“Facts are stubborn, but statistics are more pliable.”
- Mark Twain
There is a solution for everything. It is the statistics you have to watch out for.
Ha.
Statistics: The discipline that concerns the collection, organization, analysis, interpretation, and presentation of data.
Note: In an ideal world the above definition would be wonderful. The Operative Word is Ideal. Unfortunately, it is oftentimes not the reality. This week’s Mindful Monday is dedicated to addressing this hiding-in-plain-sight problem. Thus dramatically influencing both business and personal lives collectively.
Law of Large Numbers: In statistics and probabilities it states “As a sample size grows, its mean gets closer to the average of the whole population”. This is due to the sample being more representative of the population (and not the individual) as the sample becomes larger.
Garbage In Garbage Out: The concept that flawed, biased or inaccurate information input produces the same relevant result output, garbage.
Problem: An inability to achieve a pure solution caused by either an accidental or deliberate disregard for the sanctity of the information. Garbage In, Garbage Out.
Solution: An action or process of solving a problem.
-The Back Story-
What is the point? What is the solution? There cannot be a void. In physics, nature abhors a vacuum.
Statistics (and their representative charts, graphs, and such) were created for the purpose of helping humanity, but alas. Through misuse, ill-intent, disregard, and self-servingness the intended solution has now often become the problem.
As humanity progresses and the complexity of technology increases, statistics now are easily and oftentimes a reflection of the human, the individual or the organization, becoming subordinate to the purity of the information.
Facts have progressively become highly susceptible to manipulation and misuse through previously esteemed vehicles to include statistics, graphs, charts, and such, offering the tempting illusion of precision and certainty.
Statistics-Graphs-Charts
Destabilization of Information Sanctity
Problems & Solutions
Relevance
Relevance is often assumed when a statistic or graph is introduced.
Do not assume. Objectively confirm the relevance and sources.
Cumulative Graphing
Cumulative Reporting may often be subject to selected bias. Accurate Data gives way to optics and visual pleasure.
If it appears (looks) too good to be true….
Visual Bias
The brain loves a shortcut, and when given an image to explain numerical data, the visual bias kicks in.
See Above.
Representative Sample
The Law of Large Numbers states “As a sample size grows, its mean gets closer to the average of the whole population."
Determine the size and representativeness of the sample.
Measurement Error
Does a dimpled or hanging chad count?
Assess the method of measurement for accuracy.
Correlations vs Causation
Correlation does not equate to causation. Two data points on a graph or chart, infers a causal relationship, which is not necessarily the case.
Determine whether causation exists.
The Mean and The Maverick
Often, when a statistic calls for an average, it is in actuality reporting a mean (the total of the sample divided by the number sampled). If it is a small enough sample, and there is a large enough maverick (one of the samples is exceedingly large as compared to the others), it throws off the average to the point that it’s really not an average at all.
Determine if a maverick is skewing the data.
“Torture the data and it will confess anything.”
- Ronald Harry Coase - Nobel Prize Laureate, Economics
Comments