SPSS one sample t test displays incorrect CI bounds for diff of means

I ran a simple one sample t-test. All of the results are correct, except for the 95% Confidence Interval of the difference. The lower and upper bounds are both negative. It seems to me that each of the bounds should have the same magnitude, but with different signs; so the lower bound should be negative and the upper bound should be positive. For sure, 0, indicating no difference between the means should reside inside of this confidence interval.
N = 10; Mean = 77.7; Std Dev = 11.46056; St Error = 3.62415
Here is the output:

One-Sample Test
Test Value = 89

t = -3.118; df = 9; Sig. (2-tailed) = .012
; Mean Difference = -11.30000

95% Confidence Interval of the Diff : Lower = -19.4984 ;
Upper = -3.1016

Please let me know what the numbers represent in the output because it does not represent the 95% Confidence Interval of the difference of the means. Thank you!


Leave a Reply