Row percentages use each commute group as the denominator. Among short-trip commuters, 60/150=40% attended. Among long-trip commuters, 40/50=80% attended. Column percentages answer another question. Of all 100 attendees, 40 percent had long commutes. The statement “80 percent of long-trip commuters attended” cannot be reversed into “80 percent of attendees had long trips.”

Before calculating a cell percentage, find which row or column totals 100 percent. Wording gives the clue. “Of long-trip commuters” places that group in the denominator. “Of attendees” places all attendees in the denominator. Row and column percentages can both be correct while answering different questions. A tempting wrong answer often reports the correct number for the wrong denominator.

Counts and rates also answer different questions. City A may record 500 burglaries among one million residents, while City B records 100 among 100,000. City A has the larger count. City B has the higher rate. Divide by the population at risk before comparing probability. A group can have a high rate and a small count when the group itself is small.

Percentage points and percent change are not the same. If attendance rises from 20 percent to 30 percent, the increase is 10 percentage points. Relative to the original 20 percent, it is a 50 percent increase. Subtraction gives the point change. Division by the original value gives relative change. The question’s wording determines which calculation belongs.

Measures of center can conceal a distribution. If nine workers earn 30,000 and one executive earns 1,000,000, the mean is pulled upward by the extreme salary. The median remains 30,000 because it is the middle ordered value. A mean is not false, but it may be a poor description of a highly skewed typical experience.

Graphs require a frame check. A vertical axis that begins at 95 rather than zero can make a move from 98 to 100 look dramatic. Unequal intervals, changed category widths, missing years, and three-dimensional shapes can also distort visual comparison. Read axis labels and scale before judging the size of a change. A truncated axis may be useful for showing a small difference, but it should not be mistaken for a large absolute shift.

Statistical significance, when reported, means a pattern would be unlikely under a stated chance model. It does not show that the effect is large, important, unbiased, or causal. A very large sample can make a tiny difference statistically significant. Interpretation still requires effect size, design quality, and a substantive mechanism.

Before doing arithmetic, say the comparison in words. “Attendance is 12 percentage points higher” describes a change from 40 percent to 52 percent. Calling that a 12 percent increase is wrong because the relative increase is 30 percent of the original 40. A question may include both numbers. Write the old value, the new value, and the denominator before choosing.

Read before calculating. Use this order: title, cases, variables, denominator, arithmetic, claim. Many table errors begin when a reader calculates before deciding what the number is supposed to describe.

Quick review: Name the denominator, distinguish count from rate, distinguish percentage points from percent change, inspect the axis, and keep a descriptive table from becoming a causal claim.

A research claim is a chain. The question defines the relationship, operationalization creates measures, design produces a comparison, ethics governs how people are treated, and validity sets the boundary of the conclusion. Chapter the related chapter uses that chain to study shared meanings, norms, symbols, and cultural difference. The methods in this chapter will help you ask how evidence supports a claim about culture without mistaking one familiar practice for a universal rule.

Run the full audit before practice. State the claim and population in one sentence. Name how each concept became a variable, how the cases entered the study, and what comparison created the result. Then identify the strongest remaining threat and write the narrowest accurate conclusion. This routine forces the entire research chain into view.

Watch the lesson connection

Sociology Research Methods gives you a second explanation of the ideas surrounding this lesson. As you watch, pause when the lesson concept appears and explain how the example fits.

Try the idea yourself

Write one original example, one close nonexample, and one observation that would help you tell them apart. That small exercise turns a definition into a sociological tool you can use in daily life.