possumDiv             package:robustbase             R Documentation

_P_o_s_s_u_m _D_i_v_e_r_s_i_t_y _D_a_t_a

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

     Possum diversity data: As issued from a study of the diversity of
     possum (arboreal marsupials) in the Montane ash forest
     (Australia), this dataset was collected in view of the management
     of hardwood forest to take conservation and recreation values, as
     well as wood production, into account.

     The study is fully described in the two references. The number of
     different species of arboreal marsupials (possum) was observed on
     151 different 3ha sites with uniform vegetation. For each site the
     nine variable measures (see below) were recorded. The problem is
     to model the relationship between 'diversity' and these other
     variables.

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

     data(possumDiv)

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

     Two different representations of the same data are available:

     'possumDiv' is a data frame of 151 observations of 9 variables,
     where the last two are factors, 'eucalyptus' with 3 levels and
     'aspect' with 4 levels.

     'possum.mat' is a numeric (integer) matrix of 151 rows
     (observations) and 14 columns (variables) where the last seven
     ones are 0-1 dummy variables, three ('E.*') are coding for the
     kind of 'eucalyptus' and the last four are 0-1 coding for the
     'aspect' factor.

     The variables have the following meaning:

     _D_i_v_e_r_s_i_t_y main variable of interest is the number of different
          species of arboreal marsupial (possum) observed, with values
          in 0:5.

     _S_h_r_u_b_s the number of shrubs.

     _S_t_u_m_p_s the number of cut stumps from past logging operations.

     _S_t_a_g_s the number of stags (hollow-bearing trees).

     _B_a_r_k bark index (integer) vector reflecting the quantity of
          decorticating bark.

     _H_a_b_i_t_a_t an integer score indicating the suitability of nesting and
          foraging habitat for Leadbeater's possum.

     _B_A_c_a_c_i_a a numeric vector giving the basal area of acacia species. 

     _e_u_c_a_l_y_p_t_u_s a 3-level 'factor' specifying the species of eucalypt
          with the greatest stand basal area.  This has the same
          information as the following three variables

     _E._r_e_g_n_a_n_s 0-1 indicator for Eucalyptus regnans

     _E._d_e_l_e_g_a_t_e_n_s_i_s 0-1 indicator for Eucalyptus deleg.

     _E._n_i_t_e_n_s 0-1 indicator for Eucalyptus nitens 

     _a_s_p_e_c_t a 4-level 'factor' specifying the aspect of the site.  It
          is the same information as the following four variables.

     _N_W-_N_E 0-1 indicator

     _N_W-_S_E 0-1 indicator

     _S_E-_S_W 0-1 indicator

     _S_W-_N_W 0-1 indicator

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

     Eva Cantoni (2004) Analysis of Robust Quasi-deviances for
     Generalized Linear Models. _Journal of Statistical Software_ *10*,
     04, <URL: http://www.jstatsoft.org/>

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

     Lindenmayer, D.~B., Cunningham, R.~B., Tanton, M.~T., Nix, H.~A.
     and Smith, A.~P. (1991) The conservation of arboreal marsupials in
     the montane ash forests of the central highlands of victoria,
     south-east australia: {III}. The habitat requirements of
     leadbeater's possum _gymnobelideus leadbeateri_ and models of the
     diversity and abundance of arboreal marsupials. _Biological
     Conservation_ *56*, 295-315.

     Lindenmayer, D.~B., Cunningham, R.~B., Tanton, M.~T., Smith, A.~P.
     and Nix, H.~A. (1990) The conservation of arboreal marsupials in
     the montane ash forests of the victoria, south-east australia,
     {I}. Factors influencing the occupancy of trees with hollows,
     _Biological Conservation_ *54*, 111-131.

     See also the references in 'glmrob'.

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

     data(possumDiv)
     head(possum.mat)

     str(possumDiv)
     ## summarize all variables as multilevel factors:
     summary(as.data.frame(lapply(possumDiv, function(v)
                                  if(is.integer(v)) factor(v) else v)))

