What people do with a pay number

75% of visits run more than one calculation, and 71% of those change the income between runs. Among the 798 visits that ended in a different income range from the one they began in, 506 ended higher and 292 ended lower. People are mostly not modelling a pay cut. They are pricing the raise, the offer or the extra hours in front of them.

How this was tested. One visitor trying twenty figures should not count twenty times, so the headline gives each visit a single vote: where it ended against where it began. Counting every individual change instead gives 59% (1,278 up, 903 down). The form opens on a $20 hourly wage, so a visitor who picks a province before typing their own wage records that default first; ignoring each visit’s first change gives 55% (798 up, 640 down). The direction holds under all three counts. Income is stored in seven ranges, so a change that stays inside one range is invisible here.

Hourly workers enter the shift they actually work. Of 3,109 schedules that visitors changed from the form’s default, 13% start before 7 a.m. and 8% start between 6 p.m. and 6 a.m. Statistics Canada reports how many hours people work; we know of no public Canadian dataset of the hour a shift begins.

How this was tested. The form opens on a 9-to-5 with a 30-minute break. 1,699 of the 4,808 shift calculations carry exactly that schedule, and a real nine-to-five cannot be told apart from a schedule nobody touched. Those rows are therefore left out above, which removes genuine nine-to-fives too — so read the figures as the spread of start times among people who adjusted their schedule. Keeping every row instead gives a floor: at least 46% of all shift calculations do not start at 9 a.m., and at least 5% start between 6 p.m. and 6 a.m.

209 of 1,871 visits calculated take-home pay for more than one province. Interprovincial migration appears in official statistics after the move has happened. This is what it looks like while someone is still deciding.

How this was tested. A visit counts when two or more different provinces appear in its calculations. Some of those visitors are curious rather than moving, and some are correcting a wrong first choice, so this is an upper reading of intent to relocate, not a count of movers.

16% of all the calculations were done in Chinese. The calculator offers ten languages, and the rest of the traffic spreads thinly across French, Korean, Spanish, Punjabi, Hindi, Tagalog, Ukrainian and Vietnamese. Working out what a job pays after tax is one of the first things anyone does on arriving in Canada, and a good deal of it is not happening in English or French.

How this was tested. This is the language the page was displayed in: either one the visitor chose from the menu, or the language their phone or computer is set to. Unlike the shift and income figures it involves no form default — a device language is a real setting either way. It is not a measure of what anyone speaks at home or how well they read English; a bilingual person may simply prefer their own language for a page full of numbers. Thinly-served languages will also be understated, because a language nobody knows the calculator offers is a language nobody arrives in.

Of 4,808 schedules entered by hourly workers, 28% cover at least one weekend day. Taken with the start times above, the picture is of a calculator being used mostly by people whose week is not the standard one.

How this was tested. The form opens on Monday to Friday with both weekend days switched off, so a weekend day is only ever present because somebody switched it on. That makes this figure a floor and never an inflated one — the opposite of the trap in finding 2, where the default worked the other way. Anyone who works weekends but left the days alone is counted here as not working them.