I realize this video is from 11 years ago so I doubt ill get a speedy response but... How would you analyze the data if for example you have male and female and 3 different drug dosages and you are scoring their behavior across 12-24-36 hours. So time is within subjects and dose and sex are between subjects
Hi... Thanks! Aren't there 12 students (6 in each group) undergoing 3 dosages rather than 36 students (I guess I heard that in the video). So, isn't it better to go for repeated measures?
+Akshay Maggu Its 36 students total Researchers wanted to test a new anti-anxiety medication. They measure the anxiety of 36 participants on one of three different dosages of the medication: 0 mg, 50 mg, 100 mg. Participants are also divided based on what school they attend, which researchers hypothesize will also affect anxiety levels. Anxiety is rated on a scale of 1-10, with 10 being “high anxiety” and 1 being “low anxiety.” Use α = 0.05 to conduct your analysis. Copy the data to Excel and sort for SPSS.
That is correct. They should be categorial. However, the data he has is non-parameteric (which means he has to run another test, other than ANOVA). So in order to conduct an ANOVA test (which is a parametric test) he assumes that the data is interval scale and not categorical which allows him to run such a test. This is something that researchers can do to run such a test, but there are also equivalent tests you can do instead like Mann-Whitney Rank test or the Kruskal-Wallis H test, for instance, if you want to examine your data without making this assumption. the only difference is that parametric tests are more robust and can bare the changes you're making.
Thank you so much for going over in detail how to enter the variables!! You saved me! lol
Thanks a lot! Your explanation helped me with my finals!
I love this presentation. You helped me so much.
Thank you! This was very helpful!
This was really helpful, thanks!
Thank you!!!
Very helpful. Many Thanks
Thanks.
thanks a lot
Nice !
thanks :)
Is the null hypothesis school and dosage WILL affect anxiety level?
Thnx.... Buddy
your welcome
This was a very helpful video, except for 4:37 where you call anxiety an independent variable, but is clearly a dependent variable.
In two way anova can I use only one fixed factor?
and if Levene's is significative?
I realize this video is from 11 years ago so I doubt ill get a speedy response but... How would you analyze the data if for example you have male and female and 3 different drug dosages and you are scoring their behavior across 12-24-36 hours. So time is within subjects and dose and sex are between subjects
Hi... Thanks! Aren't there 12 students (6 in each group) undergoing 3 dosages rather than 36 students (I guess I heard that in the video). So, isn't it better to go for repeated measures?
+Akshay Maggu Its 36 students total
Researchers
wanted to test a new anti-anxiety medication. They measure the anxiety of 36 participants on one of three
different dosages of the medication: 0 mg, 50 mg, 100 mg. Participants are also
divided based on what school they attend, which researchers hypothesize will
also affect anxiety levels. Anxiety is rated on a scale of 1-10, with 10 being
“high anxiety” and 1 being “low anxiety.” Use α = 0.05 to conduct your analysis.
Copy the data to Excel and sort for SPSS.
assumptions testing???
To skip the introduction: ruclips.net/video/G-YexMVQAGc/видео.html
shouldn't 'school' and 'dosage' be categorical variables instead of scale?
I also have the same question. They definitely should not be "scale".
Exactly my thought!
EDIT: if you see this 5 years later...
ya it should be Nominal right?
That is correct. They should be categorial. However, the data he has is non-parameteric (which means he has to run another test, other than ANOVA). So in order to conduct an ANOVA test (which is a parametric test) he assumes that the data is interval scale and not categorical which allows him to run such a test. This is something that researchers can do to run such a test, but there are also equivalent tests you can do instead like Mann-Whitney Rank test or the Kruskal-Wallis H test, for instance, if you want to examine your data without making this assumption. the only difference is that parametric tests are more robust and can bare the changes you're making.
Assumption testing? Waste of time on entering the data too, probably already know how to do that if you are doing factorial anova...