dp                  package:gmvalid                  R Documentation

_D_e_a_t_h _p_e_n_a_l_t_y _e_x_a_m_p_l_e _o_f _S_i_m_p_s_o_n'_s _p_a_r_a_d_o_x

_D_e_s_c_r_i_p_t_i_o_n:

     Well known example to illustrate Simpson's paradox. The data set
     shows that ignoring victim's race lead to a different conclusion
     than including victim's race in the analysis.

_U_s_a_g_e:

     data(dp)

_F_o_r_m_a_t:

     A data frame with 326 observations on the following 3 variables.

     '_D_e_f_e_n_d_a_n_t_s._R_a_c_e' a factor with levels 'white' and 'black'

     '_V_i_c_t_i_m_s._R_a_c_e' a factor with levels 'white' and 'black'

     '_D_e_a_t_h._P_e_n_a_l_t_y' a factor with levels 'yes' and 'no'

_S_o_u_r_c_e:

     Radelet ML (1981) _Racial characteristics and the imposition of
     the Death penalty._ American Sociological Review, 46(6):918-927.

_E_x_a_m_p_l_e_s:

     data(dp)
     ## Graphical model analysis shows that 'defendants' race' is 
     ## independent from 'death penalty' given 'victims' race'.
     gm.analysis(dp,program="coco",recursive=TRUE)

