Frequently Asked Questions

The most likely reason is that your PDF is scanned — created by photographing or scanning a physical document. A scanned PDF contains no actual text, only an image of text. Screen readers cannot read images, so a student using one receives a completely blank document.

How to check: Open the PDF and try to highlight a word. If you can select text, it's digitally created. If nothing selects when you click and drag, it's a scanned image.

These are two different problems with different fixes:

• Scanned PDF — Contains no actual text, only an image of text. Screen readers return a blank page. Severe
• Untagged PDF — Created digitally but exported without the accessibility structure screen readers use to navigate. Text is present but may be read out of order or skipped. Major

The most common cause of untagged PDFs is using Print → Save as PDF instead of the correct export method. Always export using File → Save As → PDF → More Options → "Best for electronic distribution and accessibility."

Almost always, this is an export method issue. Using Print → Save as PDF strips all accessibility tags from the file — even if the source Word document was perfectly structured. This is the single most common cause of untagged PDFs in Canvas.

Go back to the original Word document, run the Accessibility Checker ( Review → Check Accessibility), fix any issues, and re-export using File → Save As → PDF → More Options → "Best for electronic distribution and accessibility."

Try opening the PDF directly in Microsoft Word ( File → Open → select the PDF). Word will attempt to convert it using OCR. Review the output carefully — OCR is imperfect, especially for documents with columns, tables, or handwriting. Correct any errors, run the Accessibility Checker, and re-export as an accessible PDF.

If Word can't produce a usable result, try Adobe's free online OCR tool at acrobat.adobe.com. Note the free tier has monthly limits and doesn't produce a fully tagged PDF — use it as a bridge while you work toward a Word-based fix.

If neither works, contact IDT. Some documents need a different approach entirely.

You are responsible for the accessibility of content you place in Canvas, including third-party materials. Your options depend on the document:

• Check whether an accessible version is available through Xavier's library — many publisher documents are available through library databases in accessible formats.
• If the document is text-based (not scanned), try opening it in Word to convert and re-export it.
• If it can't be modified or replaced, add a Library Reference in Ally so students can request an accessible version through Xavier's library. Click the Ally gauge next to the file and look for the Library Reference option in the feedback panel.
• Contact IDT if you're unsure how to proceed with a specific document.

Word's built-in heading styles — particularly Heading 2 and Heading 3 — use blue and gray font colors that frequently fail the WCAG 4.5:1 contrast ratio requirement. This isn't something you've done wrong; it's Word's default behavior.

The fix: after applying a heading style, select the heading text and change the font color to Automatic (black) in the Home tab font color dropdown. Do this for all headings before exporting to PDF. You can verify any color combination at webaim.org/resources/contrastchecker.

For a student using a screen reader, headings are the primary way to navigate a page or document. Screen readers can generate a list of all headings and let the user jump between them directly — similar to a live table of contents. Without headings, a student must listen to every line of content from top to bottom to find what they need.

Well-structured headings also benefit sighted students: pages are easier to navigate and less cognitively demanding to read.

Bold text looks like a heading visually, but a screen reader reads it as plain body text with no structural significance. If a student navigates by headings, bolded "section labels" are completely invisible to them.

Always apply a heading style from the paragraph format menu in the Canvas RCE or the Styles panel in Word — not the bold button. The heading style is what gives the text structural meaning that screen readers can use.

Start your content headings at Heading 2. Canvas automatically applies Heading 1 to the page title — adding a second H1 in your content creates a duplicate, which Ally flags as a heading structure error.

From there: use Heading 3 for subsections within a Heading 2 section, Heading 4 for subsections within Heading 3, and so on. Never skip levels — don't jump from Heading 2 to Heading 4.

Previous course templates may have used banner-style images with text embedded directly in the image as section headers. Because screen readers cannot read text that is part of an image, Ally flags these as accessibility issues. To quickly resolve the flagged images directly within your existing course we recommend replacing each 'heading image' with a correctly formatted text heading in the Canvas Rich Content Editor (RCE).

For each heading image in your course:

1. Starting at the top of the page or assignment, delete the heading image and any redundant heading text that appears alongside it.
2. Re-add the heading text and set it to Heading 2 using the paragraph format menu in the Canvas RCE.
3. For each subsequent heading, delete the heading image and any redundant heading text.
4. Re-add the heading text and set it to the appropriate heading level. (See the Accessibility Training for Canvas for guidance on choosing the correct heading level).
5. When all heading images have been replaced, navigate to the Files area of your course and delete any unused images that are no longer referenced in the course.

If you would like hands-on support working through this process, schedule a consultation with IDT.

An empty heading is a blank line that has a heading style applied to it. This often happens when content is deleted but the heading formatting is left behind, or when a heading style is accidentally applied to a blank line.

