Towards a typology of student migration: Illustrations from student record data for the United Kingdom

Transcription

1 Towards a typology of student migration: Illustrations from student record data for the United Kingdom Author: Neil Bailey Neil.Bailey@soton.ac.uk Division of Social Statistics and Demography - University of Southampton Abstract Around two and half million people were attending an institute of higher education in the United Kingdom in the 2010/2011 academic year, which equates to around 4.1% of the total population. Surprisingly, given the importance of higher education very little work has been conducted on the migratory patterns of students attending institutes of higher education in the UK. This paper puts forward a typology that can be used to categorise the different migration transitions that a person can undertake in order to attend a higher educational institution. With the use of the student migration typology and the Student Record Dataset of the Higher Educational Statistics Agency, which contains detailed information on every student recorded as attending an institute of higher education in the UK, illustrations of the migratory patterns of those attending higher education in the UK during the 2010/11 academic year with a focus on the local authority (LA) and county level geography are provided. Using the typology of student migration the data indicate that around 37% of students are student migrants, 9% were local students, while 35% stayed in the LA but studied elsewhere suggesting they distance learn or commute. When analysing LAs; Oxford, Cambridge, Nottingham, Leeds and Sheffield all stood out as areas with large student populations with high proportions of students in there population. Introduction Migration as a process involves three key variables: the migrant, the origin and the destination (Dennett & Stillwell, 2010). With reference to this paper, the migrant refers to a student, the origin refers to the domicile 1 of that student and the destination refers to the location of termtime address of the student or the location of the higher educational institution (HEI) attended. Internal migration estimates for England and Wales published in September 2012, indicated that there were an estimated 2.59 million moves between Local Authorities (LA) 2 in the year ending June 2011, which represented a 1% reduction on the levels estimated for the year ending June 2010 and an estimated five per cent of the mid-2010 England and Wales 1 The students place of permanent residence prior to undertaking a course at a Higher Educational Institution. 2 Local Authority (LA) is a generic term for any level of local government in the UK. In geographic terms LAs therefore include English counties, non-metropolitan districts, metropolitan districts, unitary authorities and London boroughs; Welsh unitary authorities; Scottish council areas; and Northern Irish district council areas (Office for National Statistics, 2011a). The dataset from HESA contains 408 LAs in the UK; 328 in England, 32 in Scotland, 22 in Wales, and 26 in Northern Ireland. 1

2 population moved to a different LA in the year to mid-2011 (Office for National Statistics, 2012). Demographers have long recognised that one of the major attributes affecting an individual s propensity to migrate is age and that persisting regularities appear in empirical age-specific migration schedules (Cadwallader, 1992; Dennett & Stillwell, 2010; Raymer, Abel, & Smith, 2007; Raymer, Bonaguidi, & Valentini, 2006; Rogers & Castro, 1981). The Office for National Statistics (2012) internal migration statistics mirror the typical age-specific patterns, with young adults most likely to migrate. Around one in five of those living in England and Wales in mid-2010 aged migrated to a different LA, which constitutes around six per cent of all the migration moves in year ending mid-2010, while another peak can be seen amongst those aged 22. The Office for National Statistics (2012, p. 3) internal migration report states that the peaks in internal migration at young adult ages can largely be explained by moves to and from university or other higher education institutions. Students are a highly mobile part of society and a large amount of turnover of individuals is associated with student areas. This significant amount of population turnover associated with student areas can have a profound effect on university towns and cities (Duke-Williams, 2009). Students have a large impact on the local economy, while HEIs also play an important role in the areas that they are located in. According to Universities UK (2010) report into how universities engage with local communities, the average annual spend by a full time English domiciled student on living costs was 6,496, which equates to expenditure of around 7.9 billion a year for English domiciled students alone. Universities educate many professionals such as architects, engineers and health professionals, and such professionals directly and indirectly support the local region. Universities UK (2010) state that for every 100 university jobs a further 100 are created in the wider economy by a knock-on process, whereby it is estimated that higher education generated 669 thousand jobs throughout the UK economy through direct and knock-on effects in The effects of students migrating to university are not only economic but also cultural and social. Universities and their students enrich their local communities through a variety of activities and facilities. The majority of universities in the UK, with the help and support of student unions, encourage community involvement in sport, exhibitions, drama and many other activities which helps deliver social inclusion, reduce crime and improve health. Another closely related to strand of research is the work investigating trends of gentrification in urban areas across the UK and how student migration is impacting on this process. D. P. Smith and Holt (2007) focus upon the relationships between higher education students and contemporary urban change in the UK and consider the process of studentification in urban locations closely situated to HEI s. Ley (1996, p. 181) also discusses this phenomenon stating that: emergent youth ghettos nestled symbiotically around inner city university campuses, and the sheer numbers of students could not help but introduce a distinctive sub-culture. It is important to understand that there is no formal legal obligation to register a change of address in the UK. Therefore, the measurement of internal migration around the UK is extremely difficult to quantify and inherent with uncertainty and underreporting. One of the 2

