Having used SPSS in my undergrad, I did not expect there to be so many aspects that we didn't go over during our course. I'm glad these videos are offering a refresher and a new challenge.
sorry to be offtopic but does anybody know of a method to get back into an Instagram account?? I stupidly lost the account password. I would appreciate any tips you can offer me.
OMG! So happy to find you teaching advanced stats. I am getting my doctorate and taking stats. I watch your psych videos!!! Thank you for creating this content to help fill in the gaps that my stats book doesn’t address!!!
Dr. Grande- As I am familiar with SPSS- this video refreshed my memory about how to organize the plots. Interesting, SPSS is very straight forward and is error controlled, if your information is put in right. Thanks so much for the refresher!
Thank you for the video Dr. Grande. SPSS is extremely intimidating when trying to navigate all the tables and graphs created, and understanding what it all means. It was helpful to have a break down of what the numbers mean and what the graphs are showing within this video.
Before this class I have never heard of the program SPSS. When I researched what the program did I was pleasantly surprised at what it was created to do.After watching Dr. Grande's walk-through of how to calculate if the distribution is normal in a study I feel more confident about using SPSS in the future.
Normality tests can be measured in SPSS to determine whether they are normal or not. This particular example used 90 participants. From this one has to determine whether the pretest and posttest scores normally distributed. There were two different types of tests that were used Shaprio-Wilk test and Kolmogorov-Smirnov test. These were used to determine the validity. This was a bit confusing but I understand the general idea.
Toniette Morda I found it somewhat confusing as well, but got the basic concept. I think further practice would help! But overall, certainly a useful program in determining normality.
This is another video where I feel a bit in the dark. I like how you connect the KS and SW test to the histogram and how the graphs will look as they should regarding being normally distributed as per their results. I found it useful that you explained that the point of this is that researchers want to know if a distribution is normal, and what a typical data setup in SPSS would look like.
Having the step by step instructions is important for me to be able to utilize SPSS effectively, but this video also teaches about how to apply normality tests. I do think I would understand applying the results better if I was given an example of what the data was and what the actual null hypothesis was, but as far as reading the results and the graphs, this video was helpful in teaching me normality tests in SPSS.
SPSS seems like it is a cool program and it is really interesting to see the difference between it and Excel. The names of the tests are interesting and I really liked learning how to approach them. This video was very helpful and was a good introduction to SPSS.
The video was easy to follow and Dr. Grande makes this seem so simple. This is all new to me but seeing the Normality test in SPSS gives me a better understanding. I will need more information on some of the concepts, Kolmogorov-Smirnov and Shapiro-Wilk tests, discussed.
I remember when I used SPSS in the past, plugging in the correct numbers and keeping everything consistent, was difficult for me, because numbers and math isn't always easy for me. I'll really need to watch this again to become more familiar and comfortable with this program.
I have worked wit SPSS a lot in the past and I still never really understood it. I liked how this video broke down the different parts and really explained what was going on in SPSS at each part.
This video yielded many new concepts that I was unfamiliar with. However, the null and rejecting the null hypothesis is important in determining statistical significance. The SPSS system seemed to be user friendly and program features are easy to find. Very inflightfull video.
***** I used SPSS a lot in undergrad and yes it is very user friendly once you learn where all the buttons are! It will become a great tool for you to use for research.
***** It is definitely good that I can now grasp what kind of software SPSS is! I had heard references to it previously, but it didn't mean anything to me so it was nice to see how it operates.
***** I liked how Dr. Grande showed us SPSS because it did seem to be user friendly. When using SPSS in the past I did not think it was user friendly at all, but that is probablyy because I did not know what I was doing. After watching this video I believe I might be able to work with SPSS a little better.
Since I don't have first hand use with the SPSS and have only learned of it by name and never in detail this video was helpful in laying out where to find information on the spread sheet. I am not ashamed to say that many of the videos have been or will be viewed many times. Overall it did help me understand some of the terms used.
Danielle Duboski I also have never used SPSS, and I have never even heard of it before. I am glad that this portion of the class is in video form so we continue to go over it until we feel comfortable with the information. You will not be alone in viewing certain videos multiple times!
Great video. Before doing any sort of analysis, it is important to be very familiar with the data set so that you understand what you hope to achieve and what the most appropriate process is to determine the relationship between the results that were attained and how the data is interpreted and then analyzed.
