第四十三篇 Drudgery to Strategy—A Statistical Metamorphosis
你好，我是小编H。请对以下文章有翻译兴趣的组员留下你的预计完成时间，并发短信息联系小编H，以便小编登记翻译者信息以及文章最终完成时的奖惩工作。感谢支持翻译组！本文由xy_persist翻译 校稿：wangchunchun
Drudgery to Strategy—A Statistical Metamorphosis苦役到战略——统计的质变
A strategy of experimentation can point you toward success一个能给你指出成功之路的实验策略
by Lynne B. Hareand Mark Vandeven
作者： Lynne B. Hare，Mark Vandeven
Think back to your Stats 101 course. You entered the first session laden withapprehension—induced by survivors’ horror stories—and your worst fears wereconfirmed. Early on, the professor said, "OK, boys and girls, today we’regoing to discuss t-tests and confidence intervals." And you sat therethinking you’d rather visit the dentist.
回想一下你上过的统计课程，当你满怀担忧的去上第一节课时——课堂上还有恐怖故事的幸存者——你最恐怖的担忧被证实了。最初，教授说：“好吧，孩子们，我们将要讨论t-验证和置信区间。”你坐在那里会想，这还不如去看牙科医生呢。
Tools andoccasional toy (artificial) problems characterize many introductory statisticscourses. To be sure, the professors are gifted, enthusiastic lecturers; manyhave great humor and human kindness in their veins. But still, the class dealtwith statistical tools suitable for fixed occasions.
工具和时不时采用的手工道具是先很多前介绍的统计过程的特征。实际上，教授们都是天分的富有热情的讲师，他们的血液里也流淌着幽默感和仁慈，但是这门课程需要针对固定的场合选择适合的统计工具来解决问题。
At semester’s end,you knew about a bunch of tools you could resurrect if the right occasion everarose.
It never did.学期结束时，你了解了大量工具的用法，你能够在正确的时机灵活应用，而以前是不行的。
There’s no getting around the fact that statistics is a difficult subject. Thethinking is different from that of many other disciplines: It acknowledgesuncertainty, whereas others profess determinism. It runs counter to everythingwe are taught in algebra: "Solve for x, a fixed but unknownquantity."统计学是一门非常难的学科这是毫无争议的事实，它的思维是不同于其他学科的：统计学承认不确定性，而其他学科则要求确定性。这跟我们代数学到的是完全相悖的：求解x，是一个定值但是未知。
Further, statistical applications require following mathematical formulas ofwhich there are relatively few in political science, English, history orphilosophy. Even pure mathematics claims closure (not always true), which isabsent in statistics. The semester usually ends beforethe professor scratches the surface. And because the first semester lacks ahappy ending, many students are reluctant to proceed to a second.另外，统计应用需要遵循数学模型，这跟政治、语文、历史或哲学是不太相同的，尽管数学宣称已经结束了（但并不是总是如此），这些在统计学中是没有的。在学期结束的时候，教授通常会揭开谜底，但是因为第一个学期的课程缺乏乐趣，大多数学生不愿意再来一次了。
On the rise上升期Over the last twodecades, we’ve noticed statistical stock is rising at some companies. They maybe in the minority, but some top executives have begun to see statistics’strategic value.在过去20年里，我们已经注意到在一些公司里的统计股票在不断上升，他们可能只是少数，但是一些高管已经开始看到统计学的战略价值。
We believe this occurred because those executives witnessed very positive(read: dollars to the bottom line) results, especially when scientists andengineers used statistical methods to guide projects from highly uncertainproject beginnings to solid, successful and sustainable products and processes.To their internal statisticians, executives are saying, "I want more ofthat. Make it happen."我们之所以相信这些情况的发生是因为高管亲自目睹了积极的结果（阅读：美元底线），尤其是当科学家和工程师们应用统计方法将具有高度不确定性的项目转化为完整可信的，成功的合适的产品和流程的时候。高管们会对他们内部的统计学专家说：“我还想看到更多，来吧，让奇迹发生吧。”
"Egad," says the internal statistician. "Now what do I do? Allthat my colleagues learned from their statistics course was not to take anotherone. The only statistical methods they use come at the end of a project, whenthey compare the new prototype against the current product. And now I have tochange the organizational culture so people not only use the methods, but alsouse them upstream in the development process for guidance. I’d rather visit thedentist and have root-canal surgery!"
