International student migration research update: August 2017

Since 2016, Office for National Statistics (ONS) has published three updates on its research work on international student migration and has worked with other government departments on new analysis based on administrative and survey data sources. Today we published a summary note that aims to explain to users of our published international migration statistics what we have learned from our joint work on international student migration so far. This technical paper accompanies that summary and sets out in more detail important findings from our analysis.

This technical paper updates on our progress towards developing a better understanding of international student migration to and from the UK since our last update in April this year. We have been working closely with other government departments¹ on new analysis based on administrative data sources, as well as collaborating with the Centre for Population Change and Universities UK on a new Survey of Graduating International Students. This includes analysis with the Home Office on visa, Migrant Journey and Exit Checks data; and analysis of graduating international students’ post-study intentions.

The focus of this article is non-EU international students as they account for approximately 70% of all international student immigration and are subject to immigration control on how long they can stay in the UK. Due to these requirements, there are more available data sources to better understand non-EU student migration compared with EU student migration. The Home Office routinely publishes statistics on student visas, including a comparison for non-European Economic Area (EEA)² migration of trends shown by the International Passenger Survey (IPS).

In previous research updates, we have explained how the IPS shows a “gap” between the numbers of immigrants arriving to study and the numbers of emigrants who had previously immigrated to study. Since 2012, when this new measure of former student emigration was introduced, the “gap” has averaged around 100,000 (and was 73,000 in provisional estimates for the year ending December 2016). In the year ending March 2017, the “gap” was approximately 56,000 for non-EU students.

There is considerable interest in international student migration and the “gap” and concerns have been raised in the past about this implied “gap”. It has been suggested that the IPS may be implicitly overestimating the number of students remaining in the UK after their initial period of leave to study.

We recognise there are areas for improvement in the publication of international student migration statistics, as outlined in the recent Office for Statistics Regulation (OSR) report. We have been working on these improvements since 2016 as part of a programme of work to better understand what international students do at the end of their studies, and have published three updates on this. In February 2017 we set out plans more generally to develop data sources for international migration. Further detail will be published in September 2017.

Notes for: About this technical paper

1. Office for National Statistics (ONS) has been working collaboratively with the Home Office, Cabinet Office, Department for Education, and the Treasury to better understand the contribution that international students make to net migration.

2. Non-European Economic Area (EEA) here is defined as those who are not EU nationals or nationals of Iceland, Norway, Liechtenstein or Switzerland (Switzerland is not in the EEA, but have the same rights to live and work in the UK as EEA nationals).

International students coming to the UK are included in immigration, emigration and net migration figures, based on the International Passenger Survey (IPS) in line with international definitions. The IPS estimates migration flows – the number of people entering and leaving the UK for 12 months or more. Latest estimates indicate that total long-term net migration was 246,000 in the year ending March 2017; total immigration for study was 139,000 and emigration of former students was 59,000¹.

Estimates of visits to and from the UK lasting less than 1 year for study are also published by Office for National Statistics (ONS).

Different rules currently apply to EU and non-EU citizens, and specifically the movements of non-EU citizens are governed by the immigration rules and the terms of the visa they are granted.

Not all international students arrive to study in the UK and then leave at the end of their initial period of study. For non-EU students, some may apply to extend their studies or some may go on to work in the UK for a number of years but then eventually leave. Others may extend their stay in non-study visa categories (for example, to marry) and in due course, following a period of eligibility, then apply for permission to stay permanently (indefinite leave to remain).

When a student leaves the UK after the end of their initial period of leave to study they may have lived in the UK for more than a year, therefore adding to the long-term UK resident population. These behaviours do mean that students do contribute to long-term immigration both initially for the duration of their stay and over a longer term for those who stay permanently.

In our previous updates² we have explained some of the “gap” between immigration and emigration figures for students, but acknowledged part of this “gap” can’t be fully explained because some student outcomes may not be possible to measure with existing or even new data sources.

