The 1,000 commonest forenames in the Philippines, covering 74,581,757 people. The file states no collection date — but multiplying its two count columns together recovers a population of 106,009,905, which dates it itself.
Mary is carried by 2,229,082 people — one Filipino in 48, and 2.1% of the whole country. The second commonest name, Maria, has 2.55 times fewer.
Cumulative share of the covered population as the list lengthens. The curve is steep at the start and long in the tail: the thousandth name, Melecio, still has 16,692 bearers.
Of the 74,581,757 people this list covers.
28,765,796 people — 27.14% of the whole country.
Half the list carries four-fifths of its people. The other half is the tail.
This is the most useful thing in the dataset and it is not one of its columns. Incidence is a count of bearers; frequency is one in N people. Multiply them and you recover the population the file was compiled against — which has to be the same number in every row, and is.
Median across all 1,000 rows.
105,713,344 to 106,995,936, which is the frequency denominator being rounded to a whole number and nothing else.
The Philippine population passed 106 million around then. The file gives no date; this is inferred from it, and a check fails if the two columns ever stop agreeing to within 2%.
People per first letter, not names per first letter — the two give different orders. 3 letters of the alphabet begin no name in this list.
11,628,779 people. Reynaldo, Ronaldo, Rosario, Roberto, Rowena and so on.
Mary and Maria between them carry more than any other pair on the list.
Not second, as this page previously said. The top four initials cover 55.57% of everyone in the list.
The list holds more female names than male ones and fewer female bearers, which means each male name is doing more work. It says nothing about the sex ratio of the country — it is about how widely names are shared.
Carrying 35,729,816 people, or 68,711 each.
Carrying 38,851,941 people, or 80,942 each.
A male name on this list is carried by 1.18 times as many people as a female one.
Each row carries the share of bearers of the stated gender. For most names that is 99% or 100%. For 37 of them it is below 90%, and those are the interesting ones.
Of 1,000, where fewer than 90% of bearers share the recorded gender.
The closest to even in the file: 51% male and 49% not.
Bearers, at rank 707. It may be a real forename, an initial recorded as one, or an artefact of how the source extracted names. It is named here rather than charted as a name.
More is missing here than on most pages in this project, so it is worth being blunt about it.
One snapshot. No name can be shown rising or falling, and nothing here supports a claim about naming fashions. A name common among the living is not a name common among the newborn, and this file cannot separate the two.
National totals only. Nothing distinguishes Ilocos from Mindanao, or a name popular in 1960 from one popular in 2015, though the Spanish-derived names near the top — Maria, Antonio, Reynaldo — are the kind that skew older.
Philippine surnames are a much stranger subject: most were assigned administratively from a catalogue in 1849, which is why surnames cluster geographically in a way forenames do not. None of that is in this file.
The source does not say how incidence was counted or from what. The internal agreement between its two columns is strong evidence they were derived consistently, which is not the same as evidence that they are right.
One small file, and one cross-check it makes possible.
Incidence times frequency must recover the same population in every row. A check fails if any row falls more than 2% from the median, which allows for the denominator being a rounded whole number and nothing more. All 1,000 rows pass, from 105,713,344 to 106,995,936.
A list ordered by rank must have incidence falling as rank rises. A name further down with more bearers would mean one of the two columns had been misread, and it is the kind of fault that produces a chart which still looks reasonable.
Ninety per cent, written once in the fetcher, used for the flag and for the count the page quotes. A check asserts the flagged list and the threshold have not drifted apart.
The page says the file dates to roughly 2019-2020 because the population it implies is 106,009,905. It does not say the file is from 2019, because the file does not say that.
The list covers 70.35% of the implied population. Shares are given either of the covered people or of the implied total, and each is labelled, because those two denominators differ by almost a third.
Every figure on this page traces to one of these, through a CSV in
data/ph-names/. Each is checked against its source query on every
build.
The 1,000 most common Philippine forenames with the number of bearers, a 1-in-N frequency, and the gender split for each — One snapshot with no stated collection date, no time dimension and no regional split. Multiplying incidence by frequency implies a population base of about 106.0 million, which dates it to roughly 2019-2020 — an inference from the data, not a claim by the source.
I wrote a companion post — “One In 48 Is Called Mary, And The File Told Me Its Own Age” — in simple, everyday words.