内部统计专家说：“天哪！现在我该怎么办？我的同事们从他们的统计课程中学到的知识并不能再创造一件这样的事情，他们在项目结束时用到的唯一的统计方法是他们将设计的新样品跟现有的产品进行对比，现在我不得不改变组织文化，这样大家不仅使用统计方法，而且能将统计方法向前延伸，能够用统计方法给研发过程作指导。我宁可去牙科医生那儿进行根管治疗。
Well, this is notan appeal for long lines at dentists’ offices. Positive results can emerge fromsome lessons learned, if only by osmosis, in Stats 101. An appeal to intuitionreduces resistance to the notion that the t-test used to compare a sample meanto some hypothesized value is analogous to signal-to-noise, for example. Build on this to show that a reduction of noise makesit easier yet to hear the signal. Build on it further to show that the test canbe expanded to compare two sample means as in Stats101, chapter 3, but it’sstill signal to noise.
嗯，这并不是说要倡导在牙科门诊排起长队。从统计101学到的一些教训中我们能够得到积极的结果，比如说，用t-检验来对比样本均值和假设值来减少阻抗的想法可以跟信噪比的例子进行类比。在这个基础上表明，噪音的减少会使信号更容易被听到；再进行进一步的测试表明这个检验方法可以扩大到两个样本的对比，就像统计101第三章中所提到的，仍然是信号和噪声的对比。
It is a bit of aleap from there to compare more than two treatment means, but it can be shownthat the guiding principle remains akin to signal to noise. Then, if you wantto compare multiple treatment means, wouldn’t it be wise to economize byrunning half the experimental treatment combinations on one material and theother half on another? Doing that introduces a two-way classificationpainlessly—almost. The discussion with colleagues dredges up their unpleasantmemories, true, but at least they form a base for intuitive appeal.
从这个结果到对比两个以上数据处理的均值是有一个巨大飞跃的，但是能够看出，知道原则仍然是信号和噪声。那么，如果你想对比多重处理的均值，如果把一种材料进行半实验处理组合而另一半与另一种材料对比是不是会更经济更明智呢？这样做一个几乎无痛苦的双向分类介绍，跟同事进行这样的讨论确实会消除他们不愉快的记忆，说真的，至少他们会给他们形成一个直观的吸引。
If you can studymultiple treatments and multiple materials, why not add multiple speeds? Nowyou have a three-way classification. Conceivably, you could add other factors,such as ingredient levels, machines and locations. Whoa! Wait a minute andthose experiments will get big in a hurry. Big experiments cost big bucks, andthey are logistically hard to control. Fu get about it!
如果你能学会进行多重处理机多重材料分析，为什么不加入多重速度？现在你有了三阶分类，你可以很便捷的添加其他因子，比如说配料，设备，地点。噢！等一下，这样的话实验会变得很大，大的实验需要花费很多钱，而且逻辑上来讲很难控制，那么就抛弃它吧！
To the rescue comethe two-level designs and their fractions. Don’t do multiple factor levels. Examineonly two levels per factor, and make those levels the extremes."Absurd," you say," some of those levels could be best!"重新规划2水平的结构设计，不要进行多因子多水平的设计。每个因子只验证两个水平，而且要选择极端的水平进行验证，你会说：“荒谬，一些水平可能是最好的了。”
"Right,"we say in appropriately humble rejoinder, "but we don’t even know whichfactors are important right now. Let’s experiment to find which factors aremost likely to drive success. We can hone in on levels in subsequentexperiments."我们很赞同的说：“对，但是我们现在不知道哪个因子才是最重要的，让我们进行实验来确认哪些因子是最容易成功的，我们会对其展开单独的多水平实验。”
This is the greatleap or paradigm shift—we move from comparing treatment or material A and B toestimating some thing called an "effect." The question has changedfrom "What factor level is best?" to "Which factors are mostimportant?"
