The other day I was having an interesting conversation with a successful entrepreneur friend of mine and we were discussing how hard it is to get a startup exit let alone more than one. I have had three in 35 years. One within my first 10 years, and then another one not until 23 years later. This friend of mine has had a few exits as well and at one point he joked and said “I guess we are 6-Sigma’s”. We both laughed but that joke had me thinking all night. So when I couldn’t sleep I pulled my lovely R and Rstudio out and tried to figure out what Sigma am I. First we should start with what a Sigma is. It’s the eighteenth letter in the Greek alphabet, but for the purposes of this discussion it is used in statistics as a representation of Standard Deviation (SD). A “1-Sigma” is one standard deviation from the norm (norm is also called mean or average). When plotted against a normal distribution, a “bell-shaped curve” (see Figure 2), or on a control chart (see Figure 1), it shows how much variance a particular data item is from the average. Figure 1 – From Wikipedia on Walter Shewhart The 68-95-99.7 rule is another way of looking at variation. In a normal distribution, it is postulated that things that are true 68% of the time are considered 1-Sigma events. Things that are true 95% of the time are considered 2-Sigma events and the three-Sigma rule implies that heuristically nearly all values lie within three standard deviations of the mean (3-Sigma). Figure 2 – From Wikipedia on 68-95-99.7 Rule You might have also heard the term Six Sigma. Six Sigma, by pure definition, is just 6 standard deviations from the norm. However, companies like Motorola (1986) and General Electric (1995) used the term as a quality initiative. The initiative is widely known through it’s banner slogan “Defects Per Million Opportunities (DPMO)”. The Six Sigma quality initiative has a goal of not producing more than 3.4 defects per million opportunities. In common IT parlance, you could loosely consider this an service level of 99.9966. In reality this number is actually closer to a 4.5-Sigma, but Motorola determined that due to long term drift, they need to add something called the 1.5 Sigma Shift. The bottom line is that Sigma is used often as a term to differentiate data points. Next time your friendly aunt tells you that you are a 1 in a million, you can thank her for putting you in an group with 6,999 other people (based on a world population of 7 billion). This group would collectedly be known as, at least from your aunt’s perspective, a group of 5-Sigma’s. So back to my original question, my aunt’s opinion aside, “What Sigma am I?”. In order to figure this out, I first needed to figure out what other people’s Sigma is (see Figure 3). Figure 3 – My Table I created my table chart below based on a chart I used from Wikipedia (see Figure 4) and a constant value of 7 billion of the world population. I cheated and took the frequency ratios and used them in the following R calculations to estimate total population and percentages. x = 7000000000 # World Population y = c(3,7,22,81,370,2149,15787,1744278,26330254,506797346)format(x/y, scientific=FALSE)# Get the number per 7 billion print(100 * (x-(x/y)) / ((x-(x/y)) + (x/y)),digits=12) Figure 4 – From Wikipedia on 68-95-97.5 Rule Let’s look at some of my data points here…
So this leaves me with my original question. If my dearly departed Aunt Genie were still alive and if she had studied statistics, she would probably call me a 3-Sigma. On a good day my wife might consider me 6-Sigma and lower than a 2 on my bad days. My 16 year old son used to think I was a 6-Sigma, but now he is too cool so I’m probably just a solid 3 from him. And God bless all 12 year old kids because mine thinks (at least for a couple more years) that I’m a 10-Sigma. The reason I know this is because every night he says to me “Daddy I love you infinity”.
- About The Authors
John WillisJohn Willis has worked in the IT management industry for more than 35 years. Currently he is an Evangelist at Docker Inc. Prior to Docker Willis was the VP of Solutions for Socketplane (sold to Docker) and Enstratius (sold to Dell). Prior to to Socketplane and Enstratius Willis was the VP of Training & Services at Opscode where he formalized the training, evangelism, and professional services functions at the firm. Willis also founded Gulf Breeze Software, an award winning IBM business partner, which specializes in deploying Tivoli technology for the enterprise. Willis has authored six IBM Redbooks for IBM on enterprise systems management and was the founder and chief architect at Chain Bridge Systems. Follow John on Social Media Jump to SectionHow do you calculate sigma from standard deviation?How Does Standard Deviation Relate to Six Sigma?. Calculate the mean of the data set (x-bar or 1. ... . Subtract the mean from each value in the data set.. Square the differences found in step 2.. Add up the squared differences found in step 3.. What is the standard deviation for Six Sigma?Six Sigma does not refer to a process operating within 6 or 4.5 standard deviations of the desired state. To achieve Six Sigma, a process must not produce more than 3.4 defects per million opportunities, where a defect is defined as anything outside of the customer specification.
What is the value when it is 2 standard deviations above the mean in a normal probability distribution?The empirical rule, or the 68-95-99.7 rule, tells you where most of the values lie in a normal distribution: Around 68% of values are within 1 standard deviation of the mean. Around 95% of values are within 2 standard deviations of the mean. Around 99.7% of values are within 3 standard deviations of the mean.
Is standard deviation 1One standard deviation, or one sigma, plotted above or below the average value on that normal distribution curve, would define a region that includes 68 percent of all the data points. Two sigmas above or below would include about 95 percent of the data, and three sigmas would include 99.7 percent.
|