For non-EU students, we provide additional evidence in this article for:

• the numbers who stay after their initial period of leave to study in the UK and go on to further study, work or stay for other reasons

• the numbers who depart after their initial period of leave to study and don’t return within 12 months, or return within 12 months after a break

• students’ plans after their initial period of leave to study, and how certain they are of these plans

• a small number who have an unknown outcome

Notes for: Background information

1. Following the Office for Statistics Regulation (OSR) report published on 27 July 2017, The quality of the long-term student migration statistics, former student emigration estimates are now labelled as Experimental Statistics.

2. Our previous updates include Population Briefing: International Student Migration – What do the statistics tell us? (January, 2016); An update on international student migration statistics (November 2016); and International student research update (April 2017).

4.1. Latest figures

Latest provisional figures¹ indicate that in the year ending March 2017, long-term immigration for study was estimated to be 139,000 (+/- 18,000)², of whom 93,000 (+/- 15,000) were non-EU nationals. Long-term emigration by former students was estimated to be 59,000 (+/- 10,000), of whom 37,000 (+/-7,000) were non-EU nationals.

Since the 1990s, immigration for study by long-term migrants (those who stay for a year or more) has more than doubled from 56,000 (+/- 10,000) in 1991 to 136,000 (+/-18,000) in 2016. Much of this increase has happened since the early 2000s, reaching a peak of 238,000 (+/- 17,000) in 2010. Since then, immigration for study has fallen. Figure 1 shows these trends for long-term non-EU international immigration for study between 1991 and 2016.

A number of operational and policy changes were made to UK immigration rules³, which will be reflected in the fall in immigration for study since 2011. Much of this decrease has been concentrated in the further education sector as a result of tightening of the immigration system, following concerns raised regarding parts of that sector, and a new system designed to ensure better compliance by sponsoring institutions. Further detail on the statistics has been published by the Home Office, including regularly publishing a comparison of trends in immigration measured by the International Passenger Survey (IPS) and numbers of student visas granted.

4.2. Non-EU nationalities dominate, but the mix of nationalities has changed

Currently, around 70% of those immigrating to the UK for formal study are non-EU nationals and of these, more than four-fifths now apply for higher education; that is, to study at university rather than in the further education sector.

Since 2010, as well as a higher proportion of students arriving to study in the higher education sector, there has also been a change in the mix of non-EU nationalities immigrating to study in the UK.

In 2016, Chinese nationals accounted for approximately 49,700 (or 35%) of those granted Tier 4 student entry clearance visas of 1 year or more (main applicants); in 2010 they accounted for around 28,400 (15%). Indian students accounted for approximately 9,600 visas granted in 2016 (or 7%), but in 2010 they accounted for around 1 in 5 of visas granted (40,500).

Figure 2 shows the change in the number of student visas granted for 1 year or more between 2010 and 2016. These visas will include both students intending to study at higher education institutions and those studying in further education institutions or other education sectors. The numbers fell from 2010 to 2012 and have since been broadly flat, as reported by the Home Office.

4.3. International student journeys are complex and their intentions after their initial period of leave to study may change

The International Passenger Survey (IPS) provides estimates of two discrete groups of international students – those who are entering the UK to study long-term and those who are leaving the UK long-term who state they previously immigrated to study. We would therefore expect a difference between the two estimates because not all international students arrive to study in the UK and then leave immediately after their initial period of leave to study⁴.

The IPS estimates are based on a person’s intention to migrate, not actual behaviour. In reality students’ intentions or plans may change, which means that what they tell the IPS as they leave the UK may not reflect what they actually do. For example, they may say that they will return to the UK within 12 months, but don’t. Some students may have understated the length of time they have been living in the UK, particularly if they had short trips abroad during their studies.

Former students are identified in the IPS as a long-term emigrant⁵ if their answers to the IPS meet all of the following criteria if interviewed when they depart the UK:

• they have been living in the UK for 12 months or more

• they originally immigrated to study in the UK

• they intend to leave the UK for 12 months or more

Analysis of a small sample of student journeys⁶ using Home Office travel event data indicates the complexity of students’ journeys from arrival to study to eventual departure. Students selected in this sample spent both short and longer periods of time abroad during their studies in the UK. This analysis of a small number of cases suggests that international students spent periods of less than 12 months in the UK punctuated with short periods away during the main holiday periods (Christmas and summer).