这是一个很大的飞跃——我们从对比处理或材料A和B来评估“影响”。问题就从“哪个因子水平是最好的？”变成了“哪些因子是最重要的？”
"You mean I have to do more than one experiment?" you ask. Yes. Thefirst one or two identify which factors or combinations of factors areimportant. The next experiments help identify best levels among the importantfactors. It is a strategy for experimentation, and it has a higher success ratethan competitive approaches to experimentation such as spray and pray, tryeverything, and try your best hunch and hope you get lucky.
你会问：“你的意思是我需要做不只一个实验？”是的，第一个或者第二个实验识别了哪些因子或因子组合式最重要的。下面的适应会帮助你识别重要因子的最佳水平。这就是实验策略，这要比以前的那些诸如喷雾和祈祷，尝试每一种情况，努力去靠运气发现希望的实验有更高的成功率，
We see other advantages to the strategy of experimentation, especially as it isused to drive decision making upstream in the research process. For one, thecompetitive method of giving it your best shot and testing at the end simplytells you if you were successful. If you weren’t, you don’t know why.
我们还能看到实验战略的其他优势，尤其是在研发过程中我们用实验法来推动决策的时候。比如说，在竞赛中如果你获得了成功，只能在比赛结束时简单的给出你最好的尝试和测试结果，但是你没有成功的话，你不会知道是为什么。
The strategy of experimentation puts data behind your directional decisions soyou know, early in the experimental process, what path to take; there are noblind alleys. You gain product and process knowledge along the way.
实验策略在实验过程的初期不会直接给出能够给你想当然进行决策导向的数据，让你知道该选择哪条路径，因此没有盲区，在整个实验过程中你既得到了产品也了解了工艺知识。
The strategy of experimentation also provides trade-offs. If practicalconstraints block the path to using one combination of factors and their levelsfor success, there may be another combination that comes close. The data pointyou in the direction toward success.
实验策略还能帮助你进行权衡。如果成功的路径能够采用数个因子及因子水平的组合，那么就可能还有另外的组合能够更容易获得成功。这些数据会给你指出一条成功之路。
We’ve found that those who use the strategy are successful and they feelliberated. To go back to the old methods would be like having more oralsurgery.
我们发现那些采用这一战略的公司都获得了成功，而且获得了思想解放。我们在回想过去的那些方法更像是多了几个口腔外科。
Drudgery to Strategy—A Statistical Metamorphosis苦役到战略——统计的质变
A strategy of experimentation can point you toward success一个能给你指出成功之路的实验策略
by Lynne B. Hareand Mark Vandeven
作者： Lynne B. Hare，Mark Vandeven
Think back to your Stats 101 course. You entered the first session laden withapprehension—induced by survivors’ horror stories—and your worst fears wereconfirmed. Early on, the professor said, "OK, boys and girls, today we’regoing to discuss t-tests and confidence intervals." And you sat therethinking you’d rather visit the dentist.
回想一下你上过的统计课程，当你满怀担忧的去上第一节课时——课堂上还有恐怖故事的幸存者——你最恐怖的担忧被证实了。最初，教授说：“好吧，孩子们，我们将要讨论t-验证和置信区间。”你坐在那里会想，这还不如去看牙科医生呢。
Tools andoccasional toy (artificial) problems characterize many introductory statisticscourses. To be sure, the professors are gifted, enthusiastic lecturers; manyhave great humor and human kindness in their veins. But still, the class dealtwith statistical tools suitable for fixed occasions.
工具和时不时采用的手工道具是先很多前介绍的统计过程的特征。实际上，教授们都是天分的富有热情的讲师，他们的血液里也流淌着幽默感和仁慈，但是这门课程需要针对固定的场合选择适合的统计工具来解决问题。
At semester’s end,you knew about a bunch of tools you could resurrect if the right occasion everarose.
