這個 ≠ 那個：英語精讀講義

Have you ever wondered how related the number of people that drown in pools each year 表達脈絡

你有沒有想過，每年在游泳池中溺斃的人數

is to how many Nicolas Cage films are released the same year?

與同一年上映的尼可拉斯·凱吉電影數量究竟有多大關聯？

Turns out there is a 66.6% correlation between the two. 片語搭配

結果發現，兩者之間的相關性高達66.6%。

Or how about the fact that there's a 99.26% correlation between the divorce rate in Maine

再比如，緬因州離婚率與另一項指標的相關性高達99.26%，

and the per capita consumption of margarine, 用詞

這項指標就是每人平均人造奶油消耗量；

or a 99.79% correlation between spending on science, space, and technology,

又比如，科學、太空與科技支出和另一項指標的相關性高達99.79%，

and the number of suicides by hanging, strangulation, and suffocation.

這項指標竟是上吊、勒頸和窒息自殺的人數。

Just because there is a correlation between two variables, 文法

僅僅因為兩個變數之間存在相關性，

doesn't mean that one causes the other.

並不表示其中一個導致了另一個。

This assumption is a logical fallacy, 用詞

這種假設是一種邏輯謬誤，

and yet we're drawn to headlines like 片語搭配

但我們還是會被這樣的標題吸引：

people who have more sex make the most money.

「性生活越頻繁的人，賺的錢越多。」

Unfortunately, this does not always equal that.

遺憾的是，兩者並不總能畫上等號。

Even though it may look like it,

即使看起來可能如此，

cheese consumption probably isn't related to how many people die tangled in their bed sheets.

起司消耗量大概和有多少人被床單纏住而死沒有關係。

But sometimes it isn't so obvious.

但有時，問題並沒有那麼顯而易見。

Numerous studies found that menopausal women taking hormone replacement therapy had a lower than average incidence of heart disease,

多項研究發現，接受荷爾蒙替代療法的更年期女性，心臟病發生率低於平均值，

leading doctors to believe that hormone replacement could protect against heart disease. 文法

這讓醫師認為，荷爾蒙替代療法可以預防心臟病。

However, when women underwent randomized controlled trials, 文化背景

然而，當女性接受隨機對照試驗時，

they found that hormone replacement therapy actually increased the risk of heart disease.

結果卻發現，荷爾蒙替代療法實際上會提高心臟病風險。

When the original data was finally reanalyzed,

等到原始資料終於被重新分析時，

it was found that women who took the therapy were of a higher socioeconomic group 文法

人們發現，接受這種療法的女性屬於社經地位較高的族群，

with a better diet and exercise regime.

飲食與運動習慣也更健康。

This was the real cause behind decreased risk of heart disease.

這才是她們心臟病風險較低的真正原因。

Another case found that those who used night lights as a kid were more likely to develop myopia, 文法

另一項研究發現，小時候使用小夜燈的人更容易罹患近視，

but there's actually a strong link between parental myopia and the development of child myopia, 片語搭配

但其實，父母近視與孩子近視之間存在很強的關聯，

so in reality, myopic parents were simply more likely to leave a light on in their child's bedroom. 文法

所以實際情況只是：近視的父母更可能讓孩子臥室的燈一直亮著。

These are examples of a lurking variable, 用詞

這些都是「潛伏變數」的例子：

where A does not cause B,

並不是A導致B，

but rather C causes them both. 表達脈絡

而是C同時導致了A和B。

It's a little like taking people who have lung cancer and thinking, 表達脈絡

這就像看到一群肺癌患者，然後心想：

"Hey, they're all carrying lighters in their pockets,

「嘿，他們口袋裡都放著打火機，

so lighters must cause cancer." 文法

所以打火機一定會致癌。」

While not realizing that smoking is the confounding variable.

卻沒發現，吸菸才是那個混淆變數。

And scientists work hard in their studies to try and avoid this.

