The presenter is the typical presenter whom I expect to watch. He precisely explained the topic. He used time effectively - only important points are explained without boring.
Thank you Todd. Great explanation. I was struggling with SPSS. I was about to quit the couse in statistics, Thanks God I found you !!! MERCI du fond du coeur. From Québec
Great videos. I spent 4 hours in class on this but got more out of these videos in 30 minutes. Very informative with great presentation. Thank you sir!!
What a wonderful comment...thank you so much. I am glad that the videos are helpful to you. Please share with your classmates if the videos might help them, too.
So true, my statistics teachers were way too theoretical that it's so difficult to comprehend for beginners. These videos contain practical examples, which makes it much easier.
your explanation is just GREAT, it is clear and answer all my questions I get from watching the others tutorials. I am grateful for this tutorial, I can't thank you enough for this.
Thank you so much for the videos, it is really helpful. I hope all instructors can watch these videos before they give courses. It is a much clearer and greater explanation than my course instructor.
You're very welcome! I have heard from other instructors, especially those new to teaching stats that my videos give them a head start with their own teaching. I love it. It's all about making stats easier to understand so more students will learn.
Dear Todd, thank you for the very clear explanation! I love how you explicitly explain all three different ways to find out if the differences are statistically significant. I have a question and I would be very grateful for your answer: The video said that *the 95 % confidence interval is a range in which the sample mean* (here: the mean difference) *falls 95 % of the time, if we repeat the experiment numerous times*. If I understand correctly, this doesn't make much sense because we would calculate a separate confidence interval for each sample, i.e. there is no one confidence interval within which 95 % of the sample means fall. Also, this statement doesn't make any inferences about the true parameter. Is it meaningful to talk about confidence intervals without an inference to the population parameter? This is the way I understand confidence intervals: a 95 % confidence interval means that if we take a large amount of samples (e.g. 100 samples), and we calculate a statistic (e.g. a sample mean) and estimate a confidence interval for each of those samples, then 95 % of those confidence intervals would contain the true parameter. Is this correct or have I misunderstood something? Thank you so much for your help!
Hello Todd, Thank you so much for these amazing videos! Listening to your very clear explanations, examples etc, you made me actually like SPSS. A massive thank you!!!
You are so welcome! Love hearing from people who find the videos helpful. If all goes well, you will see an updated version of this series pretty soon. Working on SPSS 29. Cheers!
Learned so much from this. Is there like a follow up video on how to interpret and place these data in a graph form? This is for a thesis/research paper.
This video may help...it goes into more detail about the t test. You can also find a playlist on my Chanel. Good luck. ua-cam.com/video/VWEC9TIU1qs/v-deo.html
Sir it's quite helpful but how did u fill the answers from questionnaires into the data view. if I have 15 people in a survey and I have 17 questions on both type of my questionnaires pre & post,both have MCQ type of questions,marked by participants differently..some people marked a.) some marked b.)...so how do we enter these answers of people onto the data view...did u calculate the mean for all the questions first & then enters their means into first two rows of data view? if u did so,then where did u enter the sample size?? PLEASE CLARIFY AS SOON AS POSSIBLE...I'm in a mist of completing my dissertation but don't know how to fill answers market by people on the data view sheet of spss. We can calculate the mean easily by spss because the entries of the questions & sample size are quite easy...but I don't understand how to fill the answers of questionnaire on the data view to analyse the paired t test.Pls reply😳🙏🙏
Thank you so much for the video! However, I had one quick question and will really appreciate if you can help me with it. I ran a paired sample t test but somehow obtain two significance values (one-sided p and two-sided p) and I had no idea why. I asked my teacher but she seemed to have no clue as well.
Is it ok if I get sign.value 0.000 in paired samplest test.. watever data i collected from the survey that is savings before and after covid.. after entering the data I am getting sign.value as 0.000..is it fine?
Hello doctor. If I want to see the change in sales before and after the covid19. Then should I use paired sample t test or any other test. Can you please suggest.
You will definitely use paired or repeated measures. The repeated t test would allow you to compare two time points. A repeated measures ANOVA would allow you to compare multiple times (months of average sales). You might also look into a time series analysis if you want to track changes across the year. Good luck
i mean for example at first time i did measure100 cases. and find the mean. after that for internal validation i measured another50data and compared the mean with previous one
So the means are only statistically significantly different if they pass all three of the tests? So for my data it passed two out of the three, so does that mean it is not statistically significant?
