Testing the Assumptions for Two-Way ANOVA Using SPSS

This video describes how to test the assumptions for two-way ANOVA using SPSS.

[@ashleymogle2700]

A two-way ANOVA is an extension of the one-way ANOVA, but now there are two independent variables, therefore, it can be a bit more challenging to sift through the data and make sense of your findings. This video was helpful in not only refreshing me on what exactly a two-way ANOVA is, but it also showed me how to use SPSS in an easy to follow manner.

[@ivybyun2791]

I see how each week's videos build on our previous knowledge, and examining the assumptions is clearer now. I appreciate your discussion on the practical applications of how to treat outliers.

[@chelseareeve36]

I like that you included discussion of what each of the assumptions are with using a two-way ANOVA to make this concept easier to understand. In addition, since the data did not fit into the basic assumptions, I like that you talked about where one could go with the data after the fact.

[@graysenmicah869]

sorry to be so offtopic but does anyone know a tool to log back into an Instagram account..? I was stupid forgot the login password. I appreciate any assistance you can give me.

[@terrychandler5759]

@Graysen Micah instablaster ;)

[@graysenmicah869]

@Terry Chandler Thanks for your reply. I got to the site thru google and I'm in the hacking process atm. Looks like it's gonna take quite some time so I will reply here later with my results.

[@graysenmicah869]

@Terry Chandler It worked and I actually got access to my account again. I am so happy! Thanks so much you really help me out !

[@terrychandler5759]

@Graysen Micah Glad I could help :D

[@jacklinskibicki6098]

Although I am not familiar with many of these terms or concepts, I felt this video helped to explain the process of testing assumptions in SPSS. I also found it helpful that the video explained how address the violations found within each assumption, such as the presence of outliers and lack of homogeneity of variance that could be discussed in the limitations section when presenting research.

[@aliciazahn1718]

Jacklin Skibicki It was helpful to get a rundown of the assumptions and which tests are used to test those assumptions. I can see this being helpful with the Analysis portion of the final paper.

[@amandavalentino8182]

Dr. Grande- Thank you for the informative video about the two-way ANOVA using SPSS. As i am familiar with SPSS, this helped me understand and remember what is needed.

[@dr.deborahbannan5051]

I found this helpful, but had to break it down into several segments for practical use. I appreciated how you went back and explained the steps.

[@Adrianao74]

Great tool, very useful to have a better understanding of how to test assumptions for 2-way ANOVA using SPSS. Thank you Dr. Grande.

[@marisamaguire8136]

I really appreciate Dr. Grande’s explanation and demonstration on how to properly test the assumptions for two-way ANOVA when using SPSS. As someone who is unfamiliar with SPSS and has little knowledge of the concepts, I find the demonstrations to be helpful as they give a clear and concise walkthrough of the information. Similar to Chelsea, I also found the discussion of the three assumptions when using two-way ANOVA to be beneficial in understanding the concepts further.

[@vanessakent8700]

This was a lifesaver in completing the assignment in Chapter 4. It did go at a fast clip so I had to pause it at each step to make sure I understood it. Some of the variables were already preset on your screen.

[@chelseaswift9252]

This was really helpful! I'm writing my masters dissertation and I violated the Levenes and I've been stuck for days but I feel like I understand the solution now, thank you!

[@gabijajarasiunaite4512]

Can you share the solution ?

[@sotiriagilou510]

Thank you for your so clear explanations!!!Greetings from Greece

[@simonehenry469]

Dr. Grande did a good job describing the assumptions.

[@brittanyhoffman8068]

Thank you for sharing this video Dr. Grande. There seems to be a lot of information displayed when plugging in all the data and it is a bit confusing. I appreciate the box and whiskers graph in clearly displaying the outliers within the data.

[@mandabear52289]

I thought this video did a great job of explaining the assumptions of a Two-Way ANOVA. I had forgotten about all the assumptions so I was thankful for the reminder. I also loved how he explained what to do if a assumption was violated. Good video!

[@jacklinskibicki6098]

Mandy Moore I agree, and also found it useful to know what to do if an assumption is violated, specifically the importance of addressing such violations within one's limitations section of a research proposal.

