Past performance is not an indicator of future returns. That, at least, is the advice given to investors. But can the likelihood of a person committing a crime be predicted by looking at his record? The answer, according to a team of clinical forensic psychologists, is that it cannot. Not only is risk prediction unreliable but, when applied to individuals rather than groups, the margins of error are so high as to render any result meaningless.

   Making assumptions about individuals from group data is generally only reasonably safe when the variation within the group is small. Despite this, risk assessments are routinely used to help decide who should be locked up, who should undergo therapy and who should go free. Risk prediction is also set to be used to assess the threat posed by people ranging from terrorist suspects to potential delinquents.

   Stephen Hart, of Simon Fraser University in British Columbia, Canada, and colleagues decided to determine how accurate the tests of risk assessment are when applied to individuals rather than groups. Typically the tests work by assigning a score to people depending on factors such as their age, the history of their relationships, their criminal past and the type of victims they have chosen. If someone's score places him in a group in which a known proportion has gone on to commit a crime on release from detention, then the risk that person will prove a recidivist is thought to be similar to the risk for the group as a whole.

   The paper published by Dr Hart and his colleagues in last month's issue of the British Journal of Psychiatry focused on two popular tests that follow this logic. The first was a 12-item test designed to assess risk for general violence over periods of seven to ten years. The second was a ten-item test designed to assess risk for violence and sexual violence over periods of five to 15 years. The researchers have also assessed other tests used for predicting sexual offences and domestic violence.
 
   They found that variations between members of the groups were very large. In one of the tests, for example, the standard estimate of the chances of members of the group sexually reoffending was put at 36% within 15 years. They calculated that the actual range was between 30% and 43% of the group, with a 95% confidence level. But calculating the average probability for a group is much easier than calculating the same probability for any individual. Thus, using standard methods to move from group inferences to individual ones, they calculated that the chance of any one person reoffending was in the range of 3% to 91%, similarly with a 95% confidence level. Clearly, the seemingly precise initial figure is misleading.。

   The principle is not peculiar to psychology. It has been recognised by statisticians for decades. They call it the ecological fallacy (although this term captures broader subtleties, too). Medicine has also been confounded by statistically based procedures. Indeed, the technique is only really useful when the successes and failures are aggregated. A life-insurance company, for instance, could wrongly predict the life span of every person it insured but still get the correct result for the group.

   注（1）：本文選自Economist, 06/02/2007

   注（2）：本文習(xí)題命題模仿對(duì)象為2004年真題Text 4。

   1． What do the clinical forensic psychologists think of risk prediction?

    [A] Risk prediction fails in the stock market.

    [B] Risk prediction is always effective when being applied to groups.

    [C] Risk prediction is not dependable when it comes to individual behavior.

    [D] Rist prediction should be widely used in all fields.

   2． We can learn from the text that tests of risk assessment are _______.

    [A] longitudinal

    [B] very tricky

    [C] convincing enough

    [D] unreasonable

   3． Traditional view and that of Dr. Hart on risk assessment are _______.

    [A] identical

    [B] similar

    [C] complemetary

    [D] opposite

   4． According to Dr. Hart, using standard methods to predict individuals _______.

    [A] is as easy as using them for groups

    [B] yields ineffective statistics

    [C] can help attain precise results

    [D] might be influcenced by confidence level

   5． Which of the following statements is TRUE?

    [A] Ecological fallacy is a psychological phenomenon.

    [B] Statisticians have been dealing with ecological fallacy for a long time and have almost succeeded in figuring out a solution.

    [C] The statistics from a life insurance company on the longevity of a group of people should be reliable.

    [D] Risk prediction is proved to be of zero practical value.