You should go back and check those numbers. Because of the interlocking math used to calculated each, the significance value and confidence interval interpretations should always agree. It may be that you were off by a degree of freedom with the critical value, especially if the p was really close to .05. Other considerations would be if you were using a one-tail test, or an alternative p value such as .01. Good luck
For a paired t test the assumption of normality is for the difference scores. So if you create a variable which is Posttest - Pretest, those differences should be approximately normally distributed. Hope that helps.
Yes, you can use a paired t test for past and present responses...assuming that you are using the same scale measurement for both times and that you can identify whose past response goes with what present response. You need to be able to pair the responses for each respondent. Good luck!
Very much appreciated for these fantastic series videos! i have a question, could we conduct an independent t-test to determine whether the profitability means for small-sized farms were significantly different from the medium-sized farms. or should we use other methods thanks
@@ResearchByDesign Thank you so much for your prompt reply, Dr. Tod! in fact, I use chi-square and The results of tests seem unreliable as several of my cells have an expected frequency of less than 5 (for each test). The t-test looks inappropriate due to the nature of my data if I am not mistaken. I am wondering how could I compute the mean profitability based on the categorical data that I have. Can you explain how you compute the mean profit of these farms? or use the best method? Your reply would be highly appreciated.
Can I use this to compare year over year volumes after an intervention ? ie 2018 volume of CPR compared to 2019 CPR after residents are introduced in 2019?
hy if we want to compare the two means of same sample measured by automated method and manual method so whic test i will applay ? this one ? if yes then how ?
What you describe is the Paired Samples (or repeated measures) t test. You are measuring the same sample twice using different methods. Those methods should be the same if they are valid. This video will give you the details: ua-cam.com/video/VWEC9TIU1qs/v-deo.html
Hello sir, one question please, you told not to worry about negetive value but when you replaced both variables lower and upper confidence value also changed.In first case it was significant because was negetive and less than 0 and after they turned into a positive value passing 'zero'. Totally confused.
The negative sign for the t value only tells you that the mean of the first group was smaller than the mean of the second group. If you switched the coding on the groups (i.e. 1 = 2 & 2 = 1, now), then the t value would be exactly the same only without the negative sign (i.e. positive). For the confidence intervals, if both values are negative (or both are positive), then they are on the "same side of zero", the interval does not include 0. In that case, the result is statistically significant. So...you can ignore the negative sign for t values because you will report the actual means, but you should note the negative signs when interpreting a confidence interval because they tell you if the interval includes zero (when one is positive and the other is negative).
Here is a playlist that includes paired samples t tests. There are other playlists for other types of analyses. Hope you find them useful. ua-cam.com/video/uMLPR8aTYgQ/v-deo.html
Suppose: I want to check the fresh weight of a fungus (Z) and then dried for two days or three days then check the dry weight, can I use paired sample t-test for the different weight of the same fungus
Absolutely...your fresh weight would go in first, then the dried weight would go in second (like a posttest). The t test would tell you whether the weight was statistically significantly different and the Cohen's d would tell you the effect size for drying over two (or three) days.
You COULD use a paired t test to measure before and after measurements of either temperature or of Co2. You COULD NOT use a paired t test to compare temperature to Co2 because those are on different measurement scales. You COULD use correlation to study the relationship between temperature readings and Co2 levels. Good luck
you can use ANOVA... temperature would be the dependant variable. you can classify temperatures into medium, high and low levels of CO2 and then compare means of temperature for these categories.. correct me if i am wrong
welcome! then you find if the difference in means of the 3 categories of temperature, classified according to dif co2 levels (low, medi, high) are statistically significant or not. refer to ANOVA videos for more.
You would need a variable, probably a nominal variable with the product preferred by each customer, then you would need to record whether the customer was male or female. The best analysis would be a cross tabulation, perhaps also with a chi-square. Good luck
Thanks for the comment...yes, I have all the videos submitted so that they can be used in classrooms. That way more people have a chance to learn. Cheers...and please keep watching
Machine Learning would be a great topic. I have a huge list of videos I want to do, however, so it will have to wait a while. Too little time, I am afraid. Thanks for the interest, though.
I'll assume that you are asking about the paired samples t-test and that your CV is .05...in that case the tests is not statistically significant. I would report t(19) = , p = .25, ns. You can also report a Cohen's d, if you like.
