Negative values in Cohen's Kappa results?

#1
Hi,

I have performed Cohen's Kappa test in spss on my categorical data - on 15 cases - and I have got so negative values and some values that haven't worked at all. Any thoughts on how to analyze them would be very helpful.
Cheers
Brionny

Here they are:
HFACS Category Sub-Category Cohen’s Kappa % Agreement

Level 3: Unsafe Supervision Supervisory Violation - 80.0
Failed to Correct a Known Problem .815 93.3
Planned Inadequate Operations - 93.3
Inadequate Supervision 1.000 100
Level 2: Preconditions for Unsafe Acts Technology Environment .762 93.3
Physical Environment .474 73.3
Personal Readiness 1.000 100
Crew Resource Management -.098 80.0
Physical/Mental Limitation -.119 66.7
Adverse Physiological States .634 93.3
Adverse Mental States .762 93.3
Level 1: Unsafe Acts of Operators Routine Violations - 100
Exceptional Violations -.071 86.7
Perceptual Errors .762 93.3
Skill-based Errors .762 93.3
Decisional Errors .727 86.7
 
#2
Here's a bit more data to help ya understand:

I have two judges who have analysed 15 cases. Each case was subjected
to a classification framework with 16 categories. If the judge
percieved a category to be present in a case, it was marked with a 1.
If not, a zero was recorded for the absent category. I don't know how
to work out the cohen's kappa value because in some instances a
category will not be marked as present by a coder at all e.g.

Category: Physical Environment
Rater 1 Rater 2
0 0
0 0
0 1
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0

My next question would be: how do i compile the reliability scores so
that I get the overall agreement on the 16 categories? Is this
possible?
I'm maths illiterate so any help would be better in laymans terms ;p
Thanks again
Brionny