Premium
Relationship between Number of Independent Variables and Number of Observations in Plant Analysis Calibration Studies 1
Author(s) -
Walker W. M.,
Carmer S. G.,
Peck T. R.
Publication year - 1969
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1969.00021962006100020043x
Subject(s) - polynomial regression , calibration , mathematics , quadratic equation , regression analysis , statistics , regression , linear regression , transformation (genetics) , biology , biochemistry , geometry , gene
Data were presented indicating that the number of data points in plant analysis calibration studies should be 5 to 10 times as large as the number of potential variables in the regression model. This criterion was valid for the quadratic polynomial and the square root transformation of the quadratic polynomial in these studies. This kind of information should assist researchers in planning the number of plant samples to be obtained for calibration studies after a regression model has been selected.