A formula to correct for the effect of errors of measurement on the correlation of initial values with gains. · Study