irish traveller population in england

Can we store analytics cookies on your device?

Analytics cookies help us understand how our website is being used. They are not used to identify you personally.

You’ve accepted all cookies. You can change your cookie settings at any time.

• Ethnicity facts and figures homepage Home

Gypsy, Roma and Irish Traveller ethnicity summary

Updated 29 March 2022

1. About this page

2. the gypsy, roma and traveller group, 3. classifications, 4. improving data availability and quality, 5. population data, 6. education data, 7. economic activity and employment data.

• 8. Home ownership data data
• 9. Health data

This is a summary of statistics about people from the Gypsy, Roma and Traveller ethnic groups living in England and Wales.

It is part of a series of summaries about different ethnic groups .

Gypsy, Roma and Traveller (GRT) is a term used to describe people from a range of ethnicities who are believed to face similar challenges. These groups are distinct, but are often reported together.

This page includes:

• information about GRT data and its reliability
• some statistics from the 2011 Census
• other statistics on the experiences of people from the GRT groups in topics including education, housing and health

This is an overview based on a selection of data published on Ethnicity facts and figures or analyses of other sources. Some published data (for example, on higher education) is only available for the aggregated White ethnic group, and is not included here.

Through this report, we sometimes make comparisons with national averages. While in other reports we might compare with another ethnic group (usually White British), we have made this decision here because of the relatively small impact the GRT group has on the overall national average.

The term Gypsy, Roma and Traveller has been used to describe a range of ethnic groups or people with nomadic ways of life who are not from a specific ethnicity.

In the UK, it is common in data collections to differentiate between:

• Gypsies (including English Gypsies, Scottish Gypsies or Travellers, Welsh Gypsies and other Romany people)
• Irish Travellers (who have specific Irish roots)
• Roma, understood to be more recent migrants from Central and Eastern Europe

The term Traveller can also encompass groups that travel. This includes, but is not limited to, New Travellers, Boaters, Bargees and Showpeople. (See the House of Commons Committee report on Tackling inequalities faced by Gypsy, Roma and Traveller communities .)

For the first time, the 2011 Census ethnic group question included a tick box for the ethnic group ‘Gypsy or Irish Traveller’. This was not intended for people who identify as Roma because they are a distinct group with different needs to Gypsy or Irish Travellers.

The 2021 Census had a ‘Gypsy or Irish Traveller’ category, and a new ‘Roma’ category.

A 2018 YouGov poll found that 66% of people in the UK wrongly viewed GRT not to be an ethnic group, with many mistaking them as a single group (PDF). It is therefore important that GRT communities are categorised correctly on data forms, using separate tick boxes when possible to reflect this.

The 2011 Census figures used in this report and on Ethnicity facts and figures are based on respondents who chose to identify with the Gypsy or Irish Traveller ethnic group. People who chose to write in Roma as their ethnicity were allocated to the White Other group, and data for them is not included here. Other data, such as that from the Department for Education, includes Roma as a category combined with Gypsy, with Irish Traveller shown separately.

The commentary in this report uses the specific classifications in each dataset. Users should exercise caution when comparing different datasets, for example between education data (which uses Gypsy/Roma, and Irish Traveller in 2 separate categories) and the Census (which uses Gypsy and Irish Traveller together, but excludes data for people who identify as Roma).

Finally, it should be noted that there is also a distinction that the government makes, for the purposes of planning policy, between those who travel and the Gypsy, Roma and Traveller ethnicities. The Department for Communities and Local Government (at the time, now the Department for Levelling Up, Housing and Communities) planning policy for traveller sites (PDF) defines "gypsies and travellers" as:

"Persons of nomadic habit of life whatever their race or origin, including such persons who on grounds only of their own or their family’s or dependants’ educational or health needs or old age have ceased to travel temporarily, but excluding members of an organised group of travelling showpeople or circus people travelling together as such."

This definition for planning purposes includes any person with a nomadic habit, whether or not they might have identified as Gypsy, Roma or Traveller in a data collection.

The April 2019 House of Commons Women and Equalities Select Committee report on inequalities faced by Gypsy, Roma and Traveller communities noted that there was a lack of data on these groups.

The next section highlights some of the problems associated with collecting data on these groups, and what is available. Some of the points made about surveys, sample sizes and administrative data are generally applicable to any group with a small population.

Improving data for the Gypsy, Roma and Traveller populations, as well as other under-represented groups in the population is part of the recommendations in the Inclusive Data Taskforce report and the key activities described in the ONS response to them. For example, in response to recommendation 3 of the report, ONS, RDU and others will "build on existing work and develop new collaborative initiatives and action plans to improve inclusion of under-represented population groups in UK data in partnership with others across government and more widely".

Also, the ONS response to recommendation 4 notes the development of a range of strategies to improve the UK data infrastructure and fill data gaps to provide more granular data through new or boosted surveys and data linkage. Recommendation 6 notes that research will be undertaken using innovative methods best suited to the research question and prospective participants, to understand more about the lived experiences of several groups under-represented in UK data and evidence, such as people from Gypsy, Roma and Traveller groups.

4.1 Classifications

In some data collections, the option for people to identify as Gypsy, Roma or Traveller is not available. Any data grouped to the 5 aggregated ethnic groups does not show the groups separately. Data based on the 2001 Census does not show them separately as there was no category for people identifying as Gypsy, Roma or Traveller. As part of our Quality Improvement Plan, the Race Disparity Unit (RDU) has committed to working with government departments to maintain a harmonised approach to collecting data about Gypsy, Roma and Traveller people using the GSS harmonised classification. The harmonised classification is currently based on the 2011 Census, and an update is currently being considered by the Office for National Statistics (ONS).

In particular, RDU has identified working with DHSC and NHS Digital colleagues as a priority – the NHS classification is based on 2001 Census classifications and does not capture information on any of the GRT groups separately (they were categorised as White Other in the 2001 Census). Some of these issues have been outlined in the quarterly reports on progress to address COVID-19 health inequalities .

Research into how similar or different the aggregate ethnic groups are shows how many datasets are available for the GRT group.

Further information on the importance of harmonisation is also available.

4.2 Census data

A main source of data on the Gypsy and Irish Traveller groups is the 2011 Census. This will be replaced by the 2021 Census when results are published by the ONS. The statistics in this summary use information from Ethnicity facts and figures and the Census section of ONS’s NOMIS website.

4.3 Survey data

It is often difficult to conclude at any one point in time whether a disparity is significant for the GRT population, as the population is so small in comparison to other ethnic groups.

Even a large sample survey like the Annual Population Survey (APS) has a small number of responses from the Gypsy and Irish Traveller ethnic group each year. Analysis of 3 years of combined data for 2016, 2017 and 2018 showed there were 62 people in the sample (out of around 500,000 sampled cases in total over those 3 years) in England and Wales. Another large survey, the Department for Transport’s National Travel Survey, recorded 58 people identifying as Gypsy or Traveller out of 157,000 people surveyed between 2011 and 2019.

Small sample sizes need not be a barrier to presenting data if confidence intervals are provided to help the user. But smaller sample sizes will mean wider confidence intervals, and these will provide limited analytical value. For the 2016 to 2018 APS dataset – and using the standard error approximation method given in the LFS User Guide volume 6 with a fixed design factor of 1.6 (the formula is 1.6 * √p(1 − p)/n where p is the proportion in employment and n is the sample size.) – the employment rate of 35% for working age people in the Gypsy and Traveller group in England and Wales would be between 16% and 54% (based on a 95% confidence interval). This uses the same methodology as the ONS’s Sampling variability estimates for labour market status by ethnicity .

A further reason for smaller sample sizes might be lower response rates. The Women and Select Committee report on the inequalities faced by Gypsy, Roma and Traveller communities noted that people in these groups may be reluctant to self-identify, even where the option is available to them. This is because Gypsy, Roma and Traveller people might mistrust the intent behind data collection.

The RDU recently published a method and quality report on working out significant differences between estimates for small groups using different analytical techniques.

4.4 Administrative data

While administrative data does not suffer from the same issues of sampling variability, small numbers of respondents can mean that data is either disclosive and needs to be suppressed to protect the identity of individuals, or results can fluctuate over time.

An example of this is the measure of students getting 3 A grades or better at A level . In 2019 to 2020, no Irish Traveller students achieved this (there were 6 students in the cohort). In 2017 to 2018, 2 out of 7 Irish Traveller students achieved 3 A grades, or 28.6% – the highest percentage of all ethnic groups.

Aggregating time periods might help with this, although data collected in administrative datasets can change over time to reflect the information that needs to be collected for the administrative process. The data collected would not necessarily be governed by trying to maintain a consistent time series in the same way that data collected through surveys sometimes are.

4.5 Data linkage

Linking datasets together provides a way of producing more robust data for the GRT groups, or in fact, any ethnic group. This might improve the quality of the ethnicity coding in the dataset being analysed if an ethnicity classification that is known to be more reliable is linked from another dataset.

Data linkage does not always increase the sample size or the number of records available in the dataset to be analysed, but it might do if records that have missing ethnicity are replaced by a known ethnicity classification from a linked dataset.

An example is the linking of the Census data to Hospital Episode Statistics (HES) data and death registrations by the ONS. The ethnicity classifications for GRT groups are not included in the HES data, and are not collected in the death registrations process at the moment. So this data linking gives a way to provide some information for Gypsy and Irish Travellers and other smaller groups. The report with data up to 15 May 2020 noted 16 Gypsy or Irish Traveller deaths from COVID-19.

RDU will be working with ONS and others to explore the potential for using data linking to get more information for the GRT groups.

4.6 Bespoke surveys and sample boosts

A country-wide, or even local authority, boost of a sample survey is unlikely to make estimates for the GRT groups substantially more robust. This is because of the relatively small number in the groups to begin with.

Bespoke surveys can be used to get specific information about these groups. The Department for Levelling Up, Housing and Communities list of traveller sites available through their Traveller caravan count statistics can help target sampling for surveys, for example. Bespoke surveys might be limited in geographical coverage, and more suitable for understanding GRT views in a local area and then developing local policy responses. An example of a bespoke survey is the Roma and Travellers in 6 countries survey .

Another method that could be useful is snowball sampling. Snowball sampling (or chain-referral sampling) is a sampling technique in which the respondents have traits that are rare to find. In snowball sampling, existing survey respondents provide referrals to recruit further people for the survey, which helps the survey grow larger.

There are advantages to snowball sampling. It can target hidden or difficult to reach populations. It can be a good way to sample hesitant respondents, as a person might be more likely to participate in a survey if they have been referred by a friend or family member. It can also be quick and cost effective. Snowball sampling may also be facilitated with a GRT community lead or cultural mediator. This would help bridge the gap between the GRT communities and the commissioning department to encourage respondent participation.

However, one statistical disadvantage is that the sampling is non-random. This reduces the knowledge of whether the sample is representative, and can invalidate some of the usual statistical tests for statistical significance, for example.

All data in this section comes from the 2011 Census of England and Wales, unless stated otherwise.

In 2011, there were 57,680 people from the Gypsy or Irish Traveller ethnic group in England and Wales, making up 0.1% of the total population. In terms of population, it is the smallest of the 18 groups used in the 2011 Census.

Further ONS analysis of write-in responses in the Census estimated the Roma population as 730, and 1,712 people as Gypsy/Romany.

Table A: Gypsy, Roma and Traveller write-in ethnicity responses on the 2011 Census

Source: Census - Ethnic group (write-in response) Gypsy, Traveller, Roma, GypsyRomany - national to county (ONS). The figures do not add to the 57,680 classified as White: Gypsy/Traveller because Roma is included as White Other, and some people in the other categories shown will have classified themselves in an ethnic group other than White.

An ONS report in 2014 noted that variations in the definitions used for this ethnic group has made comparisons between estimates difficult. For example, some previous estimates for Gypsy or Irish Travellers have included Roma or have been derived from counts of caravans rather than people's own self-identity. It noted that other sources of data estimate the UK’s Gypsy, Roma and Traveller population to be in the region of 150,000 to 300,000 , or as high as 500,000 (PDF).

5.1 Where Gypsy and Irish Traveller people live

There were 348 local authorities in England and Wales in 2011. The Gypsy or Irish Traveller population was evenly spread throughout them. The 10 local authorities with the largest Gypsy or Irish Traveller populations constituted 11.9% of the total population.

Figure 1: Percentage of the Gypsy or Irish Traveller population of England and Wales living in each local authority area (top 10 areas labelled)

Basildon was home to the largest Gypsy or Irish Traveller population, with 1.5% of all Gypsy or Irish Traveller people living there, followed by Maidstone (also 1.5%, although it had a smaller population).

Table 1: Percentage of the Gypsy or Irish Traveller population of England and Wales living in each local authority area (top 10)

28 local authorities had fewer than 20 Gypsy or Irish Traveller residents each. This is around 1 in 12 of all local authorities.

11.7% of Gypsy or Irish Traveller people lived in the most deprived 10% of neighbourhoods , higher than the national average of 9.9% (England, 2019 Indices of Multiple Deprivation).

81.6% of people from the Gypsy or Irish Traveller ethnic group were born in England, and 6.1% in the other countries of the UK. 3.0% were born in Ireland and 8.3% were born somewhere else in Europe (other than the UK and Ireland). Less than 1.0% of Gypsy or Irish Traveller people were born outside of Europe.

5.2 Age profile

The Gypsy or Irish Traveller ethnic group had a younger age profile than the national average in England and Wales in 2011.

People aged under 18 made up over a third (36%) of the Gypsy or Irish Traveller population, higher than the national average of 21%.

18.0% of Gypsy or Irish Traveller people were aged 50 and above , lower than the national average of 35.0%.

Figure 2: Age profile of Gypsy or Irish Traveller and the England and Wales average

Table 2: age profile of gypsy or irish traveller and the england and wales average, 5.3 families and households.

20.4% of Gypsy or Irish Traveller households were made up of lone parents with dependent children , compared with 7.2% on average for England and Wales.

Across all household types, 44.9% of Gypsy or Irish Traveller households had dependent children, compared with an average of 29.1%.

8.4% of Gypsy or Irish Traveller households were made up of pensioners (either couples, single pensioners, or other households where everyone was aged 65 and over), compared with 20.9% on average.

All data in this section covers pupil performance in state-funded mainstream schools in England.

At all key stages, Gypsy, Roma and Irish Traveller pupils’ attainment was below the national average.

Figure 3: Educational attainment among Gypsy, Roma, Irish Traveller and pupils from all ethnic groups

Table 3: educational attainment among gypsy, roma, irish traveller and pupils from all ethnic groups.

Source: England, Key Stage 2 Statistics, 2018/19; Key Stage 4 Statistics, 2019/20; and A Level and other 16 to 18 results, 2020/21. Ethnicity facts and figures and Department for Education (DfE). Figures for Key Stage 2 are rounded to whole numbers by DfE.