To fix: open the page in the Canvas RCE, find the blank line with the heading style, click into it, and change the paragraph format back to "Paragraph." Save the page. This is one of the quickest fixes in Ally — nearly 3,000 Canvas pages in Xavier courses have this issue.

Tables without a designated header row are flagged because screen reader users have no context for what the columns contain — they hear data values with no labels.

In Word: Click inside the table, select the first row, go to the Table Design tab, check the Header Row box. Then right-click the row → Table Properties → check "Repeat as header row at the top of each page."

In Canvas (RCE): Click inside the table, select the top row, open Row Properties or Cell Properties, and set the row type to Header.

Almost always this happens when list items are typed manually — using hyphens, dashes, or numbers as plain text — rather than using the list formatting buttons in the Canvas RCE toolbar.

Screen readers announce properly formatted lists by telling the user how many items are in the list before reading them. Manually typed lists are read as plain body text with no list context.

Fix: Select the list content, delete the manual characters, and reformat using the bulleted or numbered list button in the RCE toolbar.

Previous course templates may have used banner-style images with text embedded directly in the image as section headers. Because screen readers cannot read text that is part of an image, Ally flags these as accessibility issues. To quickly resolve the flagged images directly within your existing course we recommend replacing each 'heading image' with a correctly formatted text heading in the Canvas Rich Content Editor (RCE).

For each heading image in your course:

1. Starting at the top of the page or assignment, delete the heading image and any redundant heading text that appears alongside it.
2. Re-add the heading text and set it to Heading 2 using the paragraph format menu in the Canvas RCE.
3. For each subsequent heading, delete the heading image and any redundant heading text.
4. Re-add the heading text and set it to the appropriate heading level. (See the Accessibility Training for Canvas for guidance on choosing the correct heading level).
5. When all heading images have been replaced, navigate to the Files area of your course and delete any unused images that are no longer referenced in the course.

If you would like hands-on support working through this process, schedule a consultation with IDT.

Alt text is a written description attached to an image that communicates its content or purpose to students who can't see it. Screen readers read alt text aloud in place of the image. Without it, a screen reader user hears only "image" — or the filename — and receives no information about what the image contains or why it's there.

Alt text also helps students on slow connections (images may not load), students with cognitive processing differences, and any student accessing content in a non-visual format.

Standalone image files — .jpg, .png, .gif — uploaded directly to Canvas are scanned by Ally as their own items, separate from any page they appear on. If a file has no alt text at the file level, Ally flags it regardless of whether it also appears on a page with a nearby description.

To fix, click the Ally gauge next to the file, type an alt text description in the feedback panel, and click Add. No re-upload needed.

Describe what the image communicates, not just what it shows. The goal is to give a student who can't see the image the same information a sighted student gets from looking at it.

• Keep it under 125 characters for simple images
• Don't begin with "image of," "photo of," or "graphic of" — screen readers already announce it's an image
• For charts and graphs, summarize the key finding — not every data point
• For images with text overlaid, include all the text from the image
• For linked or button images, describe the function: "Submit assignment," not what it looks like

No — but they need to be marked as decorative rather than left with a blank or missing alt text field. Marking an image as decorative tells the screen reader to skip it intentionally. A blank alt text field is ambiguous.

The test: If this image were removed, would any student lose information they need for the course? If no, it's decorative.

In Canvas RCE: Check the "Decorative image" box in the image options panel. In Word: Right-click the image → Edit Alt Text → check "Mark as decorative."

Canvas cannot add alt text to images inside uploaded files. You must add it in the source document before uploading.

In Word: Right-click the image → Edit Alt Text → type your description in the Alt Text panel. Mark as decorative if applicable. Re-export as an accessible PDF and re-upload.

In PowerPoint: Right-click the image → Edit Alt Text → enter your description. Repeat for every image on every slide.

Ally has detected that text or visual elements inside the image don't meet the WCAG 4.5:1 contrast ratio. Common examples: infographics with light text on white backgrounds, charts with gray data labels, slides saved as images with low-contrast titles.

Canvas can't fix contrast inside an image. You need to edit the source file — in PowerPoint, Canva, or your design tool — increase the contrast of text and key elements, then export and re-upload.

If editing the source isn't possible, consider providing the same information as text directly on the Canvas page so all students have equal access to the content.

Screen readers can generate a list of all links on a page. When links say "click here" or display raw URLs, that list becomes meaningless — the user can't tell where any link goes without reading the surrounding content.

Replace vague link text with a description of the destination: "Assignment 1 submission page," "Xavier Library research database," or "Week 3 reading: introduction to qualitative methods (PDF)."

To edit in Canvas: Select the link text in the RCE → click the link icon → update the display text → save.