Core dictionary (coreCIF) version 2.4.5



This definition has been superseded and is retained here only for archival purposes. Use instead '_refine_ls_restrained_S_gt'


   The least-squares goodness-of-fit parameter S' for observed
   reflections after the final cycle of least-squares refinement.
   This parameter explicitly includes the restraints applied in the
   least-squares process.  See also _refine_ls_goodness_of_fit_

        {sum { w [ Y(obs) - Y(calc) ]^2^ }                   }^1/2^
        {         + sum~r~ { w~r~ [ P(calc) - P(targ) ]^2^ } }
   S' = { -------------------------------------------------- }
        {            N~ref~ + N~restr~ - N~param~            }

   Y(obs)   = the observed coefficients
              (see _refine_ls_structure_factor_coef)
   Y(calc)  = the calculated coefficients
              (see _refine_ls_structure_factor_coef)
   w        = the least-squares reflection weight
              [1/square of standard uncertainty (e.s.d.)]

   P(calc)  = the calculated restraint values
   P(targ)  = the target restraint values
   w~r~     = the restraint weight

   N~ref~   = the number of reflections used in the refinement
            (see _refine_ls_number_reflns)
   N~restr~ = the number of restraints
            (see _refine_ls_number_restraints)
   N~param~ = the number of refined parameters
            (see _refine_ls_number_parameters)

   sum     is taken over the specified reflections
   sum~r~  is taken over the restraints

The permitted range is 0.0 -> infinity

Related item:

Type: numb

Category: refine