6.1 Primary education

In the 2018 to 2019 school year, 19% of White Gypsy or Roma pupils, and 26% of Irish Traveller pupils met the expected standard in key stage 2 reading, writing and maths . These were the 2 lowest percentages out of all ethnic groups.

6.2 Secondary education

In the 2019 to 2020 school year, 8.1% of White Gypsy or Roma pupils in state-funded schools in England got a grade 5 or above in GCSE English and maths, the lowest percentage of all ethnic groups.

Gypsy or Roma (58%) and Irish Traveller (59%) pupils were the least likely to stay in education after GCSEs (and equivalent qualifications). They were the most likely to go into employment (8% and 9% respectively) – however, it is not possible to draw firm conclusions about these groups due to the small number of pupils in key stage 4.

6.3 Further education

Gypsy or Roma students were least likely to get at least 3 A grades at A level, with 10.8% of students doing so in the 2020 to 2021 school year. 20.0% of Irish Traveller students achieved at least 3 A grades, compared to the national average of 28.9%. The figures for Gypsy or Roma (61) and Irish Traveller (19) students are based on small numbers, so any generalisations are unreliable.

Due to the impact of the COVID-19 pandemic, the summer exam series was cancelled in 2021, and alternative processes were set up to award grades. In 2020/21 attainment is higher than would be expected in a typical year. This likely reflects the changes to the way A/AS level grades were awarded rather than improvements in student performance.

6.4 School exclusions

In the 2019 to 2020 school year, the suspension rates were 15.28% for Gypsy or Roma pupils, and 10.12% for Irish Traveller pupils – the highest rates out of all ethnic groups.

Also, the highest permanent exclusion rates were among Gypsy or Roma pupils (0.23%, or 23 exclusions for every 10,000 pupils). Irish Traveller pupils were permanently excluded at a rate of 0.14%, or 14 exclusions for every 10,000 pupils.

6.5 School absence

In the autumn term of the 2020 to 2021 school year, 52.6% of Gypsy or Roma pupils, and 56.7% of Irish Traveller pupils were persistently absent from school . Pupils from these ethnic groups had the highest rates of overall absence and persistent absence.

For the 2020 to 2021 school year, not attending in circumstances related to coronavirus (COVID-19) was not counted toward the overall absence rate and persistent absence rates.

Data in this section is from the 2011 Census for England and Wales, and for people aged 16 and over. Economic activity and employment rates might vary from other published figures that are based on people of working age.

47% of Gypsy or Irish Traveller people aged 16 and over were economically active, compared to an average of 63% in England and Wales.

Of economically active people, 51% of Gypsy or Irish Traveller people were employees, and 26% were self-employed. 20% of Gypsy or Irish Traveller people were unemployed, compared to an average for all ethnic groups of 7%.

7.1 Socio-economic group

Figure 4: socio-economic group of gypsy or irish traveller and average for all ethnic groups for people aged 16 and over, table 4: socio-economic group of gypsy or irish traveller and average for all ethnic groups for people aged 16 and over.

Source: 2011 Census

31.2% of people in the Gypsy or Irish Traveller group were in the socio-economic group of ‘never worked or long-term unemployed’. This was the highest percentage of all ethnic groups.

The Gypsy or Irish Traveller group had the smallest percentage of people in the highest socio-economic groups. 2.5% were in the ‘higher, managerial, administrative, professional’ group.

15.1% of Gypsy or Irish Traveller people were small employers and own account workers. These are people who are generally self-employed and have responsibility for a small number of workers.

For Gypsy or Irish Travellers, who were 16 and over and in employment, the largest group worked in elementary occupations (22%). This can include occupations such as farm workers, process plant workers, cleaners, or service staff (for example, bar or cleaning staff).

The second highest occupation group was skilled trades (19%), which can include farmers, electrical and building trades. The Gypsy or Irish Traveller group had the highest percentage of elementary and skilled trade workers out of all ethnic groups.

7.2 Employment gender gap

The gender gap in employment rates for the Gypsy or Irish Traveller group aged 16 and over was nearly twice as large as for all ethnic groups combined. In the Gypsy or Irish Traveller ethnic group, 46% of men and 29% of women were employed, a gap of 17%. For all ethnic groups combined, 64% of men and 54% of women were employed, a gap of 10%.

This is likely to be due to the fact that Gypsy or Irish Traveller women (63%) were about 1.5 times as likely as Gypsy or Irish Traveller men (43%) to be economically inactive, which means they were out of work and not looking for work.

7.3 Economic inactivity

There are a range of reasons why people can be economically inactive. The most common reason for Gypsy or Irish Travellers being economically inactive was looking after the home or family (27%). This is higher than the average for England and Wales (11%). The second most common reason was being long term sick or disabled (26%) – the highest percentage out of all ethnic groups.

8. Home ownership data

Figure 5: home ownership and renting among gypsy or irish traveller households and all households, table 5: home ownership and renting among gypsy or irish traveller households and all households.

Source: England, 2011 Census

In 2011, 34% of Gypsy or Irish Traveller households owned their own home, compared with a national average of 64%. 42% lived in social rented accommodation, compared with a national average of 18%.

In 2016 to 2017, 0.1% of new social housing lettings went to people from Gypsy or Irish Traveller backgrounds (429 lettings).

In 2011, a whole house or bungalow was the most common type of accommodation for Gypsy or Irish Traveller households (61%). This was lower than for all usual residents in England and Wales (84%).

Caravans or other mobile or temporary homes accounted for 24% of Gypsy or Irish Travellers accommodation, a far higher percentage than for the whole of England and Wales (0.3%).

The percentage of people living in a flat, maisonette or apartment was 15% for both Gypsy or Irish Travellers and all usual residents in England and Wales.

In 2011, 14.1% of Gypsy and Irish Traveller people in England and Wales rated their health as bad or very bad, compared with 5.6% on average for all ethnic groups.

In 2016 to 2017, Gypsy or Irish Traveller people aged 65 and over had the lowest health-related quality of life of all ethnic groups (average score of 0.509 out of 1). The quality of life scores for the White Gypsy or Irish Traveller ethnic group are based on a small number of responses (around 35 each year) and are less reliable as a result.

Ethnicity facts and figures has information on satisfaction of different health services for different ethnic groups. For the results presented below, the Gypsy or Irish Traveller figures are based on a relatively small number of respondents, and are less reliable than figures for other ethnic groups.

In 2014 to 2015 (the most recent data available), these groups were the most satisfied with their experience of GP-out-of-hours service , with 75.2% reporting a positive experience.

In 2018 to 2019, they were less satisfied with their experience of GP services than most ethnic groups – 73.0% reported a positive experience.

They were also among the groups that had least success when booking an NHS dentist appointment – 89.0% reported successfully booking an appointment in 2018 to 2019.

The Gypsy or Irish Traveller group were also less satisfied with their access to GP services in 2018 to 2019 – 56.9% reported a positive experience of making a GP appointment, compared to an average of 67.4% for all respondents.

Publication release date: 31 January 2022

Updated: 29 March 2022

29 March 2022: Corrected A-level data in Table 3, and All ethnic groups data in Table 4. Corrected the legend in Figure 1 (map).

31 January 2022: Initial publication.

Gypsy Roma and Traveller History and Culture

Gypsy Roma and Traveller people belong to minority ethnic groups that have contributed to British society for centuries. Their distinctive way of life and traditions manifest themselves in nomadism, the centrality of their extended family, unique languages and entrepreneurial economy. It is reported that there are around 300,000 Travellers in the UK and they are one of the most disadvantaged groups. The real population may be different as some members of these communities do not participate in the census .

The Traveller Movement works predominantly with ethnic Gypsy, Roma, and Irish Traveller Communities.

Irish Travellers and Romany Gypsies

Irish Travellers

Traditionally, Irish Travellers are a nomadic group of people from Ireland but have a separate identity, heritage and culture to the community in general. An Irish Traveller presence can be traced back to 12th century Ireland, with migrations to Great Britain in the early 19th century. The Irish Traveller community is categorised as an ethnic minority group under the Race Relations Act, 1976 (amended 2000); the Human Rights Act 1998; and the Equality Act 2010. Some Travellers of Irish heritage identify as Pavee or Minceir, which are words from the Irish Traveller language, Shelta.

Romany Gypsies

Romany Gypsies have been in Britain since at least 1515 after migrating from continental Europe during the Roma migration from India. The term Gypsy comes from “Egyptian” which is what the settled population perceived them to be because of their dark complexion. In reality, linguistic analysis of the Romani language proves that Romany Gypsies, like the European Roma, originally came from Northern India, probably around the 12th century. French Manush Gypsies have a similar origin and culture to Romany Gypsies.

There are other groups of Travellers who may travel through Britain, such as Scottish Travellers, Welsh Travellers and English Travellers, many of whom can trace a nomadic heritage back for many generations and who may have married into or outside of more traditional Irish Traveller and Romany Gypsy families. There were already indigenous nomadic people in Britain when the Romany Gypsies first arrived hundreds of years ago and the different cultures/ethnicities have to some extent merged.

Number of Gypsies and Travellers in Britain

This year, the 2021 Census included a “Roma” category for the first time, following in the footsteps of the 2011 Census which included a “Gypsy and Irish Traveller” category. The 2021 Census statistics have not yet been released but the 2011 Census put the combined Gypsy and Irish Traveller population in England and Wales as 57,680. This was recognised by many as an underestimate for various reasons. For instance, it varies greatly with data collected locally such as from the Gypsy Traveller Accommodation Needs Assessments, which total the Traveller population at just over 120,000, according to our research.

Other academic estimates of the combined Gypsy, Irish Traveller and other Traveller population range from 120,000 to 300,000. Ethnic monitoring data of the Gypsy Traveller population is rarely collected by key service providers in health, employment, planning and criminal justice.

Where Gypsies and Travellers Live

Although most Gypsies and Travellers see travelling as part of their identity, they can choose to live in different ways including:

• moving regularly around the country from site to site and being ‘on the road’
• living permanently in caravans or mobile homes, on sites provided by the council, or on private sites
• living in settled accommodation during winter or school term-time, travelling during the summer months
• living in ‘bricks and mortar’ housing, settled together, but still retaining a strong commitment to Gypsy/Traveller culture and traditions

Currently, their nomadic life is being threatened by the Police, Crime, Sentencing and Courts Bill, that is currently being deliberated in Parliament, To find out more or get involved with opposing this bill, please visit here

Although Travellers speak English in most situations, they often speak to each other in their own language; for Irish Travellers this is called Cant or Gammon* and Gypsies speak Romani, which is the only indigenous language in the UK with Indic roots.

*Sometimes referred to as “Shelta” by linguists and academics

New Travellers and Show People

There are also Traveller groups which are known as ‘cultural’ rather than ‘ethnic’ Travellers. These include ‘new’ Travellers and Showmen. Most of the information on this page relates to ethnic Travellers but ‘Showmen’ do share many cultural traits with ethnic Travellers.

Show People are a cultural minority that have owned and operated funfairs and circuses for many generations and their identity is connected to their family businesses. They operate rides and attractions that can be seen throughout the summer months at funfairs. They generally have winter quarters where the family settles to repair the machinery that they operate and prepare for the next travelling season. Most Show People belong to the Showmen’s Guild which is an organisation that provides economic and social regulation and advocacy for Show People. The Showman’s Guild works with both central and local governments to protect the economic interests of its members.

The term New Travellers refers to people sometimes referred to as “New Age Travellers”. They are generally people who have taken to life ‘on the road’ in their own lifetime, though some New Traveller families claim to have been on the road for three consecutive generations. The New Traveller culture grew out of the hippie and free-festival movements of the 1960s and 1970s.

Barge Travellers are similar to New Travellers but live on the UK’s 2,200 miles of canals. They form a distinct group in the canal network and many are former ‘new’ Travellers who moved onto the canals after changes to the law made the free festival circuit and a life on the road almost untenable. Many New Travellers have also settled into private sites or rural communes although a few groups are still travelling.

If you are a new age Traveller and require support please contact Friends, Families, and Travellers (FFT) .

Differences and Values

Differences Between Gypsies, Travellers, and Roma

Gypsies, Roma and Travellers are often categorised together under the “Roma” definition in Europe and under the acronym “GRT” in Britain. These communities and other nomadic groups, such as Scottish and English Travellers, Show People and New Travellers, share a number of characteristics in common: the importance of family and/or community networks; the nomadic way of life, a tendency toward self-employment, experience of disadvantage and having the poorest health outcomes in the United Kingdom.

The Roma communities also originated from India from around the 10th/ 12th centuries and have historically faced persecution, including slavery and genocide. They are still marginalised and ghettoised in many Eastern European countries (Greece, Bulgaria, Romania etc) where they are often the largest and most visible ethnic minority group, sometimes making up 10% of the total population. However, ‘Roma’ is a political term and a self-identification of many Roma activists. In reality, European Roma populations are made up of various subgroups, some with their own form of Romani, who often identify as that group rather than by the all-encompassing Roma identity.

Travellers and Roma each have very different customs, religion, language and heritage. For instance, Gypsies are said to have originated in India and the Romani language (also spoken by Roma) is considered to consist of at least seven varieties, each a language in their own right.

Values and Culture of GRT Communities

Family, extended family bonds and networks are very important to the Gypsy and Traveller way of life, as is a distinct identity from the settled ‘Gorja’ or ‘country’ population. Family anniversaries, births, weddings and funerals are usually marked by extended family or community gatherings with strong religious ceremonial content. Gypsies and Travellers generally marry young and respect their older generation. Contrary to frequent media depiction, Traveller communities value cleanliness and tidiness.

Many Irish Travellers are practising Catholics, while some Gypsies and Travellers are part of a growing Christian Evangelical movement.

Gypsy and Traveller culture has always adapted to survive and continues to do so today. Rapid economic change, recession and the gradual dismantling of the ‘grey’ economy have driven many Gypsy and Traveller families into hard times. The criminalisation of ‘travelling’ and the dire shortage of authorised private or council sites have added to this. Some Travellers describe the effect that this is having as “a crisis in the community” . A study in Ireland put the suicide rate of Irish Traveller men as 3-5 times higher than the wider population. Anecdotal evidence suggests that the same phenomenon is happening amongst Traveller communities in the UK.

Gypsies and Travellers are also adapting to new ways, as they have always done. Most of the younger generation and some of the older generation use social network platforms to stay in touch and there is a growing recognition that reading and writing are useful tools to have. Many Gypsies and Travellers utilise their often remarkable array of skills and trades as part of the formal economy. Some Gypsies and Travellers, many supported by their families, are entering further and higher education and becoming solicitors, teachers, accountants, journalists and other professionals.

