One of the primary troubles is confounders, which can lead researchers to make mistaken assumptions about causes. For example, suppose breakfast skippers have a personality trait that makes them more likely to gain weight than breakfast eaters. If that’s the case, it may look as if skipping breakfast causes weight gain even though the cause is the personality trait.
混淆因素是主要问题之一,它们能让研究人员们对原因做出错误的猜想。举个例,倘若不吃早餐者更容易增重是源于其性格。如果情况便是如此,那么事实也许看起来是不吃早餐导致人增重,即使罪魁祸首是性格特征。
In analyzing the results of observational studies, scientists make statistical adjustments to minimize the potential confounding factors that they can measure — age, alcohol consumption, exercise, employment, and the like. But the adjustments are imprecise, and there is no guarantee that the groups are not different in some other unmeasured way. Because of those weaknesses, many scientists prefer randomized controlled trials, which they often say provides the "gold standard" in evidence.
在分析观测试验的结果时,科学家们会对数据进行调整,力图减少他们所能测量到的潜在混淆因素,像是年龄、酒精消耗量、锻炼、就业率等。但这些调整是不精确的,而且也没人能保证这样分组在其他某种未经测量的角度来看是一致的。因为这些不足,许多科学家们更倾向于做随机控制实验。他们总称这种实验方式能提供黄金标准作为证据。
【囧研究:不吃早饭真的会长胖吗?!】相关文章:
最新
2020-09-15
2020-09-15
2020-09-15
2020-09-15
2020-09-15
2020-09-15