An evolving literature evaluates the inferential and behavioral implications of measurement error (ME) in agricultural data. We synthesize findings on the nature and sources of ME and potential remedies. We provide practical guidance for choosing among alternative approaches for detecting, obviating, or correcting for alternative sources of ME, as these have different behavioral and inferential implications. Some ME biases statistical inference and thus may require econometric correction. Other types of ME may affect (and shed light on) farmers’ decision-making processes even if farmers’ responses are objectively incorrect. Where feasible, collecting both self-reported and objectively measured data for the same variable may enrich understanding of policy-relevant agricultural and behavioral phenomena.

中文翻译：

---

农业数据中测量误差的推断和行为影响


不断发展的文献评估了农业数据中测量误差 （ME） 的推理和行为影响。我们综合了关于 ME 的性质和来源以及潜在补救措施的发现。我们为在检测、消除或纠正 ME 的替代来源的替代方法中进行选择提供了实用指导，因为这些方法具有不同的行为和推理意义。一些 ME 偏倚统计推断，因此可能需要计量经济学校正。其他类型的 ME 可能会影响（并阐明）农民的决策过程，即使农民的反应在客观上是不正确的。在可行的情况下，收集同一变量的自我报告和客观测量数据可能会丰富对与政策相关的农业和行为现象的理解。