3 major downsides to the lack of legislation in place in the UK to register internal migrations is that there is no dedicated administrative source that was designed or created for the purpose of measuring internal migration flows. Instead, proxy data, such as the registration and reregistration of patients in the National Health Service (NHS) is used. Young adults, especially young males, can be slow to change their registration with their doctor when they move. This is partially attributable to the fact that younger people are less likely to visit their local doctor due to their more desirable health compared to older people. With these doctor registers being one of the key components to measuring internal migration in the UK, this is a particular issue with regards to the quality and accuracy of these internal migration estimates. As a result, an enhanced knowledge of the volume, pattern and trends of migration to institutes of higher education will increase the knowledge on a group of people who are recognised as causing uncertainty within migration and population estimates. It is therefore important to understand, at a local level, the patterns of internal migration that contribute to these large scale population changes and how they redefine the areas in which they decide to settle. An increased understanding on how young people are influenced in their migration choice where they are migrating from, where to and why are these patterns occurring will benefit not only the institutions themselves but local and national government, planning authorities, the national population statistics, as well other entities, such as retailers in the surrounding economy. As mentioned previously, to measure the patterns of migration transitions of people to higher educational institutions, there are three variables (domicile, term-time address and institute address) which can be analysed. However, to truly understand the real migrations transitions that are taking place then these three variables need to be analysed simultaneously. Therefore, the aim of this paper is to propose a typology that can be used to categorise the different migration transitions that a person can undertake in order to attend a higher educational institution. The paper first sets out the previous research conducted on student migration and then introduce the dataset being used for the analysis. The paper then introduces the proposed typology that categorises the different transitions of student migration. Finally, the paper ends by illustrating the use of the typology by answering two substantive main questions while analysing the Student Record Dataset of the Higher Educational Statistics Agency, which contains detailed information on every student recorded as attending an institute of higher education in the UK. Firstly, what are the main migratory patterns and trends of those attending higher education in the UK during the 2010/11 academic year, with a focus on the local authority and county level geography. Secondly, how does the classification of students using the typology differ as a result of the geographical level (LA versus County) used for the analysis and which geographical level provides the best results. Previous Research on Student Migration 3

4 The majority of previous research on student migration tended to differentiate between the migration patterns of people moving to a HEI and the migration behaviour of graduates. Many authors have tended to focus primarily on the migratory behaviour of university graduates. Faggian, McCann, and Sheppard (2006, 2007a), for example, analysed the employmentmigration behaviour of 13,753 UK graduates with a particular focus on ethnicity and the sequential migratory patterns of 76,000 Scottish and Welsh students. The same authors (Faggian, McCann, & Sheppard, 2007b) also modelled the migratory behaviour of some 380,000 UK university graduates by controlling for a range of variables related to human capital and local economic conditions. Faggian and McCann (2009) focused on the migratory patterns of high quality British graduates from their university to the location of their first employment. P. W. Smith, Raymer, and Giulietti (2010) combined certain aspects from multiple sources of data to provide a time series of detailed migration flows. These detailed migration flows were cross-classified by origin, destination, age, sex and economic activity in which students were one of the categories. Mak and Moncur (2003), investigated the migration transitions of first-year college students in the USA. The work analysed state-level data to examine the economic determinants of interstate migration of college-bound first years. There has also been a limited amount of previous research that has focused on migrations to institutes of higher education. As mentioned before, D. P. Smith and Holt (2007) focused on how concentrations of students in parts of university towns and cities have led to changes in those areas. Allinson (2006) analysed the impact of rising student numbers on the settlements in which they reside although with little focus on the patterns and trends of their origin areas, or their characteristics. Patiniotis and Holdsworth (2005) conducted research on the transition of young people to university and found that the UK students have traditionally been moving away from their parental home to attend a HEI, however this trend has recently been seen to be decreasing. Christie (2007) investigated the mobility decisions of students going into higher education in the UK, particularly the decision not to migrate and stay living at the home for their studies. Again, this research was conducted on students attending a HEI but the primary focus was not on the actual patterns of migration themselves. Duke-Williams (2009), on the other hand, used 2001 census data to examine the migration flows associated with a set of electoral wards selected on the basis of having a high concentration of students. The paper found that the majority of in-migrants to these wards were new students entering the system of higher education and the majority of out-migrants were recently graduated students leaving the system. Duke-Williams concluded that since there have been significant changes in the arrangements of student funding, that there was a scope for further study using student data from HESA to look at more recent trends, which is one of the aims of the current paper. The HESA Student Record Data has a large amount of detailed data on every student registered at a HEI in the UK and very little previous research has taken advantage of this data source and produced detailed analysis of student migration behaviour in the UK. As previously mentioned, due to the multiple variables that can be used to analyse student migration, there is a need to create a typology that can be used to categorise the possible transitions that can be made by a student to attend a HEI. There has been no previous research that attempts to 4

