Inquiring minds want to know: so how did it go?

Google makes semantic level query assumptions by default now- that's part of the function that allows google to suggest queries as you type.  To address your question more directly, it depends on the individual terms and whether or not Google is giving complete, partial, or no equivalence to those terms.  Non-profit and nonprofit are treated nearly the same in organic ranking- the top results are identical and the remaining results have more variance. You can experiment with this idea and you'll see some of this phenomena.

As for "concrete evidence"- you aren't going to get it.  You should study the results of your individual terms' rankings because that's exactly what Google did to weight potential results... at least, it's a part of what they did.  They have a vast army of "raters" that look into the relevance of query/results and assign a value to the search engine's success/lack of success.  In the end, a load of data is collected in various categorically-structured groups of search terms. Are they using the data attached to these unique categories to craft a more relevant set of results?  I don't know if they are, but I find it hard to believe they flush all of that expensive data down the drain just to use an umbrella algorithm to spit out all query results.

In the end, I think you'll find that worrying about the minor difference between the two is wasted time if you're looking for a concrete answer.  To avoid the pitfall, stick to a single convention on each individual page, but vary the convention from page to page.  When you search one or the other, watch how Google bolds the alternate terms in the list of results..  Compare the 1st page results of the two alternate terms..  are you seeing a pattern? With the other components of SEO on-track, it effectively won't matter at all.

The purpose behind 'the schema' is to help search engines better match semantic search terms with relevant user information. Word out there is that is doesn't help in the SERPs.  See this article.  But it might and Google may have integrated it into their algo yesterday for all we know.

With it being relatively new in the search landscape and little evidence of testing out there, I'd recommend that you donate some time to the cause   It just so happens that you'll likely get your answer from doing so.  Here's how I'd do it:

1. Choose the pages you'll be measuring...  the more you use, the better the data...

2. Measure where these pages rank in your search results before making any changes.

3.. Divide those pages into thirds and apply the 'Events' schema to a third, the 'Aggregate Rating' to a third, and apply nothing to the remaining third.

4. Track and report your findings.

Obviously, there are some potential hang-ups here (ie. if you aren't ranking in the top 50 for your target keywords and are using SEOMoz to track).  After rubbing my magic 8-ball, I'm going to prophesize that you won't see a significant difference for a few reasons (or any at all).

Finally, I'd suggest that you persuade the engines to focus on the aggregate review information.  If you look around, you'll notice that reviews are being increasingly integrated as major indicators in search algos.  Events?  Not so much.

There are many issues at hand, but you get to decide which are important and which you're going to act on.  Either way, good luck!

It sounds like you're already nofollowing the links.  This will reduce the number of links on your page as the Search Engines see it, which looks to be your goal.

Assuming this is what you're aiming to do, there's no reason to hide your links.  If you don't want search engines OR users to see them, just get rid of them altogether.

The purpose behind 'the schema' is to help search engines better match semantic search terms with relevant user information. Word out there is that is doesn't help in the SERPs.  See this article.  But it might and Google may have integrated it into their algo yesterday for all we know.

With it being relatively new in the search landscape and little evidence of testing out there, I'd recommend that you donate some time to the cause   It just so happens that you'll likely get your answer from doing so.  Here's how I'd do it:

1. Choose the pages you'll be measuring...  the more you use, the better the data...

2. Measure where these pages rank in your search results before making any changes.

3.. Divide those pages into thirds and apply the 'Events' schema to a third, the 'Aggregate Rating' to a third, and apply nothing to the remaining third.

4. Track and report your findings.

Obviously, there are some potential hang-ups here (ie. if you aren't ranking in the top 50 for your target keywords and are using SEOMoz to track).  After rubbing my magic 8-ball, I'm going to prophesize that you won't see a significant difference for a few reasons (or any at all).

Finally, I'd suggest that you persuade the engines to focus on the aggregate review information.  If you look around, you'll notice that reviews are being increasingly integrated as major indicators in search algos.  Events?  Not so much.

There are many issues at hand, but you get to decide which are important and which you're going to act on.  Either way, good luck!