There have always been successful Gypsy and Traveller businesses, some of which are household names within their sectors, although the ethnicity of the owners is often concealed. Gypsies and Travellers have always been represented in the fields of sport and entertainment.

How Gypsies and Travellers Are Disadvantaged

The Traveller, Gypsy, and Roma communities are widely considered to be among the most socially excluded communities in the UK. They have a much lower life expectancy than the general population, with Traveller men and women living 10-12 years less than the wider population.

Travellers have higher rates of infant mortality, maternal death and stillbirths than the general population. They experience racist sentiment in the media and elsewhere, which would be socially unacceptable if directed at any other minority community. Ofsted consider young Travellers to be one of the groups most at risk of low attainment in education.

Government services rarely include Traveller views in the planning and delivery of services.

In recent years, there has been increased political networking between the Gypsy, Roma and Traveller activists and campaign organisations.

Watch this video by Travellers Times made for Gypsy Roma Traveller History Month 2021:

Information and Support

We have a variety of helpful guides to provide you with the support you need

Community Corner

Read all about our news, events, and the upcoming music and artists in your area

Cookies on GOV.UK

We use some essential cookies to make this website work.

We’d like to set additional cookies to understand how you use GOV.UK, remember your settings and improve government services.

We also use cookies set by other sites to help us deliver content from their services.

You have accepted additional cookies. You can change your cookie settings at any time.

You have rejected additional cookies. You can change your cookie settings at any time.

• Society and culture
• Equality, rights and citizenship

Ethnicity facts and figures: Gypsy, Roma and Irish Traveller ethnic group

An overview of facts and figures about people from the Gypsy, Roma and Irish Traveller ethnic groups, and a discussion of data quality issues.

Gypsy, Roma and Irish Traveller ethnic group: facts and figures

https://www.ethnicity-facts-figures.service.gov.uk/summaries/gypsy-roma-irish-traveller

This is a summary of statistics about people from the Gypsy, Roma and Irish Traveller ethnic groups living in England and Wales.

It is part of a series of summaries about different ethnic groups .

This is an overview based on a selection of data published on Ethnicity facts and figures or being published soon.

Related content

Is this page useful.

• Yes this page is useful
• No this page is not useful

Help us improve GOV.UK

Don’t include personal or financial information like your National Insurance number or credit card details.

To help us improve GOV.UK, we’d like to know more about your visit today. We’ll send you a link to a feedback form. It will take only 2 minutes to fill in. Don’t worry we won’t send you spam or share your email address with anyone.

Irish In Britain

Contact info, not yet a member join us, irish in britain community members, log in to your account, email address.

[email protected]

******* Forgotten your password?

Remember me

Already a community member but need to set up an account? Email us

Not yet a member? Join us

Keep up to date with everything IIB, sign up to our mailing list

Thank you for signing up to our mailing list.

Please fill out all required fields

The traveller movement.

Yvonne MacNamara (CEO)

020 7607 2002

[email protected]

http://www.travellermovement.org.uk/

40 Jeffrey's Road, Stockwell, London SW4 6QX

Both, Irish Travellers and Romany Gypsies, (hereafter referred to as Traveller communities) are documented as the most disadvantaged ethnic groups in Britain today. It is estimated that there are 300,000 of Travellers in Britain although this is thought to be an underestimate due to their lack of participation in the census. 

The Traveller Movement (TM) was established in 1999 as a second–tier community development charity, following the recognition of a gap in service provision and the marginalisation of the Irish Traveller community in Britain. 

TM gained charitable status in 2003. Following a recognised demand and its capabilities TM expanded its remit in 2008 to include work with all other Gypsy Roma and Traveller communities in Britain. 

Through commissioning and collating research, TM has become the leading charity and first point of contact regarding Traveller issues. TM seeks to empower Traveller communities by campaigning for a new policy framework for addressing exclusion and discrimination. 

TM encourages Traveller projects and other interested service providers to use TM as a channel to collect and disseminate information on legislative policies and their effect on the Traveller community. TM also promotes equality, diversity and racial harmony for the benefit of all members of British society by encouraging others to understand the culture and needs of Travellers. 

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 16 February 2017

Genomic insights into the population structure and history of the Irish Travellers

• Edmund Gilbert 1 ,
• Shai Carmi 2 ,
• Sean Ennis 3 ,
• James F. Wilson 4 , 5   na1 &
• Gianpiero L. Cavalleri 1   na1  

Scientific Reports volume  7 , Article number:  42187 ( 2017 ) Cite this article

35k Accesses

24 Citations

120 Altmetric

Metrics details

• Consanguinity
• Population genetics

The Irish Travellers are a population with a history of nomadism; consanguineous unions are common and they are socially isolated from the surrounding, ‘settled’ Irish people. Low-resolution genetic analysis suggests a common Irish origin between the settled and the Traveller populations. What is not known, however, is the extent of population structure within the Irish Travellers, the time of divergence from the general Irish population, or the extent of autozygosity. Using a sample of 50 Irish Travellers, 143 European Roma, 2232 settled Irish, 2039 British and 6255 European or world-wide individuals, we demonstrate evidence for population substructure within the Irish Traveller population, and estimate a time of divergence before the Great Famine of 1845–1852. We quantify the high levels of autozygosity, which are comparable to levels previously described in Orcadian 1 st /2 nd cousin offspring, and finally show the Irish Traveller population has no particular genetic links to the European Roma. The levels of autozygosity and distinct Irish origins have implications for disease mapping within Ireland, while the population structure and divergence inform on social history.

Similar content being viewed by others

Network of large pedigrees reveals social practices of Avar communities

Genome-wide association studies

Leveraging functional genomic annotations and genome coverage to improve polygenic prediction of complex traits within and between ancestries

Introduction.

The Irish Travellers are a community within Ireland, consisting of between 29,000–40,000 individuals, representing 0.6% of the Irish population as a whole 1 . They are traditionally nomadic, moving around rural Ireland and providing seasonal labour, as well as participating in horse-trading and tin-smithing 2 . Since the 1950’s the need for such traditional services has declined 3 , and the population has become increasingly urban, with the majority living within a fixed abode 1 . Despite this change in lifestyle, the Traveller community remains tight-knit but also socially isolated. The population has its own language 4 , known as Shelta, of which Cant and Gammon are dialects.

There is a lack of documentary evidence informing on the history of the Irish Traveller population 5 , 6 . As a result, their origins are a source of considerable debate, with no single origin explanation being widely accepted. It has been suggested that the Irish Travellers are a hybrid population between settled Irish and Romani gypsies, due to the similarities in their nomadic lifestyle. Other, “Irish Origin”, hypothesised sources of the Irish Travellers include; displacement from times of famine (such as between 1740–1741, or the Great Famine of 1845–1852), or displacement from the time of Cromwellian (1649–53) or the Anglo-Norman conquests (1169 to 1240). The Irish Traveller population may even pre-date these events, and represent Celtic or pre-Celtic isolates 4 . These models of ethnogenesis are not necessarily mutually exclusive, and the Irish Traveller population may have multiple sources of origin with a shared culture.

Consanguineous marriages are common within the Irish Traveller community 7 , 8 . Small, isolated and endogamous populations such as the Travellers are also more prone to the effects of genetic drift. The isolation and consanguinity have in turn led to an increased prevalence of recessive diseases 7 , 9 , 10 , with higher incidences of diseases such as transferase-deficient galactosaemia 11 , 12 , and Hurler syndrome 13 observed in the Traveller population relative to the settled Irish. However, the extent of autozygosity within the population has yet to be quantified; as a result it is unknown how homozygous the population is compared to other, better-studied, isolated European populations.

Previous work into the genetics of the Irish Traveller population has been conducted on datasets of relatively low genetic resolution. A recent study used blood groups to investigate the population history of the Irish Travellers 2 . Multivariate analysis of genotype data across 12 red blood cell loci in 119 Irish Travellers suggested that the population clustered closely with the settled Irish to the exclusion of the Roma. They did, however, appear divergent from the settled Irish. The authors attributed the source of such divergence to genetic drift - but were unable to determine whether any such drift was due to a founder effect, or sustained endogamy. Studies of Mendelian diseases suggest that pathogenic mutations in the settled Irish population are often the same as those observed in the Traveller population such is the case for tranferase-deficient galactosaemia (Q118R in the GALT gene 11 ) and Hurlers Syndrome (W402X, in the α-l-iduronidase gene 13 ).

Using dense, genome-wide, SNP datasets which provide much greater resolution than genetic systems studied in the Travellers to date, we set out to i) describe the genetic structure within the Traveller population, ii) the relationship between the Irish Travellers and other European populations, iii) estimate the time of divergence between the Travellers and settled Irish, and iv) the levels of autozygosity within the Irish Traveller population.

Population Structure of the Irish Travellers

In order to investigate the genetic relationship between the Irish Travellers and neighbouring populations we performed fineStructure analysis on Irish Travellers, settled Irish from a subset of the Trinity Student dataset 14 , and British from a subset of the POBI dataset 15 . A subset of the datasets were used in this analysis as we were primarily interested in the placing of the Irish Travellers within the context of Britain and Ireland, not the full structure found within Britain and Ireland. The results are presented in Fig. 1 in the form of a principal component analysis of fineStructure’s haplotype-based co-ancestry matrix (1A) and a dendrogram of the fineStructure clusters (1B).

( A ) The first and second components of principal component analysis of the haplotype-based co-ancestry matrix produced by fineStructure analysis. Individual clusters are indicated by colour and shape. Individual Irish Travellers are indicated with black bordered shapes, with cluster shown in Legend. ( B ) The full fineStructure tree with the highest posterior probability, with cluster size and name, and broad branches shown.

We observe that 31 of 34 of the Irish Travellers cluster on the Irish branch, indicating a strong affinity with an Irish population ancestral to the current day “Traveller” and “settled” populations ( Fig. 1B ). One “Irish Traveller” is found within the Borders 1 cluster, and two are found within the Borders 2 cluster. These three individuals report full, or partial, English gypsie ancestry, a distinct and separate travelling population in Britain. One individual is found within the Ireland 1 cluster, and two are found within the Ireland 2 cluster. Traveller individuals within the Ireland 2 cluster report recent settled ancestry, and we have no such genealogical data on the individual grouped within the Ireland 1 cluster. Given their mixed ancestry, these individuals were excluded from subsequent F st , f 3 , and divergence estimate work.

The remaining 28 Irish Travellers in the fineStructure analysis were arranged into four clusters. These clusters were grouped on two separate branches ( Fig. 1B ), with Traveller 1 (n = 7) and Traveller 2 (n = 5) on the same branch, and Traveller 3 (n = 5) and Traveller 4 (n = 11) on a separate branch. The branch with clusters Traveller 3 and 4 , forms an outgroup to the rest of the settled Irish and Irish Traveller clusters. These two branches of Irish Traveller clusters align closely with the split of Irish Travellers observed through PCA ( Fig. S1 ). All the individuals who separate on the first principal component (henceforth “PCA group B”) are found in clusters Traveller 3 and 4 ( Fig. S2A ), and nearly all the individuals who remain grouped with the settled Irish on principle component 1 (henceforth “PCA group A”) are found in clusters Traveller 1 and 2 ( Fig. S2A ). The remaining PCA group A individuals are those Irish Travellers found in the aforementioned settled Irish or British clusters. This pattern is also repeated in the PCA ( Fig. 1A ), where members of Traveller 1 and 2 cluster with the settled Irish, where Traveller 3 and 4 individuals cluster separately.

Having identified distinct genetic groups of Irish Travellers, we investigated the correlation with Irish Traveller sociolinguistic features, specifically Shelta dialect, and Rathkeale residence ( Fig. S2B ,C, respectively). The majority of the Gammon speakers were members of clusters Traveller 1 and 2. All of Traveller 1 consisted of Gammon speakers. The majority of clusters Traveller 3 and 4 consisted of Cant speakers, where all but one individual, for whom language identity is unknown, of Traveller 4 were Cant speakers. We found that only clusters Traveller 1 and 2 contain any Rathkeale Travellers, where 4 out of 5 individuals in Traveller 2 are Rathkeale Travellers.

We next investigated population structure using the maximum-likelihood estimation of individual ancestries using ADMIXTURE ( Figs 2 and S3). For this analysis we used a subset of the European Multiple Sclerosis dataset consisting of three northern European (Norway, Finland and Germany), two southern European (Italy and Spain), and a neighbouring population (France). We categorised the POBI British as English, Scottish, Welsh, and Orcadian. We further separated out the Irish Travellers to those in PCA group A and those in PCA group B.

Shown are the ancestry components per individual for the two groups of Irish Travellers (Group A and Group B), settled Irish, British, and European populations; modelling for 4 to 6 ancestral populations.

At k  = 4–6 ( Fig. 2 ), we observe the well-described north-south divide in the European populations ( k  = 4), as well as Finland and Orkney ( k  = 5) differentiating due to their respective populations’ bottleneck and isolation. Although at lower values of k the Irish Travellers generally resemble the settled Irish profile ( Fig. S3 ), at higher values of k two components are found to be enriched within the population. Each of these components is enriched in one of the two Irish Traveller PCA groups. Individuals with more than 20% of the “red” component when k = 5 belong to PCA group B and individuals with near 100% of “blue” component all belong to PCA group A ( Fig. 2 ). The fact that even at k  = 3 PCA group B gains its own ancestral component ( Fig. S3 ) suggests strong group-specific genetic drift.

In order to investigate a possible Roma Gyspie origin of the Irish Travellers, we compared the Irish Travellers, and settled Irish to a dataset of Roma populations found within Europe 16 using PCA and ADMIXTURE. The results broadly agree, with the Irish Travellers clustering with the settled Irish in the PCA plot, and resembling the settled Irish profile in ADMIXTURE analysis (see Fig. 3 ). There was no evidence for a recent ancestral component between the Irish Traveller and Roma populations. In addition, we formally tested evidence of admixture with f 3 statistics in the form of f 3 (Irish Traveller; Settled Irish, Roma). We found no evidence of admixture either when considering all the Roma as one population, or in each individual Roma population’s case (all f3 estimates were positive).

( A ) The first and second components from principal component analysis using gcta64. ( B ) The ancestry profiles using ADMIXTURE, assuming 2 to 4 ancestral populations.

Given the apparent structure between the Travellers and the settled Irish populations, we quantified genetic distance using F st and “outgroup” f 3 statistics. F st analysis reveals a considerable genetic distance between the settled Irish and the Irish Traveller population (F st  = 0.0034, Table S1 ) which is comparable to values observed between German and Italian, or Scotland and Spain.

