1 person found this helpful
Yes, it is possible to create accessible versions of the these tables. There are several possible solutions. In the example above, you are correct that Period 1 and Period 2 need to be tagged as <TH> with Span = 2 Column. Scope also needs to be set to Column. FTE and $ also need to be tagged as TH with Scope = Column. All the cells in Column 1 (A. First Heading, 1. First Subhead, etc.) are <TH> with Scope = Row.
Because this is a complex table you need to Associate the table content with these headings using <ID> tags.
This is unfortunately where things can get a little time consuming. The process can be done within Acrobat using the Touch Up Reading Order (TOR) tool (Accessibility Tool Panel). To briefly summarize -- Select the TOR, select the Table, and click the Table Button on the TOR Menu to open up the Table Editor. The TOR Menu will disappear, and the table will be visible. You can click on individual cells, right click and select Cell Properties.
Select one of the header cells, right click and then select Table Cell Properties. There you will see the ID, which is usually something memorable like TD_08934_121. Changes this to something meaningful like PERIOD_1 and P1_FTE. The name must be unique within the file.
Click OK to save and exit the Table Cell Properties window.
Using the TOR select all of the content cells beneath that header and right click to open the Cell Properties. Acrobat will let you select more than one cell by using the Shift Key, if there are no conflicting values in Cell Properties. Note: At times you may need to work one cell at a time.
Once the Table Cell Properties is open, click the plus next to Associate Header Cell IDs and select the appropriate Heading ID that you defined above.
Repeat until all the content cells are associated with their appropriate column headings, subheadings and row headings.
That's method 1.
Method 2 would be to use the CommonLook plug-in. Adding Table Cell IDs is CommonLook's strong point, but the price is high and there is a learning curve. And I've experienced some troublesome issues involving borders and shading in some files. Still if you have Commonlook available it can be a REAL timesaver for tagging complex tables.
Lastly, there are some tricks and workarounds that would make the structure of this table simple and therefore accessible without all of this. If you are experienced working in the Tags Panel in Acrobat. In the example above you could convert the first row of content Headings Period 1 and Period 2 to Artifacts and then entirely delete that row from the table within the tags panel, Then add alternative text to the second row of headings so FTE will always read as "Period 1 FTE" or Period 2 FTE" and so forth. Some purists may object to this method, but it would work.
NOTE: Any of these methods require further checking with a screen reader to ensure the tables are recognized by tables and no mistakes were made in assigning the headings to the cells and I would also check to make sure any blank cells read as "Blank" as they may be mistagged in the tags panel by the authoring application (e.g., MS Word).
Thank you! I started doing some of this tagging yesterday and in the process came to the conclusion that the "trick" you mentioned might be a good approach at least in terms of simplifying the structure and tagging process.
To correctly process the table you may want to review the discussion in 7.5 Tables of the following document:
PDF/UA-1 Technical Implementation Guide: Understanding ISO 14289-1 (PDF/UA-1)