We have reanalyzed the existing data on Zinc abundances in damped Ly \alpha ( DLA ) absorbers to investigate whether their mean metallicity evolves with time .
Most models of cosmic chemical evolution predict that the mass-weighted mean interstellar metallicity of galaxies should rise with time from a low value \sim 1 / 30 solar at z \sim 3 to a nearly solar value at z \sim 0 .
However , several previous analyses have suggested that there is little or no evolution in the global metallicity of DLAs .
The main problem is that the effective number of systems that dominate the N ( { H I } ) -weighted mean metallicity is very small .

We have used a variety of statistical techniques to quantify the global metallicity-redshift relation and its uncertainties , taking into account both measurement and sampling errors .
Three new features of our analysis are : ( a ) an unbinned N ( { H I } ) -weighted nonlinear \chi ^ { 2 } fit to an exponential relation ; ( b ) survival analysis to treat the large number of limits in the existing data ; and ( c ) a comparison of the data with several models of cosmic chemical evolution based on an unbinned N ( { H I } ) -weighted \chi ^ { 2 } .
We find that a wider range of evolutionary rates is allowed by the present data than claimed in previous studies .
The slope of the exponential fit to the N ( { H I } ) -weighted mean Zn metallicity vs. redshift relation is -0.20 \pm 0.11 counting limits as detections and -0.27 \pm 0.12 counting limits as zeros .
Similar results are also obtained if the data are binned in redshift , and if survival analysis is used .
These slopes are marginally consistent with no evolution , but are also consistent with the rates predicted by several models of cosmic chemical evolution ( e.g. , slopes of -0.61 to -0.25 for the models of Pei & Fall 1995 , Malaney & Chaboyer 1996 , and Pei et al .
1999 ) .
The \chi ^ { 2 } values obtained for most of these models are somewhat worse than that for the exponential model because the models lie above the observed data points , but still suggest that the present DLA data could indicate some evolution of the metallicity with redshift .
Finally , we outline some future measurements necessary to improve the statistics of the global metallicity-redshift relation .