In order to further investigate sub-structure within the Irish Travellers, we performed F st analysis on the Irish Traveller PCA (n = 2) and fineStructure (n = 4) groups, comparing them to the settled Irish (see also Table S1 ). The individuals belonging to cluster PCA group B are considerably more genetically distant from the settled Irish (F st  = 0.0086), relative to PCA group A (F st  = 0.0036). This could be explained by distinct founder events for PCA groups A and B, or that PCA group B has experienced greater genetic drift. The F st estimates of the Irish Traveller clusters are higher than the PCA groups. The estimates of clusters Traveller 1, 2 , and 3 range from 0.0052 to 0.0054. However, Traveller 4 shows the highest F st value (F st  = 0.0104), suggesting this cluster of individuals is responsible for the inflation of the PCA group B’s estimate. Generally, however, these results suggest that the general Irish Traveller population does not have a very recent source, i.e. within 5 generations or so. If we perform the same F st analysis on two random groups of settled Irish see observe a F st value < 1∙10 −5 .

To inform on whether lineage-specific drift is influencing the observed genetic distances between the Irish Travellers, the settled Irish and other neighbouring populations, we performed outgroup f 3 analysis, using HGDP Yorubans as the outgroup. Such analysis can inform on whether PCA group B and Traveller 4 do indeed represent an older Irish Traveller group, or a sub-group that has experienced more intense drift. When we compare PCA groups A/B to the settled Irish we see no significant difference between the two groups (see Table S2 , A:settled f 3  = 0.1694 (stderr = 0.0013), B:settled f 3  = 0.1698 (stdrr = 0.0013), A:B f 3  = 0.1700 (stderr = 0.0013)); with similar results for the fineStructure clusters ( Table S2 ). These results suggest that PCA group B has experienced more drift than PCA group A, inflating the F st statistic, which in turn has inflated the Irish Traveller population F st . We note however that f 3 statistics may not be sensitive enough to detect differences from settled Irish to Traveller PCA groups A and B should the difference between A and B be a relatively limited number of generations.

A key question in the history of the Travellers is the period of time for which the population has been isolated from the settled Irish. In order to address this we utilized two methods, one based on linkage disequilibrium patterns and F st (which we call T F ), and one based on Identity-by-Descent (IBD) patterns (which we call T IBD ).

The T F method estimates the divergence to be 40 (±2 std.dev – obtained via bootstrapping) generations. Assuming an average generation time of 30 years the T F method estimates that the divergence occurred 1200 (±60 – std.dev) years ago. The method also estimates the harmonic mean N e for the two populations over the last 2000 years. The Irish Traveller estimate (1395, std.dev = 16 – obtained via bootstrapping) is considerably lower than the settled Irish estimate (6162, std.err = 122 – obtained via bootstrapping). However, the isolation of the Irish Travellers will artificially increase the F st value and consequently inflate the T F divergence estimate. We therefore estimated the divergence time with a different IBD-based method; as such an approach can accommodate genetic drift.

We first identified IBD segment sharing within and between the Irish Travellers and our settled Irish subset. The Irish Travellers were found to share 35-fold more genetic material IBD (in cM per pair) than the settled population ( Fig. 4A ). Specifically, a pair of Travellers share, on average, 5.0 segments of mean length 12.9 cM, compared to 0.4 segments of mean length 4.9 cM for the settled population ( Fig. 4A ; segments with length >3 cM). Additionally we compared IBD sharing within and between the two PCA groups; A and B ( Fig. 4B ). We observe a greater amount of IBD segments shared within PCA group B than PCA group A. These sharing patterns are not due to familial sharing, as we have previously removed individuals with close kinship (see Supplementary Methods 1.3 ). Sharing between settled and Traveller Irish was of similar extent to that within the settled group ( Fig. 4A ), with no significant difference between the PCA groups A and B (p = 0.12, using permutations, for the difference in the number of segments shared with the settled) ( Fig. S4 ). We used the number and lengths of segments shared within settled, within Travellers, and between the groups to estimate the demographic history of those populations, and in particular, the split time between these two groups.

( A ) The number and lengths of shared segments within Settled Irish, within Traveller Irish, and between the groups. Left panel: The mean segment length; middle panel: the mean number of shared segments; right panel: the mean total sequence length (in cM) shared between each pair of individuals. ( B ) The number and lengths of shared segments within Traveller Group A, Traveller Group B, and between the groups. The format of the figure is as in ( A ).

Briefly, we used the method developed in Palamara et al . 17 (see also Zidan et al . 18 ). We assumed a demographic model for the two populations ( Fig. 5A ), in which an ancestral Irish population has entered a period of exponential expansion before the ancestors of the present day settled Irish and Irish Travellers split. After this split, the settled Irish continued the exponential expansion, whilst the Irish Travellers experienced an exponential population contraction. We then computed the expected proportion of the genome found in shared segments of different length intervals using the theory of ref. 17 , and found the parameters of the demographic model that best fitted the data (see Supplementary Data 1.3 , Fig. 5B , and Table 1 ).

( A ) The model used for demographic inference. The two populations were one ancestral population, with size N e , T G generations ago. At this point the ancestral population started to grow exponentially until T S generations ago, where the ancestral Traveller and settled populations split from each other, with N S,T being the initial starting population size of the Traveller population. The settled population experienced continued exponential growth until the present, with a population size of N C,S . The Traveller population experienced a period of exponential contraction until the present, with a population of N C,T . ( B ) The proportion of the genome in IBD segments vs the IBD segments length. The total genome size and the sum of segment lengths were computed in cM. Left: sharing between pairs of settled Irish; middle: sharing between pairs of one settled and one Traveller individuals; right: sharing between pairs of Traveller Irish. Each data point is located at the harmonic mean of the boundaries of the length interval it represents.

The results of the model suggest the Irish Travellers and settled Irish separation occurred 12 generations ago (95% CI: 8–14). The results also support opposite trends in the effective population sizes (N e ) of the settled and Traveller Irish since that split: while the settled population has expanded rapidly, the Irish Travellers have contracted (see Table 1 ). When restricting to the 12 members of PCA group A, the split time was estimated to be 15 generations ago (95% CI: 13–18) ( Table 2 ). When restricting to the 16 members of PCA group B, the split time was 10 generations ago (95% CI: 3–14). We stress these results should be seen as the best fitting projection of the true history into a simplified demographic model, in particular given the limited sample sizes.

Runs of Homozygosity

Consanguinity is common within the Irish Traveller population, and in this context we quantified the levels of homozygosity compared to settled Irish and world-wide populations 19 . We calculated the average total extent of homozygosity of each population using four categories of minimum length of Runs of Homozygosity (ROH) (1/5/10/16 Mb). Elevated ROH levels between 1 and 5 Mb are indicative of a historical smaller population size. Elevated ROH levels over 10 Mb, on the other hand, are reflective of more recent consanguinity in an individuals’ ancestry 10 . We also include average figures for the European Roma in the Irish Traveller – European analysis. Full European Roma ROH profiles are shown in Figure S5 .

As expected, the Irish Travellers present a significantly higher amount of homozygosity compared to the other outbred populations and to the European isolates the French Basque and Sardinian, which is sustained through to the larger cutoff categories of 10–16 Mb (see Fig. 6 ). Our results for the other world-wide populations agree with previous estimates 10 , with the Native American Karitiana showing the most autozygosity, and the Papuan population showing an excess of short ROHs. Two other consanguineous populations, the Balochi and Druze show slightly more homozygosity than the Irish Travellers, and the European Roma are most similar to the Travellers for both shorter and longer ROH.

Shown, across four minimum lengths of runs of homozygosity (ROH), are the average lengths of ROH in each population. The average ROH burdens for the European Roma are the mean of means across the 13 Roma populations studied. These values are from a separate analysis, and collated with the wider European ROH values for reasons of SNP coverage between the different datasets.

These results indicate a higher level of background relatedness in the Irish Traveller population history. The high levels of ROH larger than 10 Mb in length reflect recent parental relatedness within the population. This is supported by the average F ROH5 in the Irish Travellers (F ROH5  = 0.015), which is slightly lower but comparable to the F ROH5 score found among Orcadian offspring of 1 st /2 nd cousins (F ROH5  = 0.017) 20 .

Finally, in order to explore the potential of the Irish Traveller population for studying rare, functional variation for disease purposes, we tested minor allele frequency (MAF) differences between the settled Irish and the Irish Travellers from a common dataset of 560,256 common SNPs for 36 Traveller, and 2232 settled Irish individuals. We observed 24,670 SNPs with a MAF between 0.02–0.05 in the settled Irish population. We found that 3.29% of these SNPs had a MAF >0.1 in the Irish Traveller population. We tested the significance of this observation by calculating the same percentage, but taking a random 36 settled Irish sample instead of 36 Irish Travellers. We repeated this 1000 times and found no samples (p =< 0.001) with a greater percentage than 3.29 (mean = 1.3, std.dev = 0.11). This has additional implications for disease mapping within Ireland, as a proportion of the functional variants in the settled Irish population will be observed at a higher frequency in the Traveller population.

We have, using high-density genome-wide SNP data on 42 Irish Traveller individuals, investigated the genetic relationship between the Travellers and neighbouring populations and another nomadic European population, the Roma. For the first time we have estimated a time of divergence of the Irish Travellers from the general Irish population, and have also quantified the extent of autozygosity within the population.

We report that the Irish Traveller population has an ancestral Irish origin, closely resembling the wider Irish population in the context of other European cohorts. This is consistent with previous observations made using a limited number of classical markers 2 , 4 . In both our fineStructure and ADMIXTURE analyses, the Traveller population clusters predominantly with the settled Irish. Our fineStructure tree qualitatively agrees with the topology presented by Leslie et al . 21 , although there are some differences. For example, in the tree presented here, the Irish and individuals from south-west Scotland are grouped on one branch, with the rest of Scotland and England placed on a separate branch. fineStructure tree building is sensitive to the sample size, and due to the larger proportion of Irish genomes in our analysis, compared to Leslie et al .’s analysis (300 versus 44), it is not surprising that the Irish branch is placed differently.

We observe substructure within the Irish Traveller population, identifying (via fineStructure) four genetic clusters occupied only by Irish Travellers ( Fig. 1B ). These clusters align with the broad two way split in the Irish Traveller population we observe via allele frequency based PCA ( Fig. S1 ). In addition, our fineStructure clusters reflect sociolinguistic affinities of the population, membership of the Rathkeale group ( Traveller 2), and speakers of the Cant ( Traveller 4 ) or Gammon ( Traveller1 ) dialects of Shelta ( Fig. S2 ). Our results, therefore, suggest that these groups represent genuine structure within the Irish Traveller population, rather than having by chance sampled broad family groups.

Several Irish Traveller individuals in the fineStructure analysis show an affinity either with British or settled Irish, demonstrating some genetic heterogeneity within the Irish Traveller population. This heterogeneity can be explained by recent settled ancestry or ancestry with other Travelling groups within Britain and Ireland. However, the existence of sole Irish Traveller genetic clusters suggest that there is some sub-structure within the population, and a larger follow up study is warranted to elucidate the extent of this structure, and the representative nature of the observed clusters.

It appears that the Traveller population has experienced lineage-specific drift, as demonstrated by the discordant F st and f 3 estimates between the Travellers and the settled Irish. F st estimates of Traveller to Settled Irish genetic distance are comparable to that we observed between the Ireland and Spain ( Table S1 ). However, when we estimate using f 3 statistics (which is less sensitive to lineage-specific drift) the genetic distance, is reduced, and comparable to that observed between Irish and Scots. The theory of lineage-specific drift is also supported by the IBD analysis, which demonstrates very high levels of haplotype sharing within the Traveller population. Indeed, much of the overall genetic differentiation of the Travellers from the settled Irish is driven by the high F st distance between the Irish Traveller PCA group B (specifically the Traveller 4 cluster), and the settled Irish. This suggests that some subgroups within the Irish Travellers may have experienced greater genetic drift than others.

The dating of the origin of the Irish Travellers is of considerable interest, but this is distinct from the origins of each population. We have estimated the point of divergence between the Traveller and the settled Irish population using two different methods. Our LD-based (T F ) method estimates a split 40 (±2 std.err) generations ago, or 1200 (±60 – std.err) years ago (assuming a generation time of 30 years). Our IBD-based method (T IBD ) estimates 12 (8–14) generations, or 360 (240–420) years ago. However both estimates suggest that the Irish Travellers split from the settled population at least 200 years ago. The Irish Great Famine (1845–1852) is often proposed as a/the source of the Irish Traveller population, but results presented here are not supportive of this particular interpretation. The T IBD method suggested differences between the PCA groups; whilst PCA group A seems to have split relatively early and remained relatively large, PCA group B seems to have split off more recently and quickly decline in size ( Table 2 ). This might explain the higher degrees of genetic differentiation we see in PCA group B in our F st and f 3 analyses.

An important limitation of our dating analysis is that both the T IBD and T F approaches assume a single origin source, but there may have been multiple founding events contributing to the population present today. Both methods are further limited in that they do not model for subsequent gene flow in to the population. We would also consider the T F date to be inflated, given the lineage-specific drift we and others have illustrated in the Traveller population, and its corresponding impact on F st calculation. In the case of the T IBD method, the sample size of the Irish Traveller cohort was too small to infer more complex demographic models (e.g. post-split gene flow or multiple epochs of growth/contraction for each group), due to the risk of over-fitting. A larger dataset is required to explore the possibility of dating distinct events for the Traveller clusters our analysis has resolved.

One of the hypothesised sources of the Irish Travellers is that they are a hybrid population between the settled Irish and the Roma. The results of our ADMIXTURE analysis would not support such a hypothesis, with none of the self-identified Irish Travellers showing ancestry components specific to the Roma populations. We did however detect one individual showing a significant proportion of a Roma-specific ancestral component. This individual self-reported Gypsie ancestry, and did not cluster with the clusters of sole Irish Traveller membership.

We have presented the first population-based assessment of autozygosity within the Irish Traveller population. Compared to other cosmopolitan populations, we observe within the Irish Travellers an excess of ROH and IBD segments. The ROH profile of the Irish Travellers is comparable to other consanguineous populations such as the Balochi of Pakistan and Druze of the Levant. However, of the populations we tested for ROH, the Irish Travellers were most similar to the European Roma, who are also an endogamous nomadic community. This, and the F ROH5 statistic for the Irish Travellers, agrees with previous observations of endogamy within the Irish Travellers 7 , 8 . Our homozygosity results would account for the well-documented higher prevalence of recessive disease within the Irish Traveller community 11 , 13 , 22 . The levels of homozygosity have clear importance in the medical genetics of the Irish Traveller population and together with the drift of rarer variants to higher frequencies in the Irish Travellers may greatly aid in the identification of rarer variants contributing to the risk of common disease within Ireland 23 , both for the settled and the travelling populations.