5 create or even proposes the creation of such a typology of student migration. Therefore, this paper aims to propose a typology that can be used to categorise student migration, not only in the HESA dataset, but any dataset that contains the locational information of students on any geographical level. The paper also aims to produce analysis that investigates the internal migration flows of people moving within the UK, as well as the international migration flows of people arriving from overseas, to attend a HEI in the UK. HESA Student Record Data The data source used in this paper is the Higher Education Statistics Agency (HESA) Student Record Data. HESA is the official agency for the collection, analysis and dissemination of quantitative information about higher education. It was set up by agreement between the relevant government departments, the higher education funding councils and the universities and colleges in 1993, following the White Paper Higher Education: a new framework, which called for more coherence in HE statistics, and the 1992 Higher and Further Education Acts, which established an integrated higher education system throughout the United Kingdom (Higher Education Statistics Agency, 2012b). The HESA Student Record is collected in respect of all students registered at a reporting HEI that follows courses that lead to the award of a qualification or institutional credit, excluding those registered as studying wholly overseas. The record excludes students studying overseas for the entire duration of their course, even when they are formally registered at a UK-based HE institution. Students studying overseas by distance learning are similarly excluded; unless they are funded by a UK HE funding body. The subset of the student record dataset used in this paper contained three basic variables: - Domicile of Student - Term-time Address of Student - Institution Address These variables are available at three different levels of geography; Government Office Region, County and LA. Throughout this paper only counties and local authorities are used, and the paper focuses on the 2010/11 academic year. Boyle (2009) explains that the definition of migration used in human geography is often influenced more by the data resources available and the capability of defining migration from that resources, as opposed to theoretically guided principles. With regards to defining migration in this paper there were a couple of restraints as a result of the available data. There was no capability to report migrations with regards to any change in address that would have corresponded with Lee s (1966) and Rees s (1977) definitions of migration. It was only possible to record a migration if an administrative geographical boundary was crossed. The lowest level of geography provided in the HESA dataset was LA, while the geographical level of counties 3 3 Counties were formerly administrative units across the whole UK. Due to various administrative restructurings however then the only administrative areas still referred to as counties are the non-metropolitan (shire) counties of England. The English 5

6 were aggregated from the LA information. Throughout this paper a migration was only recorded if a county or LA boundary was crossed, depending on the level of geography being analysed. There are two variables in the HESA dataset than can be used as the destination variable: term-time address of the student and institution address of the student. When analysing the institution address variable, the institution address for every individual is recorded and therefore there are no individuals with institution address recorded as unknown. However, the institution address variable does not indicate the actual student migration as this is not where the student actually lives. The term-time address variable, on the other hand, does provide the information that is desired - where does the student live while studying at the HEI. However, the term-time address variable was recorded as unknown for a number of students in the dataset. In the 2010/11 academic year, 167,553 (7%) students had term-time address recorded as unknown. These unknown cases are taken into account separately in the analysis presented in this paper. Typology of student migration When analysing student migration flows and stocks it is possible to analyse the destination variables of term-time address and institution address independently, as some basic statistics can be produced. However, analysing these variables independently raises the problems of which destination variable actually provides the desired and most accurate answers. For example, when comparing domicile and term-time address, 52.83% of all students in 2010/11 had a term-time address in a different county to their domicile, suggesting that just over half of all students in 2010/11 migrated to attend a HEI. However, when institute address was used instead of term-time address, this value increases significantly to 72.52%. This shows that a lot more students attended a HEI in a different county than those students who lived in a different county to their domicile during term-time. So which variable is most suitable? Well, the answer to this question is ambiguous and raises the point that by analysing these destination variables separately, there are several key questions that cannot be answered with any certainty. These included; how many students actually migrated to attend a HEI, how many stayed in the same county and commuted or how many stayed in the same county and attend a local HEI. Therefore, I suggest that if any of these more complex and relevant questions are required to be answered with a greater degree of certainty, then a typology should be created that interacts all three variables in unison. When analysing whether a student migrated to attend a HEI then it is necessary to create a typology where categories of the possible student migrations are defined by using the three location variables in the student record data as discussed in the data section - simultaneously. The creation of this typology allows the researcher to accurately capture the actual moves (or non-move) a student undertakes in order to attend a HEI. metropolitan counties, although no longer administrative units, are also used for statistical purposes (Office for National Statistics, 2011a). The dataset from HESA contains 94 Counties in the UK; 47 in England, 8 in Wales, 11 in Scotland, 26 in Northern Ireland, Isle of Man and Channel Islands (same as LA). 6