This behaviour was also apparent for those international students who responded to our Survey of Graduating International Students⁷. Although the results of this survey have some limitations⁸ (Annex 1), they give some clear indicative results, with most (70%) who responded saying they had travelled outside the UK in either the Christmas, Easter or summer holidays during the previous year. These trips abroad may affect a person’s perception of how long they have been living in the UK. This illustrates the complexities of capturing information from international students to identify that they have been living in the UK for 12 months or more during their studies, the first of the three criteria used to identify long-term emigrants in the IPS.

Additional data on graduating international students’ post-study intentions and how sure they are of these intentions was also collected; 75% of responding non-EU graduating students planned to stay for less than 12 months in the UK after completing their courses. Overall, 46% of non-EU graduating students stated that they were not certain about their post-study migration plans. Of the quarter of responding students stating that they intended to stay in the UK or to remain longer than 12 months, 59% also indicated they were uncertain of their intentions (Figure 3). If students are uncertain of their plans after study, this will affect how they answer the third of the criteria used to identify long-term emigrants.

Notes for: The changing nature of international student migration

1. ONS International Passenger Survey figures on long-term international migration (based on the UN definition of someone changing their country of residence for 1 year or more).

2. Figures in brackets indicate the 95% confidence interval on estimates; that is, the range the true value will lie within.

3. The August Exit Check report details changes in the immigration system.

4. For more information, please see the International student research update (April 2017).

5. The UN recommended definition of a long-term international migrant: “A person who moves to a country other than that of his or her usual residence for a period of at least a year (12 months), so that the country of destination effectively becomes his or her new country of usual residence.” ONS includes students in international migration statistics in line with this internationally agreed definition.

6. A simple random sample of 10 non-EU students from each nationality (Chinese, Malaysian, Nigerian, Bangladeshi, Pakistani, Saudi Arabian and American (USA)) was selected.

7. The Survey of Graduating International Students was conducted between 13 March and 30 April 2017. The survey targeted all graduating international students at UK universities. A total of 3,560 responses were provided and the overall response rate was 3.5%. Further detail on this survey is available.

8. The percentages presented from the survey are subject to uncertainty and are not weighted to be representative of the wider student population. Further detail on the survey and representativeness is available.

In our April 2017 update we presented a conceptual diagram to describe more fully what non-EU students may do after their initial period of leave to study. This diagram identified where data sources are available on outcomes after study; where they’ll become available; and the outcomes for non-EU students that may not be possible to measure with existing or planned data sources.

We previously identified three main outcomes for non-EU students after their initial period of leave to study:

• they emigrate long-term or depart short-term

• they remain in the UK legally – for non-EU students, this means they will have obtained an extension to their visa for study or another purpose (such as work or family reasons) and it is expected that they would normally leave the UK at some point unless given further leave to remain or allowed to settle permanently

• they have an unknown outcome as no data captured in existing sources

Using Home Office Exit Checks data we have updated this diagram to show the outcomes for non-EU students who immigrated to study long-term (1 year or more) whose visas expired during 2015 and 2016 (Figure 4). We linked a further 12 months of travel event data to examine this cohort of students who returned within 12 months of their last departure. From this we are able to identify those who emigrated (did not return within 12 months or are assumed not to have returned as the evidence of a return is inconclusive), those who returned within 12 months and stayed long-term or on a short-term visit visa and those with no initially identified departure.

Figure 4: Outcomes for non-EU students with a leave expiry date between 8 April 2015 and 7 April 2016, UK

Notes:

1. This diagram is based on Exit Checks data.

2. The number of non-EU students who immigrated to study long-term (1 year or more) is based on those with an identified visa expiry date between 8 April 2015 and 7 April 2016 in Exit Checks data.

3. Those emigrating (for more than 12 months) were identified as those students with no evidence of return within 12 months of their last departure date, after linking a further 12 months of travel event data.

4. Those who departed the UK and returned within 12 months were identified as those students who had an arrival travel event within 12 months of their last departure date, after linking a further 12 months of travel event data. If there were multiple arrival travel events then the first arrival event was taken.

5. "Not conclusive" includes those who returned for an unknown duration or where the individual had a further instance of leave but no travel movement after their last departure date.

