Surname frequencies in the 1990 and 2000 US censuses

A posting on Dick Eastman's blog has alerted me to some interesting surname data which can be found on the US census bureau website. The bureau provides lists of the most frequently occurring surnames in the 1990 and 2000 US censuses. A surname must be included at least 100 times to be appear in the list and consequently CRUWYS is not shown. I did however find listings for most of the other variant spellings. I've provided details below of the rankings. The final column for the year 2000 is a count of the occurrences of each surname in the 2000 census.

1990          Ranking   Count
CREWS        1373

CRUSE         3975

CRUISE        7221

CREW          8148

CRUCE        9146

CRUZE       21658

SCREWS    21905

CRUICE     57820

CREWE      73936

2000
CREWS       1404       23167

CRUSE        3874         8422

CREW         7871         3900

CRUISE      9754         3058

CRUCE     14457         1893

CRUZE      20468        1205

SCREWS   24243         969

CREWE     38706         537

CRUS        41060         501

CRUCES   45829         439

CREWSE  49767          396

For the year 2000 information is also provided on ethnic breakdown. As I cannot format tables properly on the blog I've uploaded the data to Google Docs. You can see the file here.

It should be noted that the surname data from the two censuses is not directly comparable. For the 1990 data a sample was taken from 6.3 million census entries, equating to approximately one fortieth of the US population. The report for 2000 uses "name responses from almost 270 million people with valid name information".

I do not include all of these variants spellings in my one-name study because the high frequency of some of these surnames, especially in the US, would make the task unmanageable. A further problem is that some surnames have developed in different ways on both sides of the Atlantic. We have some evidence for instance that in Virginia the surname Crew evolved into Crews. In the UK Crew and Crewe do not appear to be related to any of the Cruse and variant spellings. The situation is also complicated by the different ethnic composition of the American population. In the US Cruze and Crus seem to be variants of the Spanish surname Cruz. In the UK these names are rarely seen and usually appear as variant spellings in parish register entries. There is a still a long way to go before we understand the full picture of our complicated family of surnames, but the DNA project is already proving to be a very useful way of working out which spellings are related and we will learn more as more people with the many variants join the project.

© Debbie Kennett 2010