Question

I am using excel to analyze 2 sets of data and see if they are statistically the same. On paper, they look different enough that I should have a larger t test value, but they're not.

Here i have:
sample A: 0.000962, .0007925

sample B: .003, .00282

When I do a t.test in excel, (t.test(sample A, sample B, 2, 3)) I get a t value of .00365.
Seeing as how sample B's numbers are way bigger than sample A, shouldn't the t value be larger?

Thank you!

Answer 1

Sample A	Sample B
0.000962	0.003
0.000793	0.00282

t-Test: Two-Sample Assuming Unequal Variances
	Variable 1	Variable 2
Mean	0.000877	0.00291
Variance	1.44E-08	1.62E-08
Observations	2	2
Hypothesized Mean Difference	0
df	2
t Stat	-16.4432
P(T<=t) one-tail	0.001839
t Critical one-tail	2.919986
P(T<=t) two-tail	0.003678
t Critical two-tail	4.302653

The above analysis is done using excel. The absolute value of the statistic is 16.4432 and the p-value for the this test is 0.003678 which is much less than 0.05( alpha value), so that the test is rejected, that is we can say the two samples have statistically significant mean, or they are significantly different.

So, as you guessed the t value is large. I think you just confused t-value with the p-value.