Hello There, Guest!  

UK: immigrant performance by country of origin

In the review of Noah's crime and immigration preferences paper, there was some discussion of whether one can find suitable statistics for doing a regular country of origin study for the UK. It seems that the answer is "yes".

To locate data, I used the typical Google advanced search approach:

Search phrase: income foreigners denmark poland italy iraq somalia site:uk filetype:pdf OR filetype:XLS OR filetype:XLSX

This searchers for content with: income foreigner denmark poland italy iraq somalia
Filtered by filetype: PDF, XLS or XLSX
Filtered by TLD: uk (e.g. co.uk, gov.uk)

One can vary the content terms, but it's a good idea to include a diverse selection of countries as well as a some kind of immigrant related keyword: foreigners, immigrants, migrants

Job seekers, unemployment, benefits claims
There's data for 20 countries of origin here.
  • https://www.migrationwatchuk.org/briefing-paper/367
Data for 206 groups here. These are raw counts, so one has to find suitable population data.
  • https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/514682/nationality-at-point-of-nino-registration-feb-2015.xls
Found via
  • http://researchbriefings.files.parliament.uk/documents/SN06955/SN06955.pdf
This also has child benefit claims.

Closest I found was this file. Has the right table, but it's almost empty.
  • https://cy.ons.gov.uk/file?uri=/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/adhocs/005936selectedsocs113535623563and4138brokendownbyftptageindustryprivatepublicregiongrosspayqualificationnationalityandethnicityjd15/selectedsocs1135356235634138brokendownbyftptageindustryprivatepublixregiongrosspayqualificationnationalityandeithnicityjd15.xls
Arrests by citizenship, 2008 to 2012, 2013-2015:
Note that the original link is now 404, but luckily the IA had a copy. Always save raw files in supplementary materials! Never delete cases from data just because they are not used in a given study.

Prison rates by citizenship:
These numbers cover a lot more countries. See table 1.11 (in the yearly version spreadsheet). There's prisoner rate data for about 200 countries. You just need to pair it with matching population count data for the same periods. The data covers 2002 to 2016, so one can average out a lot of the randomness for the smaller groups.

Nothing found. Looks like one can perhaps get this from the 2011 census.

Population data
Noah's paper has some.
The latest version seems to be here:
  • https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/internationalmigration/datasets/populationoftheunitedkingdombycountryofbirthandnationalityunderlyingdatasheets
I have downloaded a backup of every dataset: 2001-2015.

Broad educational attainment (4 levels) by country of birth, and by citizenship. About 200 countries. By sex. Data from 2011. The same file has atching population data, also broken down by age. Looks like the data is from the 2011 census, so perhaps one can obtain these data or request more computations.
  • https://www.ons.gov.uk/ons/rel/census/2011-census/united-kingdom-submission-for-united-nations-questionnaire-on-population-and-housing-censuses--part-2-/rft-table-1.xls

Cognitive or scholastic
GCSE, GPA or the like would be suitable.

Nothing found. I tried a bunch of related terms like: education, test score, GCSE, o-level.

Live births by country of birth of mother. Total, 1st child, 2nd child, 3rd+. Also by sex. This enables one to calculate a crude fertility proxy using 3+/total. Example numbers:

Poland: .08
India: .13
Germany: .14
England: .15
Nigeria: .23
Ghana: .24
Pakistan: .41
Bangladesh: .44
Afghanistan: .45
Somalia: .52

This number is affected by age distribution of women and by the distribution of births. In weird and unrealistic cases, it can give wrong indicates with regards to total fertility. For instance, if a population has exactly 2 children per woman with no variation, their third/total rate is 0.00. Another country with a distribution of 0 births for 90% of women and 3 births for 10% of women would have a ratio of .10. The first country has a TFR of 2.00 and the second has 0.30. One can also use first/total to reduce such problems. Using all the data, these metrics correlate .68. If one only uses data for the top 100 countries, they correlate .89.

One can calculate the growth rate from longitudinal population count data which includes effects of immigration.


Anything else?
Forum Jump:

Users browsing this thread: 1 Guest(s)