Twitter Credibility Part 2: making a research plan

Credibility: qualities that someone has that make people believe or trust them
definition from http://www.macmillandictionary.com/

A week or so ago I proposed repeating a recent American research experiment looking at perceived credibility based on twitter streams. I had a lot of encouragement - including 8 twitter contacts who kindly volunteered to let me use their twitter streams as part of my experiment.

Having allowed enough time to pass that these 8 have probably forgotten they offered, and hence have no mention of the project in their stream I have taken snapshots and begun to think about the next stages, and would welcome any feedback on how I take things from here.

In the original American research, participants were asked 3 questions about the (fictitious professors) twitter streams they were asked to view.

The questions were:
This professor seems caring [strongly disagree - 1/2/3/4/5/6/7 - strongly agree]
This professor seems competent [strongly disagree - 1/2/3/4/5/6/7 - strongly agree]
This professor seems trustworthy [strongly disagree - 1/2/3/4/5/6/7 - strongly agree]

My first research question is purely to test if I can replicate the same results - namely, that when asking the 'caring' question about tweets of a personal nature, there will be a significantly higher response that viewing tweets of a more academic nature.

The tweet streams I want to use are genuine users, but anonymised, as although I suspect discerning twitter users make assumptions based on avatars, account names, number of followers etc, that is not part of the first question. Real streams contain replies and retweets which is such a fundamental part of twitter I don't think faking it is easily achievable. Here are a five of my anonymised streams:

 

So having a quick look through those 5, how would you respond for each to the questions: "this person seems caring?" "this person seems competent?" "this person seems trustworthy?"

I think all five display some degree of personal and professional tweets with numbers 1 and 4 primarily social, 2 and 5 primarily professional and 3 a mix. That's getting subjective, so do you think I really ought to edit the streams more tightly to make sure I can compare personal with professional on a less murky basis? Or more interestingly; What is it about the tweets above that lead you to make judgements about the person who posted them?

[etiquette note: please be kind, these are real people who volunteered, I'm interested in the process of making judgements not in judging these individuals!]