摘要：Inference in surveys with nonresponse has been studied extensively in the literature with a focus on the estimation phase. Propensity weighting and calibrated weighting are among the adjustment methods used to reduce the nonresponse bias. The data collection phase has come into focus more recently; the literature on adaptive survey design emphasizes representativeness and degree of balance as desirable properties of the response obtained from a probability sample. We take an integrated view where data collection and estimation are considered together. For a chosen auxiliary vector, we define the concepts incidence and inverse incidence and show their properties and relationship. As we show, incidences are used in balancing the response in data collection; the inverse incidences are important for weighting adjustment in the estimation.