科學家會在研究中努力避免這類問題。

But it gets worse when popular media takes advantage of these potentially coincidental correlations. 講解

但當大眾媒體利用這些可能只是巧合的相關性時，情況就更糟了。

And one study about the chocolate weight loss connection

有一項研究探討巧克力與減重之間的關聯，

was actually designed as a way to expose how science reporting can be sensationalized. 用詞

其實是專門設計來揭露科學報導如何被炒作得過度聳動的。

A science writer with a PhD in microbiology ran a real clinical trial 片語搭配

一位擁有微生物學博士學位的科學撰稿人進行了一項真正的臨床試驗，

where participants were assigned to three groups: 文法

將參與者分成三組：

a low carb diet,

低碳水化合物飲食組，

a low carb diet plus a 1.5 oz chocolate bar,

低碳水化合物飲食加一塊1.5盎司巧克力的組，

and a group that maintained their regular diet.

以及維持平常飲食的組。

At the end of three weeks,

三週後，

the chocolate group did lose the most weight, 文法

巧克力組確實減掉了最多體重，

but the journalists consciously used terrible science.

但記者們刻意採用了極其糟糕的科學方法。

He used 15 participants and measured 18 different measurements including weight loss,

他只找了15名參與者，卻測量了18項不同指標，包括減重幅度、

cholesterol, sleep quality, blood pressure, well-being, etc.

膽固醇、睡眠品質、血壓、身心狀態等等。

And when you use a small group of people and measure a large number of things,

當樣本人數很少，測量的項目卻很多時，

you're pretty much guaranteed to get a statistically significant result, 用詞

幾乎一定會得到一個具有統計顯著性的結果，

which Veritasium has an amazing video on here if you want to check out.

關於這個現象，Veritasium製作了一支很棒的影片，有興趣可以點這裡看看。

The result could have easily been something different, 文法

而這個結果原本完全可能是別的，

such as chocolate correlates to lower blood pressure.

例如巧克力與較低的血壓相關。

If the study had been peer-reviewed by other researchers, 講解

如果這項研究經過其他研究人員的同儕審查，

it would have been called out. 片語動詞

問題早就會被指出來。

So instead, he submitted it to a journal for a fee of 600 Euros, 片語搭配

所以，他乾脆花600歐元把論文投到一份期刊，

making up a fake institution name, the Institute of Diet and Health. 片語動詞

還捏造了一個假機構名稱——「飲食與健康研究所」。

He then sent out a press release to dozens of media publications, 講解

接著，他向數十家媒體發送新聞稿，

and very quickly getting slim by chocolate was front page news. 片語搭配

很快，「吃巧克力變瘦」就成了頭版新聞。

With that in mind, however, 表達脈絡

不過，即便如此，

we can't dismiss correlation entirely. 片語搭配

我們也不能完全否定相關性。

Correlative evidence is an essential part of science.

相關性證據是科學不可或缺的一部分。

Double-blind studies are not always possible or ethical to run, 文化背景

雙盲研究不一定總能進行，也未必符合倫理，

often leaving correlation as the best evidence available.

這往往使相關性成為當下能取得的最佳證據。

When every possible causative relationship is systematically explored,

當所有可能的因果關係都經過系統性探究時，

correlation can be used as a powerful tool 片語搭配

相關性就能成為一項強而有力的工具，

for assessing cause and effect relationships

用來評估因果關係，

and progressing science even further.

並推動科學進一步發展。

Big thanks to Tyler Vigen for providing his charts on these interesting correlations.

非常感謝Tyler Vigen提供這些有趣的相關性圖表。

You can check out his website or his book Spurious Correlations for more peculiar examples.

想看更多奇特的例子，可以造訪他的網站或閱讀他的著作《Spurious Correlations》。

And subscribe for more weekly science videos every Thursday.

別忘了訂閱，每週四收看更多科學影片。