[@aussiebreezekorb8877]

This was helpful AFTER watching the newer video you made on "Introduction of a 2-way ANOVA" I can see what you are talking about put to work in the SPSS spreadsheet.Thanks

[@grishmaprabhukhot3925]

Hi Dr. Grande, I have a question related to the test of homogeneity. If I do not have homogeneity of variances, how do I proceed? I came across this website 'laerd' and they have mentioned: If your data does not have homogeneous variances, you have four options: 1. Transform your data. 2. Carry on regardless. 3. Perform robust analysis. 4. Weighted least squares regression. I referred to the 'carry on regardless' which says: "If group sample sizes are equal or approximately equal and large, there is normality and the ratio of the largest group variance to the smallest group variance is less than 3, run the two-way ANOVA anyway because it is somewhat robust to heterogeneity of variance in these circumstances (Jaccard, 1998)." So my question is: If the group sample size of my data is equal and if my data is normally distributed, then can I carry on regardless? Your help is highly appreciated. Thank you so much!

[@richaudsley5032]

This was very helpful for me to complete my homework assignment in Chapter 4 of Mertler & Vannatta. Thank you Dr. Grande!

[@ariadnaaguero-roman5525]

This video was helpful in understanding ANOVA. I like that it's divided by levels and you can compare the data between the two groups. I like the feedback on what to do with the outliers, either have a good reason to eliminate from data or note it in the limitation section.

[@angieku5083]

It is helpful to know when we should use split file versus analyze groups collectively. Thanks for explaining how to test the assumptions thoroughly.

[@ZeroesOnes]

Thank you, you've pretty much saved my life.

[@DrGrande]

You're welcome, thanks for watching -

[@rachelfoster5463]

This video contained a lot of information. However, I feel the more Dr. Grande discusses significance and the Null Hypothesis, the more I understand those concepts. I also liked how Dr. Grande discussed what to do if you have an issue with these assumptions in your research.

[@kimberlykelly386]

Rachel Foster I, too, appreciate Dr. Grande repeatedly going over concepts even if I don't understand the context or the entire video (I'm far from that). But it helps me to grasp the basics - i.e. significant vs. non-significant values.

[@corieschoeneberg9295]

I was thankful you discussed what to do in cases in which the assumptions are violated

[@tamarabakergibbons5715]

The outliers did not meet the assumption. That was a lot of data to sift through! Thank you Dr.Grande for the video.

[@vanessagraham6473]

thanks. It was helpful to hear what to do in cases where the ideal is not reality.

[@jazzynovy5400]

Although I found the video confusing, it helped to explain the testing assumptions in SPSS.

[@simonehenry469]

Jasmine Aviles yes, i did too.

[@cherylchance904]

This is certainly a challenging analysis but it is one that is more commonly used so it is important to be sure that you can gain awareness of how to calculate the information.

[@sherrietilghman2745]

I learned something very simple from this video that statistics can sometimes be overlooked because of assumptions

[@chelseydavis4989]

This video was well organized and full of new information. I am a little more confused with the Two way Anova and the outlier process. I am not sure I would be able to reproduce this process on my own

[@danielleduboski9831]

I watched this video twice in a row, and I was following ok until the giant spread sheet opened up with all of the data. It was just overwhelming. I did see that it was kind of easy to see where the data was falling and instead of having to change over the data sets for each one it gave the data for all of the variables; which was nice.

[@MarkVanderley]

This will be helpful in completing the assignment this week thanks

[@kimberlydixon8760]

I had a harder time trying to digest this video, the instructions were easier to understand but I just thought this was a little overwhelming. I am really appreciating seeing this done in SPSS as I am still unfamiliar with the software.

[@MichaelSmith-jd5ki]

Kimberly Dixon You're not alone. For those of us with no experience with SPSS, this video is a little confusing. I still think it's useful for us to be able to see the program at work.

[@heather7927]

I found this video to be an informative visual of testing the assumption for a 2 way anova.

[@MichaelSmith-jd5ki]

Legitimate reason. That is the most important concept that I got out of this video. outliers can be removed from a data set only if there is legitimate reason to do so.

[@ariadnaaguero-roman5525]

My preference would be to just to note these outliers in limitation section.

[@adamwilson4642]

Very helpful in breaking down how to test the various assumptions. At what point does violation of the assumptions mean you should not use the results, however? For example, how many outliers, or how non-normal of a distribution before you do not risk using the results?

[@kyarapanula1002]

great point about outliers....especially in relating to a particular population and their "agenda."

[@bradleyfairchild1208]

You talked about ways to get eliminate outliers, are some of the reasons mistakes, like the person forgot to answer one of the questions? Great video and very helpful, thanks.

[@reneemendez4287]

There was A LOT of information covered in this video, and anything involving SPSS makes me curious so I end up pausing the video frequently to Google something. That being said, even though it was a lot to take it I believe I understood what a two-way ANOVA is and how it's used. I want to say that I will retain this information, but I think if I had access to SPSS then I would retain it better.

