Take it, to the limit, one more time

Published 04 September 07 03:49 PM | wayne.griffiths 

I was shocked this afternoon to discover how easy it would have been to exceed the word limit for my thesis. I took one page (full) of text from my thesis and counted the average number of words per line: 14. Then I counted how many lines there were on the page and adjusted for the "half-lines" at the ends of paragraphs: 36. I've got around 200 pages in my thesis, so

14 x 36 x 200 = 100800.

This would exceed the word limit of 100000 words! Fortunately though, I have quite a few figures in my thesis, so those pages would not be full of words. Additionally, I have over 500 equations (!) which means that the space they take up is not being used for text. So I'm under the 100K limit. This of course begs the question, how many words is an equation equivalent to?

Comments

# greg.cresp said on September 4, 2007 4:49 PM:

(Please don't take any of this as a "whose is bigger", you just raised an interesting issue)

You talking about figures (and other floats) and equations made me wonder how many pages in an average thesis would be completely full of text. A quick count of mine (~200 pages) revealed 23, but of those 18 had at least one Section or Subsection heading (which tend to take up quite a bit of space). I do have significantly fewer equations than you though, only about 350, so I'd be interested to hear your count.

As to the question of how many words an equation counts as, I have no idea.

Getting an accurate word count is tricky with LaTeX (if you have a cunning way, I'd be interested to hear it). Including comments, latex commands and the like, I get 72,000. According to WinEdt, about 60,000 of these are actually text - but I don't know how it does its count, so this might not be accurate. In any case, that's (thankfully!) a long way from the theoretical 100,800.

# wayne.griffiths said on September 4, 2007 6:39 PM:

I Googled for this problem, and one page suggested the following 'cheat' way of doing it:

1. Get your PDF

2. Press Ctrl+A to select everything

3. Copy everything to the clipboard

4. Open Word

5. Press Ctrl+V to paste into a Word document

6. Go to Tools -> Word Count

The answer it spat out was around 65,000. Admittedly, this still included a couple of pages of appendices, a table, and the bibliography, none of which I think need to be included in the tally. So, I still think I'm safe.

# greg.cresp said on September 4, 2007 8:08 PM:

I've only ever trusted that method as an upper bound, because I end up with a lot of garbage from equations, not to mention the page headers and footers (which just get included as plain text). I'm never sure how much of that Word includes in its count.

Of course as a test of whether you're over the 100K limit, an upper bound is fine.

# robyn.owens said on September 5, 2007 5:47 AM:

You can also DeTeX your LaTeX source and then run a word count on the output.

Anonymous comments are disabled

About wayne.griffiths

I started at UWA in 1997, completing a Diploma in Modern Languages (Italian) in 1999. By 2001, I had completed a Bachelor of Computing and Mathematics degree with Honours. In 2002, I worked part-time in the Department of Mathematics and Statistics. From 2003 to 2007, I studied for the qualification of PhD in Electrical Engineering at the Western Australian Telecommunications Research Institute (WATRI). My thesis title is "On A Posteriori Probability Decoding of Linear Block Codes over Discrete Channels", and it is currently under examination.