How do I add criteria to obtain two equivalent samples from my data?

I have a dataset of about 6000 words. I'd like to pull two samples of 25 words each, however the average WordLength for both samples must be the same, and all words across both samples must be different.

This is what my data looks like:

Word      Type                 CEFR    WordLength
a         indefinite article   a1      1
abandon   verb                 b2      7
ability   noun                 b2      7
able      adjective            a1      4

This is just one of the criteria I need the samples to match on; if this is straightforward to you, please consider taking a look at my other question which adds a more complicated level: Scraping Oxford5000 words and obtaining two equivalent word lists