I was watching this compilated YouTube video about work ethic, that some scientists (including Feynman) were speaking. There was this guy saying something along, “I think to have great outcomes, you actually do need to work at like three standard deviation above the mean ”
I only remember basic statistics and simple calculations of mean and standard deviation.
Mean is an average of the dataset.
Standard Deviation is the representation of how close or far are the Datasets from mean.
But I still cannot figure out what he meant.
Note: excuse my English.
In: Mathematics
Have you heard of the Bell Curve or the [Normal Distribution](https://en.wikipedia.org/wiki/Normal_distribution#Standard_deviation_and_coverage)?
3 Standard deviations above the mean, means in the top 0.27% of your sample.
Each standard deviation moves you a certain amount away from the mean. One standard deviation above means about 68% of the sample is lower than that value. Two SD means about 95% are below that value. Three SD means about 99.7% are below that value.
This person was saying that you must do work better than 99.7% of other work in that field in order for it to be truly *great* work.
3 SD above the mean would be the top 0.15% of results. Note, that is not 15%, that’s 0.15%
The mean is simply the average. Most data comes in what is known as a guassian curve, bell curve, or normal distribution. It’s a big hump centered on the mean.
Within 1 SD of the mean (above or below), you would expect to find 68% of all results. Within 2 SD of the mean, you should expect to find 95% of all results. Within 3 SD of the mean, you should expect to find 99.7% of all results. This means only 0.3% is more than 3 SD away from the mean. Half of that would be what we expect to be 3 SD above the mean.
Distributions have a characteristic of width. It is unhelpful to use total range as that can be non-representative of the distribution. Instead there is some represetative value of a distribution called deviation. There are an infinite ways to calculate such a value so someone invented a standard way to calculate a deviation. Thus was invented a calculation to produce a characteristic value called standard deviation.
When someone says a value is three sigma above the mean then the value is mean plus sigma three times. Often the distribution is Gaussian or close to it so that description has immediate meaning. Oh that’s more than 98% or the rest or similar.
To be described personally as three sigma above average it’s akin to say you must be a 1 in a thousand talent.
Honestly, from the context you’ve given, it’s along the lines of “You need to give 110% effort” and other such motivational nonsense.
To have great outcomes, you need to put in more work than 99% of everyone else. Something like that.
Latest Answers