One sample t-test using SPSS version 28

This video provides a demonstration of how to perform a one-sample t-test using SPSS version 28. I begin with some discussion of differences in available output between this version and earlier versions of SPSS, and then move on to three examples where the test is performed using fictional student data.
A copy of the data can be downloaded here:
drive.google.com/file/d/10coj...
Obtain a copy of the Powerpoint referenced in the video here:
drive.google.com/file/d/14NkT...

[@ICQwhereRU]

Thank you Mike for your videos and for providing the data set..
what kind of pen you use for writing on the output? thanks again

[@mikecrowson2462]

Hi, thank you for your comment. I use a free download called Epic Pen. You can download from here: epic-pen.com/#download
Cheers!

[@ICQwhereRU]

Thank you Mike.. I am familiar with the epic pen.

[@talzabidi1569]

Thanks so much Dr, I have few questions :
1- does t -value get effect by sample size?
2- from where do get test value ?
3- regarding accept or reject null hypothesis, is it ok to look to the sing of t value?
4- EZ is it about the magnitude of effect in our population right not in sample?

[@mikecrowson2462]

Hi Talal, thanks for visiting! Regarding your questions:
1. The observed t-value is computed as a ratio of the difference in means to the standard error. All other things being equal, the standard error will be smaller with a larger same and larger with a smaller sample. Thus, it is entirely possible to get different t-values even with the same difference in means. This speaks to one of the factors that impacts the power of your test: sample size.
2. The test value I referred to can be found in a distribution of tabled t-values (under the null hypothesis). Before we relied so much on computers, in the past we computed the observed t-value from our data and compared it against tabled values in the t-distribution. Here's an example of the t-distribution: www.tdistributiontable.com/
3. Nowadays, instead of relying on the t-distribution (such as is often found in the back of statistics textbooks) for testing our hypothesis, we use the significance level (i.e., p-value) printed out in our computer output. The p-value is actually associated with t-values from the t-distribution. The p-value is the proportion of the sampling distribution that falls at or beyond our parameter estimate. We compare this value against alpha when making a decision as to reject or maintain the null.
4. We compute an effect size based on the sample data. But ultimately we are attempting to estimate effect size as it might be in the population (just as we use the sample mean to estimate the population mean, etc.). You can think of the sample estimate as a point estimate of the population ES and the confidence interval that is provided as an interval estimate of the population ES. [Just keep in mind that the actual population ES is ultimately unknown as we do not have population data here.]

[@talzabidi1569]

@@mikecrowson2462 much appreciated.