It was nice to see the steps and how to apply it in Excel and complete the analysis. But I had a hard time focusing on the video since it was very mathematical, as my peers previously explained. I am unfamiliar with SPSS and have never used this before. Very useful in determining normality and understanding significance.
This video was very insightful being that I have never used SPSS. It seems easy enough but I have to actually use it to get more familiar with it. I believe I have a better understanding of statically significance as well as normality. The system seems to be user friendly and structured for data entry.
I have never used SPSS and to be honest I never heard of it until Dr. Grande mentioned it in class. However, watching and listening to Dr. Grande's step by step approach for normality made it appear easy to use. Testing for normality appears pretty straight forward, so I do not feel lost or like I do not grasp the concept.
Danielle Maldonado Danielle, I have never used SPSS either. Watching the video and seeing the steps does make it appear easy. I am sure it will take more time to fully grasp SPSS and I will need to watch this video again.
It's been quite a while since I have used SPSS and this was a great video explaining what all the output that SPSS gives you represents. I thought the explanation of what output tells you if it is statistically significant was great. I think that showing that one test of normality is more valid than the other was important and many would not now that. I myself had forgotten that.
I am not yet familiar with SPSS but I do appreciate some exposure to the information. Hopefully, as the course progresses I will understand this better. It is helpful to know that software can help organize the information and interpret the data, I will take away that the Shapiro-Wilks is considered more valid.
I am not very comfortable nor confident with this of statistics and distributions. This video helped me to gain more insight into the subject. Although still very skeptical, I think using videos, such as this one, will help me to understand this subject matter a lot better.
Thankyou Dr. for your clear explanation. However, i was wondering, since the post test scores are not normally distributed, can we still opt for parametric analysis test?
Hi Todd! Your video helped me analyse the normality for my data. I have a question though. My pre treatment scores are normally distributed and post treatment scores are not normally distributed. To give you the details, sample size 50, i'm testing 3 variables pre and post treatment and want to see if the scores are significantly different after therapy. What test should i use in this case, Mann-Whitney U test?
Thank you Dr Todd for this video, I would like to ask if Iam testing 5 main variables as my primary outcome, and i found only 2 had not normally distribution and the rest 3 variables are normally distributed, would i use parametric or non parametric test allover testing my primary out come??? or would i use a parametric test for normally distributed variables and non parametric for not normally distributed one?
Watching the various videos on normality testing helps to understand which tests are most relevant for what data. SPSS is becoming less scary with each video I watch.
It has been a long time since I have used SPSS but this video was a brief and helpful reminder. I think I will need further practice with this program, but it seems to remain very useful in determining normality and understanding significance.
I have never used SPSS. However the system seemed easy to navigate and merge with research statistics. I agree, an in depth tutorial would prove beneficial.
This video was very helpful in breaking down the steps to determine how to tell if there is a normal distribution . I am not that insightful when it comes to statistics, but the video definitely explained the process in a clear and concise way.
+Nakesha Tucker I agree! The video clearly went through each step which is extremely helpful when trying to actually do this. I think once we understand these tests a bit more, this video will be even more helpful. I do feel confidant that if I were asked to find out if there is a normal distribution I would be able to do just that by following these steps.
Hello Dr. Grande, I have one question. Do the results change when you include several variables in the dependent list for the normality analysis? When I include several variables and I have missing values in some of my variables the final sample size of all the variables decrease. Instead of having a sample of 78 in a variable I have less because of the missing values in other variables included in the dependent list. So my question is,. when checking normality when you have missing values, should I check normality using just a single variable on the dependent list per test?
Great video. One question. I have a dataset which involves physiological variables. All skewness and kurtosis values are within -1.96 and 1.96, apart from three variables. Also, I have some where skewness & kurtosis values are normal, but the shapiro wilk value indicates non-normality (i.e. p
Hi, thanks you for the video :) If you have a data set resembling the one in the example were the pre test is normally distributed, but the post test is not, what statistical test should we apply?
>Hey, thank you for the great video! I was wondering for questionnaires: do I first add the scores of all the items toghether and test for normality afterwards? So what I mean I have 3 Items testing one underlying construct so obviously for every participant one score for every item. At the end I need to add all the Items for the questionnaire but do I first do the normality test or after I combined the Items? Thank you so much!
