Regression results and Shapley decomposition
| . | D Scores (std.) . | Index (std.) . | Personal View . | Perceived Norm . | ||||
|---|---|---|---|---|---|---|---|---|
| Dep. Variable . | OLS (1) . | Shapley (%) (2) . | OLS (3) . | Shapley (%) (4) . | OLS (5) . | Shapley (%) (6) . | OLS (7) . | Shapley (%) (8) . |
| Respondent Type | 74.9 | 20.3 | 0.6 | 0.2 | ||||
| Male | 0.527** | 0.439** | −0.025 | −0.010 | ||||
| (0.057) | (0.053) | (0.064) | (0.053) | |||||
| Region | 9.2 | 57.8 | 46.4 | 92.9 | ||||
| Gegharkunik | 0.176** | 0.492** | 0.386** | 0.086 | ||||
| (0.065) | (0.060) | (0.073) | (0.061) | |||||
| Syunik | −0.040 | −0.236** | −0.334** | −0.558** | ||||
| (0.059) | (0.055) | (0.067) | (0.055) | |||||
| Socioeconomic Status | 16.0 | 22.0 | 53.0 | 6.9 | ||||
| Income level | 0.012 | 0.076** | −0.075** | 0.006 | ||||
| (0.021) | (0.020) | (0.024) | (0.020) | |||||
| Education level | −0.031 | −0.075** | −0.149** | −0.019 | ||||
| (0.029) | (0.027) | (0.033) | (0.027) | |||||
| Year of birth | −0.003 | −0.004† | −0.015** | −0.002 | ||||
| (0.002) | (0.002) | (0.002) | (0.002) | |||||
| Agriculture | 0.010 | −0.127 | 0.040 | −0.015 | ||||
| (0.095) | (0.088) | (0.107) | (0.088) | |||||
| Commerce and private business | −0.060 | −0.110 | 0.176† | 0.072 | ||||
| (0.084) | (0.078) | (0.094) | (0.078) | |||||
| Construction | 0.026 | −0.049 | 0.364* | 0.109 | ||||
| (0.132) | (0.123) | (0.149) | (0.123) | |||||
| Government, intergovernmental organization, or NGO employee | 0.034 | −0.126* | 0.113 | 0.087 | ||||
| (0.065) | (0.061) | (0.074) | (0.061) | |||||
| Pay by the day | 0.048 | −0.226† | −0.259† | 0.050 | ||||
| (0.131) | (0.122) | (0.148) | (0.122) | |||||
| Constant | 5.037 | 7.187† | 29.981** | 4.174 | ||||
| (4.015) | (3.733) | (4.532) | (3.743) | |||||
| N | 1,804 | 1,804 | 1,804 | 1,804 | ||||
| Overall R2 | .065 | .146 | .130 | .087 | ||||
| F Statistic Model | 11.363 | 27.904 | 24.405 | 15.534 | ||||
| . | D Scores (std.) . | Index (std.) . | Personal View . | Perceived Norm . | ||||
|---|---|---|---|---|---|---|---|---|
| Dep. Variable . | OLS (1) . | Shapley (%) (2) . | OLS (3) . | Shapley (%) (4) . | OLS (5) . | Shapley (%) (6) . | OLS (7) . | Shapley (%) (8) . |
| Respondent Type | 74.9 | 20.3 | 0.6 | 0.2 | ||||
| Male | 0.527** | 0.439** | −0.025 | −0.010 | ||||
| (0.057) | (0.053) | (0.064) | (0.053) | |||||
| Region | 9.2 | 57.8 | 46.4 | 92.9 | ||||
| Gegharkunik | 0.176** | 0.492** | 0.386** | 0.086 | ||||
| (0.065) | (0.060) | (0.073) | (0.061) | |||||
| Syunik | −0.040 | −0.236** | −0.334** | −0.558** | ||||
| (0.059) | (0.055) | (0.067) | (0.055) | |||||
| Socioeconomic Status | 16.0 | 22.0 | 53.0 | 6.9 | ||||
| Income level | 0.012 | 0.076** | −0.075** | 0.006 | ||||
| (0.021) | (0.020) | (0.024) | (0.020) | |||||
| Education level | −0.031 | −0.075** | −0.149** | −0.019 | ||||
| (0.029) | (0.027) | (0.033) | (0.027) | |||||
| Year of birth | −0.003 | −0.004† | −0.015** | −0.002 | ||||
| (0.002) | (0.002) | (0.002) | (0.002) | |||||
| Agriculture | 0.010 | −0.127 | 0.040 | −0.015 | ||||
| (0.095) | (0.088) | (0.107) | (0.088) | |||||
| Commerce and private business | −0.060 | −0.110 | 0.176† | 0.072 | ||||
| (0.084) | (0.078) | (0.094) | (0.078) | |||||
| Construction | 0.026 | −0.049 | 0.364* | 0.109 | ||||
| (0.132) | (0.123) | (0.149) | (0.123) | |||||
| Government, intergovernmental organization, or NGO employee | 0.034 | −0.126* | 0.113 | 0.087 | ||||
| (0.065) | (0.061) | (0.074) | (0.061) | |||||
| Pay by the day | 0.048 | −0.226† | −0.259† | 0.050 | ||||
| (0.131) | (0.122) | (0.148) | (0.122) | |||||
| Constant | 5.037 | 7.187† | 29.981** | 4.174 | ||||
| (4.015) | (3.733) | (4.532) | (3.743) | |||||
| N | 1,804 | 1,804 | 1,804 | 1,804 | ||||
| Overall R2 | .065 | .146 | .130 | .087 | ||||
| F Statistic Model | 11.363 | 27.904 | 24.405 | 15.534 | ||||
Notes: The results in this table are based on our representative sample, which excludes households living in communities that we have specifically targeted because of their extreme sex ratios at birth. See section A.1 of the online appendix for details about our two-stage sampling approach. D scores and index values are standardized to have unit variance so that coefficients on correlates are comparable. The variables in columns 5 through 8 are constructed from responses to the questions about the Armenian adage discussed in the main text. We encode the responses by assigning values between 1 (strongly disagree) and 5 (strongly agree). For occupations, the excluded category is “Housework or other.” Standard errors are shown in parentheses.
†p < .10; *p < .05; **p < .01