In order to achieve two functions graph (DF1, DF2) for biometric sheet, this script has been used:
_________________________________________________________________
># run the discriminant analysis
>discrminantTest #see the discriminant power of the variables
>discrminantTest$power
>#make new columns on tableM with the scores of each specimen on the first and second discrminant functions
>tableM$DF1 tableM$DF2 # to only keep the columns you need
>tableM % select(collection.number:XXX)
_________________________________________________________________
The triple "X" is my latest biometric trait which after that I received this error:
>Error in is_character(x, encoding = encoding, n = 1) :
object 'collection.number' not found
Seyed,
I think you can achieve the subset of data without the piping (i.e., %>%). Wouldn't tableM
Caleb,
Thank you indeed. Maybe I could not do it correctly. May you please kindly have a look to my data set and R script too? I have attached her. Please change the YY and ZZ in the mentioned script with your idea.
Regards
Seyed,
From viewing your data and code, I believe you may be unclear on what the select function is trying to do and I am unclear on exactly which columns you are trying to keep. So, let me attempt to make clear/understand; but help me if I'm wrong.
When you call the select function it is attempting to read columns from a dataframe that you specify—the process is done by naming the function, then the dataframe, then the column names. In your case tableM is the name of your dataframe and the names of your columns are in the first row of your xlsx file (e.g., species, Code, LogBl, BL, etc.). So, collection.number, YY, and ZZ can't be selected because they aren't column names in your dataframe (error). If you know the columns you want to keep (I don't know which columns you'd like to keep) simply call the select function and list dataframe and desired column names putting them into a new or existing dataframe. For example, tableM
Caleb,
I change the grouping with dividing to sub species (4 group) with a bit change in script. Is there any possibility error causing when we try with only two group? Kindly look at the new script please:
#install.packages("dplyr")
# load extra packages
library(xlsx) #to read excel files
# to manipulate tables
library(dplyr)
# read the spreadsheet
tableM % count(species)
# for discriminant analyses
library(DiscriMiner)
# run the discriminant analysis
discrminantTest
I don't see why it would throw an error (technical error) when only using two subspecies or grouping your current subspecies into two groups (I'm not sure what these new "species" groups would represent, biologically speaking). If you just wanted to focus on two of the subspecies, just subset your data (see link below). If you want to put your current subspecies into two groups, just create a new variable. E.g.,
tableM$newSpecies
- Badges
- Science topic
- Similar topics
- Analytical Chemistry
- Column