Skip to content
Prev 6712 / 7419 Next

GLMM: Should Integers be Levels of Dr. Alexandre F. Souza Prof. Associado a Factor?

Hello all,

Judging by some sources, it seems to me that before running a GLMM I should
make all categorical variables into factors, including numerical ones that
are integers like binary response variables, plot numbers, block levels,
sampling year and the ID of individual measured plants, even if these
variables are correctly identified as integers as in the example below. Is
this understanding correct in your opinion?

Thank you very much for any reply,

Alexandre
'data.frame': 1428 obs. of  12 variables:
 $ numero                  : int  4006 4009 4010 4020 4022 4024 4025 4027
4035 4036 ...
 $ bloco                   : int  1 1 1 1 1 1 1 1 1 1 ...
 $ transecto               : int  1 1 1 1 1 1 1 1 1 1 ...
 $ parcela                 : int  1 1 1 1 1 1 1 1 1 1 ...
 $ Tree_Height_m_Fuste_Parc: num  8.32 8.32 8.32 8.32 8.32 ...
 $ Grupos_Finais           : Factor w/ 3 levels "Disturbed Unflooded
Forests",..: 1 1 1 1 1 1 1 1 1 1 ...
 $ ano_medida              : int  2016 2016 2016 2016 2016 2016 2016 2016
2016 2016 ...
 $ Basal_Area_m2           : num  0.182 0.182 0.182 0.182 0.182 ...
 $ das_mm                  : num  185 161 167 114 164 ...
 $ reproducao              : int  3 0 0 0 0 0 0 0 0 0 ...
 $ reproducao_bin          : int  1 0 0 0 0 0 0 0 0 0 ...
 $ R_dens_total            : int  31 31 31 31 31 31 31 31 31 31 ...