In summary, we confirm an ancestral Irish origin for the Irish Traveller population, and describe for the first time the genetics of the population using high-density genome-wide genotype data. We observe substructure within the population, a high degree of homozygosity and evidence of the “jackpot effect” of otherwise rare variants drifting to higher frequencies, both of which are of interest to disease mapping and complex trait genetics in Ireland. Finally we provide important insight to the demographic history of the Irish Traveller population, where we have estimated a divergence time for the Irish Travellers from the settled Irish to be at least 8 generations ago.

Materials and Methods

Study populations.

We assembled five distinct datasets; the Irish Travellers (n = 50), the Irish Trinity Student Controls 14 (n = 2232), the People of the British Isles dataset 15 (n = 2039), a dataset of individuals with European ancestry 24 (n = 5964), individuals with Roma ancestry 16 (n = 143), and a dataset of world-wide populations 19 (n = 931). For more details of each dataset, see Supplementary Data 1.1 .

The Irish Traveller cohort and data presented here were analysed within the guidelines and regulations put forward by the Royal College of Surgeons in Ireland Research Committee, and approved by the same Committee (reference number REC 1069). A waive of informed consent was granted by this Committee under an amendment of the same ethics reference number.

Quality Control of Genotype Data

Each of the five cohorts was individually processed through a number of quality control steps using the software PLINK 1.9 25 , 26 . Only autosomal SNPs were included in the analysis. Individuals or SNPs that had >5% missing genotypes, SNPs with a minor allele frequency (MAF) <2%, and SNPs failing the HWE at significance of <0.001 were discounted from further analysis. Identity-by-Descent (IBD) was calculated between all pairs of individuals in each of the five datasets using the—genome function in plink, and one individual from any pairs that showed 3 rd degree kinship or closer (a pihat score ≥0.09) was removed from further analysis. Amongst the Irish Traveller cohort eight cryptic pairings closer than second-degree cousins were found, leaving 42 individuals for further analysis.

Individuals included from the European ancestry dataset 24 were genotyped as part of a study of multiple sclerosis (MS), which included cases. As the HLA region contains loci strongly associated with multiple sclerosis (MS) 24 , for any analyses that included the European individuals from this MS study we omitted SNPs from a 15 Mb region around the HLA gene region, starting at 22,915,594 to 37,945,593. In order to restrict the MS cohort to individuals of European ancestry, we conducted principal component analysis (PCA) with gcta64 (v1.24.1) 27 and outliers from each of the MS populations were also removed. This left the final 5964 individuals included in the MS European Cohort.

Population Structure

FineStructure 28 analysis was carried out on a combined dataset of Irish Travellers, Trinity Student Irish, and POBI British. As fineStructure is more sensitive to relatedness, instead of the previously described IBD threshold we removed one from each pair with a pihat score >0.06. Additionally we removed SNPs that were either A/T or G/C. This left a combined dataset of 34 Irish Travellers, 300 randomly chosen Irish from the Trinity Student dataset, and 828 British from the POBI dataset. The POBI samples were selected as follows; 500 individuals were chosen from England, and all 131 from Wales, 101 from Scotland, and 96 from Orkney. In order for the English individuals to be as representative as possible of English clusters identified previously 21 , the 500 consisted of; 200 randomly chosen from Central/South England, 50 randomly chosen from each of Devon and Cornwall, and 200 randomly chosen from the north of England. This final combined dataset had a total coverage of 431,048 common SNPs. Further details of the fineStructure analysis pipeline and its parameters are described in Supplementary Data 1.2 .

In order to compare to other population structure visualisation methods we also performed allele frequency-based PCA using the software gcta64 (v1.24.1) 27 . Detailed methods are provided in Supplementary Data 3 . This was applied to the same dataset as the fineStructure analysis, with the exception that we first pruned the dataset with regards to LD using plink 1.9 25 , 26 with the—indep-pairwise command, using a window of 1000 SNPs moving every 50 SNPs, with an r 2 threshold of 0.2. We also removed common SNPs that were either A/T or G/C, leaving 75,214 common SNPs.

Maximum likelihood estimation of individual ancestries was carried out using ADMIXTURE version 1.23 29 and a dataset that had been pruned with respect to LD, as recommended by the authors 29 . This was achieved using plink 1.9 25 , 26 with the—indep-pairwise command, using a window of 1000 SNPs moving every 50 SNPs, with an r 2 threshold of 0.2. For this analysis we used a combined dataset of 42 Irish Travellers, 40 randomly selected Irish individuals from the Trinity Irish cohort, 160 individuals from the POBI dataset (40 randomly chosen English, Welsh, Orcadian, and Scottish individuals), and 40 random individuals from each of the following populations within the MS European dataset; France, Germany, Italy, Norway, Finland, and Spain. The combined dataset consisted of 83,759 SNPs (after the removal of A/T or G/C variants), and 476 individuals.

ADMIXTURE analysis was carried out on k  = 2–7 populations, with 50 iterations of each k value. The iteration with the highest log-likelihood and lowest cross validation score was used for further analysis.

Inter-population fixation indexes between the populations were studied using the Weir and Cockerham method 30 and the combined dataset used in ADMIXTURE analysis. The dataset was pruned with respect to LD using the same parameters as described above, leaving 83,759 common SNPs.

Due to the suspected lineage-specific drift in the Irish Traveller population history, we additionally calculated genetic distance using “outgroup” f 3 -statistics 31 , an extension of the f-statistics framework 32 . f 3 is proportional to the shared genetic drift between two test populations and an outgroup population, and should therefore be less sensitive to the Irish Travellers lineage-specific drift than the F st statistic. We performed this analysis on the same combined dataset used in F st analysis, with the additional inclusion of 21 Yorubans from the HGDP dataset in order to act as an outgroup to the pair-wise comparisons. The combined dataset consisted of 245,594 common SNPs (after the removal of A/T or G/C variants). The outgroup f 3 statistic was calculated using the software within the admixtools package 32 using default settings.

In order to estimate a time of divergence between the Irish Travellers and the settled Irish we utilised two methods. The first, the T F method, is based on a method first described by McEvoy et al . 33 and uses linkage disequilibrium patterns between markers in discrete bins of recombination distances, and genetic distance measured by F st in order to estimate a divergence time. The second, the T IBD method, uses the sharing of Identical by Descent (IBD) segments and demographic modelling using this sharing data to estimate a time of divergence and is based on the methodology previously described in Palamara et al . 17 and applied in Zidan et al . 18 . For more details of both methods, see Supplementary Data 1.3 .

Runs of Homozygosity Analysis

ROH analysis was carried out on a merged dataset of all individuals within the Irish Traveller, Trinity Student, and POBI cohorts, and a subset of the populations found within the Human Genome Diversity Project (HGDP) dataset. The HGDP populations were chosen to be i) representative of world-wide diversity of autozygosity, and ii) to compare the levels of autozygosity of the Irish Travellers to known endogamous populations such as the Balochi and Karitiana. The combined dataset had an overlap of 193,508 common markers.

With the exception of one parameter (the gap between consecutive SNPs, see below), we followed McQuillan et al .’s methodology 20 for the ROH analysis; the window was defined as 1000 kb, moving every 50 SNPs, with 1 heterozygous position allowed and 5 missing positions allowed within the window. The run of homozygosity call criteria were defined as; 1/5/10/16 Mb minimum in length, 100 SNPs minimum within the window, the minimum marker density greater than 50 Kb/SNP. Due to the reduced SNP coverage in this dataset compared to previous analyses 10 , 20 the largest gap between consecutive SNPs before ending a run of homozygosity call was changed to 500 Kb. We calculated F ROH5 as it had previously been shown to strongly correlate with the inbreeding coefficient F PED 20 . F ROH5 was estimated for the 17 populations, as per the equation below.

where S ROH5 is the total length of ROH found in an individual where runs are >5 Mb and L auto is the total length of the autosomal genome (called as 2,673,768 kb here). The F ROH5 was averaged across the individuals to find the population mean of F ROH5 .

Relationship to European Roma

We performed several analyses in order to investigate the relationship between Irish Travellers and European Roma. Firstly, we assembled a merged dataset that included the full Irish Traveller, Trinity Student, and European Roma datasets. We additionally removed any variants that were A/T or G/C. For subsequent PCA and ADMIXTURE analysis the combined Roma dataset was pruned for LD, using a window of 1000 SNPs, moving every 50 SNPs with a r 2 inclusion threshold of 0.2 in PLINK, leaving 66,099 common SNPs.

Secondly, PCA was performed using gcta64 v1.24.1 27 , creating a genetic relationship matrix, and then generating the first 10 principal components. Thirdly we applied ADMIXTURE on a reduced combined dataset that included all Irish Traveller and European Roma individuals, but only 40 of the Trinity Student Irish. ADMIXTURE was used with the same parameters as above, modelling for 2–4 ancestral populations. Finally, we compared the levels of homozygosity between the Irish Travellers, Trinity Student Irish, and European Roma - using the full combined Roma dataset, with 148,362 common SNPs and using the parameters described above.

Thirdly, we formally tested evidence for admixture using admixture f 3 statistics 32 in the form f 3 (Traveller; Settled, Roma) using the full Trinity Irish dataset, a reduced European Roma dataset excluding the Welsh Roma (due to their outlier status in the rest of the dataset 16 ), and a reduced dataset of Irish Travellers belonging to Irish Traveller clusters identified in fineStructure analysis (see Results). This combined dataset consisted of 148,914 SNPs.

Additional Information

How to cite this article: Gilbert, E. et al . Genomic insights into the population structure and history of the Irish Travellers. Sci. Rep. 7 , 42187; doi: 10.1038/srep42187 (2017).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Abdalla S. et al. Summary of the findings of the All Ireland Traveller Health Study . School of Public Health and Population Science, University College Dublin (2010).

Relethford, J. & M. Crawford Genetic drift and the population history of the Irish travellers. American Journal of Physical Anthropology , 150 (2), 184–189 (2013).

Article   Google Scholar  

Commission of Itinerancy 1963, Report of the Commission of Itinerancy 1963. Stationery Office, Government Publication: Dublin (1963).

North, K. E., Martin, L. J. & Crawford, M. H. The origins of the Irish travellers and the genetic structure of Ireland. Annals of Human Biology , 27 (5), 453–66 (2000).

Article   CAS   Google Scholar  

McCann, M., Síocháin, S. Ó. & Ruane, J. Irish Travellers: Culture and ethnicity . Institute of Irish Studies, Queens University of Belfast: Belfast (1994).

Equality Authority. Traveller ethnicity: an Equality Authority report . Brunswick Press Ltd: Dublin (2006).

Barry, J. & Kirke, P. Congenital Anomalies in the Irish Traveller Community. Irish Medical Journal . 90 (6), 233–7 (1997).

CAS   PubMed   Google Scholar  

Flynn, M. Mortality, morbidity and marital features of travellers in the Irish Midlands. Ir Med J . 79 (11), 308–10 (1986).

Woods, C. et al. Quantification of homozygosity in consanguineous individuals with autosomal recessive disease. Am J Hum Genet 78 (5), 889–96 (2006).

Kirin, M. et al. Genomic runs of homozygosity record population history and consanguinity. PLoS One . 5 (11), e13996 (2010).

Article   ADS   Google Scholar  

Murphy, M. et al. Genetic basis of transferase-deficient galactosaemia in Ireland and the population history of the Irish Travellers. European Journal of Human Genetics . 7 (5), 549–55 (1999).

Flanagan, J. et al. The role of human demographic history in determining the distribution and frequency of transferase-deficient galactosaemia mutations. Heredity . 104 , 148–55 (2010).

Murphy, A. et al. Incidence and prevalence of mucopolysaccharidosis type 1 in the Irish republic. Arch Dis Child . 94 (1), 82–4 (2009).

Desch, K. et al. Linkage analysis identifies a locus for plasma von Willebrand factor undetected by genome-wide association. Proc Natl Acad Sci USA 110 (2), 588–93 (2013).

Article   CAS   ADS   Google Scholar  

Winney, B. et al. People of the British Isles: preliminary analysis of genotypes and surnames in a UK-control population. Eur J Hum Genet . 20 (2), 203–10 (2012).

Mendizabal, I. et al. Reconstructing the Population History of European Romani form Genome-wide Data. Curr Biol . 22 (24), 2342–9 (2012).

Palamara, P. et al. Length distributions of identity by descent reveal fine-scale demographic history. Am J Hum Genet . 91 (5), 809–22 (2012).

Zidan, J. et al. Genotyping of geographically diverse Druze trios reveals substructure and a recent bottleneck. Eur J Hum Genet , 23 (8), 1093–9 (2015).

Article   MathSciNet   Google Scholar  

Li, J. et al. Worldwide human relationships inferred from genome-wide patterns of variation. Science 319 (5866), 1100–4 (2008).

McQuillan, R. et al. Runs of Homozygosity in European Populations. Am J Hum Genet . 83 (3), 359–72 (2008).

Leslie, S. et al. The fine-scale population structure of the British population. Nature . 519 , 309–14 (2015).

Hamid, N. et al. Rare metabolic diseases among the Irish travellers: results from the All Ireland Traveller Health Study census and birth cohort (2007–2011). Rare Diseases and Orphan Drugs 1 (2), 35–43 (2014).

Google Scholar  

Zuk, O. et al. Searching for missing heritability: designing rare variant association studies. Proc Natl Acad Sci USA 111 (4), E455–64 (2014).

IMSGC and WTCCC2. Genetic risk and a primary role for cell-mediated immune mechanisms in multiple sclerosis. Nature . 476 (7359), 214–219 (2011).

Chang, C. et al. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience . 4 , 7 (2015).

Purcell, S. et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet . 81 (3), 559–75 (2007).

Yang, J. et al. GCTA: a tool for genome-wide complex trait analysis . Am J Hum Genet . 88 (1), 76–82 (2011).

Lawson, D. et al. Inference of population structure using dense haplotype data. PLoS Genet . 8 (1), e1002453 (2012).

Alexander, D. H., Novermbre, J. & Lange, K. Fast model-based estimation of ancestry in unrelated individuals. Genome Research 19 (9), 1655–1664 (2009).

Weir, B. & Cockerham, C. Estimating F-statistics for the analysis of population structure. Evolution . 38 , 1358–70 (1984).

CAS   Google Scholar  

Raghavan, M. et al. Upper Palaeolithic Siberian genome reveals dual ancestry of Native Americans. Nature . 505 (7481), 87–91 (2014).

Patterson, N. et al. Ancient admixture in human history. Genetics . 192 (3), 1065–93 (2012).

McEvoy, B. et al. Human population dispersal “Out of Africa” estimated from linkage disequilibrium and allele frequencies of SNPs. Genome Research 21 (6), 821–9 (2011).

