Email: support@essaywriterpros.com
Call Us: US - +1 845 478 5244 | UK - +44 20 7193 7850 | AUS - +61 2 8005 4826

Robustness of the Effect of Riot Exposure on Support for Separation

TABLE 2 Robustness of the Effect of Riot Exposure on Support for Separation Probit Model, unless Specified Otherwise Dependent Variable: Support Separation Panel A: Robustness to Different Subsamples (1) (2) (3) Sample Excl. Nuba North-Central only Riverine Arabs only Fighting in neighborhood .725 (.196)*** .540 (.225)** .615 (.276)** Observations 566 326 166 Panel B: Robustness to Different Modeling Approaches (4) (5) (6) Model Ignore survey design Ordered probit Multiple imputation Fighting in neighborhood .384 (.132)*** .409 (.178)** .518 (.168)*** Observations 729 768 929 Panel C: Robustness to Effects of Residential Patterns (7) (8) (9) Robustness check Incl. respondent estimate of Southern neighborhood share Incl. respondent estimate, code ‘‘don’t know’’ as zero Excl. if respondent estimate is 10% or higher Fighting in neighborhood .553 (.234)** .639 (.195)*** .695 (.288)** Observations 553 675 309 (10) (11) (12) Robustness check Incl. PAU-level average respondent estimate Incl. PAU-level average, code ‘‘don’t know’’ as zero Incl. Southern share of PAU-level sample Fighting in neighborhood .535 (.160)*** 519 (.158)*** .562 (.178)*** Observations 729 729 729 Note: Models include gender, age, employment and self-employment indicators, asset index, relative wealth, logged years of education, logged years of father’s education, and AU and region of origin indicators. Due to sample limitations, model (3) does not include AU indicators. **p # .05, ***p # .0