Correlation and fit model

kazmo — Fri, 07 Apr 2023 15:29:12 GMT

Hello everybody,

I have got an odd request: We are currently working on a data analysis of medication dose and blood levels for 5 different medications. I am new to the research group and my supervisor instructed me on continuing the analysis of a colleague.

I have just received the template for the analysis and I have a hard time figuring out what I am supposed to do.

I know it is a big question, but I hoped to find some answers to what exactly we are looking at here. So far I understood that we are correlating dose with blood level by medication and e.g. sex with fit y by x. For spearman's I used a multivariate analysis (y -> med dose and medblood level by meds and e.g. sex) and then calculated Spearman with a non-parametric test. I was told to use a test model like fit model but other than that I'm utterly clueless.
Thanks to everyone in advance!

"Correlation between med dose and med blood level within meds and by patient subgroups (e.g. sex, ethnicity, weight)"

Re: Correlation and fit model

P_Bartell — Sat, 08 Apr 2023 12:05:24 GMT

When you wrote, '...utterly cluless...' I'm going to take that literally wrt to both knowledge of statistical methods and JMP. If my supposition is accurate, I suggest starting here with the JMP "Statistical Thinking for Industrial Problem Solving' online course: Statistical Thinking for Industrial Problem Solving

You'll learn enough statistics to 'do' what you need to do...and JMP at the same time. IMO, there are way too many issues at play here to give guidance in an online forum such as this.

One small but illustrative example: "How was the response data collected? By a designed experiment? Happenstance/observational data?" The answer matters wrt to both analysis methods and any conclusions/decisions/actions that come out of the effort.

topic Re: Correlation and fit model in Discussions

Correlation and fit model

Re: Correlation and fit model