How to do a Paired Samples t Test in SPSS (11-7)

Using a dog training example, we compare two types of reinforcement training using a Paired Samples t-test. We work through the five steps of hypothesis testing, and conduct the Paired Samples t-test in SPSS. We compute Cohen’s d, we interpret the results, and write them up in proper APA style.
Because Cohen’s d effect size is susceptible to correlation between the group scores (this is true of any repeated measures design), we discuss for to use the Dunlap et al. (1996) correction formula for Cohen’s d. This correction is included in an Effect Size Calculator that you can download below.

This video teaches the following concepts and techniques:
Paired Samples t-test in SPSS
IBM SPSS Statistics software
Cohen’s d and the Dunlap correction for effect size

Link to a Google Drive folder with all of the files that I use in the videos including the Effect Size Calculator for t Tests and datasets. As I add new files, they will appear here, as well.

[jackiehuang]

Best SPSS tutorial video I have ever seen. Thank you so much!

[aliakhan]

Lucid explanation...well planned...well organised...Great work!!!

[nikoloztarielashvili]

very helpful video! good explanation, everything is clear. thank you!

[prashant]

What is cohens d... Never heard of it.. Why is it important?

[tastyfoodbuzz]

I have a doubt ..When you conclude in apa style that clicker training is more effective than traditional training as we were doing two tailed test where we can only conclude that they are different by rejecting null hypothesis if we were using one tailed test then we can mention like one is more effective than other what's the difference I couldn't understand

[akakigeo]

As I understand, paired samples T-test is used traditionally for before and after design. Here we have two separate groups of dogs assigned to either clicker or food training program. They, as I understand, are not related except for the fact that they are simply dogs :) So my question would be why we are using paired samples t test here? Why are they related if none of the dogs had both training programs passed? ​ @Research By Design

[c.danieldemagalhaesfilho]

Could you comment on the p value for the correlation? and what it means then if you do find a statistically significant correlation but a non significant paired t test

[Evelina_Fani]

Hello!I have a question for you. You used the same data in two methods, paired sample t-test and on the previous video independet sample t-test and you concluded on different things. With the paired samle t-test you say that the difference in the means is significant but with the independent sample t-test there was no significant difference. How is that possible?Am I missing something?Thank you for your time!

[aliakhan]

Sir @
Research By Design, I am new to this stats thing, Could you please answer my questions?
First, What if the t value is negative? how we are going to compare it with the critical value? and my second question is you have said "We also see the 95% confidence interval for the difference between paired sample means shows a lower bound of +1.78 and an upper bound of 2.322. Because this confidence interval does not cross 0". But sir here both +1.78 and 2.32 greater than 0 that means they have crossed zero. how you have interpreted that they have not crossed zero?

[praveentiwari]

Sir if my two tailed value is 0.053 so is it significant???? What so we consider?

[syaidatulafiqah]

do this test can be use if we have more that one pair?

[bear]

What if I want a one-tailed paired sample test?? Is there no way??

[c.danieldemagalhaesfilho]

Could you comment on the p value for the correlation? and what it means then if you do find a statistically significant correlation but a non significant paired t test