Analysis of Covariance (ANCOVA) - SPSS (part 1)
Vložit
- čas přidán 5. 07. 2011
- I demonstrate how to perform an analysis of covariance (ANCOVA) in SPSS. The first part of the series is relevant to the ANCOVA tested through the conventional approach to doing so by getting SPSS to estimate adjusted means through the GLM univariate utility. In the second part of the series, I demonstrate the exact correspondence between ANCOVA and multiple regression.
NB: The results of the analysis in this series found that males appear to have larger cranial capacities than females, even after controlling for the effects of body size. However, it should be important to emphasize that research has found that there are little to no general mean differences in IQ between males and females. Furthermore, there is neuroanatomical research to suggest that female brains appear to have more neurons per cubic cm than male brains. Thus, the difference in cranial capacity/brain size between the sexes may be counteracted by the differences in neuronal density.
These videos have greatly assisted me in my Doctoral class (Quantitative Analysis). Thank you!
Love your videos. You not only explains how to do. You also explain what it is for the role analysis, each step and all results. This is amazing! You were the one who helped me most in the analysis of my PhD. Thank you very much and congratulations. I would love if you could make a video explaining how to perform an ANCOVA in SPSS, using "random factor". Could you do that?
This video is a huge help, thank you! One bit of advice I might give is to split the video parts up by steps. So when you are looking at the composite of body size measures, you could make that one part and include it in the title. At least for my learning style (where I jump around a bit, take lots of breaks and come back to stuff) that would be very helpful. (But absolutely no shade. 11 years later, you're really saving my butt!) ^^
Thanks! If you mean 3 categorical variables with two levels, then you would use the same process described in this video. If you get a significant interaction, you will likely have to break down the effects with a couple of charts to help you understand/describe them. If one or more of your categorical variables have three or more levels, you would have to use 'effect coding'. I don't have a video on that, yet.
Your whole SPSS channel is phenomenal, and I've learned more statistics in a weekend than I did in all of college. One question still lingers, especially after watching the two-way ANOVA and ANCOVA videos: how would you pursue testing 3 categorical independent variables for main effects & interactions on a dependent variable?
Your videos are wonderful. I need one on repeated measures with covariates. Please!!!
can i do this with nonparametric tests?? My sample is not normally distributed. How do i adjust for covariates in such a sample?
Thanks for the insight.
Sounds like I'll do some reading on effect coding in SPSS, as 1 of those 3 categoricals has 3 levels.
I'd like to take this example in consideration when studying these tests for my exam, could you pls provide me of your dataset used in this example? Thanks in advance!
Thanks for your videos. I have two covariates variables and one of them is categorical. Can I run ANCOVA as explained in your video using SPSS? If yes, do I need to check all the assumptions of ANCOVA similar to having continous covariate? Many thanks!
what analysis im going to use if There are five treatments. 3 phase which are not equal.
Thank you for your amazing videos.
I am still confused about what would fit with my research. I have a study of 40 participants and two treatment groups taking three different tests at three time points (pre-, post-, delayed) and these tests were graded by three raters on a scale comprised on two sections; namely; content and form and at the end overall score is provided.
Half of participants worked in one treatment while the other half was assigned to the other treatment. I initially conducted 2 x 2 ANOVA for analysing the students’ overall within and between subjects effects looking at performance at the three tests (pre-test, post-test and delayed test) and everything was fine. However, I found that the two groups are not equivalent at the pre-testing stage (baseline) even though they were randomly assigned, so I was advised to use ANCOVA instead. So would you please help me get an answer to the following concerns:
1. In my research, I want to see whether there were differences between the participants’ overall writing performance at the different time points
2. I also want to check for any differences in the writing performance in areas of form/content between the two groups.
3. I also would like to look to ( within subject effects) whether there were differences in the improvement in the form/content areas in the two treatments.
4. I also want to check for any within participants’ effects in the overall writing performance at the different time points.
Please help me and provide me with guidance on this matter.
Particularly, What type of ANCOVA should I use; Should I use one-way ANCOVA, two-way ANCOVA or Repeated measures ANCOVA.
Many thanks again for such wonderful and helpful videos .
great!! Thanks
Thank you for the nice videos.