Dr. Grande, I have been watching your videos for months now and have recommended your site to cohort members. Would it be possible for you to create a video demonstrating a 3-way MANCOVA. Your two-way has been a great help. My problem is that I have a 240 sample but normality is non-normal. I have tried everything (transformations etc.). Any suggestions? Regardless, I would love a video.
Thanks It was very useful! just a question, in the interpretation how can l explain the statistic value? Or, can l just say that the distribution isn't normal because the significance is lower than 0.05? (Mine is 0). I've already done the q-q plots and everything seems congruent (non normal distribution). Thanks in advance :D
Dr Grande i have a group of 30 participants who performed the experiment into 3 conditions. My dilemma is that Shapiro wilk showed that two of the groups' data is normally distributed and the other is not normally distributed. So I am split about which tests shall I perform: shall I perform the One way ANOVA or shall I use the Kruskal Wallis Test?
Great video thanks. I have a question. I noticed you only had 1 score for your pre and post test. What if you have two groups (control and intervention) who filled in the pre and post survey, but these surveys included 5 likert questions. Would you need to do a normality test for each separate question for both the pre and post test ? Thanks
Dear dr, I have 6 dependent variables (i.e. glomerular filtration rate), and by doing normality tests, the results show that only one of them is normally distributed. For two variables I got conflicting results from KS and SW normality tests. Which one is more valid? Also, I checked their normality for each level of independent variable (3 levels) in order to conclude which test (ANOVA or Kruskal Wallis) to use for comparing data between groups. Namely, for variables that were not normally distributed in the preliminary analysis, I obtained p-values greater than 0.05 for 2 of 3 levels of independent variable. How to interpret these results and choose a valid test?
This may be a silly question, but does the decision as to which type of inferential statistical test to use depend more on normality of the data (ie parameteric vs non parametric) or other factors? For example, if we are deciding to compare means of two independent groups, and whether to use two sample t test or wilxocon rank sum test, the normality test will be the most important determining factor or other factors would weigh more? Thank you!
I am struggling with research methods and theory at uni errrgh stats is another language!!!!!!!!!!! i need to maintain a distinction average, and you are helping me do that! thank you
Great little reminder on rejecting or accepting the null hypothesis based on the alpha level. I do not recall learning about the Shapiro-Wilk value, so watching this prompted me to dig further online to learn about it. I wish the scope of this course was longer or larger because I would have liked to use SPSS again to do statistics. I feel like one can learn more about statistics if they are actually carrying out the equations and running the tests rather than just watching them or reading about them.
Jessica Muzzin I so enjoyed using SPSS (excel is not as user-friendly in my opinion) and would be nice to go back and learn it again.I can understand why not considering how expensive the software is, though. Due to that, unless you're frequently running statistics, you don't get to us SPSS often which is all the more reason to need to learn on it again. There are a lot of tests, a lot of values, and a lot of names to remember so, you're right, it is a lot easier understanding statistics by learning it on software like this.
Can I ask one question ? How can we choose the test? I have known from my friends that Shapiro-wilk test is for small sample size. If so, what is the size of sample that makes them different? Thank you
if my data is not normally distributed and I want to transform it, should I transform dependent and independent using the same way, I mean do I have to use Log10 for both or I can use Log10 for one and SQRT for the independent as it suitable??
Infomative viedo that gave further insight into distribution and null hypothesis. The viedo also gave step by step instructions in using the excel spreadsheet.
Regina Ames I agree that this video was informative, I am not exactly excel savvy so being able to see step by step instructions on how to use the spreadsheet was a refresher and opened my eyes to how simple it can be to utilize.
If this is the case, that pretest and posttest have different Shapiro wilk values , which test we have to apply parametric test or non-parametric test?
When taking into consideration normality, how reliable would you consider a quick inspection on the histogram? Is this just considered more of a visual reference to validate the sig. data?
It is very interesting. I have a question, If one variable is not normally distributed and the other one yes, I should use parametric or non parametric test to compare “means/medians” between both groups (pre/post)?
Hi Dr Grande, thnks for your video! I was a bit surprised that you did not make a seperate testing of normality for the intervention and control groupo (post test). Can you explain why please! Thanks so much in advance!
if the data is not normal distributed than which test were used for the estimation. please help me because i am doing first time estimation through spss
SPSS is so much better than excel when it comes to this kind of application. It is much more user friendly and clear, in my opinion. With that being said, I would have liked a refresher on just the basics in this video. Like, even what "normality" even is. I think our first class should have presented these basic terms so that we could follow these videos easier. Research- and math in general- do not come easy to everyone. Nor has everyone taken research class. I took my research class four years ago and I am having trouble remembering what I learned. I liked the step-by-step rundown of how to do this in SPSS. It is very clear and easy to follow. However, I can't remember what skewness and kurtosis are, so I'll need to look those up and maybe watch this video again.
