Normality of the data

[Jasvir Kaur]

Respected Sir and Group members,

1.To what extent it is crucial for the data to be normally distributed for Stepwise Multiple Regression Analysis?  PP Plot shows my data to be nearly normal, but Kolmogorov shows it is not, What should I do? 

2. My scales are Likert scale. I have read that normality is not a condition for Likert scaled data, could you please provide me the reference for the same? 

Many Thanks in advance.

Jasvir kaur

Research Scholar

[Neeraj Kaushik]

Dear Jasvir

Here are my inputs:

1. After data collection, plz go for the Data Processing. Plz check that there shd not be any missing values, improper codes, outliers etc After that you may go ahead with the Multiple Regression without worrying about the normality assumption.

Plz note in Regression, there is an assumption of normality of residuals not of variables.

2. On Likert scale data, it is very unlikely that the data are normally distributed. So, do not worry for go ahead with further process of applying EFA, CFA or SEM.

Best wishes

--
Protocols of this Group:
 
1. Plz search previous post in group before posing the question.
2. Don't write query in someone's post. Always use the option of New topic for the new question. You can do this by writing to dataanalys...@googlegroups.com
3. Its better to give a proper subject to your post/query. It'll help others while searching.
4. Never write Open ended queries. This group intend to help research scholars NOT FOR WORK THEM.
5. Never write words like URGENT in ur posts. People will help them when they are free.
6. Never upload any info about National Seminars/Conferences. Send such info on personal emails. And feel free to share any RESEARCH related info.
7. No Happy New Year, Happy Diwali, Happy Holi, Happy B'day, Happy Anniversary etc. allowed on this group.
8. Few months back there was a facility for asking & sharing the Research Papers. Now there is no provision of asking for the research paper here.
 
Let’s make a better research environment.
---
You received this message because you are subscribed to the Google Groups "DataAnalysis" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataanalysistrai...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/9cb0c8b1-e56a-4ab0-accc-4e5881488888n%40googlegroups.com.

[Jasvir Kaur]

Respected Sir, 

Thank you very much for your valuable inputs. 

My next query is, after data processing, dealing with Outliers and Multicollinearity etc, I would like to mention about the assumptions of the regression, So what should I mention about the Normality? 
I agree with the point that Likert scaled data is not normal most of the times, but Sir do I need to say this with respect to some reference? (Like for the Durbin Watson acceptable values we have references to quote)

I am asking these points for my PhD thesis :)

Regards,
Jasvir. 

[Neeraj Kaushik]

No, no mention of normality is reqd for the Regression results

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/c622ed68-93c6-49f5-83b3-0938f07b6d01n%40googlegroups.com.

[Jasvir Kaur]

Right Sir, 

Thank you very much Sir for your precious time and valuable input.

Best Regards,

Jasvir

[Jasvir Kaur]

Respected Sir, 

Am I  have total 8 IVs and one DV. Multiple Regression shows 5 variables as significant predictors. 
For the Three variables which are not significant predictors, the VIF and tolerance values are given by SPSS in the *EXcluded variables* table. This table gives different orderly arrangements of IVs, Which arrangement am I supposed to follow to know the IF and tolerance of remaining three (Not significant variables)

[Neeraj Kaushik]

If you are using Stepwise Regression then just report about the significant vars

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/7ae0dddc-0d8c-4bd7-a9c4-cf6d9db3ddben%40googlegroups.com.

[Jasvir Kaur]

Right Sir. Thank You very much.

[SANESH K P]

let me know how to check the normality of residuals

[Neeraj Kaushik]

Run Regression, save residuals and then use 1 sample KS Test to check normality of residuals.

May refer https://youtu.be/DzyA20ILmV8

To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/cfd67ee3-f58f-4efa-a9f5-fbbdd351db4en%40googlegroups.com.