# Web Survey Bibliography

Collecting information from sampled units over the Internet or by mail is much more cost‒efficient than conducting interviews. These methods make self‒enumeration an attractive data‒collection method for surveys and censuses. Despite the benefits associated with self‒enumeration data collection—in particular Internet-based data collection—self‒enumerationcan produce low response rates compared to interviews. To increase response rates, non‒respondents are subject to a mixedmode of follow‒up treatments, which influence the resulting probability of response, to encourage them to participate. Because response occurrence is intrinsically conditional, we preliminary record response occurrence in discrete intervals, and we then characterize the probability of response by a discrete time hazard. This approach facilitates examining when a response is most likely to occur and how the probability of responding varies over both time and follow‒up treatments. Weuse regression analysis to investigate the effect of mixed‒mode on the response probability. Factors and interactions arecommonly treated in regression analyses, and have important implications for the interpretation of statistical models. The nonresponse bias can be avoided by multiplying the sampling weight of respondents by the inverse of an estimate of the response probability. Estimators and associated variance estimators of model parameters as well as of parameters of interest are studied. We take into account correlation over time for the same unit in variance estimation. The problem of optimal resources allocation within stages of the survey design is also investigated.Collecting information from sampled units over the Internet or by mail is much more cost‒efficient than conducting interviews. These methods make self‒enumeration an attractive data‒collection method for surveys and censuses. Despite the benefits associated with self‒enumeration data collection—in particular Internet-based data collection self‒enumerationcan produce low response rates compared to interviews. To increase response rates, non‒respondents are subject to a mixedmode of follow‒up treatments, which influence the resulting probability of response, to encourage them to participate. Because response occurrenceis intrinsically conditional, we preliminary record response occurrence in discrete intervals, and we then characterize the probability of response by a discrete time hazard. This approach facilitates examining when a response is most likely to occur and how the probability of responding varies over both time and follow‒up treatments. Weuse regression analysis to investigate the effect of mixed‒mode on the response probability. Factors and interactions are commonly treated in regression analyses, and have important implications for the interpretation of statistical models. The nonresponse bias can be avoided by multiplying the sampling weight of respondents by the inverse of an estimate of the response probability. Estimators and associated variance estimators of model parameters as well as of parameters of interest are studied. We take into account correlation over time for the same unit in variance estimation. The problem of optimal resources allocation within stages of the survey design is also investigated.