Jessie Howard I definitely agree here!! I would love a run down of the basics again because I feel that I am falling just short of understanding and if I could only review those basic concepts again that I would be able to understand these more advanced concepts so much better.
If I have in first group 30participants, second-39, third-13 and fourth-19 participants... I've measured some parameteres of oxidative stress and I am going to do 1fAnova test if parameters are normaly distributed. So, my question is, when I do test of normality, do I watch Kolmogorov-Smirnov or Shapiro-Wilk test???? I have more than 50 participants in total, but less than 50 in each group. So I am in big dilemma. Please help :)
I have 3 dependent variables. Two of them are in range of normal skewness value i.e. +1 to -1 and have kurtosis in range of +3 to -3, but the third remaining dependent variable is not in normal range of skewnes or kurtosis. I want to transform that variable with square root transform to run parametric tests. So the question is, Can I transform that one variable only and run parametric test on the variables or I should transform all three variables before doing test? should I transform all three variables together even the two of them are already normally distributed? will it create problems to transform only one non normal variable?
if the pre-test is normally distributed and the post-test is not. how to determine the statistical tool to use in determining the relationship between the two tests. please enlighten me thanks in advance
Yes , this video is very useful. Shoqran !(=Thanks ! ) What if the data didn't meet the KS or Shpiro requirements ? can the SPSS suggest some cases to be removed ?
I appreciated the brevity of this video as well :D Not knowing what SPSS is, it seems like a good tool for statistical things. Testing to see if a distribution is normal would be really helpful, so that's good. I really liked how much data SPSS gives out. This video helped me solidify some of the terms and their uses. Not to mention the statistically significant stuff. You threw me off when you said that the pretest was not statistically significant, but once you reminded us that the Alpha value was .05 it all came together. I'm getting there with understanding this stuff.
Jeff Pincin I'm starting to get it too, Jeff! Videos like this are really helpful because of the visual aspect. Getting to see what the distribution of the pre- and post- tests made it easier to grasp, in my opinion.
Jeff Pincin I have never used SPSS either but this video was very informative. The visuals really help bring these concepts all together. I'm not totally clear on everything research related but I am starting to get a better understanding.
Having used SPSS in my undergrad, I did not expect there to be so many aspects that we didn't go over during our course. I'm glad these videos are offering a refresher and a new challenge.
sorry to be offtopic but does anybody know of a method to get back into an Instagram account??
I stupidly lost the account password. I would appreciate any tips you can offer me.
I always like to give credit where it is due, and this supported me on my fourth chapter dissertation, thank you
OMG! So happy to find you teaching advanced stats. I am getting my doctorate and taking stats. I watch your psych videos!!! Thank you for creating this content to help fill in the gaps that my stats book doesn’t address!!!
I appreciate how well you break all the steps down. It's slowly coming together for me.
Dr. Grande- As I am familiar with SPSS- this video refreshed my memory about how to organize the plots. Interesting, SPSS is very straight forward and is error controlled, if your information is put in right. Thanks so much for the refresher!
Thank you for the video Dr. Grande. SPSS is extremely intimidating when trying to navigate all the tables and graphs created, and understanding what it all means. It was helpful to have a break down of what the numbers mean and what the graphs are showing within this video.
Before this class I have never heard of the program SPSS. When I researched what the program did I was pleasantly surprised at what it was created to do.After watching Dr. Grande's walk-through of how to calculate if the distribution is normal in a study I feel more confident about using SPSS in the future.
This video helped me a lot in better understanding some basics in my Statistics class in my doctorate program. Great job!
Dr. Grande, this was great! Thank you!
I've never used SPSS before, so this video is great. It is brief and a good introduction to what SPSS looks like and how some of the functions work.
Normality tests can be measured in SPSS to determine whether they are normal or not. This particular example used 90 participants. From this one has to determine whether the pretest and posttest scores normally distributed. There were two different types of tests that were used Shaprio-Wilk test and Kolmogorov-Smirnov test. These were used to determine the validity. This was a bit confusing but I understand the general idea.
