Dear Community,
My data table has ~5000 rows with three columns: OTU ID, condition, Relative abundance
I performed a one way Y - X analysis with the designations: condition (X), Relative Abundance (Y) and OTU ID (By)
In the one-way analysis report I have to manually check each of the OTU ID for significance effect of condition in e.g. ttest or ANOVA
Is there a way to make JMP print a new column in my workbook with the p-value for the desired statistical test without going through the entire report (there are hundreds or even thousands of tests to look through)
Thank you
JMP generally makes it very easy to do this kind of thing, though (if you want to use a script) the details will depend on which test you want to use (more specifically, on the structure of the report that the platform generates). The script below is a simple example:
NamesDefaultToHere(1);
// Sample Data
dt1 = Open("$SAMPLE_DATA/Big Class.jmp");
// One way ANOVA with a 'By' group
ow = dt1 << Oneway( Y( :height ), X( :age ), Means( 1 ), Mean Diamonds( 1 ), By( :sex ));
// Find the table you want in the report, do a right-click and select 'Make Combined Data Table'
dt2 = Report(ow[1])[TableBox(2)] << makeCombinedDataTable;
dt2 << setName("Oneway ANOVA Results");
But just because something is easy does not make it good practice. If you are doing lots of statistical tests, I strongly recommend that you read about Response Screening.
JMP generally makes it very easy to do this kind of thing, though (if you want to use a script) the details will depend on which test you want to use (more specifically, on the structure of the report that the platform generates). The script below is a simple example:
NamesDefaultToHere(1);
// Sample Data
dt1 = Open("$SAMPLE_DATA/Big Class.jmp");
// One way ANOVA with a 'By' group
ow = dt1 << Oneway( Y( :height ), X( :age ), Means( 1 ), Mean Diamonds( 1 ), By( :sex ));
// Find the table you want in the report, do a right-click and select 'Make Combined Data Table'
dt2 = Report(ow[1])[TableBox(2)] << makeCombinedDataTable;
dt2 << setName("Oneway ANOVA Results");
But just because something is easy does not make it good practice. If you are doing lots of statistical tests, I strongly recommend that you read about Response Screening.
Thank you I basically went with response screening using my "By" column as a grouping variable and it printed a table with P-values for each grouping variable that I was able to filter for <0.05
This immensely helped!!
However... what is the default statistical test that is being run, i.e. what does the p-value correspond to if I dont select any options and just run the dialog?
Adding to @ian_jmp , just right-click on one of the tables with p-values and select Make Into Combined Table.
Hi Mark,
This did not work for me, it made a data table with only five of the several hundred tables, also the p-values were not organized properly