Here you go ... both of these sites allow you to make tag cloud images with any text that you can cut & paste. Most (but not all) pdfs that you download from JSTOR or other library resources work just fine (although you will get better results if you omit the reference list, as those do tend to muck it up some).
Tag Cloud is probably the easiest to use - and lets you save as a pdf file:
But Wordle is way more fun - check out the randomize button!
If you are not offered an option to save your image you can use the print screen button (look at top row of your keyboard - unless it is a Mac, you will probably have a PrtScn button; if you have a Mac, I don't know how but I bet google does). Whatever is on your screen when you hit the button gets copied to your clipboard and can be pasted.
I use Paint Shop Pro, so I paste there and crop to the selection I want ... but you can paste into Word and crop there, or pretty much any image program will do.
All three of these word clouds used the Sisters In Spirit 2010 Report: What Their Voices Tell Us. To view larger, click on the thumbnail.