P-value is 0.33, t-stat is 0.98.
They haven’t indicated significance level or confidence interval, so based on comparing t-stat with p-value, I think null hypothesis should be rejected.
Why does the answer say “fail to reject null hypothesis”?
You cannot compare a t-statistic to a p-value; they’re not remotely the same thing. That would be tantamount to comparing how fast you’re going to how long you’ve been driving: not remotely the same thing.
You compare the calculated t-statistic to a critical t-value, or the p-value to your chosen α. You commonly choose α to be something like 10% or 5% or 1%; a p-value of 33% would lead to failing to reject the null in all of these cases.
Thanks boss, indeed of course. What was I thinking.
The lack of info about significance level tripped me. In the answer they say “in this case the p-value 0.33 is high; so we fail to reject the null hypothesis”.
But high compared to what? The t-table wasn’t given, but if they had given it then we could possibly conjecture that at p=0.1 (largest as per cfa appendix table), df=2, t-critical is 1.886. Which is greater than 0.98, so null hypothesis can’t be rejected. For higher p (like 0.33), t-critical will be lower, so we can’t say for sure whether it will be > 0.98 right?
Correct.
Well, now it’s correct.
bewildered.