Exotic Brome-Grasses in Arid and Semiarid Ecosystems of the Western US

169

Table 6.4

Results of two multiple logistic regressions with a backward stepwise elimination conducted for each of the four dependen

t variables

Is widely introduced

Is a crop weed

Is a ruderal weed

Is an environmental weed

Wald Chi Sq.

Pr >Chi Sq.

Odds ratio point est.

Wald Chi Sq.

Pr >Chi Sq.

Odds ratio point est.

Wald Chi Sq.

Pr >Chi Sq.

Odds ratio point est.

Wald Chi Sq.

Pr >Chi Sq.

Odds ratio point est.

High max. awn- length

5.5659

0.0183

1.194 (1.030–1.384)

Early year

8.9630

0.0028

0.971 (0.952–0.990)

13.6144

0.0002

0.963 (0.944–0.983)

Low mean seed mass

4.1653

0.0413

0.738 (0.551–0.988)

Low min. precip.

4.7191

0.0298

0.996 (0.992–1.000)

5.6404

0.0176

0.933 (0.987–0.999)

High max. awn length

4.6317

0.03514

0.0915 (4.6317–0.0314)

Early year

12.7009

0.0004

0.974 (0.959–0.988)

5.9404

0.0148

0.976 (0.956–0.995)

13.6144

0.0002

0.963 (0.944–0.983)

Wide temp. range

4.4334

0.0352

1.130 (1.008–1.256)

The same independent variables were used as those used in the univariate regressions with the exception of section. In one se

t of multiple regressions, minimum

temperature, maximum temperature, minimum precipitation, and maximum precipitation were used. In the other, temperature range a

nd precipitation ranges

were used. Fifty-four species were used for the multiple regressions and correlations were considered signifi

cant if the

p -value (Pr > Chi Sq.) was less than 0.05

6 Attributes That Confer Invasiveness and Impacts Across the Large Genus Bromus...