Derivation of Dqn for a Diploid Organism
Using the usual relative finesses,  (1 - s) of individuals with the aa genotype survive for every one with the AA, and (1 - s*h) individuals having the Aa genotype survive for each AA individual. In the second case, the factor h is >= 0 and <= 1. This is the dominance factor. If h = 1, the A allele is dominant; if h = 0, the a allele is dominant; and for intermediate values of h, we have incomplete dominance.
 If we have the frequency of the a allele in generation n as qn, the frequency of the a allele in the following generation (qn+1), will be given by:
   qn+1  = 1/2 * (1 - sh) *2 pnqn / [ 1 - 2shpnqn - sqn2 ] + (1 - s) * qn2 / [ 1 - 2shpnqn - sqn2 ]
            = [(1 - sh) *pnqn + (1 - s) * qn2 ] / [ 1 - 2shpnqn - sqn2 ]
             = [(1 - sh) *pnqn + (1 - s) *(1 - pn) * qn ] / [ 1 - 2shpnqn - sqn2 ]              (Replaced qnwith 1 - pn.)
             = [(1 - sh) *pnqn + (1 - s) * qn -  (1 - s) *  pnqn ] / [ 1 - 2shpnqn - sqn2 ]  (Expanded (1 - s) *(1 - pn) * qn ).
             = [(1 - sh - 1 + s) *pnqn + (1 - s)*qn ] / [ 1 - 2shpnqn - sqn2 ]                   (Grouped  pn*qn products. ).
   qn+1   = [(1 - h)*spnqn + (1 - s)*qn ] / [ 1 - 2shpnqn - sqn2 ]                               (Canceled 1 - 1 and factored s.)

Since Dq = qn+1 - q n,
    Dq = [(1 - h)*spnqn + (1 - s)*qn ] / [ 1 - 2shpnqn - sqn2 ]  - q n                                 (Expanded qn+1. )
    Dq = [(1 - h)*spnqn + (1 - s)*qn - qn*(1 - 2shpnqn - sqn2 )] / [ 1 - 2shpnqn - sqn2 ]  (Lowest common factor for q n.)
    Dq = [(1 - h)*spnqn + (1 - s)*qn - qn + 2shqnpnqn + sqnqn2 ] / [ 1 - 2shpnqn - sqn2 ]  (Expanded q n product.)
    Dq = [(1 - h + 2hqn)*spnqn + (1 - s)*qn - qn + sqnqn2 ] / [ 1 - 2shpnqn - sqn2 ]        (Grouped spnq n products.)
    Dq = [(1 - h + 2hqn)*spnqn + qn - sqn - qn + sqnqn2 ] / [ 1 - 2shpnqn - sqn2 ]          (Expanded (1 - s)*q n.)
    Dq = [(1 - h + 2hqn)*spnqn - sqn + sqnqn2 ] / [ 1 - 2shpnqn - sqn2 ]                         (Canceled q n- q n .)
    Dq = [(1 - h + 2hqn)*spnqn - sqn + sqn*(1 - pn)qn ] / [ 1 - 2shpnqn - sqn2 ]            (Substituted (1 - p n ) for q n.)
    Dq = [(1 - h + 2hqn)*spnqn - sqn + sqn*qn- sqn*pnqn ] / [ 1 - 2shpnqn - sqn2 ]       (Expanded sq n*(1 - p n)*q n.)
    Dq = [(1 - h + 2hqn -qn)*spnqn - sqn + sqn*qn] / [ 1 - 2shpnqn - sqn2 ]                  (Grouped spnq n products.)
   Dq = [(1 - h + 2hqn -qn)*spnqn - sqn + s*(1 - pn)*qn] / [ 1 - 2shpnqn - sqn2 ]        (Substituted (1 - p n ) for q n.)
   Dq = [(1 - h + 2hqn -qn)*spnqn - sqn + sqn - s*pnqn] / [ 1 - 2shpnqn - sqn2 ]        (Expanded s*(1 - p n)*q n.)
   Dq = [(1 - h + 2hqn -qn)*spnqn - s*pnqn] / [ 1 - 2shpnqn - sqn2 ]                          (Canceled -sq n + sq n .)
   Dq = [(1 - h + 2hqn -qn - 1)*spnqn] / [ 1 - 2shpnqn - sqn2 ]                                    (Grouped spnq n products.)
   Dq = [(-h + 2hqn -qn)*spnqn] / [ 1 - 2shpnqn - sqn2 ]                                              (Canceled 1 - 1 and factored s.)
  Dq = [(-h + (2h - 1)qn)*spnqn] / [ 1 - 2shpnqn - sqn2 ]                                              (Grouped q n products.)
  Dq = -spnqn*(h + (1 - 2h)qn) / [ 1 - 2shpnqn - sqn2 ]                                              (Rearranged terms and signs.)

If we choose to ignore the denominator (as Haldane did), we have:

  Dq = -spnqn*(h + (1 - 2h)qn)

Summary : The change in q due to 1 generation of selection is called delta q (Dq) is given by the equation:
   Dq = -spnqn*(h + (1 - 2h)qn) / [ 1 - 2shpnqn - sqn2 ]
where s is the selection coefficient, h is the dominance factor, p n is the frequency of the more fit (increasing) allele, and q n is frequency of the less fit (decreasing) allele.