Performed a paired sample t-test but not sure how to interpret the results.?

I did a 10 question pre %26amp; post survey of the same 10 people. The mean difference is -1.3, Degrees of freedom= 9, standard error of the mean of d= .86, t-statistic for paired data = -1.51, Critical value for alpha .05 would be 2.262.


Now I am not sure how to know whether I accept or reject the null hypothesis. Not even sure what the null hypothesis should be.


My research is whether or not medication improves the school performance of ADHD children. Pre-Survey (before meds) and after survey (while on meds).


I have all these numbers and don't know how to make since of them. Guess I should've paid more attention in that statistics class!


Please help.

Performed a paired sample t-test but not sure how to interpret the results.?
It's very difficult to answer what the null hypothesis should be. If you don't know what you're comparing, then why do the test? I assume you are comparing pre and post test results. Your null hypothesis would most likely be that the test results are the same for both tests. It also sounds like you're preforming a one-sided test on what could be a two-sided test. You would reject any statistic that is above 2.262 or below -2.262 and this would give you an alpha of .05 (.025 on each tail). Basically it looks like you found a two-sided significant value (2.262) but think you're doing a one-sided test. If you're looking to see if there's a significant difference period, then you'd use the two-sided test. If you want to see if the second test scored better, then use the one - sided. (It'd be a lot easier to explain this if I knew exactly how you did the math). If you go above the significant value or below it then you reject the null. Regardless, this won't change your result in this case since your value is -1.51. Basically, any statistic outside of alpha means you reject the null because you have a statistically significant reason to say that the data in the first test is different than the data in the second test. You are making sure there is a significant difference in the data in order to say the medication made a change. Based on your results, you aren't seeing a change since the results. You would not reject the null hypothesis. This means that, based on your results, this drug does not alter test results in patients with ADHD. And yes, you should have paid more attention in statistics class. This stuff really isn't that difficult but you seem so lost. If you plan on doing more of this type of thing I'd suggest taking a stats class again or at least getting a stats textbook and look at it. Also, make sure this is the correct test and that you did it correctly. Based on what you've said, I'm assuming that your resulting t-statistic is correct and that you're using the correct test. It sounds like you have (comparing two results given what it's for) but make sure.
Reply:Based on what you have stated, your null and alternative hypotheses are:


H0: No difference between Pre and Post results.


H1: Post results show better results (medication improves performance)


I presume that the 10 people were the subjects of the experiment and their scores were recorded.


This is a one-tailed test. The critical value at alpha =0.05 is 1.833 (or -2.262 at alpha=0.025). Based on this, the null hypothesis is rejected. That is, medication has improved the subjects' performance.