HELP!!! SPSS 16.0 question DDD:

[Lenka]

Okay, I need your help as in now

I run a validity test on one of my independent variable
The independent variable is Extent of User Control and consist of only 3 questions, one was deleted after reliability test (bigger Cronbach's Alpha) and therefore it consist of only 2 questions

Then I click
Analyze > Scale > Reliability Analysis [Statistics: checked Item, Scale, and Scale if item deleted]
Then click ok

this is what it said
"The value is negative due to a negative average covariance among items. This violates reliability model assumptions. You may want to check item coding"

I don't know why is that and what that means. What item coding? I don't think I put it wrong, I really thought I've put the same answer that my respondents put on their questionnaire D:

Please help describing to me in a way you would describe it to kids (that means, please use simple English as best as you could)

Thank you so much beforehand~!!

[daniyagami]

I found an answer for you cause I'm too curious person and I was dying to know what are you talking about.
SPSSX-L archives -- June 2007 (#503)

Don't ask me anything, I have no idea.

[Lenka]

yep, I found that answer along with other more understandable answer on the forum SPSSX
but I still can't figure out (especially that one answer) so that's why I asked to describe the answer to me like you describe it to kids

thanks dani, but I still got no clue nor answer D:

Okay, so apparently I did it wrong, so I did it all over again
I've run the validity test and had erased the questions that has r counted <= r table: 0,361 (df = 28)

after checking on all of the corrected item-total correlation, I still get that notice since now 3 of my independent variables only have 2 questions. That "The value is negative due to a negative average covariance among items. This violates reliability model assumptions. You may want to check item coding" notice is back again.

Help.....