fbpx
Size of letters 1x
Site color
Image
Additionally
Line height
Letter spacing
Font
Embedded items (videos, maps, etc.)
 

Coding as an Important Skill of Critical Thinking

19/ 03/ 2019
  Anna Grygorash. Coach of the EY Business Academy Not everything that counts can be counted, and not everything that can be counted counts. William Bruce Cameron I offer you to do a little test. You should estimate how likely it is that A would be greater than B on one occasion given that A was greater than B on another occasion in four situations. If you say “50 percent”, that means there is no relationship between them. If you say “90 percent,” you are saying that there is an extremely strong relationship between behavior on one occasion and behavior on another. Commit yourself: write down your answer for each of the questions below. If advanced staff Ivanov makes less mistakes while preparing the report for the 2nd quarter than advanced staff Petrov, what is the likelihood that Ivanov will make less mistakes than Petrov in the report for the 3rd quarter? If processing operator Klymenko does more operations per day than processing operator Kozachenko during the month (operations are similar and scope is allocated between employees in a similar way), what is the likelihood that she will do more operations in the next month? If Manager Zaytsev seems friendlier than manager Medveded on the first project meeting, what is the likelihood that he will seem friendlier on the next meeting? If senior Fedorova behaves more honestly than senior Mykolayeva in the first twenty situations in which you observe them (presenting work results at quarterly performance appraisals honestly, not transferring responsibility for job fails to her colleagues, etc.), what is the likelihood that Fedorova will behave more honestly than Mykolayeva in the second twenty situations in which you observe them? “Isn’t it evident?” – you could ask. It is much easier and reliable to make a forecast based on 20 – 30 occasions than based on just one occasion. The only thing we should account for is the difference between reliability of forecasts regarding people traits (like honesty) and abilities (like data analysis for the report). The feature of conclusions from this test is a considerable difference in our guesses regarding abilities and regarding traits. Let us transfer our likelihood percentage estimates into correlation (relationship between values). Percentage Estimate Correlation Examples of correlation 50% 0   70% 0,59 0,5 corresponds to the degree of association between IQ and performance on the average job for Americans 75% 0,71 0,7 corresponds to the association between height and weight 90% 0,95   So, how much is the difference? – You will find the answer in the picture below. It is evident that accuracy of forecasting abilities is much higher than accuracy of forecasting traits. Why are these levels of accuracy so different? It’s all in the coding. For most abilities we know what the units are for measuring behavior and we can actually give them numbers (number of mistakes in the report, number of operations per day). But what are the appropriate units for judging friendliness - smiles per minute, “Good vibes” per social exchange? How can we avoid the mistake evaluating someone’s traits? We will never find exact measuring units. The most effective way is to remind ourselves that a person’s behavior can only be expected to be consistent from one occasion to another if the context is the same. And even then, many observations are necessary to have much confidence in our prediction. What follows from this? -  We should make a difference to what we can code (i.e. transfer into numbers and measure correlation) and what we can not. If we can’t code or assign numbers to the event or behavior in question offhand, try the exercise of attempting to think of a way to code for it. The sheer effort it would take to do this is likely to alert us to the fact that we are susceptible to overestimating consistency of the event or behavior. Source: Richard E. Nisbett. Mandware. Tools for Smart Thinking. Farrar, Straus and Giroux. 2015.

Anna Grygorash

Coach of the EY Business Academy

Not everything that counts can be counted, and not everything that can be counted counts.

William Bruce Cameron

I offer you to do a little test. You should estimate how likely it is that A would be greater than B on one occasion given that A was greater than B on another occasion in four situations.

If you say “50 percent”, that means there is no relationship between them. If you say “90 percent,” you are saying that there is an extremely strong relationship between behavior on one occasion and behavior on another.

Commit yourself: write down your answer for each of the questions below.

  1. If advanced staff Ivanov makes less mistakes while preparing the report for the 2nd quarter than advanced staff Petrov, what is the likelihood that Ivanov will make less mistakes than Petrov in the report for the 3rd quarter?
  2. If processing operator Klymenko does more operations per day than processing operator Kozachenko during the month (operations are similar and scope is allocated between employees in a similar way), what is the likelihood that she will do more operations in the next month?
  3. If Manager Zaytsev seems friendlier than manager Medveded on the first project meeting, what is the likelihood that he will seem friendlier on the next meeting?
  4. If senior Fedorova behaves more honestly than senior Mykolayeva in the first twenty situations in which you observe them (presenting work results at quarterly performance appraisals honestly, not transferring responsibility for job fails to her colleagues, etc.), what is the likelihood that Fedorova will behave more honestly than Mykolayeva in the second twenty situations in which you observe them?

“Isn’t it evident?” – you could ask. It is much easier and reliable to make a forecast based on 20 – 30 occasions than based on just one occasion. The only thing we should account for is the difference between reliability of forecasts regarding people traits (like honesty) and abilities (like data analysis for the report).

The feature of conclusions from this test is a considerable difference in our guesses regarding abilities and regarding traits.

Let us transfer our likelihood percentage estimates into correlation (relationship between values).

Percentage Estimate

Correlation

Examples of correlation

50%

0

 

70%

0,59

0,5 corresponds to the degree of association between IQ and performance on the average job for Americans

75%

0,71

0,7 corresponds to the association between height and weight

90%

0,95

 

So, how much is the difference? – You will find the answer in the picture below.

It is evident that accuracy of forecasting abilities is much higher than accuracy of forecasting traits. Why are these levels of accuracy so different?

It’s all in the coding. For most abilities we know what the units are for measuring behavior and we can actually give them numbers (number of mistakes in the report, number of operations per day). But what are the appropriate units for judging friendliness – smiles per minute, “Good vibes” per social exchange?

How can we avoid the mistake evaluating someone’s traits? We will never find exact measuring units. The most effective way is to remind ourselves that a person’s behavior can only be expected to be consistent from one occasion to another if the context is the same. And even then, many observations are necessary to have much confidence in our prediction.

What follows from this? –  We should make a difference to what we can code (i.e. transfer into numbers and measure correlation) and what we can not. If we can’t code or assign numbers to the event or behavior in question offhand, try the exercise of attempting to think of a way to code for it. The sheer effort it would take to do this is likely to alert us to the fact that we are susceptible to overestimating consistency of the event or behavior.

Source: Richard E. Nisbett. Mandware. Tools for Smart Thinking. Farrar, Straus and Giroux. 2015.

If you have found a spelling error, please, notify us by selecting that text and pressing Ctrl+Enter.

Start
in the Telegram bot
Read articles. Share in social networks

Spelling error report

The following text will be sent to our editors: