Skip to main content

Table 7 Weighted OLS and FE estimates comparing the impact of diversity on wages for different groups. Dependent variable: log of weekly wages

From: The regional impact of cultural diversity on wages: evidence from Australia

Explanatory variables

Skilled

Unskilled

Urban residents

Rural residents

OLS

FE

OLS

FE

OLS

FE

OLS

FE

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Diversity index

0.722***

−0.046

0.437***

0.222

0.428***

0.036

0.648**

−0.027

 

(0.111)

(0.234)

(0.089)

(0.245)

(0.078)

(0.164)

(0.276)

(0.309)

Demographic controls

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Human capital controls

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Time dummiesa

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Individual effects

No

Yes

No

Yes

No

Yes

No

Yes

Observations

10,345

9759

9779

9170

17,758

17,321

2397

2143

Adjusted R-squared

0.460

0.237

0.612

0.337

0.623

0.389

0.545

0.281

First stage for diversity index

        

Predicted diversity index

0.599***

0.240***

0.625***

0.241***

0.629***

0.264***

0.499***

−0.399***

 

(0.021)

(0.044)

(0.024)

(0.046)

(0.016)

(0.033)

(0.050)

(0.110)

Instrument tests

        

Underidentification testb

398.33

24.01

344.38

24.17

586.48

52.60

41.89

11.49

p value

0.000

0.000

0.000

0.000

0.000

0.000

0.000

0.000

Weak identification testc

813.93

29.61

706.87

27.76

1453.03

64.92

97.93

13.15

  1. Values are estimated for the HILDA wave 4–11 respondents aged 16–45 years. Robust standard errors are in parentheses and are adjusted for clusters by individuals. The reported sample sizes differ from the original (n = 44,634) due to sample weighting and lagging. Demographic controls include age and age-squared in all models, plus gender and marital status in the OLS models. Human capital controls include education, weekly hours worked, and job tenure, in all models, plus English language skill in OLS
  2. Significance values indicate *p < 0.1; **p < 0.05; ***p < 0.01
  3. aTime dummies are partialled out in the FE models
  4. bBased on the Kleibergen-Paap rk LM statistic
  5. cBased on the Kleibergen-Paap rk Wald F statistic: Stock-Yogo weak ID test critical value maximal IV sizes range as follows: 16.38 (10 %), 8.96 (15 %), 6.66 (20 %), and 5.53 (25 %)
Back to article page