0todd0000 / spm1d

One-Dimensional Statistical Parametric Mapping in Python
GNU General Public License v3.0
61 stars 21 forks source link

Interpretation of SPM two-tailed inference #167

Closed ghost closed 3 years ago

ghost commented 3 years ago

Hello Todd,

I am starting to use the SPM and I have a question about the interpretation of the two-tailed inference. I am currently doing a study on the influence that jerking may have on the risk of ankle sprains. To do this I am conducting a gait study and comparing data on jerk variations in a group with ankle instability and a control group. My hypothesis is that subjects suffering from ankle instability will present greater variations in jerk than the control group (without pathology)

Here is an example of the result I obtain for the variations of the jerk in a sagittal plane (flexion-extension of the foot) following a t-test (paired t-test) (X axis = % gait cycle, Y axis = Jerk value m.s-3)

Despite reading several topics or articles, I admit that my understanding of the test results remains unclear
My question is about the interpretation of exceeding the positive or negative threshold.

Is there a difference in the interpretation of a positive or negative threshold crossing? Each time the threshold is reached, does this mean that the unstable group obtains greater jerk variation than the control group?

untitled

Thank you in advance for the time you will take to answer my question

0todd0000 commented 3 years ago

Hello, I noticed that the issue is now closed. Just in case this was an accident:



Is there a difference in the interpretation of a positive or negative threshold crossing?

No, in both cases t values that large are expected to occur relatively infrequently if there is truly no effect.



Each time the threshold is reached, does this mean that the unstable group obtains greater jerk variation than the control group?

The t value is conceptually (meanA - meanB) / variance , so is positive when meanA is greater than meanB, and vice versa.