6. "No initially identified departure" in the Exit Checks data, which may be due to non-matching of individuals, departure via the Common Travel Area or other reasons, as well as those who left before April 2015.

In the following sections we compare Exit Checks data for non-EU students on study visas of 12 months or more, International Passenger Survey (IPS) estimates of all departing non-EU former students (previously resident in the UK for 12 months or more), and analysis from a Survey of Graduating International Students. Exit Checks data provide evidence of departures based on actual behaviour, the IPS provides estimates of departures based on intentions, whereas the Survey of Graduating International Students provides an indication of post-study plans and certainty of their plans.

5.1. How many non-EU students depart after their initial period of leave to study?

To better understand how many non-EU students departed the UK after their initial period of leave to study, we examined Home Office Exit Checks data for two cohorts of sponsored (Tier 4) students whose study visa or extension of leave within the UK (to study) expired between 8 April 2015 and 7 April 2016, and the 8 April 2016 and 7 April 2017 respectively. The Exit Checks programme commenced on 8 April 2015 and collected data on travellers departing and arriving in the UK. Further detail on the quality of data collected under the Exit Checks programme has been published by the Home Office.

This experimental analysis of Exit Checks data used a specially created dataset of approximately 366,400 non-EU students on long-term study visas (that is, for 1 year or more) whose leave to remain expired within the two cohort time periods. Long-term study visas were chosen as these were considered to be the group most likely to identify themselves as long-term migrants in the International Passenger Survey (IPS).

This analysis will therefore exclude the most recent arrivals who haven’t yet completed their studies, as well as older cohorts of students whose visa expired before April 2015 (who departed earlier or who did not obtain extensions of leave to remain further). The figures reported here for Tier 4 sponsored study will be different from those reported in the Home Office Exit Checks report. The Home Office figures exclude those whose entry clearance visa (or extension of leave) expired but were granted extensions of leave to remain (or settlement) whilst in the UK, allowing them to stay further. Our analysis includes this group.

The number of sponsored (Tier 4) students in the 2015 and 2016 cohort with a visa or extension to leave expiry date between 8 April 2015 and 7 April 2016 and a known status was approximately 193,100. This number was lower for the 2016 and 2017 cohort with a visa or extension to leave expiry date between 8 April 2016 and 7 April 2017, and a known status (170,300)¹.

A very large majority of students were identified as having departed before their visa or extension of leave expired (Figure 5) or as staying legally by applying for and being granted further leave to remain in the UK (either to study or for other purposes). The proportion departing in time for the 2015 and 2016 cohort was 64% (approximately 124,000) and for the 2016 and 2017 cohort was 69% (117,700). A further 26% (50,300) of the 2015 and 2016 cohort had applied for and been granted leave to extend their stay in the UK (whether to study or for other reasons). This proportion was also 26% for the 2016 and 2017 cohort (44,100).

Analysis of the 12 non-EU nationalities with the largest numbers of citizens whose study visas or extensions expired in 2016 and 2017 (including those who obtained further extensions of leave) highlights differences in the relative numbers initially identified as departing after their initial period of leave to study and extending leave to remain (Figure 6).

Thai, Chinese, Indian and North American (US) students were more likely to depart before their study visas or extensions expired, whereas Russian, Bangladeshi, Pakistani and Saudi Arabian students were more likely to extend their leave to remain.

Through linkage of a further 12 months of travel event data to the Exit Checks data, we are able to quantify for those who were initially identified as departing in 2015 and 2016:

• approximately 77% (99,100) didn’t return within 12 months and therefore would be indicated to be long-term emigrants using the UN definition

• 15% (18,900) returned on a short-term visit visa, based on their first recorded return in Exit Checks data, and departed again within 12 months (these would also be indicated to be long-term emigrants using the UN definition)

• 6% (7,600) returned on a long-term visa (12 months or more), based on their first recorded return in Exit Checks data, and would therefore not be indicated to be long-term emigrants using the UN definition

• for the remaining 3% (4,000), it wasn’t conclusive from the data that was automatically matched whether they returned or not

5.2. How does this compare to International Passenger Survey estimates of long-term emigration and other departures?

Exit Checks departure and length of time away from the UK are based on actual behaviour, whereas the International Passenger Survey (IPS) records intended length of departure. In Section 4.3 we explained how former students are identified as a long-term emigrant in the IPS by intended length of time away from the UK (12 months or more).

