What does it mean when I observe values that contradict reported statistics?

When observed values differ from a reported statistic in a statistical report, it usually means there’s a discrepancy between what you’re seeing in real life or in a sample of data and what the report claims. This can happen for several reasons, and it’s crucial to understand these differences when interpreting statistics in everyday life. 

Let’s explore some possible explanations for these differences using real-world examples: 

  1. Sampling error: Sometimes, a statistic is based on a sample of the population rather than the entire population. For example, let’s say a news article reports that 60% of the people in a city prefer chocolate ice cream over vanilla. You might then notice that among your friends, more prefer vanilla. This discrepancy could be due to sampling error, which occurs when the sample isn’t perfectly representative of the whole population.
  2. Measurement error: This occurs when the data collected isn’t accurate, often due to issues with the data collection process. For instance, a blog post might report that a new exercise routine leads to a 10-pound weight loss in one month. However, if the weighing scales used during the study were faulty, the reported statistic would not accurately represent the true effect of the exercise routine.
  3. Outliers: These are extreme values that can heavily influence the reported statistic. Imagine a social media post claiming that the average salary for a particular profession is $80,000 per year. You may observe that most people in that profession earn around $60,000, with only a few earning over $200,000. The presence of these outliers could inflate the reported average salary, making it seem higher than what most people actually earn.
  4. Misinterpretation: Sometimes, discrepancies arise due to misinterpretation of the reported statistic. For example, an advertisement might claim that a product is “99% effective,” but this may refer to its effectiveness under ideal conditions, not in everyday use. In such cases, the difference between the observed values and the reported statistic is due to a misunderstanding of the context in which the statistic was measured.
  5. Outdated data: A reported statistic might be based on outdated data, which may not be relevant anymore. For example, a product rating from two years ago might not accurately reflect the current quality of the product due to improvements or changes in manufacturing. 


When encountering differences between observed values and reported statistics, it’s essential to consider these factors and critically evaluate the information presented. Being aware of potential discrepancies can help you make more informed decisions based on the statistics you encounter in everyday life.