   篇章剖析

    本文主要就如何科學(xué)進(jìn)行犯罪預(yù)測(cè)展開(kāi)論述。第一段首先提出了風(fēng)險(xiǎn)預(yù)測(cè)能夠有效預(yù)測(cè)犯罪動(dòng)機(jī)的問(wèn)題；第二段對(duì)風(fēng)險(xiǎn)預(yù)測(cè)做出了簡(jiǎn)要說(shuō)明，指出其局限性，且這種方法有些被濫用；第三、四段主要介紹了哈特博士針對(duì)以上問(wèn)題進(jìn)行的實(shí)驗(yàn)；第五段介紹了實(shí)驗(yàn)結(jié)果，說(shuō)明了風(fēng)險(xiǎn)預(yù)測(cè)針對(duì)集體的有效性遠(yuǎn)遠(yuǎn)超過(guò)了針對(duì)個(gè)人的情況；第六段則進(jìn)一步對(duì)風(fēng)險(xiǎn)預(yù)測(cè)進(jìn)行闡發(fā)性論述，提出了生態(tài)學(xué)謬論的概念。

   詞匯注釋

   margin [`mB:dVin] n. 頁(yè)邊的空白，差數(shù)            confidence level 置信度

   assessment [E`sesmEnt] n. 評(píng)估，估價(jià)              probability [9prCbE`biliti] n. 概率

   routine [ru:`ti:n] adj. 例行的；常規(guī)的               inference [`infErEns] n. 推論

   therapy [`WerEpi] n. 治療                         initial [i`niFEl] adj. 比較初的, 初始的

   delinquent [di`liNkwEnt] n. 失職者, 違法者          fallacy [`fAlEsi] n. 謬誤, 謬論

   proportion [prE`pC:FEn] n. 比例, 均衡             confound [kEn`faund] vt. 使混淆, 挫敗

   detention [di`tenFEn] n. 拘留, 禁閉                 aggregate [`Agrigeit] v. 聚集, 合計(jì)

   psychiatry [sai`kaiEtri] n. 精神病學(xué), 精神病治療法    insure [in`FuE] vt. 給...保險(xiǎn)

   domestic [dE`mestik] adj. 家庭的, 國(guó)內(nèi)的

   難句突破

   If someone's score places him in a group in which a known proportion has gone on to commit a crime on release from detention, then the risk that person will prove a recidivist is thought to be similar to the risk for the group as a whole.

   主體句式  If someone's score places him in a group， then the rist is thought to be similar to…

   結(jié)構(gòu)分析  這個(gè)句子分為逗號(hào)前后的兩個(gè)部分，其中的每一個(gè)部分包含一個(gè)定語(yǔ)從句。前半個(gè)句子中，in which引導(dǎo)的定語(yǔ)從句修飾group；而后半句中，that person will prove a recidivist 作為定語(yǔ)從句修飾risk。

   句子譯文  如果一個(gè)人的分?jǐn)?shù)屬于出獄后犯罪率很高的一組，那么此人是慣犯的幾率就大體上和這組的整體幾率相同

   題目分析

   1．C. 細(xì)節(jié)題。根據(jù)文章第一、二段的內(nèi)容，風(fēng)險(xiǎn)預(yù)測(cè)對(duì)于個(gè)人行為的預(yù)測(cè)能力錯(cuò)誤率極大，幾乎完全不可靠。

   2．A. 細(xì)節(jié)題。文章第四段中舉了兩個(gè)測(cè)試的例子，這兩個(gè)測(cè)試都是長(zhǎng)達(dá)多年的跟蹤性試驗(yàn)。

   3．D. 細(xì)節(jié)題。文章第二段中指出，傳統(tǒng)上人們習(xí)慣用風(fēng)險(xiǎn)評(píng)估來(lái)對(duì)個(gè)人的行為作出判斷，而哈特博士的實(shí)驗(yàn)結(jié)果則表明這種預(yù)測(cè)起到很大的誤導(dǎo)作用，可見(jiàn)雙方的意見(jiàn)是相反的。

   4．B. 細(xì)節(jié)題。根據(jù)文章第五段，如果將群體的標(biāo)準(zhǔn)方法用于個(gè)體，比較終得到的結(jié)果是具有誤導(dǎo)性的。

   5．C. 細(xì)節(jié)題。C選項(xiàng)的信息來(lái)自文章第六段的比較后一句話，A life-insurance company, for instance, could wrongly predict the life span of every person it insured but still get the correct result for the group.即“例如一個(gè)人壽保險(xiǎn)公司也許不能預(yù)測(cè)每個(gè)人的壽命長(zhǎng)度，但它可以對(duì)群體做出準(zhǔn)確預(yù)測(cè)�！薄�

   參考譯文

   過(guò)去的付出并不代表未來(lái)的收獲，這至少可以給投資者一些啟示。但一個(gè)人的犯罪動(dòng)機(jī)能通過(guò)他過(guò)去的記錄來(lái)預(yù)測(cè)嗎？一個(gè)臨床法醫(yī)心理學(xué)家小組的回答是否定的。風(fēng)險(xiǎn)預(yù)測(cè)不僅站不住腳，而且即使是只用于個(gè)體而不是群體，其誤差之大也導(dǎo)致結(jié)果失去意義。

   只有當(dāng)一組數(shù)據(jù)內(nèi)部變化程度相對(duì)較小時(shí)，據(jù)此做出的對(duì)個(gè)體的判斷才會(huì)比較可靠。盡管如此，人們還是例行公事地用風(fēng)險(xiǎn)評(píng)估來(lái)幫助決定誰(shuí)應(yīng)該坐牢、誰(shuí)應(yīng)該接受治療以及誰(shuí)可獲得自由。風(fēng)險(xiǎn)預(yù)測(cè)同樣用于區(qū)分評(píng)估從來(lái)自嫌疑恐怖分子到潛在罪犯的各種威脅恐嚇。

   加拿大不列顛哥倫比亞省西蒙弗雷澤大學(xué)的斯蒂芬·哈特和他的同事決意要找出當(dāng)這些測(cè)試用于個(gè)體而非群體時(shí)，其準(zhǔn)確性到底有多大。這些測(cè)試先根據(jù)個(gè)人年齡、社交經(jīng)歷、犯罪史、以及所選擇的受害者類型等因素給人們算出一個(gè)分?jǐn)?shù)。如果一個(gè)人的分?jǐn)?shù)屬于出獄后犯罪率很高的一組，那么此人是慣犯的幾率就大體上和這組的整體幾率相同。

    哈特博士和他的同事在上月的《英國(guó)精神病學(xué)雜志》上發(fā)表了一篇論文，主要介紹了按照這一邏輯進(jìn)行的兩個(gè)有名實(shí)驗(yàn)。第一個(gè)實(shí)驗(yàn)包括12項(xiàng)內(nèi)容的測(cè)試，用來(lái)評(píng)估7至10年間的普通暴力風(fēng)險(xiǎn)；第二個(gè)實(shí)驗(yàn)包括10項(xiàng)相關(guān)內(nèi)容的測(cè)試，用來(lái)評(píng)估5－15年間的暴力和性暴力風(fēng)險(xiǎn)。研究人員也評(píng)估了用來(lái)評(píng)估性侵犯和家庭暴力的其他一些實(shí)驗(yàn)。

   他們發(fā)現(xiàn)每組成員間的差別很大。比如在其中一項(xiàng)測(cè)試中，15年內(nèi)再次進(jìn)行性侵犯可能性標(biāo)準(zhǔn)估計(jì)是36% 在95%的置信度下，他們得到的實(shí)際范圍在30%和43%之間。但計(jì)算群體的平均概率要比計(jì)算個(gè)體的概率容易得多。因此，如果將群體的標(biāo)準(zhǔn)方法用于個(gè)體，他們計(jì)算出一個(gè)人是慣犯的幾率在同樣95%的置信度下為3%－91%。很明顯，看似準(zhǔn)確的原始數(shù)據(jù)卻起到了誤導(dǎo)的作用。

    這個(gè)原理不僅僅用于心理學(xué)，也已統(tǒng)計(jì)學(xué)承認(rèn)達(dá)數(shù)十年之久。他們稱其為生態(tài)學(xué)謬論（盡管這個(gè)術(shù)語(yǔ)還有更廣泛和精細(xì)的意義）。醫(yī)學(xué)也被基于統(tǒng)計(jì)學(xué)的程序弄的一團(tuán)糟。實(shí)際上，這個(gè)技術(shù)只有在成功和失敗都綜合起來(lái)的時(shí)候才真正有用。例如一個(gè)人壽保險(xiǎn)公司也許不能預(yù)測(cè)每個(gè)人的壽命長(zhǎng)度，但它可以對(duì)群體做出準(zhǔn)確預(yù)測(cè)。