7 The typology is derived by categorising an individual as a result of the geographical location of the three locational variables. If the geographical location for one or more of the variables is different to the other(s), then a geographical boundary has been crossed and some sort of migration has occurred. As a result, due to the different possible combinations of the three locational variables, five main categories of student movements were created and these are explained in detail in Table 1. In this paper, the geographical boundaries used are counties and LAs, however this typology can be used to analyse any other geographical level of data on any dataset, such as for UK data, the constitutional countries, government office regions, electoral wards, census output areas and so forth. With the use of these student categories it is now possible to make accurate and detailed analysis of the student migration patterns and trends across the UK. Instead of just being able to identify areas with large numbers of students enrolled at institutions or recorded as residing within an area during term-time, it is now possible to locate areas with large numbers of student migrants, local students, commuting students and distant learners. This enables for a much more detailed analysis of an areas student demographics which can inform local government, local planners and institutions of higher education alike. In the following sections of the paper the HESA Student Record data will be analysed using a variety of techniques including the typology created above to provide an in-depth and accurate report into the current trends and patterns of student population and migration in the UK in the 2010/11 academic year. This report will firstly focus on the county geographic level followed by the LA level. Secondly, the focus will shift to analyse how these student trends and categorisations differ between county and LA and link this back to the modifiable area unit problem. 7

8 Table 1: Typology of Student Migration: Categories, Descriptions and Diagrams Student Category Description Diagram 1 Student Migrant - Migrates to attend a HEI = Domicile in different geographical area as term-time address. - Term-time address is in same geographical area as institution address, therefore the student is assumed to live close to the institution. 2 Local Student - No migration to attend a HEI = Domicile in same geographical area as term-time address. - Domicile, term-time address and institution address the same, suggests student did not migrate to attend a HEI as they attend a local HEI (in same geographical area). 3 Commuter/Distance Learner - No migration to attend a HEI = Domicile in same geographical area as term-time address. - Institution address is in different geographical area as domicile and termtime address which suggests the student commutes to the HEI or is distance learning. 4a Migrant Commuter/Distance Learner (4b = Term-time address unknown ) 5a Migrant Commuter/Distance Learner attending local HEI (5b = Term-time address unknown ) - Migrates to attend a HEI = Domicile in different geographical area as term-time address. - All three addresses are in different geographical areas, therefore they migrate to different term-time address but still commute to the HEI or distance learns. - Migrates to attend a HEI = Domicile in different geographical area as term-time address. - Domicile and institution address in same geographical area but term-time address is different, therefore they migrate away to live and commute back to the same area as the domicile to attend the HEI or they distance learn. Notes: = Geographical Area Border = Domicile = Term-time address = Institute Address = Migration = Commute 8

9 Student Population Percentage of Total Population The main patterns and trends of student migration across the UK The large and detailed dataset collected by HESA contains geographical and characteristic variables on just fewer than 9 million students attending one of the 165 HEIs across the UK between 2007/8 to 2010/11. There were 2,562,100 students in the 2010/11 academic year, which accounted for 4.1% of the total 2010 mid-year population (Office for National Statistics, 2011b). The majority of students in the HESA dataset - 63% - were aged between the ages of 18 and 24 and, out of all people aged 18 to 24 years in the United Kingdom in 2010, 24% were attending a HEI (Higher Education Statistics Agency, 2012a; Office for National Statistics, 2011b). The total student population and corresponding percentage of the UK population are shown in Figure 1. Over the last 15 years, the student population has steadily increased from 1.7million in the academic year 1995/96 to 2.6million in 2010/11. Therefore, with the ever increasing numbers of students in the UK and the fact that student are representing a larger proportion of total population than ever before, it is important to have an understanding of where these students are geographically distributed across the UK. Figure 1: Student population in the UK ,000,000 2,500,000 2,000,000 1,500,000 1,000, , % 4.0% 3.5% 3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% Total Student Population Student Population % Source: Higher Education Statistics Agency (2012a); Office for National Statistics (2011b) Student patterns by county The geographical locations of the 2.6 million students that attended an institute of higher education in the UK in the academic year 2010/11 by county are shown in Figure 2 [For reference a labelled map of the UK counties can be found in Appendix A]. Here, the termtime address is used to define where students are geographically located. This variable is used as it has been assumed that this best depicts a student s location when studying as opposed to institute address, however it must be considered that this variable contains around 7% of values as unknown, as mentioned previously. Also labelled on this map are the 20 Russell Group Universities which represents the research-led institutions. They are found in all four 9

10 constituent countries and in every major city of the UK. The majority of the student populations in the UK clustered into the major cities with regards to total population. This was not a surprising finding as the major cities are the locations of the largest and most reputable universities. The four counties with a student population of over 80,000 contain of London, Birmingham, Manchester and Leeds, as well as being home to seven of the 20 Russell Group Universities. The top ten counties for total student population and proportion of population that are students are shown in Table 2. These values supplement the data mapped in Figures 2 but now also portray a slightly different picture when proportion of population is analysed. The county with the highest proportion of students was Lothian, the county that contains Edinburgh - Scotland s Capital and the county with the second highest proportion of students was South Glamorgan, the county containing Cardiff the Capital of Wales. The county of Inner London was 4 th and Belfast 9 th. Therefore, when considering the number of students as a proportion of the total population, the four capital cities of the four constituent countries in the United Kingdom were still prominent in the top nine. However, when the other large cities that was in the top 15 for total student population in the UK are considered, they were no longer as dominant when considering the number of students as a proportion of the total population. Counties with relatively small populations such as Herefordshire and Worcestershire, Coleraine, West Glamorgan and Dyfed have large proportions of students and as a result were highlighted as student locations whereas they were not when considering just crude numbers. While counties containing famous university cities such as Oxfordshire and Avon (University of Oxford and University of Bristol) are also more pronounced as homes to large numbers of students when reviewing the proportions of students as opposed to the crude numbers. The only county in the top ten for both measures was Inner London and therefore, it can be said that Inner London is greatly influenced by student populations, as they make up a large proportion of the area s population and also represent crude large numbers of the population. 10

