Bringing the Power of Statistics into Focus
In 1907, Mark Twain associated statistics with lies. I’m here to debunk Mark Twain’s debunking of statistics. Statistics in the hands of understanding users are not lies but clear truths.
Statistical tools are like X-Ray Machines. First, the X-ray technician needs to correctly capture the X-ray image and then the doctor needs to accurately read the X-ray image. If the X-ray images are of poor quality or the doctor doesn’t know how to read the X-ray, then there can be a misdiagnosis. But a misdiagnosis doesn’t mean that the X-ray machine “lied". Similarly, statistics require knowledge and understanding in order to garner a clear picture, but statistical tools never "lie" to us.
You see a good statistician needs the knowledge to capture accurate data, and you, the analyst, needs to understand how to interpret statistics and that is what this article is all about.
FIVE FACTORS FOR A DEEPER UNDERSTANDING OF STATISTICS:
- STATISTICAL DATA IS COLLECTED FROM SAMPLES: Statistical data is collected from a sample and a sampling only provides an image of the total population. Just like X-ray images can vary slightly image-to-image, it doesn’t mean that the bone changed. Just the images are different from each other. In the same way, statistical sample data can be slightly different each time it is measured and calculated. This changing data happens because statistics are derived from only a limited number of samples. Every sample set yields slightly different results.
- UNDERSTAND THE POPULATION THAT IS BEING MEASURED: When looking at statistics, one needs to understand what the statistician is measuring. For example, if you are looking at a statistic that characterizes the bounce height of tennis balls, you will need to know more about the tennis balls that were measured during the study. Were they practice balls or tournament balls? Were the balls all made at the same factory? Over what period of production were they made? And do the tennis ball samples represent the same models and raw materials? When looking at statistics, you need to know what exactly was sampled.
- HOW THE SAMPLES ARE COLLECTED MATTERS: In imaging, lighting matters. Shadows can hide a lot of useful information in an image, but good lighting provides an accurate depiction of the subject. Similarly, in statistics, it is important to collect samples correctly. Poor sampling methods create dark spots in the data. But samples that are collected using broad and random methods provide the most accuracy. If we look again at the tennis ball bounce analysis; were the samples collected randomly from inventory? How were they transported and stored? Are the tennis balls all of about the same age? Sample collection methods and handling are important aspects of statistical data integrity.
- HOW SAMPLES ARE MEASURED MATTERS: Measuring samples is like taking a picture. If the camera moves during the process, then the image will be blurry. Measurement represents a data snapshot. Consider the accuracy of the tennis ball bounce measurements. You may want to know some things about the measurement process. For example; was a Standard Test Method used for the testing? Was the testing room temperature controlled? How long were the tennis balls sampled out of the vacuum packaging before testing? And did they measure the bounce to an exact fraction? Consistent and accurate measurement makes for consistent and accurate statistical results.
- SAMPLE SIZE MATTERS: I liken pixel count in imaging to sample size in statistics. The more pixels that an image has, the clearer the pictures and the more samples collected in a statistical study, the more accurate the data. According to the Central Limit Theorem, the higher the sample size, the lower the standard error will be. Sports stats are typically 100% accurate because 100% of the data is measured. But in scientific statistics, there is more error because only a small percentage of a population can actually be measured. So, the more data, the more accurate the statistics.
Statistics don’t lie. Statistics simply supply images and image clarity depends on just a few key factors. Understanding these factors enlightens us and brings us closer to the truth.
Stay connected with news and updates!
If you want some weekly T4T wisdom coming straight to your inbox for your reading pleasure - look no further! Join our mailing list to receive the latest blogs and updates.
Don't worry, your information will not be shared.
We hate SPAM. We will never sell your information, for any reason.