[@gousechang9893]

Hi Dr Todd if two way ANOVA data cannot meet normality test, what should i do for non parametric analysis

[@alessandrabarberis1513]

That was very useful. I have a question- I am doing a two-way ANOVA and I tested assumptions following this video however on Laerd statistics it guides you to calculate residuals and test assumptions using those. Is there a reason why here you used raw data rather than calculated residuals? I am not sure which way I am meant to use for my data. Thank you!!

[@rebeccameece013]

I was a little lost in the video; I'm not sure why ANOVA confuses me sometimes but it does. I think the process of it is what I don't get. I like the explanation of the null hypothesis and the assumptions though!

[@carolkirby9558]

The outlier information is helpful. But what happens when you cannot eliminate an outlier? Do you select another design since it violates one of the underlying assumptions or include it and explain it?

[@aliciazahn1718]

I may have just not totally absorbed this from the one-way ANOVA video but are the assumptions the same for one-way and two-way ANOVAs? It was interesting to see how easy it was to violate the assumptions and Dr. Grande mentions that this can happen pretty easily with a two-way ANOVA. I can see how it would be helpful to see which cases are outliers so that they can be further investigated to see if there is a legitimate reason why they should not be included in the results.

[@lachusity]

very good video, i have a question though, is it necessary to measure the normality separately for every level of the dependent variable?

[@DrGrande]

Yes, technically, normality should be tested for each combination of the levels of the independent variables.

[@lachusity]

thank you, i often work with small sample sizes and some subsets only contain 2-3 data points. (i guess that questions the usefulness of the whole procedure as it lacks robustness)

[@wolfie6727]

Thanks for the great video. I really benefitted from it. I wonder what non parametric test one can resort to in case none of the assumption are met??

[@bethanyelstrom3566]

What I got out of this video is that the purpose of a two-way ANOVA is to understand if there is an interaction between the two independent variables on the dependent variable. Other than that, this video was very complex. There are so many steps to follow to get an output and it brings up many charts that I have never seen before so it is hard to comprehend this information.

[@jhoward1129]

Bethany Elstrom Thats about all I got, too. The more these videos stack, the less I think I understand it. I like to understand the reasoning why I am doing this, and I'm not sure I fully understand that.

[@daniellemaldonado631]

Bethany Elstrom Agreed! This video was complex and typically with Dr. Grande's videos, using SPSS, they appear to be user friendly. This one not so much.

[@elizabethadair5545]

You mentioned that you use Shapiro-Wilk over Kolmogorov-Smirnov. Field (2009) uses Kolmogorov-Smirnov. Can you discuss the reasons one may choose one over another? I also would like a response to Angie Ku's question.

[@DrGrande]

+Elizabeth Adair As with all areas of statistics there are varying opinions. In general, however, most researchers tend to agree that the Shapiro-Wilk is more powerful than the Kolmogorov-Smirnov, meaning it correctly rejects the null hypothesis of normality more of the time.

[@jpincinjr]

The process was confusing, but it makes sense that you need to test the assumptions. You wouldn't want to do anything starting with incorrect information.

[@user-xn2hf9re8r]

Could you increase volume your end as despite maximising mine you are always very quiet. Why SW test here Todd? thanks

[@BinaGodiva]

Thank you Dr. Grande. :-)

[@terihourihan6896]

thank you Dr. Grande

[@robertodonati1037]

This video was a bit difficult to understand at the beginning, but when I watched a second time it became easier to get the idea.

[@Over1BV]

Thank you Sir. vis itpro duc tsi te dotb logs potd otc om

[@israeladeogoabejoye9531]

Good one.

[@timf6969]

so are you saying that the outliers have violated the homogeneity in the data set?

[@TheBengalDragon]

I have SPSS 25, the one as of the day dec 4 2017, but I can't do Levene's test, on my Anova which has 2 factors and 2 levels to each factor. I always get an error message. Please help.

[@alexandersaloustros7200]

me too!!

[@idewagederatdwiyanaputra2641]

Dr. Grande, thank you for this two-way ANOVA tutorial. I have a question for you and also for others who might help. How may I conduct Two-Way ANOVA for unequal sample sizes? I have two Independent variables; Pair collaboration (two levels; pair vs individual) and self-efficacy level (two levels; high vs low self-efficacy). Please help me. Thank you.

[@thewigginsfam]

This was informative, thanks!

