A Test of Reducing Callbacks and Not-At-Home Bias in Personal Interviews by Weighting At-Home Respondents
检验了商业调查中加权在家受访者以减少回访和纠正不在家偏差的方法，发现加权虽在部分情况下提高准确性，但会增大估计方差，整体效果存疑。
Many commercial survey research organizations weight answers from at-home respondents to reduce callbacks and compensate for not-at-home bias. After identifying conditions necessary for a weighting technique to work, the authors compare weighted and unweighted results from a single call on two criteria: unweighted results from three calls and 1980 census data. Though weighting gives more accurate results than not weighting the one-call responses in eight comparisons, it distorts the data in four comparisons. Any benefits are offset by the increased variance of the estimates due to weighting. Overall, the results raise questions about the effectiveness of the technique for commercial research.