11 Figure 2: Student population by county and the 20 Russell Group Universities 2010/11 Source: HESA Student Record Data 2010/11 Note: Term-time Address variable used to show student s geographical location. Open University students not included. 11

12 Table 2 - Top 15 counties by the proportion of the total population that are students Rank County Name Student Pop. County Name % Students 1 st Inner London 199,696 Lothian 13.34% 2 nd Outer London 164,751 South Glamorgan 8.28% 3 rd West Midlands 109,888 Hereford and Worcester 6.54% 4 th Greater Manchester 100,923 Inner London 6.48% 5 th West Yorkshire 87,104 Oxfordshire 6.00% 6 th Strathclyde 79,437 Coleraine 5.58% 7 th Hampshire 66,233 Tyne And Wear 5.10% 8 th South Yorkshire 57,830 West Glamorgan 4.99% 9 th Tyne And Wear 57,140 Belfast 4.90% 10 th Merseyside 56,542 Leicestershire 4.85% Source:Higher Education Statistics Agency (2012a),Office for National Statistics (2011b) Note: Term-time Address variable used to show student s geographical location. Open University students not included. So far the analysis of the student record data has only focused on the geographical location of all students within the UK and no focus has been made on whether or not these students migrated to that geographical area in order to attend a HEI. The following section of work will investigate how the student categories - as defined using the typology of student migration discussed previously - differ across time and across the UK counties. The numbers and proportions of students in each student category for each academic year available in the dataset are shown in Table 3. The proportions in each category are relatively constant over the four-year period, with the exception of the increase of those in the student migrant category. It can be seen that the increase in the number of student migrants can be attributed to the decline in the numbers of students in the term-time address recorded as unknown categories (4b and 5b). Over the four year period the numbers of students having their term-time addresses recorded has increased. The impact of this decline in unknown term-time address has only seen an increase in student migrants, this implies that a large proportion of the students with unknown term-time address actually reside in the same county as the HEI they attend. Table 3: Total number and proportion of students by student category and academic year Student Category 2007/ / / / Student Migrant 755,551 35% 864,853 38% 929,919 40% 968,816 41% 2 - Local Student 553,215 26% 595,703 26% 615,659 26% 600,967 26% 3 - Commuter/Distance Learner 453,163 21% 474,909 21% 490,425 21% 493,271 21% 4a - Migrant Commuter/Distance Learner 93,496 4% 96,181 4% 100,237 4% 104,123 4% 4b - Migrant Commuter/Distance Learner 231,778 11% 163,076 7% 152,460 7% 140,421 6% (Term-time address unknown) 5a - Migrant Commuter/Distance Learner 12,700 1% 12,984 1% 13,317 1% 13,689 1% attending local HEI 5b - Migrant Commuter/Distance Learner 67,479 3% 42,765 2% 33,417 1% 23,519 1% attending local HEI (Term-time address unknown) Total 2,167,382 2,250,471 2,335,434 2,344,806 12

13 Source: Higher Education Statistics Agency (2012a) Note: County Level - Data does not include students registered with the Open University In the 2010/11 academic year, 968,816 students (41%) migrated from their domicile county to attend a HEI, while their corresponding term-time address was in the same county to that of the HEI attended (Category 1). The top ten geographical destinations of student migrants are shown in Table 4. Inner London received the highest inflow of these students (95,827); this is not surprising as Inner London is home to many HEIs including four Russell Group Universities. Out of the 95,827 students that migrated to Inner London to attend a HEI, 58.1% arrived from outside of the UK. With international students excluded, the largest sender of students to Inner London was Outer London, accounting for 16.57%, followed by Kent (3.95%), Hampshire (3.68%), Surrey (3.65%) and Essex (3.57%).West Midlands received the second highest inflow of category one students with 54,853. Again the highest inflows of these students were from overseas accounting 44.68%. With international students excluded, the highest inflows came from Outer London (12.96%), Leicestershire (4.12%), Inner London (3.98%), Hertfordshire (3.42%) and Staffordshire (3.32%). Table 4: Numbers and proportions of students by county that are student migrants [Cat.1] County Top 10 Counties by Num. of Student Migrants County Top 10 Counties by % of Student Migrants Inner London 95, % Cambridgeshire 84.38% 17,209 West Midlands 54, % Fife Region 78.37% 7,475 West Yorkshire 48, % Cornwall and 74.85% 2,670 Isles of Scilly Greater 46, % Durham 72.21% 11,811 Manchester Hampshire 41, % North Yorkshire 66.84% 15,122 Nottinghamshire 37, % Hampshire 62.97% 41,492 South Yorkshire 36, % Oxfordshire 61.49% 26,620 Tyne And Wear 33, % Berkshire 60.95% 8,042 Avon 30, % Nottinghamshire 59.70% 37,024 Outer London 30, % Dorset 58.38% 12,521 Source: Higher Education Statistics Agency (2012a); Office for National Statistics (2011b) Note: Data does not include students registered with the Open University The ten counties with the highest proportions of all students that are Student Migrants are also shown in Table 4. Cambridgeshire had a total of 20,394 students in 2010/11 and a very large proportion of these, 84.38%, migrated to the county to attend a HEI. This was an extremely large proportion of its total students and this could be an effect of the University of Cambridge being one of the most prestigious and top rated universities in the UK and the World, which attracts people to migrate to the county. Of the 17,209 students that migrated to Cambridgeshire, 33% came from outside the UK and 14% from London; however the remaining 8,989 are from a relatively equal spread of the remaining origins across the UK, with numbers declining with distance away from Cambridgeshire. It is also important to note that the only two counties that were in the top 10 for number of student migrants and proportion of all students that were student migrants were Hampshire and Nottinghamshire. 13

