In a recent paper we have studied the redshift power spectrum P ^ { S } ( k, \mu ) in three CDM models with the help of high resolution simulations .
Here we apply the method to the largest available redshift survey , the Las Campanas Redshift Survey ( LCRS ) .
The basic model is to express P ^ { S } ( k, \mu ) as a product of three factors

P ^ { S } ( k, \mu ) = P ^ { R } ( k ) ( 1 + \beta \mu ^ { 2 } ) ^ { 2 } D ( k, \mu ) . Here \mu is the cosine of the angle between the wave vector and the line of sight .
The damping function D for the range of scales accessible to an accurate analysis of the LCRS is well approximated by the Lorentz factor

D = [ 1 + { 1 \over 2 } ( k \mu \sigma _ { 12 } ) ^ { 2 } ] ^ { -1 } . We have investigated different values for \beta ( \beta = 0.4 , 0.5 , 0.6 ) , and measured the real space power spectrum P ^ { R } ( k ) and the pairwise velocity dispersion \sigma _ { 12 } ( k ) from P ^ { S } ( k, \mu ) for different values of \mu .
The velocity dispersion \sigma _ { 12 } ( k ) is nearly a constant from k = 0.5 to 3 h { Mpc } ^ { -1 } .
The average value for this range is 510 \pm 70 { { km s ^ { -1 } } } .
The power spectrum P ^ { R } ( k ) decreases with k approximately with k ^ { -1.7 } for k between 0.1 and 4 h { Mpc } ^ { -1 } .
The statistical significance of the results , and the error bars , are found with the help of mock samples constructed from a large set of high resolution simulations .
A flat , low-density ( \Omega _ { 0 } = 0.2 ) CDM model can give a good fit to the data , if a scale-dependent special bias scheme is used which we have called the cluster-under-weighted bias ( Jing et al .
) .