摘要
We examine the conditions under which descriptive inference can be based directly on theobserved distribution in a non-probability sample, under both the super-population and quasirandomisation modelling approaches. Review of existing estimation methods reveals that thetraditional formulation of these conditions may be inadequate due to potential issues of undercoverage or heterogeneous mean beyond the assumed model. We formulate unifying conditions that are applicable to both types of modelling approaches. The difficulties of empiricallyvalidating the required conditions are discussed, as well as valid inference approaches usingsupplementary probability sampling. The key message is that probability sampling may still benecessary in some situations, in order to ensure the validity of descriptive inference, but it can bemuch less resource-demanding given the presence of a big non-probability sample.
