BLOG

Every dataset asks a simple question: will you let one number move the story? The mean listens to everyone… including the loudest guest. The median keeps its balance even when someone shouts. Neither is “better.” They’re different kinds of fairness. If you’re setting policy where a few extreme values would distort resources, lean median. If […]

1) Sort, Then Chart  A sorted table makes the chart inevitable. If your bars are jumbled, your point is too. Sort by the message you want to deliver and the design mostly designs itself. 2) Baseline on Purpose  Choose the baseline to tell the story… just be honest about it. Baseline choice is narration; clear disclosure keeps

A scatterplot is where relationships come to tell the truth. Start with four questions: direction (up, down, none), form (straight, curved), strength (tight, loose), and outliers (the rebels). Don’t rush to a trendline; earn it by describing the shape first. If the dots stack, add a touch of jitter or adjust transparency. Scale both axes

A histogram is a rumor about shape… and the bin width writes the rumor. Too wide and you smooth away the plot twist. Too narrow and everything is noise. Try three bin widths and pick the one that matches the decision you need to make. Name your unit (seconds, points, dollars) and label axes like

Before the chart, make the table… because a good frequency table is an X-ray. It shows the bones: categories, counts, percents, and a total that actually totals. Sort once for the “what’s most common?” story; sort again for “what should we watch?” Add a percent column to unmask unequal group sizes. If you’re tracking a

Pie charts are honest only when the parts make a whole and the slices are few. If you have more than five categories, you don’t have a pie; you have a pizza nobody ordered. Use a pie when the message is share of a fixed total (“Which bucket dominates?”). Order slices by size and label

A bar chart is a decision in costume. You chose categories, counts (or percents), an order, and a baseline. Each choice nudges the reader toward action. Start by sorting the bars to make the point obvious. If the labels are long, rotate the chart and go horizontal… readability beats tradition. Use percent when group sizes

1) Toy Numbers First  Before trusting a formula, feed it toy numbers that make the answer obvious (all 10s, or a single 1 among 0s). If the spreadsheet surprises you there, it’ll betray you later. 2) Sum or Average? Pick the Story  Sum tells you the pile. Average tells you the pace. If you want effort, sum.

Scope of inference has two doors. Random selection answers “Who can we speak about?”… it buys you generalization to the larger population. Random assignment answers “Why did it happen?”… it buys you causal language inside your study. Different keys, different locks. If you have neither, be humble. If you have one, say which door it

“Random sample” sounds like a noun, but it’s a choice. Convenience saves time… and buys bias. Stratified sampling protects the small groups you can’t afford to miss. Cluster sampling trades precision for practicality when travel (or time) is the bottleneck. None of these are magic; all of them are honest when you name them. The