Toniette Morda I found it somewhat confusing as well, but got the basic concept. I think further practice would help! But overall, certainly a useful program in determining normality.
This is another video where I feel a bit in the dark. I like how you connect the KS and SW test to the histogram and how the graphs will look as they should regarding being normally distributed as per their results. I found it useful that you explained that the point of this is that researchers want to know if a distribution is normal, and what a typical data setup in SPSS would look like.
Having the step by step instructions is important for me to be able to utilize SPSS effectively, but this video also teaches about how to apply normality tests. I do think I would understand applying the results better if I was given an example of what the data was and what the actual null hypothesis was, but as far as reading the results and the graphs, this video was helpful in teaching me normality tests in SPSS.
This video and the graph helped me understand when to reject and accept the null hypothesis to test if the sample was normally distributed or not.
Wow! Thank you! That really crystallized several areas i was having trouble grasping. Much appreciated.
SPSS seems like it is a cool program and it is really interesting to see the difference between it and Excel. The names of the tests are interesting and I really liked learning how to approach them. This video was very helpful and was a good introduction to SPSS.
The video was easy to follow and Dr. Grande makes this seem so simple. This is all new to me but seeing the Normality test in SPSS gives me a better understanding. I will need more information on some of the concepts, Kolmogorov-Smirnov and Shapiro-Wilk tests, discussed.
Great Video with simple explanation. Clearly articulated the KS Test & Shapiro Test :). Thanks Prof
I remember when I used SPSS in the past, plugging in the correct numbers and keeping everything consistent, was difficult for me, because numbers and math isn't always easy for me. I'll really need to watch this again to become more familiar and comfortable with this program.
Why is stats in class so absurdly inaccessible, and why can all my profs not teach like you? Thank you, sir, for helping me out.
Dr. Grande, you are a Grandissimo. Thanks a lot.
I have worked wit SPSS a lot in the past and I still never really understood it. I liked how this video broke down the different parts and really explained what was going on in SPSS at each part.
Thank you so much! Your videos really help me out with my master’s thesis!
This video yielded many new concepts that I was unfamiliar with. However, the null and rejecting the null hypothesis is important in determining statistical significance. The SPSS system seemed to be user friendly and program features are easy to find. Very inflightfull video.
***** I used SPSS a lot in undergrad and yes it is very user friendly once you learn where all the buttons are! It will become a great tool for you to use for research.
***** It is definitely good that I can now grasp what kind of software SPSS is! I had heard references to it previously, but it didn't mean anything to me so it was nice to see how it operates.
***** I liked how Dr. Grande showed us SPSS because it did seem to be user friendly. When using SPSS in the past I did not think it was user friendly at all, but that is probablyy because I did not know what I was doing. After watching this video I believe I might be able to work with SPSS a little better.
this is the first time i actually witnessed how this is done. it is a very detailed software
Since I don't have first hand use with the SPSS and have only learned of it by name and never in detail this video was helpful in laying out where to find information on the spread sheet. I am not ashamed to say that many of the videos have been or will be viewed many times. Overall it did help me understand some of the terms used.
Danielle Duboski I also have never used SPSS, and I have never even heard of it before. I am glad that this portion of the class is in video form so we continue to go over it until we feel comfortable with the information. You will not be alone in viewing certain videos multiple times!
Great video. Before doing any sort of analysis, it is important to be very familiar with the data set so that you understand what you hope to achieve and what the most appropriate process is to determine the relationship between the results that were attained and how the data is interpreted and then analyzed.
I agree. For my research I have been watching a lot of these videos to figure out the best analysis I should be using.
Thank you so much for such an easy and brilliant interpretation.
It was nice to see the steps and how to apply it in Excel and complete the analysis. But I had a hard time focusing on the video since it was very mathematical, as my peers previously explained. I am unfamiliar with SPSS and have never used this before. Very useful in determining normality and understanding significance.
Great teaching. Thanks a million Dr.Grande.
This video was very insightful being that I have never used SPSS. It seems easy enough but I have to actually use it to get more familiar with it. I believe I have a better understanding of statically significance as well as normality. The system seems to be user friendly and structured for data entry.
Thank you, this was very helpful
I have never used SPSS and to be honest I never heard of it until Dr. Grande mentioned it in class. However, watching and listening to Dr. Grande's step by step approach for normality made it appear easy to use. Testing for normality appears pretty straight forward, so I do not feel lost or like I do not grasp the concept.
