四类方法中，首先是自然观察，是为了避免人在实验室中被观察，会“不自然”：
One way of getting
around these problems is naturalistic observation, in which researchers
watch participants’ behavior in real-world settings without trying to
manipulate their actions (Angrosino, 2007).
这不光是观察人了，在动物学家研究动物习性的时候，也经常采用。
也正是因为“自然”，所以具有较高的“外部信度”，怎么高呢？可以进行一定程度的推广：
the extent to which our findings generalize to real-world settings (Neisser & Hyman, 1999).
这个一定程度，我猜想，应该是基于类似的场景等条件 —— 而不一定是高于实验室实验（实验设计？嗯，应该是）：
Some psychologists contend that naturalistic designs almost always have higher external validity than laboratory experiments, although there’s actually not much research support for this claim (Mook, 1983).
但是，问题在于自然观察具有较低的“内部信度”，就是因果关系的推断：
Still, no research design is perfect, and naturalistic designs have their disadvantages too.
In
particular, they tend to be low in internal validity, that is, the
extent to which we can draw cause-and-effect inferences (Roe & Just,
2009).
为什么这么说？作者提到两点，一个是无法操纵变量，以及人们意识到被观察，行为会改变：
In contrast, in naturalistic designs we have no control over these variables and need to wait for behavior to unfold before our eyes.
Naturalistic
designs can also be problematic if people know they’re being observed
because this awareness can affect their behaviors.
但是，我想，如果采用自然实验，也就是黄金法则的变种，是可以实现因果推断的 —— 作者估计只是忘记了，或者不太熟悉。
相比于自然观察，其对象是“人们”，个案研究，对象就是个体了：
In a case study, researchers examine one person or a small number of people, often over an extended period of time (Davison & Lazarus, 2007).
An investigator could spend 10 or even 20 years studying one person with schizophrenia, carefully documenting her childhood experiences, job performance, family life, friendships, and the ups and downs of her mental problems.
There’s no single “recipe” for a case study.
既然没有统一的“处方”，就可以通过：
- 观察一个人
- 调查问卷，以及
- 重复性质的访谈 等等
在这种情况下：
Case
studies can be helpful in providing existence proofs, which are
demonstrations that a given psychological phenomenon can occur.
证明存在性 —— 虽然这个证明路径，也很有局限性，但是在一些罕见病的研究中，还是很有帮助，毕竟在一定的时间维度上，难以找到很多类似的案例。
此外，个案研究很可能有助于形成关于核心问题的“假设” —— 通过症状，由点及面，然后发现真问题，当然，还是很有局限性，“假设”要小心被验证之后，才能进行大规模推广，要记得：
As we discovered previously in the text,the plural of anecdote isn’t fact(see LO 1.2a).
趣闻轶事的重复加总，不能说是“事实”。
网友评论