[@jhoward1129]

I have trouble understanding HOW to do something if I don't first understand WHY. I don't think that Dr. Grande explains anywhere in any video what some of the basic concepts are (what even is an ANOVA?) and why we do them. What am I trying to accomplish by doing this? My understanding is thither is an interaction between two independent variables and one indecent variable. I am not sure is that is right, and I wish Dr. Grande would start the videos like this. This video would be helpful for someone who already knows the basics, which I do not think I do.

[@rebeccameece013]

Jessie Howard I would agree with that. Sometimes I feel like I don't really know or understand the reason why. I think it would be beneficial for the videos to start this way as well, like a summary of what the reasoning is before it is started. It might make the video make more sense over all and I may have a better understanding.

[@kimtoogood-hughes3143]

What is the non-parametric test one can use if the assumptions are violated?

[@thomasstarr6433]

Adam - after reviewing the video I had very similar questions. Also, when is it valuable to transform the date? Can you run the ANOVA and use the results anyway making note as to the limitations? Or would using this data as it is invalidate the results?

[@strongestavenger1787]

Could you do the Welch’s test too for this to correct for the violation or is that just for one way anova?

[@susanwilder7616]

There was a lot to digest with this video. It wasn't super complicated, but I was slightly confused at the end of the video

[@jpincinjr]

Susan Wilder It would definitely take more than one time watching the video in order for it to make complete sense for me.

[@armyman9907]

If the assumption of homogeneity of variance violates. Which test can be used?

[@elizareyes267]

Doc, how to perform two way kruskal wallis in spss? My data is not normal

[@kimberlykelly386]

A big takeaway for me was that you can exclude data IF there is a legitimate reason. For example, if your outliers are due to participants attempting to manipulate the results for whatever reason, you can delete the data. If you can't determine the reason for the outliers, you can continue with the analysis so long as you make note of the outliers in your limitations subsection of the discussion.

[@mandabear52289]

I thought that this was important to note as well. Just because it's an outlier doesn't mean that you delate it. You must have a legitimate reason for doing so. Always making sure to note outliers in your discussion.

[@reginaames3038]

Mandy Moore I agree that was great information to know that you can exculed what is not needed.

[@auroraborealisz]

what happens if one of the factors has p value thats below 0.05?

[@chaliendalew3826]

I am about prepare manuscript, I have to use correlation and ANOVA to analyze my data, fortunately, my data doesn't meet for both assumption, what shall I do?

[@domdollaable]

While I found this video very informative, I was and still am very confused by the material presented here. To clear up any confusion I attempted to open excel to play with it as I watched the video but still found myself lost. It is very hard to replicate this info. without any real data. I am also confused as to when to implement this process. When is it useful to do these steps?

[@angelaliu7087]

What else can you do if the homogeneity of variance assumption is not met?

[@kimeeshareedwalker2602]

It's becoming clearer each time I watch intently. The language is what confuses me. I have to make it a habit to apply these terms as I conduct research to increase my understanding. The difficult part is under standing the numbers once the configurations are complete.

[@toniettemorda6247]

Interesting video! However, I was getting lost a bit. I am having trouble understanding the content which may be why this video is confusing.

[@susanwilder7616]

Toniette Morda I'm right there with you. I think that if we had hands on exercises, rather than just watching the videos, it would be easier to grasp.

[@wendyteal4204]

What do you do if your data violates the 3rd assumption?

[@Lucy-gl2nu]

i have 5 dvs and a few of the scores are 0.00 on shapiro/kolmo what shall i do? i been told to do a anova either way not a non parametric test as not been taught one... what shall i do ?

[@reginaames3038]

Interesting video, though the step by step instructions are great it was a bit much for me considering we are not actually using this information for class work.

[@ashleydorsey6745]

This video was a little difficult for me to follow. From what I gather the AVOVA analyzes data and there are three assumptions that relates to the ANOVA. After that explanation I was completely lost. Following the spreadsheet confused me even more.

[@kimberlydixon8760]

***** I found this video a little harder to follow as well.

[@sannyuharris2827]

Thank you!

[@daniellemaldonado631]

So, I will be honest...I had a hard time with this video! Data is analyzed using ANOVA, results are interpreted, and the interpretation has to be called into question if the assumptions were never checked, right? Confused!

[@thecorgisquad9861]

Carrying out the ANOVA function in SPSS seemed a little more confusing than doing the same calculation in Excel, which is surprising to me. Typically, I find myself being more comfortable with understanding SPSS, but all of the steps were a little heavy and confusing.

[@reneemendez4287]

Jessica Muzzin I feel the same way about SPSS simply because of how easy Dr. Grande makese it seem in these videos. Excel seems to take longer, but I cannot say for sure because I have only ever tinkered with Excel and never SPSS.

