TrustFlow results for joshwriting

I tried out TrustFlow II for LiveJournal. The following people not on the friends list for joshwriting are close by:

• devrose, ricevermicelli (300 - 350)
• sarcastic_kitty, crazyprogrammer, saraken, sacz4e (350 - 400)
• anonshadow, imaginatrix, nonnycat, tinkerbellt, celticflamerhia, windowful (400 - 450)

( More results below the cut... )

Created by ciphergoth; hosted by LShift.

TrustFlow II: Who is closest to your friends list?

Using results to prove validity of a question

I have been having a discussion in which I am failing to explain what is to me an essential detail of how questionnaires work.

The issue is how accumulated responses relate to the question asked.

Laurel asks "Do you like the number 5?"

50 people say they like it. 50 people say they don't.

Hardy takes Laurel aside and says "Your question was biased and affected the results."

Laurel responds "If it was that biased, the results would not have been so balanced."

Hardy retorts, "If it had not been that biased, the results would not have been balanced at all!"
************

My assertion is that neither side has truth here and they cannot know what the impact of the possible bias is.

A) Am I being at all clear?

B) Am I right?