The presenter is the typical presenter whom I expect to watch. He precisely explained the topic. He used time effectively - only important points are explained without boring.
Thank you so much. Glad that you enjoyed it!
Thank you Todd. Great explanation. I was struggling with SPSS. I was about to quit the couse in statistics, Thanks God I found you !!! MERCI du fond du coeur. From Québec
Glad it helped! And de rien from Missouri !
Great videos. I spent 4 hours in class on this but got more out of these videos in 30 minutes. Very informative with great presentation. Thank you sir!!
What a wonderful comment...thank you so much. I am glad that the videos are helpful to you. Please share with your classmates if the videos might help them, too.
These are extremely great vids because Dr. Todd starts from the simple to the complex.
So true, my statistics teachers were way too theoretical that it's so difficult to comprehend for beginners. These videos contain practical examples, which makes it much easier.
My current teacher is a good person but not very good at communicating, so thank you for being both. I learnt a lot.
your explanation is just GREAT, it is clear and answer all my questions I get from watching the others tutorials.
I am grateful for this tutorial, I can't thank you enough for this.
You're very welcome!
Whenever I become bore, I come to do these statistical analysis, I enjoy these lectures, very good teacher, explain everything clearly.
Thanks a lot, There aren't that many teachers who can teach statistics and you teach it amazingly
Thank you so much for the videos, it is really helpful. I hope all instructors can watch these videos before they give courses.
It is a much clearer and greater explanation than my course instructor.
You're very welcome! I have heard from other instructors, especially those new to teaching stats that my videos give them a head start with their own teaching. I love it. It's all about making stats easier to understand so more students will learn.
What an absolute ripper of a video. I wish my Lecturer was as clear cut as you.
Great teaching, thank you
Best explained. I have got some clue from the beginners video.I spent 2 hours to attend it.
Dear Todd, thank you for the very clear explanation! I love how you explicitly explain all three different ways to find out if the differences are statistically significant.
I have a question and I would be very grateful for your answer: The video said that *the 95 % confidence interval is a range in which the sample mean* (here: the mean difference) *falls 95 % of the time, if we repeat the experiment numerous times*.
If I understand correctly, this doesn't make much sense because we would calculate a separate confidence interval for each sample, i.e. there is no one confidence interval within which 95 % of the sample means fall.
Also, this statement doesn't make any inferences about the true parameter. Is it meaningful to talk about confidence intervals without an inference to the population parameter?
This is the way I understand confidence intervals: a 95 % confidence interval means that if we take a large amount of samples (e.g. 100 samples), and we calculate a statistic (e.g. a sample mean) and estimate a confidence interval for each of those samples, then 95 % of those confidence intervals would contain the true parameter. Is this correct or have I misunderstood something?
Thank you so much for your help!
I appreciate all of the series so much😭😭
These videos are interesting, perfectly presented and very useful.
Hello Todd, Thank you so much for these amazing videos! Listening to your very clear explanations, examples etc, you made me actually like SPSS. A massive thank you!!!
You are so welcome! Love hearing from people who find the videos helpful. If all goes well, you will see an updated version of this series pretty soon. Working on SPSS 29. Cheers!
The BEST TEACHER!
Thanks, Dr.Todd. You are a great teacher.
Excellent presentation . I was struggling with how to explain SPSS output tables . Thanks a lot . God bless you.
So glad to hear that. Very happy that the video could be helpful. Thanks for watching!
Thank you very much sir. I have learnt a lot from your lecture.
One of the best so far
Great, I really learned in these minutes more learning in my lecture . Thank you.
It's very helpful useful and clear for beginners researchers like me.
Thank you very much professor for your help
Glad to hear that
I have learnt a lot from this video. Thank you
Very brief and instructive, thank you for making SPSS this easy
You're very welcome!
Your explanation so good. It helped me to run a data in SPSS. Thank you so much
You're welcome! Great to hear it
What does it mean when std deviation of posttest is lower than pretest although mean different is different from post to pre?
Excellent explanation! I will follow all the videos, thank you!
Thank you so much for the comment. I hope that you find the other videos helpful, as well.
Learned so much from this. Is there like a follow up video on how to interpret and place these data in a graph form? This is for a thesis/research paper.
