This one weird trick will help you spot clickbait - Jeff Leek & Lucy McGowan

譯者: Lilian Chiu 審譯者: Helen Chang

One simple vitamin can reduce your risk of heart disease.

一種簡單的維他命就可以 減少你得到心臟病的風險。

Eating chocolate reduces stress in students.

學生吃巧克力可以減少壓力。

New drug prolongs lives of patients with rare disease.

新藥品延長罕見疾病病人的壽命。

Health headlines like these are published every day,

每天都可以看到 像這類的健康標題被刊出，

sometimes making opposite claims from each other.

有時，這些標題的主張還彼此相反。

There can be a disconnect between broad,

這些廣泛、吸引注意力的標題

attention-grabbing headlines and the often specific,

和它們所報導的那些通常很明確、

incremental results of the medical research they cover.

漸進式的醫療研究結果 之間會有斷層。

So how can you avoid being misled by grabby headlines?

所以，我們要如何避免 被聳動的標題誤導？

The best way to assess a headline’s credibility

評估標題可信性的最佳辦法

is to look at the original research it reports on.

就是去看它報導的原始研究。

We’ve come up with a hypothetical research scenario

我們針對前述這三則標題分別

for each of these three headlines.

提出了一個假設性的研究情境。

Keep watching for the explanation of the first example;

繼續看第一個例子的說明；

then pause at the headline to answer the question.

接著在標題出現後 按暫停，回答問題。

These are simplified scenarios.

這些都是簡化的情境。

A real study would detail many more factors and how it accounted for them,

真實的研究會詳細說明更多因子 以及它們與研究的關係，

but for the purposes of this exercise,

但為了做這項練習，

assume all the information you need is included.

就假設所需的資訊已全被納入。

Let’s start by considering the cardiovascular effects

咱們先來思考一下 一種維他命 Healthium

of a certain vitamin, Healthium.

對心血管的影響。

The study finds that participants taking Healthium

該研究發現吃 Healthium 的受試者的

had a higher level of healthy cholesterol than those taking a placebo.

高密度脂蛋白膽固醇（HDL 好膽固醇）

比吃安慰劑的要來的高。

Their levels became similar to those of people with naturally high levels

前者的膽固醇值很接近那些

天生就具有高水平好膽固醇的人。

of this kind of cholesterol.

Previous research has shown that people with naturally high levels

先前的研究顯示

天生就具有高水平好膽固醇的人

of healthy cholesterol have lower rates of heart disease.

比較不會罹患心臟疾病。

So what makes this headline misleading:

那麼，這標題是怎麼誤導的：

"Healthium reduces risk of heart disease."

「Healthium 降低心臟疾病的風險」？

The problem with this headline is that the research didn’t actually investigate

這個標題的問題是， 研究並沒有真正去調查

whether Healthium reduces heart disease.

Healthium 是否會減少心臟疾病。

It only measured Healthium’s impact

它只是測量了 Healthium

on levels of a particular kind of cholesterol.

對於特定種類膽固醇水平的影響。

The fact that people with naturally high levels of that cholesterol

那種膽固醇天生就很高的人

have lower risk of heart attacks

比較少有心臟病，

doesn’t mean that the same will be true of people

這並不一定表示

用 Healthium 來提高 膽固醇水平的人亦會如此。

who elevate their cholesterol levels using Healthium.

Now that you’ve cracked the case of Healthium,

現在，你破解了 Healthium 的案例，

try your hand at a particularly alluring mystery:

試試看這個特別誘人的謎：

the relationship between eating chocolate and stress.

吃巧克力和壓力之間的關係。

This hypothetical study recruits ten students.

這項假設性研究招募了十名學生。

Half begin consuming a daily dose of chocolate,

其中一半學生開始 每天吃一定劑量的巧克力，

while half abstain.

另一半則避吃巧克力。

As classmates, they all follow the same schedule.

他們都是同學，所以課表一樣。

By the end of the study, the chocolate eaters are less stressed

在研究尾聲，吃巧克力的人

比不吃巧克力的人壓力小。

than their chocolate-free counterparts.

