Counting the Past
Censuses, price series, parish registers and shipping lists let you ask questions no diary can answer, such as how many and how much. They also carry the assumptions of the clerks who filled them in.
What a learner can do afterwards
- Convert a historical wage or price into today's terms and state what the conversion assumes
- Name two questions a serial record can answer that a single document cannot
- Identify one systematic bias in a historical record, such as who was left off a census and why
1 · Read
Lists from the past answer how many and how much. Parish books, tax rolls, ship logs, and census sheets used one form year on year. Sums and drifts across years show what no lone letter proves.
To set an 1850 wage beside one today, pick a wage, find the price mark of both years, and split. Then own the catch: baskets of goods differ, so the fit stays rough.
Each list mirrors its clerk. A parish logged christenings, not births, so unchristened babes slip past. A noon mart poll misses night hands, and a tax roll skips the poor. Name the lost group before you quote the count.
Two lines can climb as one with no tie between them. Bread cost and thefts may both track one lean harvest. Check order, seek a path, and test some third driver before you claim a cause.
Long lists show drifts no diary can, but each count hides its clerk, its gaps, and its baskets, so test ties before you trust them.
More to read
6.4: Sampling Bias
Counting the past starts with lists someone else made: parish registers, tax rolls, shipping logs, and census sheets. Because the same form was filled year after year, you can track totals and trends no single letter could show. The trap is trusting the count too fast. Every list reflects its clerk, who decided the categories and did the chasing. Ask who had reason to hide, and your totals start to speak honestly.
Identifying bias in samples and surveys
Samples go wrong when some people are easier to reach than others. A survey of market stalls at noon misses night workers, just as a poll of landlines once missed the young. The bias is systematic, so bigger samples do not fix it. Historians meet the same flaw in old records. A census that skips migrants or a tax roll that skips the poor is a biased sample of the past. Name the missing group before quoting the number.
1.2 Primary Sources
Each historical document was made for a purpose that shaped what it kept. A parish recorded baptisms, not births, so unbaptized infants vanish from the books. A port log tracked taxed cargo, not smuggled goods. Treat every source as a witness with interests. Ask who made it, for whom, and what they gained by each entry. The answers tell you what the numbers can and cannot prove.
6.6: Causation
Two trends can rise together without one causing the other. Bread prices and thefts might climb in the same hard year because both follow a bad harvest. Blaming the prices for the thefts skips the real driver. Counting well means testing links, not just spotting them. Look for a mechanism, check the order of events, and ask what third factor could move both lines at once.
2 · Watch
Take it off screen
Where it sits
Learn first
Optional connections
This opens up
Where this leads
Jobs that lean on this skill. Follow one to see everything it is built on.
8 questions wait behind this lesson, each with its answer explained. Every answer feeds the sky: stars light as they are learned, and dim when it is time to come back.