@@dr.sumeetbakshi When i am using median for two likert scale questions which i have marked as ordinal in my items, its giving me Scale data and not nominal. Should i change it manually to nominal?
What if you’re trying to do this with 2 questionnaires which are measuring the IV but one is 5 point and the other is 3 point. I need to separate into high , low and med to see it’s affect on the DV
Hi Dr Sumeet Bakshi, thank you for this video. If we have questionnaire with two or three variables and all the variables have multiple items using Likert scales and we want to measure the correlation between those variables, how do we do it? Do we need to have median or mean?
@Dr. Sumeet Bakshi When i am using median for two likert scale questions, its giving me Scale data and not nominal. Should i change it manually to nominal?
You should not change anything manually till the time you are not pretty sure. Check if data is normal or skewed. If your data is normal means (Mean= Median= Mode ), go for parametric tests. For selection of test th-cam.com/video/gTGpDgUWIF0/w-d-xo.html
Please Help me . I'm analysing a questionnaire with likert scale and I want to combine it with ROA from second Data. ROA is dependent variable is a ratio that measure the performance of business. It takes 0.0766 number or etc. But when I correlate with questionnaire I get a negative correlations. 😑
Thank you!! I have demonstrated how to compute mean and median both. You can choose any one. However, as I mentioned median makes more sense than mean. Hope this helped! Stay Connected!!
@@dr.sumeetbakshi Thank you so much Dr! Would normaility of data influence whether we choose mean or median? I came across papers that would use mean when data are normally distributed and median if not normally distributed
When data is normal, means there is no skewness, it’s ok to use mean. Here value of mean, median and mode is equal( bell shaped curve ) But in case of non-normal distribution, mean does not reflect all data points accurately ( due to 1 or 2 skewed values, mean changes drastically). In this case, median is preferred.
Awesome, highly appreciated. This remains controversial and still has many questions one can ask.
Thank You 😊
Yes, this topic needs an unfold discussion!!
@@dr.sumeetbakshi When i am using median for two likert scale questions which i have marked as ordinal in my items, its giving me Scale data and not nominal. Should i change it manually to nominal?
Thanku so much ma'am! Greatful !!
You’re most welcome 🤗
Stay Connected!!
Thanks a ton ma'am
What if you’re trying to do this with 2 questionnaires which are measuring the IV but one is 5 point and the other is 3 point. I need to separate into high , low and med to see it’s affect on the DV
Hello and thanks for this video
Can we use KS normality test for likert scale?
❤
Hi Dr Sumeet Bakshi, thank you for this video. If we have questionnaire with two or three variables and all the variables have multiple items using Likert scales and we want to measure the correlation between those variables, how do we do it? Do we need to have median or mean?
If we calculate median, then how to proceed further for Mann Whitney u test? Please help.
"The numeric argument required for the function specified was not supplied" This is coming when I dp this, What to do now?
@Dr. Sumeet Bakshi When i am using median for two likert scale questions, its giving me Scale data and not nominal. Should i change it manually to nominal?
You should not change anything manually till the time you are not pretty sure.
Check if data is normal or skewed.
If your data is normal means (Mean= Median= Mode ), go for parametric tests.
For selection of test
th-cam.com/video/gTGpDgUWIF0/w-d-xo.html
Please Help me . I'm analysing a questionnaire with likert scale and I want to combine it with ROA from second Data. ROA is dependent variable is a ratio that measure the performance of business. It takes 0.0766 number or etc. But when I correlate with questionnaire I get a negative correlations. 😑
It simply means two variable are negatively correlated. For details, please check this video
th-cam.com/video/Osg5PLmevFc/w-d-xo.html
Hi
If you managed to analyze your data on your research please assist on how you proceeded cause I'm stuck here with the same issue.
Great video, I am curious as to why you compute the mean and then you choose median?
Thank you!!
I have demonstrated how to compute mean and median both. You can choose any one. However, as I mentioned median makes more sense than mean.
Hope this helped!
Stay Connected!!
@@dr.sumeetbakshi Thank you so much Dr!
Would normaility of data influence whether we choose mean or median? I came across papers that would use mean when data are normally distributed and median if not normally distributed
@@dr.sumeetbakshi Why does mediam make more sense than mean? I am quite curious since I usually encounter papers focusing on the mean. Thank you!
As median makes more sense than mean when data is categorical. In Likert scale, data is ordinal/ categorical with orders.
When data is normal, means there is no skewness, it’s ok to use mean. Here value of mean, median and mode is equal( bell shaped curve )
But in case of non-normal distribution, mean does not reflect all data points accurately ( due to 1 or 2 skewed values, mean changes drastically). In this case, median is preferred.