All names are classified by a unique rank based on popularity and then by alphabetical order. The most popular name is given the 1st rank, where popularity is determined by the number of occurrences of the name in the MyNameStats.com database. The rank is determined alphabetically for names with the same number of occurrences, from A (1st rank) to Z. No two names will have the same rank.
This shows the percentage of names that are less popular than the specified name. For example the last name Smith is the most popular and obviously all the other names are less popular. 100% of the names are less popular thus the percentile rank for Smith is 1.00. Now take a name such as Smiszek which is far less popular than Smith, it has an estimated population of 42, and a percentile rank of 0.60. This can be interpreted as, the name Smiszek is more popular than 60% of all names.
Why use percentile rank instead of a simple rank (ie:
The national rank is interesting when it comes to the most popular names like Smith, Johnson, Wilson and Miller etc...
One can see that Miller is ranked 6 and Wilson is ranked 10. These names will all have a percentile rank of 1,
or very close to 1, so percentile does not tell us much. As names move down in rank,
the national rank become less meaningful. Take a name like Smoot, it is ranked 4,191st nationally and surprisingly,
it also has percentile rank of 1.00 (actually it is slightly less than 1 but due to rounding it displays as 1.00).
So we can conclude that although there is a difference in popularity between Miller and Smoot the
difference is relatively small when compared to the very large number of less popular names (99%).
Another great thing about percentile rank is that it allows us to compare different distributions. For example, compare the popularity of the name Johannes as a first name to this same name as a last name. The national rank for Johannes as first name is 3,469 and as a last name it is 8278th. So it looks like there is a sizable difference. This may lead us to conclude that Johannes is used more as a first name. But national rank is not a good measure as there are far more last names than first names. Let's compare the percentile ranks, as a first name it is 0.99 and as a last name 0.99, thus we conclude that the name is equally popular as a first or last name.
To estimate the population a sample proportion is calculated using data from the MyNameStats.com database and it is multiplied by the US population estimate provided by the US Census Bureau for July 2017. An estimation range is provided for the users to show the degree of certainty of the population estimate. Popular names such as Smith have a relatively high degree of certainty resulting in a narrow range, in the order of less than 1%. Whereas, less common names have a lower degree of certainty and a wider range. The population estimates are calculated using a confidence interval of 95%.
This is the percentage of people with the specific name, times one thousand. Since there are relatively few people that have any one name, the percentage is almost always less than 1%. For example the most popular last name Smith is held by only 0.82% of the population and the name Hall, the 20th most popular name, is held by 0.2% of the population. By multiplying these numbers by 1000, the figures are a little easier to read and compare, Smith's proportion per 100k is 823.9 and Hall's is 214.8.
Additionally proportion per 100k is a standardized measure of popularity that allows users to compare the frequency of occurrence without the bias of population size. This is most relevant when comparing names among states. States with large populations will often have more people with any given name than a smaller state despite the fact that the concentration of people with that name may be higher in a smaller state. Take the name Smith as an example, there are 208k people named Smith in California and only 102k in Pennsylvania. Whereas there are 799.7 Smith's for every 100k Pennsylvanians and only 536.9 Smith's for every 100k Californians.
US Census Bureau Statistics
We have taken data published by the US Census Bureau and we have calculated the population and proportion per 100k statistics for family name.
Every ten year the Census Bureau releases a report showing the occurence of last names collected during the Census. The published data includes a count of only the top decile of family names (that is the top 10%). So the list excludes the less popular names. We show this data here to provide a benchmark to which one can compare statistics from MyNameStats.com's own data.
First vs. Last
A ratio that compares the occurrences of the name used as a first name versus a last name. This metric is calculated from the MyNameStats.com data. The vertical bar indicates the relative value. The further the bar is to the left more it is used as a first name.