What’s wrong with this headline:

這個標題有什麼問題：

"Eating chocolate reduces stress in students"

「吃巧克力會減少學生壓力」

It’s a stretch to draw a conclusion about students in general from a sample of ten.

樣本只有十個人，卻把結果套用到 所有學生，這就是過度擴大。

That’s because the fewer participants are in a random sample,

原因是，在隨機樣本中的 受試者人數越少，

the less likely it is that the sample will closely represent

該樣本就越無法代表

the target population as a whole.

整個目標整體。

For example, if the broader population of students is half male and half female,

比如，如果廣大的學生母體中 有一半是男性，一半是女性，

the chance of drawing a sample of 10

抽出十個人做為樣本，

that’s skewed 70% male and 30% is about 12%.

此樣本有 12% 的可能性 會偏向七成男性、三成女性。

In a sample of 100 that would be less than a .0025% chance,

若是一百人的樣本，

此機率則不到 0.0025%，

and for a sample of 1000,

若樣本中有一千人，

the odds are less than 6 x 10^-36.

此機率不到 6 x 10^-36。

Similarly, with fewer participants,

同樣的，當受試的人數很少時，

each individual’s outcome has a larger impact on the overall results—

個別受試者的結果 對整體結果的影響會比較大，

and can therefore skew big-picture trends.

因此可以造成整體趨勢的偏差。

Still, there are a lot of good reasons for scientists to run small studies.

不過，科學家還是有很多 好理由去進行小型研究。

By starting with a small sample,

從小樣本開始，

they can evaluate whether the results are promising enough

他們可以評估結果是否夠理想，

to run a more comprehensive, expensive study.

再來進行更全面、廣泛的研究。

And some research requires very specific participants

有些研究會需要非常特定的受試者，

that may be impossible to recruit in large numbers.

可能無法讓很多人參與。

The key is reproducibility—

關鍵在於可重複性：

if an article draws a conclusion from one small study,

如果文章從一項小型研究得出結論，

that conclusion may be suspect—

那結論可能不可信；

but if it’s based on many studies that have found similar results,

但如果文章的根據是 很多項發現類似結果的研究，

it’s more credible.

就會比較可信。

We’ve still got one more puzzle.

我們還有一個謎要解。

In this scenario, a study tests a new drug for a rare, fatal disease.

這個情境中的研究，是在測試 治療致命罕見疾病的新藥。

In a sample of 2,000 patients,

樣本是兩千名病人，

the ones who start taking the drug upon diagnosis

被診斷出此疾病後 就開始吃這種藥的人，

live longer than those who take the placebo.

比吃安慰劑的人活得更久。

This time, the question is slightly different.

這次，問題有點不同。

What’s one more thing you’d like to know before deciding if the headline,

你還需要知道哪一項資訊， 才能判斷這個標題是否合理：

"New drug prolongs lives of patients with rare disease", is justified?

「新藥延長了罕見疾病患者的生命」？

Before making this call,

在做決定之前，

you’d want to know how much the drug prolonged the patients’ lives.

你需要先知道，這種藥 能延伸病人多長的壽命。

Sometimes, a study can have results that,

有時研究的結果可能在科學上有效，

while scientifically valid, don’t have much bearing on real world outcomes.

但對現實世界的結果影響不大。

For example, one real-life clinical trial of a pancreatic cancer drug

比如，胰藏癌藥物的實際臨床試驗

found an increase in life expectancy— of ten days.

發現可以延長壽命「十天」。

The next time you see a surprising medical headline,

下次看到驚人的醫療相關標題時，

take a look at the science it’s reporting on.

先看一下它所報導的科學。

Even when full papers aren’t available without a fee,

即使需要付費才能 看到完整研究論文，

you can often find summaries of experimental design

你通常仍然能在免費取得的摘要中

and results in freely available abstracts,

找到實驗設計的概要以及結果，

or even within the text of a news article.

或甚至在新聞文章的內文中找到。

It’s exciting to see scientific research covered in the news,

見到新聞報導科學研究讓人振奮，

and important to understand the studies’ findings.

而重要的是要了解研究的發現。