14 Local students have been defined as students with the domicile, term-time address and institute address in the geographical area. This indicates that these students either remained at the parental home or moved house, but remained in the same geographical area and attended a local university. There are several benefits for students in this category, the main one being the financial savings of remaining in the parental home offsetting the increasing cost of tuition and living. In 2010/11, 600,967 (26%) of all students were classified as local when using the county level of geography, which tells us that just over 1 in 4 of all students attended a HEI in the same county as there domicile and term-time address. The counties with the largest numbers of local students and the counties with the highest proportions of local students within their county are shown in Table 5. The county with the largest number of local students was Inner London with 62,433 (21.91%). Again this is not surprising given Inner London s total population and the number of HEI s in the county and also the high cost of living in London may have an impact on this trend. According to the Mayor for London (2003) report, the cost of living in London is 17% higher than in Edinburgh, and 23% higher than in Manchester and the majority of this is attributable to the very high price of housing in the UK capital. As a result of the very high housing costs it is not surprising that many students attending London universities are students who resided in the area prior to attending the HEI and presumably decided to continue living in the parental home. Strathclyde has a large crude number of local students as well as a high proportion of all students in the county. 63% of all students in Strathclyde have the same domicile, termaddress and institute address that suggests they have remained in the area to attend the local HEI and quite possibly remained in the parental home. One of the reasons for this large proportion of people staying in Strathclyde to attend a HEI may a result of the Scottish Government subsidising tuition fees for Scottish students attending a Scottish HEI. Other areas with a large proportion of local students include Suffolk, Cleveland, Hereford and Worcester. This may be a result of the areas being highly rural and people attending agricultural HEI s in the area, or a result of their relatively poor connectivity with the rest of the UK resulting in students staying local instead of migrating to attend a HEI. 14

15 Table 5: Numbers and Proportions of Local Students by LA [Cat.2] County Top 10 Counties by Num. of County Top 10 Counties by % of Local Students Local Students Inner London 62, % Suffolk 67.86% 3,767 Strathclyde 52, % Strathclyde 62.78% 52,357 Region Region West Midlands 43, % Cleveland 43.51% 11,863 Greater 35, % Hereford And 38.25% 3,861 Manchester Worcester Outer London 33, % Western Islands 37.78% 2,672 West Yorkshire 30, % Kent 36.98% 13,731 Lancashire 19, % Staffordshire 34.99% 11,502 Tyne And Wear 19, % Northamptonshire 34.91% 5,188 Merseyside 17, % Humberside 34.08% 7,948 Hampshire 15, % Greater 33.56% 35,819 Manchester Source: Higher Education Statistics Agency (2012a) Commuters/distance learner 4 (category 3) represent those students that did not make a migration to attend a HEI their domicile and term-time addresses are the same however, unlike local students, commuters/distance learners attended a HEI in a different geographical area to their domicile and term-time address. In 2010/11, 493,271 (21%) of all students had a term-time address in the same county as their domicile but attended a HEI in a different county. Of these students a large proportion (24.31%) attended HEIs in London, while 5.48% of all category 3 students attended a HEI in London and lived in the neighbouring counties of Berkshire, Essex, Hertfordshire, Kent and Surrey. This strongly suggests that these students commuted to take advantage of the cheaper housing costs of living outside of London itself. The county with the second highest number of students attending a HEI within its boundaries but residing elsewhere was Lancashire. 30,331 (6.21%) of students attended a HEI in Lancashire but resided in another county, 52.72% of these lived in either Merseyside (Liverpool) or Greater Manchester. This suggests a surprising and opposite trend to that seen with London. In this case it suggested that students remained in the urban areas of Greater Manchester and Merseyside and travelled to a HEI in Lancashire, it is unclear why this was the case. Finally, when looking at the proportion of all students in a county that are category 3 the county of Belfast stands out % of all students attended a HEI in Belfast but did not live there. Shropshire had the second highest proportion of all students that commuted/distance learned with 59.95% of all students attending a HEI in Shropshire but lived outside the county. From the analysis of the first three student categories it can be said that out of all students in the 2010/11 academic year, 41% undertook a traditional migration to attend a HEI while 47% did not make a migration and either attended the local university, commuted or were distance 4 A distance learner can be simply defined as a person who partakes in a system of education delivery in which the majority of learning takes place with the learner and the teacher separated by space and/or time, the gap between the two being bridged by technology. A distance learner is one who experiences the majority (80+%) of their learning off-campus at a distance from the teacher and consequently has limited face-to face interaction with their teachers and peers (Tynan, 2010, p. 2). 15

