Skip to main content

Table 8 Accuracy of OCW-based regressions compared to those of other studies

From: Occipital condyle width (OCW) is a highly accurate predictor of body mass in therian mammals

Taxonomic scope

OCW

Other regression equations

N

r2

%PE

%SEE

N

r2

%PE

%SEE

Variable

Ref.

All Therians

401

0.982

31.55

45.92

80*

0.985

28.83

41.98

Astragalus (Li1)

[58]

    

80*

0.980

34.01

49.68

Astragalus (Li7)

[58]

    

80*

0.980

37.17

55.53

Astragalus (Li4)

[58]

All Therians (controlling for shape)

404

0.985

28.18

41.37

69*

0.979

32.36

47.21

Calcaneus (CA2)

[93]

    

69*

0.973

38.67

55.46

Calcaneus (CA5)

[93]

    

69*

0.973

38.92

55.42

Calcaneus (CA10)

[93]

     

187

0.955

50.60

84.78

Humerus length

[87]

     

200

0.982

27.73

47.86

Humerus circum.

[87]

     

188

0.944

66.77

98.01

Femur length

[87]

     

200

0.983

31.16

45.93

Femur circum.

[87]

Australidelphian Marsupials

32

0.972

30.54

47.44

38

0.980

27

39

Total jaw length

[158]

    

38

0.979

28

40

CBL

[158]

    

38

0.967

35

51

M1-4 length

[158]

Carnivora + Large Dasyuromorphia

86

0.973

23.59

34.74

72

0.96

36

53

HBL

[39]

    

72

0.95

47

66

CBL

[39]

    

72

0.95

42

61

Occiput-orbit length

[39]

Ungulates (except proboscideans)

61

0.962

27.47

39.95

91

0.975

31.42

45.54

HBL

[66]

    

94

0.967

34.51

53.02

p4-m3 length

[66]

    

94

0.963

36.34

55.82

m1-3 length

[66]

Rodents

96

0.971

29.35

40.80

36

0.940

30.65

15.70

CBL

[31]

    

36

0.890

43.36

21.50

P4-M3 length

[31]

    

34

0.890

44.62

21.80

m1 length

[31]

Rodents†

58

0.953

10.21

13.50

36

0.880

39.42

21.70

Humerus diameter

[31]

     

203*

0.964

29.55

43.33

CBL

[155]

     

93

0.948

34.55

50.14

CBL

[155]

     

75

0.920

40.10

64.45

p4-m3 length

[159]

Anthropoid primates

43

0.965

18.49

28.41

46

0.98

24

Biorbital breadth

[160]

    

46

0.97

25

Foramen magnum area

[160]

     

38

0.98

18

Humeral length

[160]

     

38

0.99

20

Anteroposterior diameter of femoral head

[160]

  1. Only linear variables are considered here as models with multiple independent anatomical variables tend to produce lower errors regardless of what measurements are considered [58, 86, 159]. When large numbers of regression equations were calculated in a given study, only the equations with the lowest %PE were used for comparison. %PE was reported for this study instead of %PEcf as many of the studies considered here did not use correction factors. * = regression equation calculated using individual specimens, instead of species averages. † = results of Moncunill-Solé et al. [119]