From the IPS, we don’t know if students who state an intention to return to the UK within a year actually do so. If these departing international students do return to the UK, they could return for further study, for work or for a different reason. When they return they will not be identified as long-term immigrants because they will have been out of the UK for less than 1 year. If they were then to leave the UK again but with an intention to leave for 12 months or more, they may be less likely to state a previous reason for visit of formal study (Table 2).

The number of former non-EU students identified as departing long-term in the Exit Checks data is higher than the IPS estimates of long-term emigration of former students. However, there are known reasons why the IPS estimates of long-term emigration of former students are lower, as we have previously reported. These include:

• international students’ intentions or plans may change, which means that what they tell the IPS as they leave the UK may not reflect what they actually do; therefore some may say they intend to depart short-term (3 to 12 months) but in fact stay away for longer (12 months or more)

• former students may not identify themselves as previous student immigrants; for example, students who extend their study visas or switch to other visa types would still be expected to leave the UK eventually (or acquire settlement); however, when they eventually leave they may not state that they originally immigrated to study in the UK

• some former students who acquire settlement may become British citizens and therefore not respond to the IPS as a foreign national

• a departing person won’t be identified as a former student immigrant if they state that they’ve not been living in the UK for 12 months or more (which may happen with short trips away)

• the IPS is not completely recording student flows, either due to sampling or other non-sampling errors not identified previously

These results indicate that the IPS figures for long-term emigration taken alone don’t provide a good indication of how many international students remain in the UK after their initial period of leave to study.

There may also be corresponding reasons why the IPS estimates of long-term immigration of students are lower than the actual number; for example, those who state an intention to migrate short-term to the UK, but switch to a long-term status and never leave, will not be identified in the Long-Term International Migration (LTIM) “switching adjustments”².

5.3. Non-EU students who remain in the UK legally

As discussed previously, not all non-EU international students arrive to study in the UK and then leave at the end of their initial study period. Some may apply to extend their studies and some may go on to work in the UK for a number of years but may eventually leave. Others may extend their stay in non-study categories (for example, to marry) and some may then apply for permission to stay permanently.

Currently, there are more data sources available on what non-EU students do after their initial period of leave to study than on EU students. Home Office data are available on extensions of stay in the UK. Extensions to stay include extensions granted to study or non-study visa categories such as work, family or other.

In 2016, a total of 44,400 grants of extensions of leave to remain were made to former students. Of these, 35,600 (80%) were extensions for further study, approximately 6,000 (14%) were extensions for work, and the remaining 2,800 (6%) were for family (for example, marriage) or other reasons (Table 3).

Data published for the Home Office publication Statistics on changes in migrants’ visa and leave status: 2015 (formerly known as “The Migrant Journey”) show how non-EU nationals changed their immigration status and the immigration routes used prior to achieving settlement in the UK.

Further evidence from the Survey of Graduating International Students³ indicates that 24%⁴ of non-EU students intend to work in the UK after their initial period of leave to study.

To better understand how many non-EU students remain in the UK legally and for how long, we examined two cohorts of students who entered the UK on a Tier 4 student visa for 1 year or more in 2009 and 2010 and their immigration status after 5 years (Figure 7).

This showed that nearly a quarter still had valid leave to remain after 5 years. Of those who had valid leave to remain, 46% were studying (either on their initial course or extended), 32% switched to a work route and a further 23% had switched to other routes including family.

For recent cohorts, the number of non-EU students reaching settlement after just 5 years is approximately 2,500, which isn’t unexpected as the study category is not a direct route to settlement⁵.

Figure 7: Immigration status, after 5 years, for two cohorts of students (main applicants) who entered the UK on a Tier 4 student visa for 1 year or more in 2009 and 2010, UK

Notes:

1. Figures are for those students (main applicants) who entered the UK on a Tier 4 student visa for 1 year or more in 2009 and 2010 and their immigration status after five years.

