X hits on this document

22 views

0 shares

11 / 13

Test Likelihood Ratio (LR) Statistics Expected Prediction Table Hosmer-Lemeshow (H-L)

Statistic and Probability

• 61.10555

P = [0.000000]

• 76.81

%

• 21.3800

2(8) = 0.0062

Table 3 – Regression results (odds ratios)

Constant

0.417659

0.406803

1.518403

AGE

-0.037674

-3.057215**

0.963027

-0.006

GENDER

0.248005

1.050976

1.281466

0.037

D1EDU

1.345380

2.289635**

3.839645

0.201

D3EDU

0.205347

0.355748

1.227951

0.031

D4EDU

-0.637212

-0.995511

0.528765

-0.095

KNOW

0.159947

2.584408**

1.173449

0.024

MAJOR

0.079689

2.647469**

1.08295

0.012

BELONG

-0.170993

-3.956992**

0.842827

-0.026

Variables Coefficient z-Statistic

** significantly different from zero at the 5 percent level

Odds Ratio

Marginal Effect

Table 3 presents the results of the logit analysis of having or not having identity dilemma. The coefficients should be interpreted as the effects of the variables on the log odds of not having identity dilemma versus not (logarithm of the probability of

having identity dilemma). The Odds Ratio column is the antilog value of the coefficient. The Marginal Effect column provides the marginal effect of a change in

on the probability of not having identity dilemma.

Being male has a positive effect on the odds ratio, but the effect is not statistically significant. Age has a statistically significant effect on the odds ratio; that is being older results in a lower probability of not having an identity dilemma. This result is

related to the assimilation process. ethnicity of the majority ethnic group similarities and shared characteristics of

The older generation has assimilated the of the country having realized the many the majority.

Having received secondary and tertiary education has no effect on the odds ratio of not having an identity dilemma. However, primary education has a positive significant effect on the odds ratio of not having an identity dilemma.

Knowledge of ethnic group has a significant positive effect, where the more knowledge an individual has on his (her) ethnic group; the higher is the probability of not having an identity dilemma. A one point increase in the knowledge score would increase the probability of not having an identity dilemma by 2.4 percent.

Identification with the majority group has a significant positive effect, where the more an individual feel identified with the majority group, the higher is the probability of not having an identity dilemma. A one point increase in the identification with the majority group score would increase the probability of not having an identity dilemma by 1.2 percent.

80

 Document views 22 Page views 22 Page last viewed Thu Oct 27 03:09:24 UTC 2016 Pages 13 Paragraphs 297 Words 5019