Skip to main content

Posts

Showing posts with the label statistics

How to Craft Credible Predictions

The ability to make reliable and insightful predictions is a valuable skill in a world flooded with information. Whether you’re an analyst, a business leader, or simply someone who wants to understand the future, crafting credible predictions is essential. We can explore the key principles to guide you in creating forecasts that are not only informative but also impactful. Focus on Your Expertise: The foundation of a strong forecast lies in an in-depth understanding of your chosen domain. Avoid venturing into areas where you lack expertise. Instead, focus on your strengths and leverage your knowledge to generate predictions that are grounded in reality. Unravelling the Underlying Dynamics: A compelling forecast goes beyond simply stating what might happen. It delves into the “why” behind the prediction. Explain the causal relationships driving the phenomenon you’re forecasting. Use historical examples to illustrate your points and demonstrate how past events inform your current predict...

Go Beyond the Numbers to Unveil the Stories Data Tells

Numbers, by themselves, are inert. They are mere digits, devoid of context or meaning. To truly understand the story they hold, we must go beyond the raw data and enter into the narrative it reveals. This requires a nuanced approach, one that goes beyond simply presenting the figures and instead seeks to illuminate the insights they offer. The first step in interpreting data is to identify the surprises it presents. What findings defy expectations? What appears counter-intuitive? These unexpected insights often hold the key to understanding the deeper implications of the data. For instance, if a study reveals a surprising correlation between two seemingly unrelated factors, it prompts us to question our assumptions and explore the underlying mechanisms at play. When comparing multiple datasets, it is crucial to highlight both similarities and disparities. By drawing attention to the points of convergence and divergence, we gain a richer understanding of the overall picture. For example...

Always cite your statistics

Last month, I was surprised to run across this on a sales page: “With 8 of 10 Canadians doing the majority of their shopping online these days...”   Statistics slipup #1: No source was listed. Since I knew the person behind that page, I emailed her asking for the source of that data. She sent a link. Statistics slipup #2: She'd misconstrued what she read.   The study cited in that link reported that 79 percent of Canadians have shopped online. That is, 8 out of 10 Canadians have at some point bought something online—a far cry from 8 out of 10 doing most shopping online.