I would like to reduce 2 associated covariates using the PCA as you showed (you combined 3 continuous variables). In my case I have 1 continuous variable (Age) and 1 ordinal variable (pubertal stage). Can I still use the Factor Analysis fonction on SPSS to create a component?
I did not do so purposely. It should be nominal; however, for the most part, it is irrelevant how you specify the 'measure' function in SPSS. The results come out the same way. I usually ignore that part of the SPSS.
Awesome Video:: ONE QUESTION: How would you extract a component for MIXED DATA -- for example: I have Socio-economic status data - INCOME ( 6 categories), Education (5 levels), and Full time work (yes or no), and Part time work (yes or no). How would I create a component factor for SES? I am using the the resulting component factor SES to do an ANCOVA [[ SES (INCOME controlled for education and work status)----->Body mass index ]] ? Any input you can provide will be appreciated.
¡Exceptional explanation! By the way, ¿how to fix "Insufficient memory to execute this command" issue? This occurs only when I'm trying to carry out an ANCOVA. Please help me! Thanks in advance.
Hello, your video is great !! I want to know how i can do ancova when my dependent variable is quantitative does not follow a normal distribution. Thanks !!
@how2stats
What is the method, if i have a pre- and post test data(dependent variable), with multiple covariates and multiple fixed factors(groups)?
Mixed-Design ANOVA. So, in SPSS: Analyze --> General Linear Model --> Repeated Measures. Specify 2 levels in the factor. Click 'Define'. You'll see three sections for variables: within, between (groups), and covariates.
Hi
You can find it, by the option OPEN in file ....
Hello. I'm so helpful with your simple explanation. But, I'm still confused about one thing that I thought ancova or anova is for more than to groups or means, while 2 means use t test. Please help me for understanding? Really want to learn. Thanks so much
Hi..the video was very helpful. What if we have two groups
1 disease group and
2 control group
and we want to compare them on some parameter suppose Blood pressure and we want to compare them by adjusting age and sex (co-variates). How do we do that? Pls help!!!
Great videos!!! Thank you! Is it possible to use these data for educational purposes? If yes, where can we get the .sav file from?
@how2stats
What are the mean achievement scores in male and female Students taught social studies using digital instructional material and non digital instructional material at pretest and post test
What about this?
Thank you for this tutorial. Two questions: 1) Is ANCOVA only relevant to normal-distributed dependent variables? (i assume this is the case in this video). 2) What is the method when it comes to dependent variable which is dichotomous? (e.g. presenting in Table1 rates. which I would like to adjust for age. How can I get is SPSS the adjusted rated? I know how to do that in Excel, but didn't find how in SPSS).
I'd consider Logistic Regression: czcams.com/video/OvQShzJ7Sns/video.html
excuse me! you change "Sex" from Nominal scale into Scale, right? hơ to change it, please? I recognized it has format as .00 with Sex column...I wonder about that.
is it still right if I put Sex as Nominal variable for all calculation?
Sex is scored 0 and 1; yes, it is a nominally scaled variable. Totally fine to analyse in this example as an independent variable.
since you are comparing two groups, why you did not use T-test? please explain
+Ahmed Al Bayati
It includes a covariant
Ed Breen Thanks
Sir why you havnt put any variable in the box in front of covariate, although in ancova there is existence of covariate.
nice 1
Dude, where is part 2?
Why did you code gender as a scale? (I see the little ruler).
A great video! As an assumption in ANCOVA, the covariate should not be different across the groups; i.e. an ANOVA using the groups (gender) as the independent variable and the covariate (body size) as the outcome, should also be non significant. Such an analysis is not conducted in the video, right?
The commonly stated ANCOVA assumption that the independent variable should be unrelated to the covariate (i.e., no difference between groups on covariate) is seriously misguided on several levels. It's simply not true, despite what you can read in some sources. I'm going to write a paper on this to hopefully help put an end to this myth!
Thanks for the quick reply! Just interested in referenced addressing this misunderstanding, to make a reference in our manuscript.
isn't the assumption for ANCOVA that independent variables must be more than 2? here there are only males and females.
burp
¡Exceptional explanation! By the way, ¿how to fix "Insufficient memory to execute this command" issue? This occurs only when I'm trying to carry out an ANCOVA. Please help me! Thanks in advance.