What allowances have been made for expected non-response in the sample design?
(H8) The first phase sample allocations were adjusted to allow for a 15% final non-response rate. Minimum sample sizes of 15 in size 1 strata and 10 in other sample strata were used. Strata were defined to so that the population forms reasonably homogenous groups within them (which supports a respondent mean imputation methodology within the estimation process). The second phase sample is a complete enumeration of the in-scope first phase. Second phase non-respondents, after standard ABS follow-up practices are exhausted, are dealt with through a combination of manual and automatic mean imputation. In what ways does the coverage provided by the frame differ from the target population? (G6) The frame may include respondents who have made an error in reporting internet usage on the ACS form, or who have ceased using the internet in the period between the ACS and answering the IT Supp. Some overcoverage (duplications) and undercoverage can occur due to time lags in receiving information a