It never did.学期结束时，你了解了大量工具的用法，你能够在正确的时机灵活应用，而以前是不行的。
There’s no getting around the fact that statistics is a difficult subject. Thethinking is different from that of many other disciplines: It acknowledgesuncertainty, whereas others profess determinism. It runs counter to everythingwe are taught in algebra: "Solve for x, a fixed but unknownquantity."统计学是一门非常难的学科这是毫无争议的事实，它的思维是不同于其他学科的：统计学承认不确定性，而其他学科则要求确定性。这跟我们代数学到的是完全相悖的：求解x，是一个定值但是未知。
Further, statistical applications require following mathematical formulas ofwhich there are relatively few in political science, English, history orphilosophy. Even pure mathematics claims closure (not always true), which isabsent in statistics. The semester usually ends beforethe professor scratches the surface. And because the first semester lacks ahappy ending, many students are reluctant to proceed to a second.另外，统计应用需要遵循数学模型，这跟政治、语文、历史或哲学是不太相同的，尽管数学宣称已经结束了（但并不是总是如此），这些在统计学中是没有的。在学期结束的时候，教授通常会揭开谜底，但是因为第一个学期的课程缺乏乐趣，大多数学生不愿意再来一次了。
On the rise上升期Over the last twodecades, we’ve noticed statistical stock is rising at some companies. They maybe in the minority, but some top executives have begun to see statistics’strategic value.在过去20年里，我们已经注意到在一些公司里的统计股票在不断上升，他们可能只是少数，但是一些高管已经开始看到统计学的战略价值。
We believe this occurred because those executives witnessed very positive(read: dollars to the bottom line) results, especially when scientists andengineers used statistical methods to guide projects from highly uncertainproject beginnings to solid, successful and sustainable products and processes.To their internal statisticians, executives are saying, "I want more ofthat. Make it happen."我们之所以相信这些情况的发生是因为高管亲自目睹了积极的结果（阅读：美元底线），尤其是当科学家和工程师们应用统计方法将具有高度不确定性的项目转化为完整可信的，成功的合适的产品和流程的时候。高管们会对他们内部的统计学专家说：“我还想看到更多，来吧，让奇迹发生吧。”
"Egad," says the internal statistician. "Now what do I do? Allthat my colleagues learned from their statistics course was not to take anotherone. The only statistical methods they use come at the end of a project, whenthey compare the new prototype against the current product. And now I have tochange the organizational culture so people not only use the methods, but alsouse them upstream in the development process for guidance. I’d rather visit thedentist and have root-canal surgery!"
内部统计专家说：“天哪！现在我该怎么办？我的同事们从他们的统计课程中学到的知识并不能再创造一件这样的事情，他们在项目结束时用到的唯一的统计方法是他们将设计的新样品跟现有的产品进行对比，现在我不得不改变组织文化，这样大家不仅使用统计方法，而且能将统计方法向前延伸，能够用统计方法给研发过程作指导。我宁可去牙科医生那儿进行根管治疗。
Well, this is notan appeal for long lines at dentists’ offices. Positive results can emerge fromsome lessons learned, if only by osmosis, in Stats 101. An appeal to intuitionreduces resistance to the notion that the t-test used to compare a sample meanto some hypothesized value is analogous to signal-to-noise, for example. Build on this to show that a reduction of noise makesit easier yet to hear the signal. Build on it further to show that the test canbe expanded to compare two sample means as in Stats101, chapter 3, but it’sstill signal to noise.
嗯，这并不是说要倡导在牙科门诊排起长队。从统计101学到的一些教训中我们能够得到积极的结果，比如说，用t-检验来对比样本均值和假设值来减少阻抗的想法可以跟信噪比的例子进行类比。在这个基础上表明，噪音的减少会使信号更容易被听到；再进行进一步的测试表明这个检验方法可以扩大到两个样本的对比，就像统计101第三章中所提到的，仍然是信号和噪声的对比。
It is a bit of aleap from there to compare more than two treatment means, but it can be shownthat the guiding principle remains akin to signal to noise. Then, if you wantto compare multiple treatment means, wouldn’t it be wise to economize byrunning half the experimental treatment combinations on one material and theother half on another? Doing that introduces a two-way classificationpainlessly—almost. The discussion with colleagues dredges up their unpleasantmemories, true, but at least they form a base for intuitive appeal.