2. These figures reflect immigration rules for study prior to 2011. Current immigration policy for study makes it harder to extend leave to remain or switch to other visa routes.

3. Figures exclude dependants.

4. Figures exclude EEA and Swiss nationals.

After 5 years, leave to remain had expired for 78% of students in the two cohorts. However, not all of these former students will be included in the International Passenger Survey (IPS) estimates of emigration flows for those whose former reason for immigration was study, because some may:

• intend to emigrate short-term (3 to 12 months) and return to the UK

• emigrate after 1 year but before the end of the 5-year analysis window

• not identify themselves in the IPS as former students

• depart via the Northern Ireland and Ireland land border (not covered by the IPS)

• be in the process of changing their immigration status

• remain after their visas have expired

It should be noted that these are cohorts of students who arrived before the tightening of immigration requirements for international students, which started from 2010. Therefore the behaviour of later cohorts of students is expected to differ from the behaviours of those arriving before these changes.

Further analysis of grants of settlement over a longer time period indicates that between 2009 and 2015, a total of 163,300 grants of settlement were made to non-EU main applicant migrants who originally arrived in the UK through a study route (this varied between just under 19,000 in 2009 and just under 29,000 in 2013). 159,500 (98%) arrived prior to 2010 before the tightening of immigration rules. These figures reflect previous rules, and current immigration policy for study makes it harder to extend leave to remain or switch to other routes, and in due course be granted settlement.

5.4 No initially identified evidence of departure

Our analysis of Exit Checks data identified that for a small proportion of non-EU students, whose visas were due to expire in 2015 and 2016, and 2016 and 2017 (7% (13,200) and 4% (6,000) respectively), there was no evidence of departure. There are a number of potential reasons why individuals are not identified as departing for this small group, including:

• incomplete matching of records in the Exit Checks data, which relies on data supplied by passengers and carriers, together with immigration system records

• departure via the Common Travel Area (not covered by the Exit Checks programme) and other coverage issues

• quality and completeness of travel event data supplied by carriers under the Exit Checks programme, especially around the start of the programme

• dual nationals (or others whose passport details change over time) using different passports arriving and departing may be difficult to match

• overstaying their visas and remaining in the UK without valid leave to remain

Further detail on the quality of data collected under the Exit Checks programme is being published by the Home Office.

5.5. Some may have left the UK via un-sampled ports or “out of hours” flights

The International Passenger Survey (IPS) is designed to ensure that coverage and interviewing hours (6am to 10pm) are sufficient to ensure that all types of passengers are included. Data collected in the IPS are weighted to overcome this.

Analysis of departing flights in the Exit Checks data for those students whose long-term study visas expired in 2016 and 2017 suggests that the majority of outbound flights (90%) departed between 6am and 10pm.

Further work will be undertaken to examine the ports that international students arrive and depart from to inform the IPS pilot exercise to understand if student journeys are missed by not interviewing between 10pm and 6am. However, the high level of coverage for outbound flights suggests that there is little likelihood of undercounting emigrants on account of out of hours flights.

Notes for: Explaining the international student migration “gap” further

1. Approximately 2,200 (1%) sponsored Tier 4 students in the 2015 and 2016 cohort did not have a known status and were excluded from this analysis. This was 800 (0.5%) for the 2016 and 2017 cohort. Status refers to whether the individual had departed before visa expiry, departed after visa expiry, had extended leave to remain or another status including no initially identified departure in the Exit Checks data (which may be due to non-matching of individuals, departure via the Common Travel Area or other reasons). Further detail on the quality of Exit Checks data is available.

2. The inflows for formal study are Long-Term International Migration estimates (LTIM). LTIM estimates include a number of adjustments, including for people who change their intentions (switchers). Please see Long-Term International Migration estimates methodology for more detail.

3. The Survey of Graduating International Students was conducted between 13 March and 30 April 2017. The survey targeted all graduating international students at UK universities. Further detail on this survey is available.

4. The percentages presented from the survey are subject to uncertainty and are not weighted to be representative of the wider student population. Further detail on the survey and representativeness is available.

5. The required residency period for settlement is normally 5 years. It should also be noted that the time spent in the study category does not generally count towards the qualifying period.

