG, PNG, HEIC, ODP, ODT, BMP, or TIFF
up to 20MB
Uploading...
To build a Schoology question bank, go to Resources, create a question bank, and add questions to it, either by typing them in or by importing a supported package. Schoology's test and quiz importer accepts Blackboard 7.1 to 9.0, Edmastery, and ExamView files, plus QTI packages as a ZIP in versions 1.2, 2.1, and 2.2. It does not accept Word, PDF, or CSV. Once the bank exists you can pull individual questions into any test or set a random draw, which is what makes the second version of a test nearly free.
The build itself is not the hard part. Deciding what goes in it is, and most banks fail because they were built one quiz at a time with no plan for reuse.
Question banks belong to the classic Tests and Quizzes material inside a course. Item banks belong to Schoology Assessments, the newer assessment type with more question types and richer reporting. They serve the same purpose, a reusable pool of questions, but they live in different parts of the platform and do not interchange freely.
For a weekly chapter check, question banks with a classic Test or Quiz are the right call: fewer clicks and the grade lands in the course gradebook where you expect it. For a department benchmark that four teachers give and someone analyzes by standard afterward, item banks and Assessments earn their extra setup. Choose based on who consumes the results, not on which one is newer.
More than one quiz needs, by a wide margin. The number that makes a bank actually useful is roughly three times the length of the test you draw from it.
| What you give | Items on the test | Bank size that works | What the surplus buys you |
|---|---|---|---|
| Weekly chapter check | 10 to 15 | 35 to 45 | A makeup version and a practice set |
| Unit test | 25 to 30 | 75 to 90 | Two full forms plus a retake |
| Semester final | 50 to 60 | 150 or more, built across the term | Random draws that differ per student |
| Department common assessment | 20 to 30 | 60 to 90, reviewed by the team | Item swaps when one question underperforms |
A bank the same size as the test is not a bank, it is a test in a different folder. The whole value is in having spare items: when a question turns out to be ambiguous, you replace it instead of rewriting it, and when a student needs a makeup, the second form already exists.
Three sources, in descending order of how well they usually work.
Your own course materials. The chapter you assigned, the lab handout, the slide deck you taught from. Questions built from what students actually read are the only ones guaranteed to be fair, and they are the reason a generic publisher bank so often feels off by a few degrees.
A publisher test bank. Fine when it exists and matches your sequence, and it imports cleanly if the publisher shipped QTI or ExamView. Check the alignment before you trust it. Publisher banks are written to the textbook's emphasis, which is not always your emphasis.
Questions written from scratch. The most accurate and by far the most expensive. Forty defensible multiple choice items with plausible distractors is most of an evening, every time, which is why banks stay small in real schools.
The practical shortcut is to generate from your own materials and then edit. Upload the chapter, ask for forty questions instead of fifteen, and spend your time reviewing rather than composing. A Schoology quiz generator workflow gets you the raw set with an answer key; you still make the judgment calls about which items are fair and which are trivia. That split, machine drafts and you decide, is what makes a full-size bank realistic on a teacher's schedule.
Yes, if it is in a format Schoology reads. QTI ZIP packages import to Resources and are tagged to your personal account rather than a course, so after importing you pull the questions into the courses that need them. Blackboard, Edmastery, and ExamView exports import directly as tests and quizzes.
What you cannot do is upload a Word document, a PDF, or a spreadsheet of questions. This surprises people every year, and it is why so many teachers end up with questions in the editor by hand. Twenty seconds per question with a ready list beside you is not a tragedy, but it does mean there is no point hunting for the Word import button. It is not hidden. It is not there.
The single habit that separates a bank you use for five years from one you abandon in March is tagging each question when you add it, not later. At minimum, record the chapter or unit it came from and the learning objective it tests. If your department reports by standard, add the standard.
Once questions carry those tags, a lot becomes cheap. You can pull every question tied to one objective for a targeted reteach quiz. You can check whether your unit test actually covers the objectives you claimed, which is a check that quietly fails more often than anyone admits. You can hand a colleague a coherent subset instead of a folder.
Retrofitting tags onto three hundred untagged questions is a job nobody has ever finished. Do it at entry, when the context is in your head anyway.
Pull from the bank twice with different draws, or build the second test from the questions the first one left behind. With a bank three times the test length, version B is a few minutes of selection rather than a night of writing. This is far more useful than shuffling question order, because a shuffled test is still the same test, and any two students who talk after the first period know it.
If you also reveal correct answers after the due date, a second version stops being a nice option and becomes necessary. Reveal answers on a single-form test and the makeup students are taking an open-book exam whether you meant them to or not.
Plenty of course material predates the current file server: photocopied packets, scanned readings, a departmental binder somebody inherited. These are still the assigned readings, and a bank that ignores them is a bank that covers half the course.
Scans work as source material as long as the page is readable, and a phone photo of a page is fine if it is shot flat and square. Where a whole archive needs converting rather than a page at a time, that is a different class of problem, and there are tools built to extract text from large batches of scanned documents for exactly that. For one chapter before Friday, uploading the scan directly is faster than any pipeline.
Repeat per unit and the bank builds itself over a semester without a single heroic weekend. By the second year you are editing rather than creating, which is the point where the whole exercise starts paying you back.
Schoology gives you the bank, the draws, and the gradebook. It does not give you the questions, and it will not import them from a Word file or a PDF. Build from your own materials, aim for three times the test length, tag at entry, and draw every test from the bank rather than writing into it. Do that for one unit and you will do it for the rest, because the second unit takes half as long as the first.
From the same family of tools