从这个结果到对比两个以上数据处理的均值是有一个巨大飞跃的，但是能够看出，知道原则仍然是信号和噪声。那么，如果你想对比多重处理的均值，如果把一种材料进行半实验处理组合而另一半与另一种材料对比是不是会更经济更明智呢？这样做一个几乎无痛苦的双向分类介绍，跟同事进行这样的讨论确实会消除他们不愉快的记忆，说真的，至少他们会给他们形成一个直观的吸引。
If you can studymultiple treatments and multiple materials, why not add multiple speeds? Nowyou have a three-way classification. Conceivably, you could add other factors,such as ingredient levels, machines and locations. Whoa! Wait a minute andthose experiments will get big in a hurry. Big experiments cost big bucks, andthey are logistically hard to control. Fu get about it!
如果你能学会进行多重处理机多重材料分析，为什么不加入多重速度？现在你有了三阶分类，你可以很便捷的添加其他因子，比如说配料，设备，地点。噢！等一下，这样的话实验会变得很大，大的实验需要花费很多钱，而且逻辑上来讲很难控制，那么就抛弃它吧！
To the rescue comethe two-level designs and their fractions. Don’t do multiple factor levels. Examineonly two levels per factor, and make those levels the extremes."Absurd," you say," some of those levels could be best!"重新规划2水平的结构设计，不要进行多因子多水平的设计。每个因子只验证两个水平，而且要选择极端的水平进行验证，你会说：“荒谬，一些水平可能是最好的了。”
"Right,"we say in appropriately humble rejoinder, "but we don’t even know whichfactors are important right now. Let’s experiment to find which factors aremost likely to drive success. We can hone in on levels in subsequentexperiments."我们很赞同的说：“对，但是我们现在不知道哪个因子才是最重要的，让我们进行实验来确认哪些因子是最容易成功的，我们会对其展开单独的多水平实验。”
This is the greatleap or paradigm shift—we move from comparing treatment or material A and B toestimating some thing called an "effect." The question has changedfrom "What factor level is best?" to "Which factors are mostimportant?"
这是一个很大的飞跃——我们从对比处理或材料A和B来评估“影响”。问题就从“哪个因子水平是最好的？”变成了“哪些因子是最重要的？”
"You mean I have to do more than one experiment?" you ask. Yes. Thefirst one or two identify which factors or combinations of factors areimportant. The next experiments help identify best levels among the importantfactors. It is a strategy for experimentation, and it has a higher success ratethan competitive approaches to experimentation such as spray and pray, tryeverything, and try your best hunch and hope you get lucky.
你会问：“你的意思是我需要做不只一个实验？”是的，第一个或者第二个实验识别了哪些因子或因子组合式最重要的。下面的适应会帮助你识别重要因子的最佳水平。这就是实验策略，这要比以前的那些诸如喷雾和祈祷，尝试每一种情况，努力去靠运气发现希望的实验有更高的成功率，
We see other advantages to the strategy of experimentation, especially as it isused to drive decision making upstream in the research process. For one, thecompetitive method of giving it your best shot and testing at the end simplytells you if you were successful. If you weren’t, you don’t know why.
我们还能看到实验战略的其他优势，尤其是在研发过程中我们用实验法来推动决策的时候。比如说，在竞赛中如果你获得了成功，只能在比赛结束时简单的给出你最好的尝试和测试结果，但是你没有成功的话，你不会知道是为什么。
The strategy of experimentation puts data behind your directional decisions soyou know, early in the experimental process, what path to take; there are noblind alleys. You gain product and process knowledge along the way.
实验策略在实验过程的初期不会直接给出能够给你想当然进行决策导向的数据，让你知道该选择哪条路径，因此没有盲区，在整个实验过程中你既得到了产品也了解了工艺知识。
The strategy of experimentation also provides trade-offs. If practicalconstraints block the path to using one combination of factors and their levelsfor success, there may be another combination that comes close. The data pointyou in the direction toward success.
实验策略还能帮助你进行权衡。如果成功的路径能够采用数个因子及因子水平的组合，那么就可能还有另外的组合能够更容易获得成功。这些数据会给你指出一条成功之路。
We’ve found that those who use the strategy are successful and they feelliberated. To go back to the old methods would be like having more oralsurgery.
我们发现那些采用这一战略的公司都获得了成功，而且获得了思想解放。我们在回想过去的那些方法更像是多了几个口腔外科。
没有找到相关结果
已邀请:
RSS Feed
0 个回复