Download references

Acknowledgements

We would like to thank the members of the Irish Traveller population who participated in this study. The work was part funded by a Career Development Award (13/CDA/2223) from Science Foundation Ireland. We would also like to thank, Eoghan O’Halloran for help with data formatting, the Irish Center for High-End Computing (ICHEC) for the provision of computing facilities and support, Dan Lawson for advice and help with fineStructure, Michael McDonagh for helpful comments and insights into linguistic groups with the Irish Travellers, and Sinead Ní Shuinéar for inquiries on groups within the Irish Travellers. SC thanks a private donation from the Barouh and Channah Berkovits Foundation. We thank Liam McGrath and Scratch Films for their support in developing this project. We thank the reviewers for their helpful comments. This study makes use of data 24 generated by the Wellcome Trust Case-Control Consortium. A full list of the investigators who contributed to the generation of the data is available from www.wtccc.org.uk . Funding for the project was provided by the Wellcome Trust under award 76113, 085475 and 090355.

Author information

James F. Wilson and Gianpiero L. Cavalleri: These authors contributed equally to this work.

Authors and Affiliations

Molecular and Cellular Therapeutics, Royal College of Surgeons in Ireland, St Stephen’s Green, Dublin, 2, Ireland

Edmund Gilbert & Gianpiero L. Cavalleri

Braun School of Public Health and Community Medicine, The Hebrew University of Jerusalem, Jerusalem, Israel

School of Medicine and Medical Science, University College Dublin, Dublin, Ireland

Centre for Global Health Research, Usher Institute for Population Health Sciences and Informatics, University of Edinburgh, Teviot Place, Edinburgh, Scotland

• James F. Wilson

MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh, Scotland

You can also search for this author in PubMed   Google Scholar

Contributions

E.G., S.C., J.F.W., and G.L.C., wrote the main manuscript, E.G. ran the analysis, with exception of TIBD, which was run by S.C. S.E. contributed to supervision of E.G. J.F.W. and G.L.C. designed the study. All authors reviewed the manuscript.

Corresponding author

Correspondence to Gianpiero L. Cavalleri .

Ethics declarations

Competing interests.

The authors declare no competing financial interests.

Supplementary information

Supplementary data (pdf 1190 kb), rights and permissions.

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Reprints and permissions

About this article

Cite this article.

Gilbert, E., Carmi, S., Ennis, S. et al. Genomic insights into the population structure and history of the Irish Travellers. Sci Rep 7 , 42187 (2017). https://doi.org/10.1038/srep42187

Download citation

Received : 23 August 2016

Accepted : 04 January 2017

Published : 16 February 2017

DOI : https://doi.org/10.1038/srep42187

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

The newfoundland and labrador mosaic founder population descends from an irish and british diaspora from 300 years ago.

• Edmund Gilbert
• Heather Zurel
• Michael S. Phillips

Communications Biology (2023)

Presentations of self-harm and suicide-related ideation among the Irish Traveller indigenous population to hospital emergency departments: evidence from the National Clinical Programme for self-harm

• Katerina Kavalidou
• Caroline Daly
• Paul Corcoran

Social Psychiatry and Psychiatric Epidemiology (2023)

The genetic landscape of polycystic kidney disease in Ireland

• Katherine A. Benson
• Susan L. Murray
• Peter Conlon

European Journal of Human Genetics (2021)

Microbiome and health implications for ethnic minorities after enforced lifestyle changes

• David M. Keohane
• Tarini Shankar Ghosh
• Fergus Shanahan

Nature Medicine (2020)

Runs of homozygosity: windows into population history and trait architecture

• Francisco C. Ceballos
• Peter K. Joshi

Nature Reviews Genetics (2018)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

  Skip navigation

• Births, Deaths & Marriages
• Crime and Justice
• Social Conditions
• Information Society
• Housing and Households
• Labour Market
• Agriculture
• Building and Construction
• Digital Economy
• Enterprise Statistics
• International Enterprises
• People in Business
• Retail and Services
• Small and Medium Enterprises
• Technology and Innovation
• Tourism and Travel
• International Accounts
• National Accounts
• Government Accounts
• External Trade
• Key Economic Indicators
• IMF Summary Data Page
• Agri-Environment
• Ecosystem Accounts
• Environment Accounts
• Water and Waste Water
• General Statistical Publications A-Z of Releases and Publications
• Show Survey

Census 2022 Profile 5 - Diversity, Migration, Ethnicity, Irish Travellers & Religion

• Irish Travellers

Census 2022 Results

This publication is part of a  series of results  from Census 2022. More thematic publications will be published throughout 2023 as outlined in the Census 2022  Publication Schedule .

The number of Irish Travellers living in the State and counted in Census 2022 was 32,949, an increase of 6% from 30,987 in the 2016 census. Irish Travellers make up less than 1% of the population so, for comparison purposes, it can be helpful to use rates per 1,000 of the population. This shows that in Census 2022, six out of 1,000 people in the State were Irish Travellers. The proportion of Irish Travellers in the population varied from county to county.

In Galway City, 21 out of every 1,000 people were Irish Travellers, in Longford, the rate was 20 per 1,000 of the population and in Offaly, it was 14 per 1,000.

Dún Laoghaire-Rathdown had the lowest number of Irish Travellers per 1,000 of the population with just under two Irish Travellers for every 1,000 people.

In Kildare and Dublin City, there were just under four Irish Travellers for every 1,000 people.

The Irish Traveller population increased in most counties, the largest rise being recorded in Offaly, up 30% to 1,174.

The Traveller population also increased by more than 200 in Cork (up 11% to 2,376), Fingal (up 17% to 1,545) and Tipperary (up 17% to 1,434).

There were drops in the number of Irish Travellers in some counties; the largest were recorded in Longford (down 13% to 913) and South Dublin (down 12% to 1,943).

Note: The analysis of Irish travellers is based on the usually resident population. The corresponding de facto figures in 2022 and 2016 were 33,033 and 31,075, respectively.

The figure for Irish Travellers has a pyramid shape as opposed to the hourglass shape of the figure for the total population. This reflects higher fertility rates and lower average life expectancy among the Irish Traveller population than in the overall population.

Children under the age of 15 made up 36% of Irish Travellers compared with 20% of the total population. At a national level, 15% of the total population was aged 65 years and over while for Irish Travellers, the equivalent figure was just 5%.

Marital Status of Irish Travellers

Overall, 45% of Irish Travellers aged 15 years and over were single, up from 40% in 2016. The proportion of married Travellers dropped from 49% in 2016 to 44% in 2022.

Irish Traveller men were more likely to be either single (47%) or married (46%) than Irish Traveller women (42% single and 42% married).

Around 10% of Irish Traveller women were separated or divorced compared with 5% of Irish Traveller men.

Irish Traveller women were also more likely to be widowed (5%) than Irish Traveller men (2%).

Over 85% of Irish Travellers aged 15 to 24 years were single while 13% were married.

The proportion that were married increased to 49% for 25 to 34 year olds.

Among Irish Travellers aged 55 to 64 years, 14% were separated or divorced compared with 8% of Travellers aged 65 and over.

Overall, 25% of Irish Travellers aged 65 and over were widowed; the figure for Traveller women aged 65 and over was 35% and 15% for Traveller men.

Long-Lasting Conditions and Difficulties

There were 8,577 Irish Travellers who reported experiencing at least one long-lasting condition or difficulty to any extent, accounting for 26% of the Traveller population. In comparison, 22% of the total population living in the State reported experiencing at least one long-lasting condition or difficulty to any extent.

Breaking this down further, 15% of Irish Travellers (4,952 people) reported experiencing at least one long-lasting condition or difficulty to a great extent or a lot compared with 8% of all people living in Ireland.

Another 11% of Irish Travellers (3,625 people) reported experiencing at least one long-lasting condition or difficulty to some extent or a little while the comparable figure for the total population was 14%.

The overall proportion of Irish Travellers experiencing a long-lasting condition or difficulty to any extent was slightly higher for men (27%) than women (25%). Looking at the total population, women (22%) were more likely to experience a long-lasting condition or difficulty to any extent than men (21%).

Of all children under the age of 15 living in the State, 4% reported experiencing at least one long-lasting condition or difficulty to a great extent compared with 7% of Traveller children.

The proportion of 15 to 29 year old Irish Travellers experiencing at least one long-lasting condition or difficulty to a great extent (13%) was more than twice that of all people in the same age cohort (6%).

Between the ages of 30 and 59, the proportion of the population experiencing at least one long-lasting condition or difficulty to a great extent was over three times higher for Irish Travellers (21%) than the total population (6%).

Among the older age cohorts, the differences were less pronounced, and Irish Travellers over the age of 80 were slightly less likely to experience a long-lasting condition or difficulty to any extent than would be expected in the overall population.

General Health

The question on general health shows that 22,050 Irish Travellers reported their general health as being good or very good (67%) while a further 3,899 Irish Travellers reported fair health status (12%).

There were 1,350 Irish Travellers reporting their health as bad or very bad, 4% of the Traveller population. This is twice as high as the proportion of the total population who reported their health as bad or very bad (2%).

The level of non-response in this question was quite high for Irish Travellers, at 17%, compared with 7% for the total population.

In the overall population, the proportion of people with good or very good health decreased slowly with age, up until the age of 70 when the decrease rate started to accelerate.

In the Irish Traveller population, the proportion of people with good or very good health decreased steadily with age up until the age of 70 at which point, the rate of decrease slowed down.

There were 5,427 Irish Travellers who were daily smokers in Census 2022, or 16% of the Traveller population compared with 9% of the total population.

Just under half of Irish Travellers had never smoked compared with 60% of the total population.

Some 9% of Travellers had given up smoking, compared with 19% for the total population.

Looking at smoking by age shows that one in three Irish Travellers between the ages of 25 and 54 were daily smokers.

Irish Traveller Households

There were 29,900 Irish Travellers living in private households in Census 2022. The majority were living in permanent housing, while 2,286 people were living in temporary housing units such as caravans and mobile homes.

The proportion of Irish Travellers living in private households who were living in caravans, mobile homes or other temporary accommodation was 8% in 2022, down from 12% in 2016.

In Fingal, 18% of Travellers were living in temporary accommodation, the highest proportion in the country in Census 2022.

In Dublin City, Kilkenny and Tipperary, 14% of Irish Travellers were living in temporary housing.

Household Size

There were 9,448 private households containing Irish Travellers. These households had an average size of 4 persons per household compared to an average size of 2.7 for the total population.

Irish Traveller households were largest in Leitrim, Roscommon and Kildare with an average size of 4.6 persons, followed by Clare with 4.5 persons per household.

The counties where the average size of Irish Traveller households was smallest were Dublin City with 3.5 persons per household and Louth, Dún Laoghaire-Rathdown and Donegal (all with 3.6 persons per household).

• Infographic
• Key Findings
• Citizenship
• Immigration
• Ethnic Group/Background
• Background Notes
• Contact Details
• Press Statement
• Publication Briefing
• Audio Files
• Census 2022 Profile 5 County Press Statements
• Census of Population 2022

Interactive Visualisation App

Explore census data from county level to neighbourhood (Small Area) level

Central Statistics Office Skehard Road, Cork T12 X00E, Ireland

• © 2023
• Freedom of Information
• Accessibility
• Data Protection & Transparency
• Privacy Statement

You can count on a rewarding career with the CSO.

Learn about our variety of roles and the benefits of working with the CSO.

• Open access
• Published: 29 April 2024

Perceptions of and preparedness for cross-cultural care: a survey of final-year medical students in Ireland

• Lesley O’Brien 1 ,
• Nicola Wassall 1 ,
• Danielle Cadoret 1 ,
• Aleksandra Petrović 1 ,
• Patrick O’Donnell 1 &
• Siobhán Neville 1  

BMC Medical Education volume  24 , Article number:  472 ( 2024 ) Cite this article

263 Accesses

4 Altmetric

Metrics details

Migration is increasing globally, and societies are becoming more diverse and multi-ethnic. Medical school curricula should prepare students to provide high-quality care to all individuals in the communities they serve. Previous research from North America and Asia has assessed the effectiveness of medical cultural competency training, and student preparedness for delivery of cross-cultural care. However, student preparedness has not been explored in the European context. The aim of this study was to investigate how prepared final-year medical students in the Republic of Ireland (ROI) feel to provide care to patients from other countries, cultures, and ethnicities. In addition, this study aims to explore students’ experiences and perceptions of cross-cultural care.

Final-year medical students attending all six medical schools within the ROI were invited to participate in this study. A modified version of the Harvard Cross-Cultural Care Survey (CCCS) was used to assess their preparedness, skill, training/education, and attitudes. The data were analysed using IBM SPSS Statistics 28.0, and Fisher’s Exact Test was employed to compare differences within self-identified ethnicity groups and gender.

Whilst most respondents felt prepared to care for patients in general (80.5%), many felt unprepared to care for specific ethnic patient cohorts, including patients from a minority ethnic background (50.7%) and the Irish Traveller Community (46.8%). Only 20.8% of final-year students felt they had received training in cross-cultural care during their time in medical school. Most respondents agreed that they should be assessed specifically on skills in cultural competence whilst in medical school (83.2%).

Conclusions

A large proportion of final-year medical students surveyed in Ireland feel inadequately prepared to care for ethnically diverse patients. Similarly, they report feeling unskilled in core areas of cross-cultural care, and a majority agree that they should be assessed on aspects of cultural competency. This study explores shortcomings in cultural competency training and confidence amongst Irish medical students. These findings have implications for future research and curricular change, with opportunities for the development of relevant educational initiatives in Irish medical schools.

Peer Review reports

Patient-centred care is linked to improved health outcomes for patients and represents a pillar of quality in healthcare delivery [ 1 ]. However, providing patient-centred care becomes more complex with increasing diversity of patient populations. Miscommunication and misunderstandings in the clinical setting can lead to patient dissatisfaction, reduced adherence to treatment regimens and poor health outcomes [ 2 ]. Cross-cultural competence is an important factor in the ability of clinicians to deliver appropriate care to patients from different sociocultural backgrounds [ 1 ]. There have been multiple definitions of cross-cultural competence developed in the literature, and for the purpose of this study cross-cultural competence can be taken to mean a shared knowledge from collective experiences of diverse groups and the integration of behaviours and attitudes by healthcare professionals to empower them to engage effectively and collaboratively with patients from these diverse backgrounds [ 3 ]. Healthcare providers skilled in cross-cultural care can improve quality of care for minority ethnic groups and help eliminate health disparities by improving communication with patients, building trust, and overcoming gaps in understanding [ 2 ]. Therefore, it can be argued that cross-cultural competence is an essential skill for clinicians and should be included in medical school curricula.

