Table 4.

Robustness checks.

Approach 1Approach 2
Model 1Model 2Model 3Model 4
Increase in Chinese in the vicinityIncrease in Koreans in the vicinityAccept Chinese or Korean as relativesAccept Chinese or Korean as relatives
First stage
Chinese population density in each region (IV)8.426*** (1.418)
Korean population density in each region (IV)11.468*** (1.226)
Chinese and Korean population density in each prefecture (IV)4.267* (2.138)4.267* (2.138)
Second stage
Chinese friends (not have = 0; have = 1)–6.521* (3.307)
Korean friends (not have = 0; have = 1)–1.084 (1.440)
chinese or korean acquaintances (none = 0; either = 1; both = 2)2.791 (2.534)–2.085 (2.169)
Survey year, relations indexYesYesYesYes
Age, educational attainment, urbanity of dwelling, income bracket, and genderYesYesYesYes
Nested within HLMYesYesYesYes
Log likelihood−4,006−4,516−1,880−2,174
N7,1597,1333,3693,373
Approach 1Approach 2
Model 1Model 2Model 3Model 4
Increase in Chinese in the vicinityIncrease in Koreans in the vicinityAccept Chinese or Korean as relativesAccept Chinese or Korean as relatives
First stage
Chinese population density in each region (IV)8.426*** (1.418)
Korean population density in each region (IV)11.468*** (1.226)
Chinese and Korean population density in each prefecture (IV)4.267* (2.138)4.267* (2.138)
Second stage
Chinese friends (not have = 0; have = 1)–6.521* (3.307)
Korean friends (not have = 0; have = 1)–1.084 (1.440)
chinese or korean acquaintances (none = 0; either = 1; both = 2)2.791 (2.534)–2.085 (2.169)
Survey year, relations indexYesYesYesYes
Age, educational attainment, urbanity of dwelling, income bracket, and genderYesYesYesYes
Nested within HLMYesYesYesYes
Log likelihood−4,006−4,516−1,880−2,174
N7,1597,1333,3693,373

Note: 1. Because the ICPP survey does not cover all prefecture-specific respondents, its population densities are broken down by region. 2. The relational indices in Models 1 and 2 are respectively for Sino-Japanese relations and Japan-South Korea relations, whereas in Models 3 and 4, the indices pertain to Japan’s relations with both China and South Korea.

Logit regression. The full results are available in Table A5 in the Supplementary Material.

Standard errors are in parentheses, *** P < .001, ** P < .01,* P < .05 (two-tailed test).

Table 4.

Robustness checks.

Approach 1Approach 2
Model 1Model 2Model 3Model 4
Increase in Chinese in the vicinityIncrease in Koreans in the vicinityAccept Chinese or Korean as relativesAccept Chinese or Korean as relatives
First stage
Chinese population density in each region (IV)8.426*** (1.418)
Korean population density in each region (IV)11.468*** (1.226)
Chinese and Korean population density in each prefecture (IV)4.267* (2.138)4.267* (2.138)
Second stage
Chinese friends (not have = 0; have = 1)–6.521* (3.307)
Korean friends (not have = 0; have = 1)–1.084 (1.440)
chinese or korean acquaintances (none = 0; either = 1; both = 2)2.791 (2.534)–2.085 (2.169)
Survey year, relations indexYesYesYesYes
Age, educational attainment, urbanity of dwelling, income bracket, and genderYesYesYesYes
Nested within HLMYesYesYesYes
Log likelihood−4,006−4,516−1,880−2,174
N7,1597,1333,3693,373
Approach 1Approach 2
Model 1Model 2Model 3Model 4
Increase in Chinese in the vicinityIncrease in Koreans in the vicinityAccept Chinese or Korean as relativesAccept Chinese or Korean as relatives
First stage
Chinese population density in each region (IV)8.426*** (1.418)
Korean population density in each region (IV)11.468*** (1.226)
Chinese and Korean population density in each prefecture (IV)4.267* (2.138)4.267* (2.138)
Second stage
Chinese friends (not have = 0; have = 1)–6.521* (3.307)
Korean friends (not have = 0; have = 1)–1.084 (1.440)
chinese or korean acquaintances (none = 0; either = 1; both = 2)2.791 (2.534)–2.085 (2.169)
Survey year, relations indexYesYesYesYes
Age, educational attainment, urbanity of dwelling, income bracket, and genderYesYesYesYes
Nested within HLMYesYesYesYes
Log likelihood−4,006−4,516−1,880−2,174
N7,1597,1333,3693,373

Note: 1. Because the ICPP survey does not cover all prefecture-specific respondents, its population densities are broken down by region. 2. The relational indices in Models 1 and 2 are respectively for Sino-Japanese relations and Japan-South Korea relations, whereas in Models 3 and 4, the indices pertain to Japan’s relations with both China and South Korea.

Logit regression. The full results are available in Table A5 in the Supplementary Material.

Standard errors are in parentheses, *** P < .001, ** P < .01,* P < .05 (two-tailed test).

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close