Table 1. Description of the variables used in the regression–tree analysis.

Variable Character† Type‡ Values
Dependent
Total deforestation (log10) Sp N 9.42 – 1300.00 (0.45 – 3.11)
Independent
Total area Sp N 24.57 – 4831.02
Residence Sp C urban, rural
Region of origin Sp C center-west, south, east, north, northeast
Pasture, “dirty” Sp N 0 – 1
Pasture “clean” Sp N 0 – 1
Secondary vegetation Sp N 0 – 1
Felled area Sp N 0 – 0.83
Crops (annual, perennial, and interplanted) Sp N 0 – 0.52
Remaining forest Sp C 1 (80-92%); 2, (50-80%), 3(20-50%), 4(<20%)
Wealth class§ Ec C (1, 2, 3, and 4)
Annual expenditures Ec N R$ 240 – 19,320
Principal activity Ec C Agriculture and ranching, commerce, urban employment, unemployed
Time of residence Temp N 1 – 29
Cattle (head per property) Phys N 0 – 1600
Cattle (head per family) Phys N 0 – 1700
Infrastructure for livestock Phys N 0 – 61
Family labor Phys N 0-7
Daily labor Phys N 0-6
Permanent labor Phys N 0-6
Labor to outside Phys N 0 – 1

† Sp = spatial; Ec = cost-effective; Phys = physical; and Temp = temporal.
‡ C = categorical; N = numerical
§ Wealth classes based on accumulated possessions: 1 - has no gas stove, refrigerator, or TV; 2 - has at least a gas stove, refrigerator, or TV; 3 - has vehicle (motorcycle, car, truck, or van) and at least one of the items above; and 4 - has two or more vehicles and at least one of the items above.