Danielle Maldonado Danielle, I have never used SPSS either. Watching the video and seeing the steps does make it appear easy. I am sure it will take more time to fully grasp SPSS and I will need to watch this video again.
Very helpful Dr. Grande. Thank you!
Interesting information. Wasn't aware of the two tests that test for normality, definitely something to remember.
Thanks a lot Dr. Todd. The lecture is very useful. Keep it up.
This video really helped me understand the steps. Thanks.
It's been quite a while since I have used SPSS and this was a great video explaining what all the output that SPSS gives you represents. I thought the explanation of what output tells you if it is statistically significant was great. I think that showing that one test of normality is more valid than the other was important and many would not now that. I myself had forgotten that.
Mandy Moore Agreed! I said pretty much the same thing. SPSS looked a lot different when I last used it. I had forgotten most of this.
+Mandy Moore Hey I loved you song Candy!
thanks! Helped me out for my thesis work!
Gentleman and a scholar, saved me a lot of pain on my diss 💪❤️
This was very helpful, especially in light of our last assignment.
I am not yet familiar with SPSS but I do appreciate some exposure to the information. Hopefully, as the course progresses I will understand this better. It is helpful to know that software can help organize the information and interpret the data, I will take away that the Shapiro-Wilks is considered more valid.
It was a great enlightenment, thank you more of this type of video.
Thank you Dr Grande, your videos inspired my channel
I am not very comfortable nor confident with this of statistics and distributions. This video helped me to gain more insight into the subject. Although still very skeptical, I think using videos, such as this one, will help me to understand this subject matter a lot better.
Thankyou Dr. for your clear explanation. However, i was wondering, since the post test scores are not normally distributed, can we still opt for parametric analysis test?
Thanks Dr. Todd, very clear and helpfull. Greeting from Thailand.
Thanks so much!!!! GREAT information!!!
Hi Todd!
Your video helped me analyse the normality for my data. I have a question though. My pre treatment scores are normally distributed and post treatment scores are not normally distributed. To give you the details, sample size 50, i'm testing 3 variables pre and post treatment and want to see if the scores are significantly different after therapy. What test should i use in this case, Mann-Whitney U test?
Thanks Dr. Grande.,That was very useful
Thank you Dr Todd for this video, I would like to ask if Iam testing 5 main variables as my primary outcome, and i found only 2 had not normally distribution and the rest 3 variables are normally distributed, would i use parametric or non parametric test allover testing my primary out come??? or would i use a parametric test for normally distributed variables and non parametric for not normally distributed one?
Watching the various videos on normality testing helps to understand which tests are most relevant for what data. SPSS is becoming less scary with each video I watch.
thank you for your clear explanation!
It has been a long time since I have used SPSS but this video was a brief and helpful reminder. I think I will need further practice with this program, but it seems to remain very useful in determining normality and understanding significance.
I have never used SPSS. However the system seemed easy to navigate and merge with research statistics. I agree, an in depth tutorial would prove beneficial.
***** I'm with you both! I need an SPSS for Dummies type of intro.
This video was very helpful in breaking down the steps to determine how to tell if there is a normal distribution . I am not that insightful when it comes to statistics, but the video definitely explained the process in a clear and concise way.
+Nakesha Tucker I agree! The video clearly went through each step which is extremely helpful when trying to actually do this. I think once we understand these tests a bit more, this video will be even more helpful. I do feel confidant that if I were asked to find out if there is a normal distribution I would be able to do just that by following these steps.
Thank you so much Sir Grande 💕
Since the pretest was normally distributed and the post test was not, do we need to apply parametric test?
Hello Dr. Grande, I have one question. Do the results change when you include several variables in the dependent list for the normality analysis? When I include several variables and I have missing values in some of my variables the final sample size of all the variables decrease. Instead of having a sample of 78 in a variable I have less because of the missing values in other variables included in the dependent list. So my question is,. when checking normality when you have missing values, should I check normality using just a single variable on the dependent list per test?
Thak you Dr. Grande! You are in did Grande in your field...
You're welcome!
Thanks a lot, it's very helpful
Hi! Thanks for the video! Would testing for multivariate normality use the same process? Thank you!
