politicalInformation          package:pscl          R Documentation

_I_n_t_e_r_v_i_e_w_e_r _r_a_t_i_n_g_s _o_f _r_e_s_p_o_n_d_e_n_t _l_e_v_e_l_s _o_f _p_o_l_i_t_i_c_a_l _i_n_f_o_r_m_a_t_i_o_n

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

     Interviewers administering the 2000 American National Election
     Studies assigned an ordinal rating to each respondent's "general
     level of information" about politics and public affairs.

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

     data(politicalInformation)

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

     A data frame with 1807 observations on the following 8 variables.

     '_y' interviewer rating, a factor with levels 'Very Low' 'Fairly
          Low' 'Average' 'Fairly High' 'Very High'

     '_c_o_l_l_e_g_e_D_e_g_r_e_e' a factor with levels 'No' 'Yes'

     '_f_e_m_a_l_e' a factor with levels 'No' 'Yes'

     '_a_g_e' a numeric vector, respondent age in years

     '_h_o_m_e_O_w_n' a factor with levels 'No' 'Yes'

     '_g_o_v_t' a factor with levels 'No' 'Yes'

     '_l_e_n_g_t_h' a numeric vector, length of ANES pre-election interview
          in minutes

     '_i_d' a factor, unique identifier for each interviewer

_D_e_t_a_i_l_s:

     Seven respondents have missing data on the ordinal interviewer
     rating. The covariates 'age' and 'length' also have some missing
     data.

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

     The National Election Studies (www.electionstudies.org). THE 2000
     NATIONAL ELECTION STUDY [dataset]. Ann Arbor, MI: University of
     Michigan, Center for Political Studies [producer and distributor].

_R_e_f_e_r_e_n_c_e_s:

     Jackman, Simon. 2009.  _Bayesian Analysis for the Social
     Sciences_.  Wiley: Hoboken, New Jersey.

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

     data(politicalInformation)

     table(politicalInformation$y,exclude=NULL)

     require(MASS)
     op <- polr(y ~ collegeDegree + female + log(age) + homeOwn + govt + log(length),
                data=politicalInformation,
                Hess=TRUE,
                method="probit")