This video may help...it goes into more detail about the t test. You can also find a playlist on my Chanel. Good luck. ua-cam.com/video/VWEC9TIU1qs/v-deo.html
This guy tells some great dad-jokes, I can tell
Sir it's quite helpful but how did u fill the answers from questionnaires into the data view.
if I have 15 people in a survey and I have 17 questions on both type of my questionnaires pre & post,both have MCQ type of questions,marked by participants differently..some people marked a.) some marked b.)...so how do we enter these answers of people onto the data view...did u calculate the mean for all the questions first & then enters their means into first two rows of data view? if u did so,then where did u enter the sample size?? PLEASE CLARIFY AS SOON AS POSSIBLE...I'm in a mist of completing my dissertation but don't know how to fill answers market by people on the data view sheet of spss. We can calculate the mean easily by spss because the entries of the questions & sample size are quite easy...but I don't understand how to fill the answers of questionnaire on the data view to analyse the paired t test.Pls reply😳🙏🙏
Thank you so much for the video! However, I had one quick question and will really appreciate if you can help me with it. I ran a paired sample t test but somehow obtain two significance values (one-sided p and two-sided p) and I had no idea why. I asked my teacher but she seemed to have no clue as well.
Thank you so much for uploading all these videos!! super helpful
Is it ok if I get sign.value 0.000 in paired samplest test.. watever data i collected from the survey that is savings before and after covid.. after entering the data I am getting sign.value as 0.000..is it fine?
what does it mean if you confidence interval does include 0? for both
If the confidence interval of the mean difference includes 0, then there is no statistically significant difference between the means.
very informative . thanks a bundle
Most welcome!
Very good, informative video.
Glad it was helpful!
what show t-calculated and t tabulated
Hello doctor. If I want to see the change in sales before and after the covid19. Then should I use paired sample t test or any other test. Can you please suggest.
You will definitely use paired or repeated measures. The repeated t test would allow you to compare two time points. A repeated measures ANOVA would allow you to compare multiple times (months of average sales). You might also look into a time series analysis if you want to track changes across the year. Good luck
@@ResearchByDesign thanks a lot for your reply.
Thanks for being available at the right time!
You are a great teacher
Hi, thanks for the video. Where can I find the distribution table? Or whether this table could be applied to all research topics?
Write on google critical t value distribution calculator
Splendid! Thank you for this :)
thanks a lot. i want to validate two categories ofh data to know the mean are the same. is it possible to compare with t test?
i mean for example at first time i did measure100 cases. and find the mean. after that for internal validation i measured another50data and compared the mean with previous one
simply ,thank you
You are welcome! Glad it was helpful
So the means are only statistically significantly different if they pass all three of the tests? So for my data it passed two out of the three, so does that mean it is not statistically significant?
You should go back and check those numbers. Because of the interlocking math used to calculated each, the significance value and confidence interval interpretations should always agree. It may be that you were off by a degree of freedom with the critical value, especially if the p was really close to .05. Other considerations would be if you were using a one-tail test, or an alternative p value such as .01. Good luck
Hi, the assumption of normality for paired t test is normality test for each score or residual of both score?
For a paired t test the assumption of normality is for the difference scores. So if you create a variable which is Posttest - Pretest, those differences should be approximately normally distributed. Hope that helps.
@@ResearchByDesign thanks so much
Thanks lecturer i understood more since hours ago.
Fascinating lesson . Thank you!
Glad you enjoyed it!
Can i use paired t test if Im going to compare and get the difference of the responses of my respondent from the past and present?
Yes, you can use a paired t test for past and present responses...assuming that you are using the same scale measurement for both times and that you can identify whose past response goes with what present response. You need to be able to pair the responses for each respondent. Good luck!
Very much appreciated for these fantastic series videos! i have a question, could we conduct an independent t-test to determine whether the profitability means for small-sized farms were significantly different from the medium-sized farms. or should we use other methods
thanks
Yes you can! Small vs. medium would be your groups and profits (in currency) would be the DV. Good luck with your project.
@@ResearchByDesign Thank you so much for your prompt reply, Dr. Tod!
in fact, I use chi-square and The results of tests seem unreliable as several of my cells have an expected frequency of less than 5 (for each test). The t-test looks inappropriate due to the nature of my data if I am not mistaken. I am wondering how could I compute the mean profitability based on the categorical data that I have. Can you explain how you compute the mean profit of these farms? or use the best method? Your reply would be highly appreciated.