The evidence presented in this report shows that:

• international student migration has changed over the last 10 years

• International students are a component of net migration

• the Home Office Exit Checks data provide a more accurate picture of what non-EU students do after their initial period of leave to study

• this knowledge has given us greater insight into the International Passenger Survey (IPS) derived statistics, which are likely to underestimate student emigration – therefore any implied student net migration figure is likely to be an overestimate

We can’t use these findings to draw conclusions about the accuracy of the total net migration figure. This is because any potential overestimate in total net migration (as a result of an underestimate of student emigration) will be offset by:

• some students stating that they originally immigrated for a different reason (such as work or accompany or join)

• the “switcher adjustment” that is applied in the production of total Long-Term International Migration (LTIM) figures¹

• any other groups of migrants with a tendency to change their intentions in the opposite way, for example, people emigrating after work or British emigrants returning earlier than expected – more research is needed to better understand how intended migration behaviour matches actual migration patterns for other groups of migrants

We intend to explore this further by:

• reviewing the switcher adjustment to ensure that the method is applying the most accurate adjustments in light of this new evidence

• continuing to investigate coverage of the IPS

• investigating Exit Checks data to assess how the migration intentions of other groups of non-EU migrants (those with work or family visas) compare with actual travel patterns

• working in partnership with colleagues across the Government Statistics Service to utilise all available administrative systems to further improve migration statistics – further detail on our plans will be published in September 2017

Notes for: Conclusions and next steps

1. The switcher adjustment is applied by using information collected from the International Passenger Survey (IPS) on people’s original intentions and actual durations when previous immigrants are departing the UK, and when previous emigrants are returning to the UK. This information is used to calculate a propensity to switch based on nationality groupings. Switching adjustments are carried out on both inflow and outflow for people who switch from a migrant to a visitor (migrant switchers) and those who switch from a visitor to a long-term migrant (visitor switcher). For more information see the Long-Term International Migration estimates methodology report.

The 2017 Survey of Graduating International Students was a collaborative effort between the Economic and Social Research Council (ESRC) Centre for Population Change (CPC) at the University of Southampton, Office for National Statistics (ONS) and Universities UK (UUK). Detailed information on post-study intentions, certainty of these intentions, travel patterns, use of public services, and working patterns whilst studying was collected from international students in UK Higher Education in their final year of study. The full technical report for the survey is available.

Sample

The target population for the Survey of Graduating International Students was all international (non-UK) students in higher education in their final year of study. Institutions took part on a voluntary basis and final year international students were invited by the institution to take part. Final year international students were able to choose whether or not to participate in the survey. Each responding student was able to enter a prize draw to win a £30 gift voucher.

Response rates

A total of 3,560 responses were provided and the overall response rate was 3.5%. Calculation of the effective response rate is complicated by the fact that there is no complete information on the numbers of sampled students at participating institutions as not all provided these data.

An approximation of the response rate was calculated by the CPC using a three-step forecasting approach (further detail is available in the survey technical report).

Response rates by nationality (EU or non-EU) and by enrolled programme of study (undergraduate (UG) or postgraduate (PG)) are shown in Table A1.

Non-response bias can be a problem where response rates are low as bias may be introduced to the survey results; that is, the characteristics of respondents to the survey may be different to the characteristics of those who did not respond.

Representativeness of survey respondents

To assess the representativeness of the survey, the characteristics of survey respondents were compared with those of all final year international students in the UK along several important dimensions (Table A2). As table A2 shows, shares of EU and non-EU nationals were fairly similar to those in the respective population. However the survey does appear to oversample postgraduate students (who accounted for 67% in the survey and 45.5% in the population).

A breakdown (Table A3) by the top seven countries of origin suggests that some countries, for example, China, were under-represented in the survey, whilst others were over-represented (US).

Caution should therefore be taken when interpreting results by individual nationality and programme of study.

The patterns and trends in international student enrolment for all UK institutions were also compared to those of participating Higher Education Institutions (HEIs) in the survey (see the full technical report for more information). The patterns observed were very similar, reassuring us that the participating HEIs were representative of all UK HEIs.