Great video. One question. I have a dataset which involves physiological variables. All skewness and kurtosis values are within -1.96 and 1.96, apart from three variables. Also, I have some where skewness & kurtosis values are normal, but the shapiro wilk value indicates non-normality (i.e. p
Hi, thanks you for the video :) If you have a data set resembling the one in the example were the pre test is normally distributed, but the post test is not, what statistical test should we apply?
Did you receive a reply?
Late night, working on data that is a mess... thanks for making this simple and clear.
>Hey, thank you for the great video!
I was wondering for questionnaires: do I first add the scores of all the items toghether and test for normality afterwards? So what I mean I have 3 Items testing one underlying construct so obviously for every participant one score for every item. At the end I need to add all the Items for the questionnaire but do I first do the normality test or after I combined the Items?
Thank you so much!
Thank you for the simple explanation :)
Dr. Grande, I have been watching your videos for months now and have recommended your site to cohort members. Would it be possible for you to create a video demonstrating a 3-way MANCOVA. Your two-way has been a great help. My problem is that I have a 240 sample but normality is non-normal. I have tried everything (transformations etc.). Any suggestions? Regardless, I would love a video.
Life saver> Thank you doctor
Thanks It was very useful! just a question, in the interpretation how can l explain the statistic value? Or, can l just say that the distribution isn't normal because the significance is lower than 0.05? (Mine is 0). I've already done the q-q plots and everything seems congruent (non normal distribution). Thanks in advance :D
Dr Grande i have a group of 30 participants who performed the experiment into 3 conditions. My dilemma is that Shapiro wilk showed that two of the groups' data is normally distributed and the other is not normally distributed. So I am split about which tests shall I perform: shall I perform the One way ANOVA or shall I use the Kruskal Wallis Test?
Great video thanks. I have a question. I noticed you only had 1 score for your pre and post test. What if you have two groups (control and intervention) who filled in the pre and post survey, but these surveys included 5 likert questions. Would you need to do a normality test for each separate question for both the pre and post test ? Thanks
Dear dr, I have 6 dependent variables (i.e. glomerular filtration rate), and by doing normality tests, the results show that only one of them is normally distributed. For two variables I got conflicting results from KS and SW normality tests. Which one is more valid?
Also, I checked their normality for each level of independent variable (3 levels) in order to conclude which test (ANOVA or Kruskal Wallis) to use for comparing data between groups.
Namely, for variables that were not normally distributed in the preliminary analysis, I obtained p-values greater than 0.05 for 2 of 3 levels of independent variable. How to interpret these results and choose a valid test?
This may be a silly question, but does the decision as to which type of inferential statistical test to use depend more on normality of the data (ie parameteric vs non parametric) or other factors? For example, if we are deciding to compare means of two independent groups, and whether to use two sample t test or wilxocon rank sum test, the normality test will be the most important determining factor or other factors would weigh more? Thank you!
Hi Dr Todd, Can you please explain to me where the value W comes from when using the Shapiro-Wilk? thank you
thanks sir for this amazing information we want more short videos about spss data
I am struggling with research methods and theory at uni errrgh stats is another language!!!!!!!!!!! i need to maintain a distinction average, and you are helping me do that! thank you
Hi Dr. Grande can we have these data so that we can practice simultaneously along with your instructions please?
thank you this helped me a lot!
I am beginning to appreciate SPSS!
did you combine the pre test and post test of the control amd experimintal group?
Great little reminder on rejecting or accepting the null hypothesis based on the alpha level. I do not recall learning about the Shapiro-Wilk value, so watching this prompted me to dig further online to learn about it. I wish the scope of this course was longer or larger because I would have liked to use SPSS again to do statistics. I feel like one can learn more about statistics if they are actually carrying out the equations and running the tests rather than just watching them or reading about them.
Jessica Muzzin I so enjoyed using SPSS (excel is not as user-friendly in my opinion) and would be nice to go back and learn it again.I can understand why not considering how expensive the software is, though. Due to that, unless you're frequently running statistics, you don't get to us SPSS often which is all the more reason to need to learn on it again. There are a lot of tests, a lot of values, and a lot of names to remember so, you're right, it is a lot easier understanding statistics by learning it on software like this.
Can I ask one question ?
How can we choose the test? I have known from my friends that Shapiro-wilk test is for small sample size. If so, what is the size of sample that makes them different?