Can I use this to compare year over year volumes after an intervention ? ie 2018 volume of CPR compared to 2019 CPR after residents are introduced in 2019?
You could use this to compare one year to the previous. If you want to compare multiple years, you would use a repeated measures ANOVA.
Great Lecture on Paired samples t Tests.
Superb presentation.
perfect teaching.thanks
hy if we want to compare the two means of same sample measured by automated method and manual method so whic test i will applay ? this one ? if yes then how ?
What you describe is the Paired Samples (or repeated measures) t test. You are measuring the same sample twice using different methods. Those methods should be the same if they are valid. This video will give you the details: ua-cam.com/video/VWEC9TIU1qs/v-deo.html
Hello sir, one question please, you told not to worry about negetive value but when you replaced both variables lower and upper confidence value also changed.In first case it was significant because was negetive and less than 0 and after they turned into a positive value passing 'zero'. Totally confused.
The negative sign for the t value only tells you that the mean of the first group was smaller than the mean of the second group. If you switched the coding on the groups (i.e. 1 = 2 & 2 = 1, now), then the t value would be exactly the same only without the negative sign (i.e. positive). For the confidence intervals, if both values are negative (or both are positive), then they are on the "same side of zero", the interval does not include 0. In that case, the result is statistically significant. So...you can ignore the negative sign for t values because you will report the actual means, but you should note the negative signs when interpreting a confidence interval because they tell you if the interval includes zero (when one is positive and the other is negative).
@@ResearchByDesign thanks a lot.Loads of love from India.
where are the other videos of the university about statistical theory do we know?
Here is a playlist that includes paired samples t tests. There are other playlists for other types of analyses. Hope you find them useful. ua-cam.com/video/uMLPR8aTYgQ/v-deo.html
that is realy helpful thanks Mr.TODD DANIEL.
How this is use in microbiology
This is very helpful. Thank you.
Thank you for the information..
Well explained. Thanks
Suppose: I want to check the fresh weight of a fungus (Z) and then dried for two days or three days then check the dry weight, can I use paired sample t-test for the different weight of the same fungus
Absolutely...your fresh weight would go in first, then the dried weight would go in second (like a posttest). The t test would tell you whether the weight was statistically significantly different and the Cohen's d would tell you the effect size for drying over two (or three) days.
Can i use paired sample t test if i want to know if there is significant difference between Temperature and Co2 measured in a house?
You COULD use a paired t test to measure before and after measurements of either temperature or of Co2. You COULD NOT use a paired t test to compare temperature to Co2 because those are on different measurement scales. You COULD use correlation to study the relationship between temperature readings and Co2 levels. Good luck
RStatsInstitute thank u so much sir! This helps me a lot..but what test can i use to compare temperature and co2?
you can use ANOVA... temperature would be the dependant variable. you can classify temperatures into medium, high and low levels of CO2 and then compare means of temperature for these categories.. correct me if i am wrong
Rishika Malik thank you for this info ! I will present my project next week
welcome! then you find if the difference in means of the 3 categories of temperature, classified according to dif co2 levels (low, medi, high) are statistically significant or not. refer to ANOVA videos for more.
learned in 10 minutes more than I learned in college course
what if i wanted to know which from malw and female wanted a certain product
You would need a variable, probably a nominal variable with the product preferred by each customer, then you would need to record whether the customer was male or female. The best analysis would be a cross tabulation, perhaps also with a chi-square. Good luck
do you have video with that?
It's very helpful! Thanks
So nice👍
Thanks 🤗
The paired t test needs a little bit elaboration. If u can explain with the weight loss therapy examplr it would be easier to understand...
Well presented videos. Subbed. Thanks.
Thanks for the comment...yes, I have all the videos submitted so that they can be used in classrooms. That way more people have a chance to learn. Cheers...and please keep watching
thanks a lot. very helpful
please add videos on machine learning
Machine Learning would be a great topic. I have a huge list of videos I want to do, however, so it will have to wait a while. Too little time, I am afraid. Thanks for the interest, though.
How to interpret data when our CI is 75%, sig. 0.250, df is 19 and mean is -3.75, please help me out to interpret it decently.
I'll assume that you are asking about the paired samples t-test and that your CV is .05...in that case the tests is not statistically significant. I would report t(19) = , p = .25, ns. You can also report a Cohen's d, if you like.
@@ResearchByDesign I got it, thanks a bunch
starts at 2:22
Done
very cinematic