Skip to main content
. 2022 Sep 1;29(1):2831–2845. doi: 10.1080/10717544.2022.2118399

Table 3.

Summary of the results of the statistical analysis and model equations for the measured responses.

Model SD R 2 R2 (adj) p value Lack of fit p value PRESS Remark
Droplet size (nm)              
 Linear 17.53 0.8807 0.8755 <.0001 <.0001 16,920.15
 Quadratic 9.25 0.9689 0.9653 <.0001 .0003 5,123.73
 Special cubic 6.95 0.9829 0.9804 <.0001 .0086 2,928.82
 Cubic 3.54 0.9959 0.9949 <.0001 .9668 8,13.19 Suggested
INS stability (%)              
 Linear 5.16 0.7581 0.7476 <.0001 <.0001 1,523.40
 Quadratic 2.70 0.9383 0.9312 <.0001 .0119 433.10
 Special cubic 2.29 0.9567 0.9505 .0001 .0659 312.15
 Cubic 1.77 0.9759 0.9703 <.0001 .4758 207.19 Suggested
INS leakage (%)              
 Linear 3.84 0.7333 0.7217 <.0001 .0107 831.44
 Quadratic 2.30 0.9103 0.8999 <.0001 .6858 324.25
 Special cubic 2.23 0.9176 0.9058 .0616 .7628 312.77
 Cubic 2.02 0.9372 0.9227 .0131 .9464 262.17 Suggested

INS, insulin; SD, standard deviation; PRESS, predicted residual error sum of squares.