cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
ChiaraC
Level I

How to perform precision analysis and ANOVA having four nested factors?

Hi!

I am quite new to JMP and I would really appreciate your input regarding the following analysis.

 

I have to establish the precision of an assay.

I performed a test that consisted in measuring my sample in x replicates, twice a day for 20 days, on 3 instruments located in 2 different sites (1 in site A, 2 in site B; site A had different environmental conditions and operators performing the test compared to site B).

 

I assume that these 4 factors are nested as follows:

  • Site
  • Instrument (nested in Site)
  • Day (nested in Instrument and Site)
  • Run (nested in Day, Instrument and Site).

I would like to have the following outputs (expressed as CV%):

  • Precision within the same run
  • Precision between runs
  • Precision between days
  • Precision within the same instrument
  • Precision between instruments
  • Precision between sites
  • Total precision.

Can this be done in JMP? What approach would you recommend me to use?

I am also supposed to analyze the data via ANOVA. Would I have to perform a four-way nested ANOVA? How can I do this in JMP?

 

Thank you very much in advance.

C.

1 REPLY 1

Re: How to perform precision analysis and ANOVA having four nested factors?

First, layout out the JMP data table so that each factor and the assay are in separate columns. You can use the Variability Chart platform or the Fit Least Squares platform to estimate the variance components with the nested model.

 

I attached a data table that is probably like your study so that you can see how to set up your data. I saved a script for the two platforms so you can study the results. The Assay data is simulated with all the nested factor random effects equal to zero.