HIGHLIGHTS
SUMMARY
STIKOM # of % Records 20 1.15 3 0.17 11 0.63 UINAMPEL_DL # of % Records 332 2.94 71 0.63 8 0.07 Secondly, the sampling method, which required a minimum of three records randomly sampled from each of the lowest hierarchies of collection in the DR, would introduce an unintended bias in DRs that had very ‘skewed` distribution of collections. In DRs with heavily skewed distribution of collection such as these, taking the same number of sample records from each collection could potentially misrepresent the characteristics of the DRs, especially those that . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.