To provide appropriate cross-cultural care, clinicians must engage in effective communication with, and provide high quality care to, patients from diverse sociocultural backgrounds [ 4 ]. Whilst there is no accepted definition of cultural competence, Betancourt et al . [ 5 ] described cultural competence training as specific efforts to enhance knowledge of sociocultural factors, health beliefs and behaviours held by patients, with an aim to develop skills to manage these factors in the delivery of equitable health care. Ultimately, training should help clinicians understand the impact of sociocultural factors on a patient’s health. However, there is variability in the methods, timing, and quality of this training [ 6 ]. Some institutions prioritise theory over practical skills, and many fail to address bias and disparities in healthcare [ 6 ]. Cross-cultural training requires standardisation to consistently produce culturally competent clinicians.

There has been a drive to improve cross-cultural care training in medicine, as studies in various countries have shown that both medical students and practising clinicians feel unprepared to deliver patient-centred, cross-cultural care [ 1 , 2 , 6 ]. Using the Cross-Cultural Care Survey (CCCS), a tool developed and validated for assessing cultural competency in medicine, Green et. al. [ 6 ] reported that final year Harvard medical students felt they lacked experience with diverse patient populations and experienced dismissive attitudes towards cross-cultural training from educators. As a result, they felt unprepared in many facets of delivering cross-cultural care. Medical students in Taiwan reported no improvement in preparedness to deliver cross-cultural care or address health inequities, as they progressed from preclinical to clinical training [ 7 ]. In Pakistan, researchers found that there was little difference between medical school year groups in their preparedness to care for patients with cultural customs and/or beliefs with the potential to affect clinical care [ 8 ]. In Switzerland, Casillas et al. [ 2 ] surveyed a group of healthcare providers, which included physicians and clinical nurses, and found that participants felt least prepared to care for patients whose religious beliefs affect treatment, and working in a department that provided some form of cross-cultural training was associated with higher levels of preparedness. Hudelson et al. [ 9 ] assessed the communication skills of both healthcare providers and local medical students when caring for migrant patients, which found that medical students scored lower than their qualified colleagues in clinical skills, intercultural communication skills and general intercultural skills.

There have been several studies conducted in the United Kingdom (UK) examining the cultural awareness of medical students and delivery of cultural competency training [ 10 ]. Studies show that UK medical students wish to be more aware of cultural differences in their patient population, and some students had not encountered any form of cultural competency training in their clinical curricula [ 11 , 12 ]. They recommended the incorporation of cultural competency training in both clinical and didactic material [ 12 ].

In the Irish context, the health of the Irish Traveller community raises particular equity concerns. This minority ethnic group faces higher mortality rates and lower average life expectancies than the general population, likely due to factors including discrimination, and access to health and social services [ 13 ]. This group was formally recognised as an indigenous ethnic minority in 2017 [ 14 ]. Diversity in Ireland has also been increasing, which adds to the complexity of delivering patient-centred care. From 2011 to 2016, the non-White-Irish population increased at a rate three times that of the White Irish ethnic majority. Preliminary results for the 2022 census revealed a population increase of 361,671, and estimated net immigration of 190,330 [ 15 ]. More recently, over 60,000 Ukrainian refugees arrived in Ireland in less than a year, with many requiring access to health services [ 16 ]. Despite these big changes, there is no research we can find that examines the preparedness of medical students in Ireland to provide cross-cultural care . The Health Service Executive (HSE), Ireland’s public health and social care service, recommends that academic institutions should integrate cultural competency training into undergraduate and postgraduate medical programmes [ 17 ]. Ireland’s changing demographics necessitate effective cross-cultural training in medicine to ensure all patients receive high quality care. Varying degrees of cross-cultural training have been employed by Irish medical institutions to provide students with skills required to navigate cross-cultural consultations. It is important to gauge the effectiveness of this training, particularly as there is no national standard in this area.

The aim of this study is to examine whether final-year medical students in Ireland feel prepared to provide high-quality care to patients from diverse cultural and ethnic backgrounds. Furthermore, this study aims to explore how these students have encountered this concept in their training thus far. In addition, we sought to explore the perceptions of students regarding ethnicity and health and identify potential areas to build on in medical school curricula. Researching the student perspective can provide medical educators with information on where students are receiving training in cross-cultural care, where students are finding greatest engagement, and areas of cross-cultural care in which students feel underprepared. Teaching basic skills required for navigating cross-cultural care consultations early in a medical student’s education establishes a foundation to build upon throughout their career and aid in the delivery of equitable healthcare for all patient cohorts.

Population and recruitment

We recruited final-year medical students, due to graduate in 2022, from both undergraduate and graduate-entry programmes in all six medical schools within the ROI: National University of Ireland, Galway (NUIG), Royal College of Surgeons in Ireland (RCSI), Trinity College Dublin (TCD), University College Cork (UCC), University College Dublin (UCD), and University of Limerick (UL). We used social media channels to distribute the survey, posting to final year medical student groups on Facebook, WhatsApp and Instagram platforms. We included a prize draw as an incentive to participate. Ethical approval was granted by the University of Limerick Faculty of Education and Health Sciences Research Ethics Committee.

Design and procedure

The Harvard cross-cultural care survey (CCCS) is a validated tool developed in the United States to assess cross-cultural competence in medicine and was adapted for the Irish context [ 6 ]. For example, questions were included regarding the Irish Traveller population. This cross-sectional survey design was used to collect both quantitative and qualitative data on four elements of cross-cultural care: i) preparedness, ii) skill, iii) training and education, and iv) attitudes. Information was collected on the medical training received by each participant, in addition to experiences outside of medical school. We assessed students during their final semester of medical school from January to June 2022. The survey was created using the online survey platform Qualtrics in line with the Harvard cross-cultural care survey layout. It was distributed via social media channels, and in addition, posters with a QR code link to the survey were placed in communal student settings in Irish hospitals.

Survey responses were stored and analysed using IBM SPSS Statistics 28.0. All components of the survey, including demographics, preparedness, skill, training and education, and attitudes were examined using frequency analyses. Fisher’s exact test was used to investigate statistically significant differences between gender and ethnicity groups, in reported preparedness and skill. Fisher’s Exact Test was chosen to provide accurate p values for low frequency samples in this study.

Participants were asked initially to disclose whether they were a final year medical student from the outset. Those who responded “no” were filtered out from the participation in the survey. Demographic data collected were analysed to investigate for gender and ethnicity difference in responses without cross-analysing two variables, e.g. institute and ethnicity or gender and ethnicity, to ensure confidentiality. Following completion of the survey, participants were offered the chance to participate in the prize draw. If accepted, participants were taken to a separate survey in which they were asked to include their email for a chance to win. No names were collected to protect confidentiality.

Pilot study

A pilot study of recently qualified doctors was carried out in 2021 to further refine the CCCS for the Irish context. The survey was distributed to recent graduates from six medical schools in the ROI via social media. The pilot study was completed by 49 participants. The reliability of the survey was tested using data collected from the pilot study. Cronbach’s alpha for the subscales within the survey were (α = 0.705–0.887) indicating acceptable reliability.

A total of 105 survey responses were collected from final-year medical students across the six medical schools in the ROI. Twenty-eight responses were excluded—four were not final year medical students and twenty-four were incomplete. There was a target population of approximately 1200 students, however the true number of students viewing the survey link is not known, therefore an exact response rate cannot be calculated.

Demographics

NUIG returned the most responses, (Table  1 (demographics), N  = 77). There was a higher proportion of female to male participants (75.3% and 24.7%, respectively). 57 participants self-identified as “ethnic majority” (74%), 19 self-identified as “ethnic minority”, (24.7%), and one did not disclose their self-identified ethnicity (1.3%).

Preparedness

Participants were asked to evaluate their perceived level of preparedness to care for patients in the contexts presented in Table  2 (Preparedness). 80.5% of participants felt prepared to care for patients in general. 50.7% felt prepared to care for patients from ethnic minorities, and 46.8% felt prepared to care for Irish Travellers specifically. Participants felt unprepared to care for new migrant patients (62.4%), patients with limited English proficiency (57.2%), and patients whose religious beliefs may affect clinical care (57.2%). There were statistically significant differences when comparing ethnicity groups. Ethnic minority participants felt more prepared to care for patients from racial/ethnic minority backgrounds (68.42% vs 43.86%, p  <  0.05 ), patients with limited English proficiency (26.32% vs 22.81%, p  <  0.05 ), and new migrant patients (26.32% vs 17.54%, p  <  0.05 ) when compared to those who did not identify this way. There were no statistically significant differences between gender groups.

Participants were *similarly prepared* to care for other minoritised patient communities, including those who identify as LGBTQIA + (54.6%) and those with disabilities (48.1%), as compared to patients from minority ethnic backgrounds (50.7%).

Participants were asked to evaluate their perceived level of skill in relation to the contexts presented in Table  3 (Skill). A high proportion of participants reported being skilled in adapting communication styles to fit a patient’s needs (80.5%) and building rapport with patients from ethnic backgrounds different to their own (76.6%). Participants reported a higher level of skill in identifying a patient’s understanding of spoken English (64.9%) compared to written English (42.9%). Participants reported being unskilled in working effectively with a medical interpreter (45.5%) and identifying religious beliefs and cultural customs that may affect clinical care (44.2%). Ethnic minority participants reported greater skill in identifying how well a patient understands verbal English than their ethnic majority counterparts (78.95% vs 59.65%, p  <  0.05 ). There were no statistically significant differences observed between gender groups.

Training and education

Participants were asked to evaluate how their educational experiences have prepared them to care for ethnic minority patients (Fig.  1 ). Participants identified experiences prior to, or outside of, the formal medical curriculum as the most useful in preparing them (36.8% “strongly agree”, 42.1% “somewhat agree”), followed by clinical electives (13.3% “strongly agree”, 40% “somewhat agree”) and formal clinical years (10.4% “strongly agree”, 41.6% “somewhat agree”). The pre-clinical education period (usually the first half of medical school training) was where a minority of students surveyed gained educational experience relevant to this topic, (6.5% “strongly agree”, 9.1% “somewhat agree”).

Training and Education: Experience. Participants were asked to self-evaluate the usefulness of the educational experiences presented above in preparing them to care for ethnic minority patients using a 5-point Likert scale from strongly disagree to strongly agree. % N  = percentage frequency of number of total participants, ( N  = 77)

Participants were asked to identify whether they had been exposed to various aspects of cross-cultural training whilst in their medical school (Fig.  2 ). Participants agreed they had practical experience caring for diverse patient populations during this time (16.9% “strongly agree”, 36.4% “somewhat agree”). A minority of participants felt they had not encountered diverse patient populations (22.1% “somewhat disagree”, 14.3% “strongly disagree”). Few agreed they had undergone cross-cultural training (9.1% “strongly agree”, 11.7% “somewhat agree”). A small majority agreed they had encountered positive attitudes to cross-cultural care amongst senior clinicians on placement, (16.9% “strongly agree”, 39% “somewhat agree”). A similar proportion encountered negative or dismissive attitudes amongst senior clinicians, (9.1% “strongly agree”, 33.8% “somewhat agree”). A majority of participants reported encountering positive attitudes towards cross-cultural care amongst their student peers (33.8% “strongly agree”, 32.5% “somewhat agree”). There were no statistically significant differences between gender or ethnic groups.

Training and Education: Exposure. Participants were asked to self-evaluate whether they had been exposed to the scenarios presented above using a 5-point Likert scale from strongly disagree to strongly agree. % N  = percentage frequency of number of total participants, ( N  = 77)

Participants were asked to evaluate how they felt their medical school had incorporated and prioritised the teaching of cross-cultural care. Many disagreed that their respective schools had incorporated cross-cultural issues into teaching (Fig.  3 A) (36.4% “somewhat disagree”, 23.4% “strongly disagree”). Similarly, they disagreed that their medical school had made the care of ethnic minority patients a priority for medical education (Fig.  3 B) (36.4% “somewhat disagree”, 40.3% “strongly disagree”).

Student Perceptions on Current Cross-Cultural Training in Medical School. Participant perceptions were assessed using a 5-point Likert scale from strongly disagree to strongly agree. A Participants were asked whether their medical school had incorporated cross-cultural issues into teaching and clinical care. Strongly disagree (23.4%), somewhat disagree (36.4%), neither agree nor disagree (10.4%), somewhat agree (22.1%), strongly agree (7.8%). B Participants were asked whether they felt their medical school makes learning about the care of ethnic minority patients a priority. Strongly disagree (40.3%), somewhat disagree (36.4%), neither agree nor disagree (7.8%), somewhat agree (9.1%), strongly agree (6.5%). % N  = percentage frequency of total participants, where N  = 77. No statistically significant differences were identified between ethnic groups or gender groups

Participant attitudes towards cross-cultural care were assessed. The majority of participants agreed that it is important to have clinical experience with diverse patient populations (Fig.  4 A) (93.5%). Furthermore, 83.2% of participants agreed that students should be assessed for their skills in cultural competence (Fig.  4 B).

Participant attitudes to cross-cultural care were assessed using a 5-point Likert scale from strongly disagree to strongly agree. A Participants were asked whether they felt it is important for medical students to have clinical experiences with a diverse mix of ethnic minority patients. “Strongly disagree” (0%), “somewhat disagree” (1.3%), “neither agree nor disagree” (5.2%), “somewhat agree” (9.1%), “strongly agree” (84.4%). B Participants were asked whether during medical school, students should be assessed for skills in cultural competence. “Strongly disagree” (5.2%), “somewhat disagree” (9.1%), “neither agree nor disagree” (2.6%), “somewhat agree” (37.75%), “strongly agree” (45.5%). Total number participants, ( N  = 77)

Finally, participants were invited to share ideas of how cross-cultural care may be further incorporated into their learning. Fourteen respondents put forward their thoughts on ways to deliver effective cross-cultural care within their curriculum. These may be considered under three broader categories—the method of delivery, the resources used, and the content delivered. Respondents felt methods of delivery should include specified lectures and/or modules on cross-cultural care, clinical sessions with patients from diverse backgrounds, and opportunities for involvement in community initiatives delivering care to minority populations. Respondents highlighted the need for learning resources that are inclusive of diverse patient populations. Finally, respondents highlighted a need for specific training in key content areas, including unconscious bias and working with medical interpreters.

This is one of the first studies in a European context to evaluate medical students’ preparedness to care for diverse patient populations using a validated survey tool and identifies areas of need to equip students to provide high-quality cross-cultural care. Previous studies have provided limited insight into the medical student perspective on aspects of cross-cultural training, usually as part of wider studies focusing on the perceptions of clinicians or schools delivering training [ 9 , 12 ].

