Multiple linear regression using R studio (Aug 2022)

This video provides a general overview of linear regression using R Studio, covering a variety of functions that you might use to obtain output of interest. I cover some basic functions for screening data and examining correlations among variables, the lm() function for performing the linear regression, and various functions for different diagnostics one might be interested in obtaining. The video is not designed to be an exhaustive or comprehensive overview of regression and all aspects of decision-making with regression.

A copy of the data (referenced in the video) can be downloaded here:

You can also download a supplemental Powerpoint here:

[drryangagnon]

Mike, again excellent work. I was preparing some materials for my multivariate students to review, and this is just what I had in mind.

[peternaumann]

Awesome recap before linear models II!!! Thankyou Sir

[remyxi]

I think it would be significantly more useful if we can have a case-study-like tutorial/walkthrough so everyone can see your analysis method and conclusion. Btw, I learned some new things from this video too nevertheless, thanks!!

[JamesMendoza-pl]

It has been a while since I last used R Studio, so I watched this a a refresher. It was super helpful. Just wondering if you have a walkthrough of a dataset that include a categorial variable.

[MA-ibmb]

Hi Crowson,
Could you please send me the link for the RStudio program you use in this video, to install to my Windows 10 laptop?

[oo-fkcn]

Get error in function vcm<-vcovHc(fit, type="Hc3")
error in match.arg(type):
'arg' should be one of "HC3", "const", "HC", "HC0", "HC1", "HC2", "HC4", "HC5"

[maryamfereidouni-txsk]

For some packages (except psych and car) I have difficulty installing. What should I do?

[galax]

get error on :
cor.test(df, use="complete.obs", methods="pearson", adjust="bonferroni")

Error in cor.test.default(df, use = "complete.obs", methods = "pearson", :
'x' must be a numeric vector