[@melissaclendaniel9811]

The purpose of the Two Way ANOVA is to compare the data from the independent variables to the dependent variables and see if there was an effect on the dependent variable. I understand this basic concept of this video, however it was difficult for me to follow, especially in SPSS.

[@rachelfoster5463]

Melissa Clendaniel I have a hard time following things in SPSS as well. I think it is easy to get confused between that and Excel when we see videos on both. I too grasped the basic concept, but got lost in the different data and tables shown.

[@ashleydorsey6745]

Melissa Clendaniel Thanks for the explanation. You explained the general purpose of ANOVA in understandable terms. I agree is was difficult to follow especially with the use of SPSS.

[@saraspong6698]

Thank you

[@taylorjohnson3730]

If there is a legitimate reason to get rid of an outlier, does that need to be documented in the narrative of results?

[@DrGrande]

+Taylor Johnson Yes, I believe this is important to explain to the consumer of the article.

[@scottmartin7764]

So happy that you answered this I had the exact same question, thank you

[@timstauffer1351]

Thanks

[@ahyaraberysalim3763]

what should we do if the data is normal but unequal. anyone know?

[@brittanyvodzak9198]

I found this video to be very complex. I think that I have difficulty with all the different steps.I also feel overwhelmed with all the tables and graphs that come up when you conduct a test. I understand the concept of this video and what information that you are trying to collect but I think all these steps are a little difficult to follow.

[@katherinebeck593]

Brittany Vodzak I agree I thought that it was very complex. I am getting a little overwhelmed with all of the steps that are needed for everything that we are learning.

[@bethanyelstrom3566]

Brittany Vodzak Well, I'm glad to see that I am not alone on this one. There was so much going on and I could not get a clear understanding of all the data that was collected. I understand the overall reason for the testing, but I could not repeat these steps.

[@CandyKaneLane]

Brittany Vodzak I struggled following this video, and I think it is because we have no experience actually conducting any of these tests using excel. We are not implementing them in the classroom so for us, it is harder to comprehend and dissect the purpose of all this.

[@paulhoard7390]

I have the same question as Adam.

[@katherinebeck593]

I think that all of these steps are a little confusing, I have a little trouble still understanding excel let alone SPSS.

[@danielleduboski9831]

Katherine Beck Yes it was a little confusing, then to add in the giant spread sheet made my head just turn off. I had to watch it a second time because after the spread sheet was opened I just stopped understanding. It was a little overwhelming, but if you had to keep switching all of the data or have all of the data on different spread sheets I think it would be harder to try and compare all of the information like that.

[@joshuakreimeyer2297]

~~Thank you!

[@CandyKaneLane]

This video was a little overwhelming for me to follow. As the step-by-step instructions are helpful, there was just a lot going on in this video that made it hard for me to follow and comprehend. I am also not familiar with using excel to this extent.

[@melissaclendaniel9811]

Candace Fernandez I felt the same way as I viewed this video. I am glad that there are step by step instructions so that I may come back and review it again so I can familiarize myself with this process.

[@toniettemorda6247]

Candace Fernandez Yes, I was a bit lost as well

[@domdollaable]

Candace Fernandez I am glad to hear I'm not the only one who was lost. The step-by-step was helpful but still confused on the process.

[@michellerobinson968]

Candace Fernandez I agree that there were alot of steps. I too found it difficult to follow. I am still becoming familiar with Excel and understand I will not comprehend all these steps.

[@thecorgisquad9861]

Candace Fernandez I also found myself confused by this video; having to stop and restart the video just to understand what was going on. I did understand slightly better because it was about SPSS instead of Excel, like previous videos.

[@chunkietv5499]

I thought what we should test is the normality of the residuals? 😳

[@barbarahoffman408]

This was a little difficult to understand. I will need to watch this again.

[@melissahansbury6522]

I agree. Watching these videos can be confusing. I think that I won't fully understand the process of doing this unless I can actually do this.

[@michellerobinson968]

The video explained the reasoning for testing assumptions using two-way ANOVA and this compares the data/interaction between the independent variable and the dependent variable. There were alot of steps involved that I did not remember.

[@babettevanhout9485]

I have a question for everybody who know the answer. What should i do when the data violates the first assumption (test of normality)?

[@DrGrande]

You may consider a data transformation. If you have a large sample size, you may want to consider proceeding with ANOVA, however, there is no universally agreed upon definition for a large sample size. It may be useful to pull a few articles where ANOVA was used when the assumption of normality was violated, and read what the authors did to address it.

[@faithstephen7477]

Thank you!