16 learning. The remaining two categories of student migration 4a, 4b, 5a and 5b only accounted for 12% of all students. Around 7% of all students had term-time address recorded as unknown. However, category 4a is still an interesting group worth investigation. Although it only represents 4.44% of all students, they represent students that did migrate to attend a HEI but their term-time address was also still different to that of the HEI. These students seemed to have migrated away from their domicile towards the location of the HEI, but still did not reside in the same county of the HEI. The majority of these students attended HEI s in London and migrated to neighbouring counties, presumably to allow them to commute to the HEI but not have to endure the living costs that are entailed by living in London. The proportions of all students in each student category for a selection of different counties are shown in Table 6. The differences in the proportions of student types between counties can suggest several things about the given area and it is also intuitive to compare the values in Table 6 to the 2010/11 values in Table 3, which show the distribution across the student categories for the whole UK. Inner London had quite a low proportion of student migrants, 7% lower than the national figure, but very high numbers of Commuter/distance learners in comparison to the other counties, 8% higher than the UK figure. Out of the six counties listed in Table 6, Greater Manchester had the highest proportion of Local Students and also had very low levels of unknowns. The county of Avon was included in the table to highlight the high levels of unknowns in comparison to the other counties; over 19% of all students in Avon had term-time address unknown, which is very high in comparison to the whole dataset (7%). Cambridgeshire had an extremely high proportion of its students in the student migrant category, 84.38% of students in Cambridgeshire migrated to the county to study which is by far the greatest proportion of counties and over double that of the national average. 16

17 Table 6: Proportions in each student categories for selected counties Student Category Source: Higher Education Statistics Agency (2012a) Inner London Greater Manchester Avon (Bristol) 1 - Student Migrant 95, % 46, % 30, % 2 - Local Student 62, % 35, % 13, % 3 - Commuter/distance Learner 84, % 14, % 14, % 4a - Migrant commuter/distance learner 22, % 2, % 1, % 4b - Migrant commuter/distance learner (Term-time address unknown) 14, % 5, % 13, % 5a - Migrant commuter/distance learner attending local HEI 3, % % % 5b - Migrant commuter/distance learner attending local HEI (Term-time address unknown) 1, % 1, % 1, % Total 284, % 106, % 74, % Student Category Hampshire Oxfordshire Cambridgeshire 1 - Student Migrant 41, % 26, % 17, % 2 - Local Student 15, % 7, % 1, % 3 - Commuter/distance Learner 6, % 6, % 1, % 4a - Migrant commuter/distance learner 1, % % % 4b - Migrant commuter/distance learner (Term-time address unknown) 1, % 2, % % 5a - Migrant commuter/distance learner attending local HEI % % % 5b - Migrant commuter/distance learner attending local HEI (Term-time address unknown) % % % Total 65, % 43, % 20, % Student patterns by local authority (LA) When analysing the location of students using the LA variable, it is possible to distinguish certain towns and cities within the UK that are heavily influenced by the HEIs that are located within them and the large numbers of student settlers that are associated with these HEIs. At the county level this was also possible; however in the majority of the counties in the UK there are more than one large settlement and more than one HEI within each county. As a result certain areas of one county might be heavily impacted by the amount of student s resident whereas other parts of the county may not be influenced by student population whatsoever. Therefore, analysing individual LAs allows for a much more detailed understanding of what areas in the UK a heavily influenced by student population and student migration. The geographical locations of the registered term-time addresses by LA of the 2.6 million students that attended a HEI in the 2010/11 academic year are mapped in Figure 3. Again, as previously mentioned, one needs to consider that this variable contains a number of unknowns. There were only 5 LAs that were resident to over 40,000 students within term-time and these LAs are labelled in Figure 3. 17

18 Figure 3: Student population by local authority Source: Higher Education Statistics Agency (2012a) Note: Term-time Address Variable used to record students geographical location. Open University students not included. 18