The self-reported preparedness points to a specific lack of experience engaging with patients from diverse backgrounds. With a majority of respondents reporting proficiency in cross-cultural skills assessed in this survey, this may reflect an under-confidence in self-reported preparedness of Irish medical students. The difference in preparedness between self-identified ethnic groups, whereby ethnic minority students reported greater preparedness in cross-cultural care, may be due to a shared experience of being minoritised in society and finding a commonality between being from cultures outside of the ethnic majority. Other studies have found similar results, with participants that identify as ethnic minority or even sexual minority reporting greater preparedness to care for patients from different cultural backgrounds and different sexual orientations [ 8 , 18 ]. It may suggest that these experiences may be collected, shared, and taught in cross-cultural education, so that future clinicians are able to better understand their diverse patients and ultimately deliver better care.

Preparedness to care for LGBTQIA + patients and patients with disabilities were included in the survey, as these groups also often face barriers to care. While few survey respondents felt prepared to care for these communities, the figures were similar to those collected in the USA and Taiwan [ 6 , 7 ]. This suggests an area to improve upon when cultivating student preparedness to care for other diverse populations, populations that are often neglected or discriminated against in the health care setting despite a potentially shared culture or ethnicity [ 18 ].

The Skills section highlighted specific areas of student concern, which can direct the development of future medical curricula on cross-cultural training. Participants in this study identified areas where they felt least skilled in delivering cross-cultural care, including identifying religious or cultural beliefs affecting clinical care and working effectively with a medical interpreter. They highlighted experiences outside the formal medical curriculum as most preparatory in building their cross-cultural competence. Research indicates that the informal or “hidden” curriculum, where students encounter a variety of patient populations and learn through direct observation, immersion, and interaction with these diverse groups, plays a crucial role in developing their cultural competence. This unintentional learning process is essential in enhancing their ability to effectively work across different cultural contexts [ 19 ]. This suggests that schools should offer and encourage elective opportunities or volunteering placements within diverse communities, as they are a rich source of cross-cultural education. Unfortunately, medical schools in Ireland are often limited by geographical location and availability of clinical placements. However, this may be an area schools can improve upon as the Irish population continues to diversify rapidly.

Despite varying cross-cultural training programs implemented in Irish medical schools, many participants in this survey felt they had not received training in cross-cultural care during their time in medical school, nor was it felt to be a priority in their curriculum. Further reinforcement of these programmes should be implemented across all years of medical school, via both theoretical and practical means. As per participant suggestions put forth in this survey, lectures, small group sessions, involvement in local community programmes, and dedicated cross-cultural clinical sessions could be implemented to enhance cultural competency. A cultural humility approach has been shown to be beneficial, which incorporates self-reflection in cross-cultural training [ 12 ]. A scoping review by Brottman et al . [ 19 ], revealed eleven cross-cultural educational methods to cultivate cultural competence, whilst Liu et al . [ 20 ], demonstrated the ways in which the hidden curriculum can influence cross-cultural competence. From these studies, multiple methods of cultivating cross-cultural competence can be utilised, and there is no method has been proven superior to another [ 19 ].

The majority of participants agreed that they should be assessed for skills in cultural competence during medical school training. Schools may look to assess this in Objective Structured Clinical Exam (OSCE) stations. There have been previous calls for greater use of objective measures of assessment of cultural competence in the literature to date [ 6 , 8 , 11 ]. A recent review by Deliz. et al. [ 21 ] found that the most commonly adopted assessment modality of cross-cultural care training in medical schools were pre- and post-training self-assessment surveys, but other forms of assessment included objective measures, namely knowledge-based tests and standardised patient encounters. It is unknown whether the medical schools listed in our study have implemented objective assessments for cultural competence among their student population.

The Attitudes section suggests that survey participants have encountered negative or dismissive attitudes towards cross-cultural care in clinical settings. This follows findings from UK studies, which revealed that ethnic minority students specifically felt isolated and subject to stereotyping by clinicians whilst on placement [ 12 , 22 ]. This suggests that clinical staff should also be exposed to cross-cultural training as role models for future health care professionals [ 19 ]. Fortunately, positive attitudes greatly outweighed negative attitudes amongst the participants’ own peer groups.

One limitation of this study was the low number of responses received, which may have been impacted by our method of recruitment and timing of our data collection. Our data were collected through indirect social media channels, therefore the number of medical students that had the potential to interact with our survey was unknown. There was a low response rate from some institutions compared to others, namely TCD, RCSI, and UCC, again likely due to method of recruitment, thus the data cannot be taken to represent all undergraduate and graduate medicine courses in the ROI. The data collection took place in the latter half of the final year, a time when student anxiety regarding final exams is high. This may have impacted the rate of participation observed in our study. While there is no public data available regarding the ethnic makeup of the medical student population in Ireland, our survey received responses representing self-identified majority and self-identified minority student perspectives. Though this may not reflect the national average, this response ratio ensured representation from both cohorts. The interpretation of data inferred from ethnicity differences cannot be overstated due to the low total number of responses. Also, it should be noted that this study asked students to self-identify as ethnic minority or majority, which may inherently pose difficulty for some.

Students engaging in a survey on cross-cultural care are likely interested in this area of medical education, which may influence the responses. This survey asked students to self-report their feelings of preparedness and skill and may not be a true reflection of their abilities. Students may feel unprepared at this stage of their career due to “imposter syndrome” or anxiety about entering the workforce, which may create a negative self-perception bias [ 23 ]. There is limited data published regarding how the schools represented in this survey implement their training in cross-cultural care. Finally, this survey tool relies on participant recall, introducing potential for recall bias as observed in similar studies [ 7 ]. Suggestions for further research include repeating this survey with alternative recruitment methods to boost response rates and collect data representative of all ROI medical institutions, assessing students’ preparedness for diverse patient populations during different stages of their medical education. The preparedness of medical students to care for patients with disabilities and/or patients from LGBTQIA + communities should be further explored. Finally, it would be advisable to assess non-hospital consultant doctors’ (NCHDs) preparedness to care for diverse patient populations in Ireland.

This was the first study assessing the perceptions of final-year medical students across Irish universities in their preparedness, skill, and attitudes towards cross-cultural care. This survey has helped to clarify the student perspective on current cross-cultural training employed by medical schools, with students reporting an unpreparedness to care for diverse patient cohorts. It highlights areas in which students do not feel adequately trained to deliver cross-cultural care. Most students have a positive perception of cross-cultural competence and feel it is important to incorporate cross-cultural competence into their education to ensure the delivery of equitable health care to diverse patient cohorts. Students expressed how they hoped to see more cross-cultural competency training, including further lectures, modules, and clinical sessions added to their curriculum. This survey has highlighted areas of medical education that students desire further training in to develop their skills in cross-cultural competence.

Availability of data and materials

The datasets generated and analysed during the current study are not publicly available due to the potential for individual privacy to be compromised but are available from the corresponding authors on reasonable request.

Abbreviations

Cross-Cultural Care Survey

Health Service Executive

Lesbian, Gay, Bisexual, Transgender, Queer, Intersex, Asexual + 

Non-Consultant Hospital Doctor

National University of Ireland, Galway

Observed Structured Clinical Exam

Royal College of Surgeons in Ireland

Republic of Ireland

Trinity College Dublin

University College Cork

University College Dublin

United Kingdom

University of Limerick

United States of America

Institute of Medicine Committee on Quality of Health Care in, A. Crossing the quality chasm: a new health system for the 21st century. Washington (DC): National Academies Press (US). Copyright 2001 by the National Academy of Sciences. All rights reserved; 2001.

Google Scholar  

Casillas A, et al. Is the front line prepared for the changing faces of patients? Predictors of cross-cultural preparedness among clinical nurses and resident physicians in Lausanne, Switzerland. Teach Learn Med. 2015;27(4):379–86. https://doi.org/10.1080/10401334.2015.1077127 .

Article   Google Scholar  

Castillo RJ, Guo KL. A framework for cultural competence in health care organizations. Health Care Manag (Frederick). 2011;30(3):205–14. https://doi.org/10.1097/HCM.0b013e318225dfe6 . PMID: 21808172.

Betancourt JR, Cervantes MC. Cross-cultural medical education in the United States: key principles and experiences. Kaohsiung J Med Sci. 2009;25(9):471–8. https://doi.org/10.1016/s1607-551x(09)70553-4 .

Betancourt JR, et al. Defining cultural competence: a practical framework for addressing racial/ethnic disparities in health and health care. Public Health Rep. 2003;118(4):293–302. https://doi.org/10.1093/phr/118.4.293 .

Green AR, et al. Measuring medical students’ preparedness and skills to provide cross-cultural care. Health Equity. 2017;1(1):15–22. https://doi.org/10.1089/heq.2016.0011 .

Lu PY, et al. Assessing Asian medical students’ readiness for diversity: localizing measures of cross-cultural care competence. Teach Learn Med. 2021;33(3):220–34. https://doi.org/10.1080/10401334.2020.1830097 .

Amanullah FS, et al. Assessing medical students’ perception of cross-cultural competence at a private University in Karachi. BMC Med Educ. 2022;22(1):534. https://doi.org/10.1186/s12909-022-03588-0 .

Hudelson P, Perron NJ, Perneger T. Self-assessment of intercultural communication skills: a survey of physicians and medical students in Geneva, Switzerland. BMC Med Educ. 2011;11:63. https://doi.org/10.1186/1472-6920-11-63 .

Dogra N, Reitmanova S, Carter-Pokras O. Teaching cultural diversity: current status in U.K., U.S., and Canadian medical schools. J Gen Intern Med. 2010;25 Suppl 2(Suppl 2):S164-8. https://doi.org/10.1007/s11606-009-1202-7 .

Stretch NDD. Developing a questionnaire to assess student awareness of the need to be culturally aware in clinical practice. Med Teach. 2001;23(1):59–64. https://doi.org/10.1080/0142159002005613 .

Nazar M, et al. Decolonising medical curricula through diversity education: lessons from students. Med Teach. 2015;37(4):385–93. https://doi.org/10.3109/0142159x.2014.947938 .

Health, D.o. All Ireland traveller health study: our Geels - summary of findings. All-Ireland Traveller Health Study. Department of Health; 2010. https://www.gov.ie/en/publication/b9c48a-all-ireland-traveller-health-study/ .

Haynes A, Joyce S, Schweppe J. The significance of the declaration of ethnic minority status for Irish travellers. Natl Pap. 2021;49(2):270–88. https://doi.org/10.1017/nps.2020.28 .

CSO, C.S.O. Census of population 2022 - preliminary results. 2022.

CSO, C.S.O. Arrivals from Ukraine in Ireland series 4. 2022.

HSE, N.S.I.O. Second national intercultural health strategy 2018–2023. 2018.

Badat A, Moodley S, Paruk L. Preparedness of final year medical students in caring for lesbian, gay, bisexual, and transgender patients with mental illness. S Afr J Psychiatr. 2023;29:1998. https://doi.org/10.4102/sajpsychiatry.v29i0.1998 .

Brottman MR, Char DM, Hattori RA, Heeb R, Taff SD. Toward cultural competency in health care: a scoping review of the diversity and inclusion education literature. Acad Med. 2020;95(5):803–13. https://doi.org/10.1097/ACM.0000000000002995 . PMID: 31567169.

Liu J, Li S. An ethnographic investigation of medical students’ cultural competence development in clinical placements. Adv Health Sci Educ Theory Pract. 2023;28(3):705–39. https://doi.org/10.1007/s10459-022-10179-7 .

Deliz JR, et al. Cultural competency interventions during medical school: a scoping review and narrative synthesis. J Gen Intern Med. 2020;35(2):568–77. https://doi.org/10.1007/s11606-019-05417-5 .

Morrison N, Machado M, Blackburn C. Student perspectives on barriers to performance for black and minority ethnic graduate-entry medical students: a qualitative study in a West Midlands medical school. BMJ Open. 2019;9(11):e032493. https://doi.org/10.1136/bmjopen-2019-032493 .

Khan M. Imposter syndrome—a particular problem for medical students. BMJ. 2021;375:n3048. https://doi.org/10.1136/bmj.n3048 .

Download references

Acknowledgements

The authors would like to thank Alisha Jaffer for her contribution to the conception and early development of this project, and Seoidín McKittrick for her time and expertise in analysing our pilot study dataset. Finally, the authors would like to thank the final-year medical students in NUIG, RCSI, TCD, UCC, UCD, and UL who shared the survey with their peers, and all the participants who completed the survey.

No funding was required for this study.

Author information

Authors and affiliations.

School of Medicine, University of Limerick, Limerick, Republic of Ireland

Lesley O’Brien, Nicola Wassall, Danielle Cadoret, Aleksandra Petrović, Patrick O’Donnell & Siobhán Neville

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the conceptualisation and design of the study. L’OB, NW, DC, and AP contributed to participant recruitment and data collection. LO’B performed statistical analysis of the results. PO’D and SN contributed to participant recruitment, and supervision of the study. LO’B and NW wrote a draft paper, which has been reviewed and revised critically by all authors. All authors approved the final version to be published and agree to be accountable for all aspects of the work.

Authors’ information

LO’B is an NCHD Intern at Beaumont Hospital, Dublin. NW is a Foundation Doctor FY1 at North Devon District Hospital. DC is a General Surgery Prelim/Interventional Radiology PGY1 at Corewell Health, Michigan State University. AP is a final year medical student at the School of Medicine, University of Limerick. PO’D is a General Practitioner and Associate Professor of General Practice at the School of Medicine, University of Limerick. SN is a Consultant General Paediatrician at University Hospital Limerick, and Associate Professor of Paediatrics at the School of Medicine, University of Limerick.

Corresponding author

Correspondence to Lesley O’Brien .

Ethics declarations

Ethics approval and consent to participate.

Informed consent was obtained from all students prior to their participation in the study once they had read and agreed to the terms of the ethical consent form. All methods were performed in accordance with the regulations and guidelines provided by the University of Limerick Faculty of Education and Health Sciences Research Ethics Committee. Approval to conduct this study was also granted by the University of Limerick Faculty of Education and Health Sciences Research Ethics Committee. Data collected included identifiers such as participant’s medical school, gender identity, and whether a participant identified as ethnic minority. Data was analysed only to investigate potential gender differences or ethnic identity differences, without cross matching two identifiers to ensure participants identities remained anonymous. Data collected is retained for 7 years in a password protected file in accordance with Irish and European Data Protection Law.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary material 1., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

O’Brien, L., Wassall, N., Cadoret, D. et al. Perceptions of and preparedness for cross-cultural care: a survey of final-year medical students in Ireland. BMC Med Educ 24 , 472 (2024). https://doi.org/10.1186/s12909-024-05392-4

Download citation

Received : 20 August 2023

Accepted : 04 April 2024

Published : 29 April 2024

DOI : https://doi.org/10.1186/s12909-024-05392-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Cross-cultural care
• Cultural competency
• Medical education

BMC Medical Education

ISSN: 1472-6920

• Submission enquiries: [email protected]
• General enquiries: [email protected]