Thank you
I gues you could not ask that question
if my data is not normally distributed and I want to transform it, should I transform dependent and independent using the same way, I mean do I have to use Log10 for both or I can use Log10 for one and SQRT for the independent as it suitable??
Infomative viedo that gave further insight into distribution and null hypothesis. The viedo also gave step by step instructions in using the excel spreadsheet.
Regina Ames I agree that this video was informative, I am not exactly excel savvy so being able to see step by step instructions on how to use the spreadsheet was a refresher and opened my eyes to how simple it can be to utilize.
thanks, this was very helpful and easy
If this is the case, that pretest and posttest have different Shapiro wilk values , which test we have to apply parametric test or non-parametric test?
When taking into consideration normality, how reliable would you consider a quick inspection on the histogram? Is this just considered more of a visual reference to validate the sig. data?
It is very interesting.
I have a question, If one variable is not normally distributed and the other one yes, I should use parametric or non parametric test to compare “means/medians” between both groups (pre/post)?
Hi Dr Grande, thnks for your video! I was a bit surprised that you did not make a seperate testing of normality for the intervention and control groupo (post test). Can you explain why please! Thanks so much in advance!
if the data is not normal distributed than which test were used for the estimation. please help me because i am doing first time estimation through spss
Thank you so much but if once the data is not normally distributed then how can we manage it?
This was very helpful thank you very much!
I'm glad you found the video helpful. Thanks for watching.
SPSS is so much better than excel when it comes to this kind of application. It is much more user friendly and clear, in my opinion. With that being said, I would have liked a refresher on just the basics in this video. Like, even what "normality" even is. I think our first class should have presented these basic terms so that we could follow these videos easier. Research- and math in general- do not come easy to everyone. Nor has everyone taken research class. I took my research class four years ago and I am having trouble remembering what I learned. I liked the step-by-step rundown of how to do this in SPSS. It is very clear and easy to follow. However, I can't remember what skewness and kurtosis are, so I'll need to look those up and maybe watch this video again.
Jessie Howard I definitely agree here!! I would love a run down of the basics again because I feel that I am falling just short of understanding and if I could only review those basic concepts again that I would be able to understand these more advanced concepts so much better.
If I have in first group 30participants, second-39, third-13 and fourth-19 participants... I've measured some parameteres of oxidative stress and I am going to do 1fAnova test if parameters are normaly distributed. So, my question is, when I do test of normality, do I watch Kolmogorov-Smirnov or Shapiro-Wilk test???? I have more than 50 participants in total, but less than 50 in each group. So I am in big dilemma. Please help :)
I have 3 dependent variables. Two of them are in range of normal skewness value i.e. +1 to -1 and have kurtosis in range of +3 to -3, but the third remaining dependent variable is not in normal range of skewnes or kurtosis. I want to transform that variable with square root transform to run parametric tests. So the question is, Can I transform that one variable only and run parametric test on the variables or I should transform all three variables before doing test? should I transform all three variables together even the two of them are already normally distributed? will it create problems to transform only one non normal variable?
if the pre-test is normally distributed and the post-test is not. how to determine the statistical tool to use in determining the relationship between the two tests. please enlighten me thanks in advance
sir, if skewness and Kurtosis values are out of acceptable range but it still has p-value greater than 5% can we say they are normally distributed?
In such case, when Pretest is normally distributed while the posttest is not normally distributed, we have to use parametric or non-parametric tests?
Yes , this video is very useful. Shoqran !(=Thanks ! )
What if the data didn't meet the KS or Shpiro requirements ? can the SPSS suggest some cases to be removed ?
I appreciated the brevity of this video as well :D Not knowing what SPSS is, it seems like a good tool for statistical things. Testing to see if a distribution is normal would be really helpful, so that's good. I really liked how much data SPSS gives out. This video helped me solidify some of the terms and their uses. Not to mention the statistically significant stuff. You threw me off when you said that the pretest was not statistically significant, but once you reminded us that the Alpha value was .05 it all came together. I'm getting there with understanding this stuff.
Jeff Pincin I'm starting to get it too, Jeff! Videos like this are really helpful because of the visual aspect. Getting to see what the distribution of the pre- and post- tests made it easier to grasp, in my opinion.
Jeff Pincin I have never used SPSS either but this video was very informative. The visuals really help bring these concepts all together. I'm not totally clear on everything research related but I am starting to get a better understanding.
Hi, when do we have to move our variables to the "factor list"? Is it always okay if we just leave it blank?