What Is Happening When People Go To Polls and Don't Vote?

I have spent a lot of time in the last week compiling information from the preliminary results of the last county-wide election. I began to notice something. The "Times Counted" was usually higher than the votes "for" or "against" COSI 14 (this also shows up in the other levies, but I didn't think the effect was as dramatic). So I took the time to write down the
"times counted", registered voters, precinct, "against" and "for". I then put these into a database using Microsoft Access, and generated a few reports. This one "Voters Grouped by Towns" I used to generate two calculate fields: "How Many Didn't Vote" (Votes-For-Against), and "Percentage Who Didn't Vote In COSI Levy" (Votes-For-Against/Votes). The five "top" precincts that had the largest percentage of voters that didn't vote either way on COSI were:
|Precinct|Votes| Levy |Against| For|How Many Didn't Vote|Percentage Who Didn'tVote|
|Tol 6K|44|COSI 14|11|28|5|11.38%|
|Tol 13G|46|COSI 14|13|28|5|10.87%|
|Tol 8D|56|18|32|6|10.71%|
|Tol 13L|118|36|71|11|9.32%|
|Tol 13B|54|16|33|5|9.26%|