19 These five LAs are all LAs of large cities within the UK. The LA with the highest stock number of students in the 2010/11 academic year was Leeds, with 54,836 students having a term-time addresses within the Leeds. Leeds is the location for two HEIs The University of Leeds and Leeds Metropolitan University in which 63,658 students are registered and with regards to the numbers of students registered at a HEI within a LA Leeds was ranked 6 th. The second highest stock of students was Birmingham with 54,607 term-time resident students. Birmingham is home to five HEIs Aston University, Birmingham City University, University College Birmingham, University of Birmingham and Newman University College in which 73,839 students are registered and therefore ranked the 4 th highest LA with regards to the number of student registered within an LA. The third largest stock of resident students was Manchester with 53,271 while Sheffield was fourth with 48,401 and Edinburgh was fifth with 40,746. These top 5 LAs with regards to term-time address of students are also LAs with very high numbers of students registered to institutions located in these LAs all top 5 for termaddress are in the top 11 for numbers registered in the LA. It is also important to consider total population size when evaluating crude values of a certain measure within one area in case the numbers are skewed by a large overall population size. The top five LAs with regards to student population are also LAs with large overall population size. Birmingham has the largest LA population in the UK of just over 1 million people, Leeds was 2 nd largest and Sheffield 4 th while all the top 5 for student numbers were in the top ten for total population estimate. Therefore it is not surprising the LAs with such a large total population will have a large number of term-time resident students and a large number of HEIs and a large amount of students registered as attending these institutions. Large settlements attract more people, more opportunities and obviously more students. The proportions of total population that are term-time resident students in a LA are mapped in Figure 4. The LAs with the two highest proportions of the total population that were resident students were the well-known and renowned university settlements of Oxford (19.81%) and Cambridge (17.61%). With regards to crude numbers of resident students, these two settlements were also ranked relatively high - 11 th and 17 th respectively and therefore it can be said that students have a significant impact on the demographic composition of these LAs. Oxford has the highest proportion of resident students than any other LA in the UK and this may be a result of the LA being the location for the joint 2 nd best university in the world, University of Oxford (The Times Higher Education, 2012), as well as Oxford s second university Oxford Brookes University. When you combine the facts that Oxford is home to two large universities and was ranked 11 th in resident student numbers and 13 th in registered students with the fact that the LA overall is relatively small in terms of overall population ranked 137 th in the UK then this results in the very high proportions of students in the LA. The LA of Cambridge was in a very similar situation to Oxford. Cambridge is home to the 7 th best university in the world, University of Cambridge (The Times Higher Education, 2012) as well as being the location for one of the campuses of Anglia Ruskin University. Again, combining these large student numbers and the relatively small size of the overall population ranked 191 st in the UK the proportion of students in the LA was very high. 19

20 Figure 4: Student population as a proportion of total population by local authority Source:Higher Education Statistics Agency (2012a), Office for National Statistics (2011b) Note: Term-time Address Variable used to record students geographical location. Open University students not included. 20

21 The LA with the third highest proportion of students was the Welsh LA of Ceredigion which contains the settlement of Aberystwyth and Aberystwyth University. The LA is sparsely populated ranking 346 th in total population size in the UK; however the Aberystwyth University attracts a significantly large amount of students in comparison to the settlements overall size. The university is the major Welsh Language university in the world and as a result attracts a large amount of students to the institution. These very high proportions of students within a LA will have a profound impact on the LAs economy, housing market and demographic composition. However, it will be of further interest to investigate how many of these students migrated to these areas to study or were already resident in the area. As people migrating from different areas will have more of an impact on the LAs composition and put more demand on the housing market than a student who is from the area, familiar with the surroundings and local customs and are highly likely to already have accommodation/housing in the area. Table 7: Top 10 LAs by student population and student proportion Rank LA Name Student Pop. LA Name % Students 1 st Leeds 54,836 Oxford % 2 nd Birmingham 54,607 Cambridge % 3 rd Manchester 53,271 Ceredigion % 4 th Sheffield 48,401 Nottingham % 5 th Edinburgh 40,746 Newcastle % 6 th Glasgow.City 38,833 Exeter % 7 th Nottingham 38,786 Southampton % 8 th Liverpool 37,150 Welwyn.Hatfield % 9 th Cardiff 36,197 Manchester % 10 th Newcastle 35,123 Cardiff % Source:Higher Education Statistics Agency (2012a), Office for National Statistics (2011b) The top ten LAs for total student population and proportion of population that are students are shown in Table 7 - these values supplement the data mapped in Figures 3 and 4. Cardiff, Manchester, Newcastle and Nottingham appear in the top 10 both student numbers and student proportions. Therefore, it is clear these LAs are areas that are greatly influenced by student populations, as they make up a large proportion of the area s population and also represent crude large numbers of the population. As in the section on the county level the focus now shifts towards whether or not these students migrated to these LAs in order to attend a HEI. The following section of work will investigate how the student categories - as defined using the typology of student migration discussed previously - differ across time and across LAs. The breakdown of all the students in dataset into the student categories for the LA level is shown in Table 8 (this is directly comparable to the county level as shown in Table 3). 21