Quick Q: Reject H0
- Thread starter gogiants
- Start date
Err, I might get this completely wrong, but: If test statistic EXCEEDS (in absolute value) critical value, REJECT H0.
I.e. z/t-statistic is: 6.4 and critical value is 1.96 -> Reject H0
Also, AFAIK, it depends on the side you need to reject on, if it's a left sided test you reject if the test statistic is smaller and if its a right sided you reject if its larger. If it's a two tailed test this does not matter, though.
Hope this doesn't add to the confusion
I.e. z/t-statistic is: 6.4 and critical value is 1.96 -> Reject H0
Also, AFAIK, it depends on the side you need to reject on, if it's a left sided test you reject if the test statistic is smaller and if its a right sided you reject if its larger. If it's a two tailed test this does not matter, though.
Hope this doesn't add to the confusion
lijodotcom
New member
- Jun 18, 2026
- 0
- 0
Is there a simple explanation as to whether it's a left sided or right sided test?
gogiants Wrote:
-------------------------------------------------------
> If the test statistic < critical value then reject
> H0
>
> Is it true for all statistics (t, z, chi-square,
> F)?
>
> Thank you so much and good luck to all!
This is incorrect.
if test statistic > critical value then reject Ho
-------------------------------------------------------
> If the test statistic < critical value then reject
> H0
>
> Is it true for all statistics (t, z, chi-square,
> F)?
>
> Thank you so much and good luck to all!
This is incorrect.
if test statistic > critical value then reject Ho
I remember this as follow:
Right sided if H0 I try to reject asks for something larger than observed mean
Left sided if H0 I try to reject asks for something smaller than the observed mean
But don't take my word for it. Until yesterday I got most of quant quite mixed up. I spend far too much time playing with statistical toys and my understanding of these things got a little rusty.
Right sided if H0 I try to reject asks for something larger than observed mean
Left sided if H0 I try to reject asks for something smaller than the observed mean
But don't take my word for it. Until yesterday I got most of quant quite mixed up. I spend far too much time playing with statistical toys and my understanding of these things got a little rusty.