Language learning experiences of postgraduate research students in the UK

Thisarticlereportsonthe ﬁ ndingsofapioneeringstudyofformal,non-formal and informal language learning experiences of postgraduate research students in the UK. The research involved a large-scale survey followed by semi-structured interviews. Quantitative and qualitative data analysis shows that the language needs and interests of postgraduate students are barely met due to limited institutional provision of formal language learning opportunities. The study con ﬁ rms some known facts, such as the composition of the student body (e.g. mostly young, fully-funded full-time students with a signi ﬁ cant proportion of international students), and reveals other possibly less-known realities, from a remarkable interest in language learning among research students to the reasons for their interest and the potential impact greater access to language learning could have on their research and beyond. Half of the respondents reported having funded provision of language tuition but one third were not even aware of the existence of a Language Centre at their institution. Student suggestions for alternative language provision are included as these can inform policy and decision-making at postgraduate level in the UK and serve as a basis for further research in this area. ﬁ cation ﬁ cation after qualitative thematic Interview data was analysed thematically and a good level of comparability observed.


Introduction
Research on language learning at universities traditionally focuses on undergraduate programmes of study (see, for instance, Pauwels 2011), with little attention to language learning at postgraduate level. However, in the past few years, there has been greater interest in looking at language learning at this level in various countries, namely, Pakistan (Ali et al. 2015), New Zealand (Sika-Paotonu and Tait 2018), Australia (Fenton-Smith and Humphreys 2015; Ryan et al. 2012), South Africa (Schulze and Lemmer 2017), Spain (Raigón Rodríguez 2015) and Sudan (Alhassan 2019), although these studies are largely concerned with English for academic purposes, and often linked to taught courses rather than research programmes. International exceptions that look at learning modern languages other than English at postgraduate level are the work of Pinto and Araújo e Sá (2016), who conducted a case study with postgraduate researchers (PGRs) in Portugal and found that languages are seen as mainly having instrumental value, and Gkaintartzi et al. (2018), whose case study revolves around the experiences of trainee teachers involved in delivering language education for refugees in Greece.
In the UK, a recent report highlighted that only 42% of participating UK Modern Languages departments are offering language-focused modules at postgraduate level (Álvarez et al. 2018).
When it comes to institution-wide language provision (IWLP), a recent survey (AULC-UCML 2018) reported that 25% of IWLP enrolments were PGRs, and around 30% of participating institutions said tailor-made IWLP courses were available to PGRs. This suggests that, albeit limited, there are some opportunities to engage with formal language learning institutionally.
The literature on postgraduate studies in the UK focuses on PGRs' needs more broadly. For instance, Smith et al. (2014) conducted a study and identified the need to cater for the PGRs' entrepreneurial skill set. Although they argue that language use matters, no reference is made to using languages other than English. Tonks and Williams (2018) highlighted the importance of more wide-ranging and flexible training as doctoral graduates are finding increasingly varied types of jobs, but their study only reveals language needs related to English, particularly in connection with thesis writing. At the same time, the Advanced Higher Education annual Postgraduate Research Experience Survey (PRES) in the UK (https://www.heacademy.ac.uk/institutions/surveys/ postgraduate-research-experience-survey) does not include any questions related to language learning. Similarly, the large scale global PhD survey by Nature (Woolston 2019) has just started to offer versions of the survey in languages other than English but still no reference is made in it to language learning. As a result, the experience of learning languages beyond undergraduate level in the UK remains largely unknown.
To fill this gap, this article reports on research undertaken with the aim of capturing the language learning experiences of PGRs during the academic year 2016-17. The study maps out formal, nonformal and informal language learning activitiesas understood by UNESCO (2011aUNESCO ( , 2011bUNESCO ( , 2011c)and builds a collective profile of their language learning needs and interests as well as perceived language learning obstacles and opportunities. As students from all academic subjects based at UK Higher Education Institutions (HEIs) were consulted, the results can contribute to understanding if, and how, PGRs' needs are being met more generally and can inform policy and decisionmaking around institutional language provision.

Formal, non-formal and informal language learning
Firstly, formal training covers learning through courses based at the HEI by IWLP units, faculties or departments. Secondly, non-formal language education includes other types of institutional provision, for instance, courses offered by funding bodies or other specialist institutes. Thirdly, informal learning involves 'learning activities that occur in the family, in the work place, in the local community, and in daily life, on a self-directed, family-directed or socially-directed basis' (UNESCO 2011b). For the purposes of this study, this includes participating in foreign film clubs and informal conversation exchanges, and using applications on personal mobile devices.

Research questions
The key research questions in this study were: (1) What are the research-related and personal language learning needs and interests of PGRs?
(2) What kinds of formal, non-formal and informal language activities do PGRs engage with?
(3) What are their perceptions and views of language learning support available?

Data collection and analysis
Two instruments were used to gather information about language learning at postgraduate research level in UK universities: a large-scale survey and semi-structured interviews. These provided both quantitative and qualitative data. In addition to valuable demographics data, the survey questions elicited responses from PGRs related to institutional language provision perceptions and experiences, their language learning needs and interests, and their language-related activities. The interviews were used to add depth on these topics beyond the limitations of the survey format by, for instance, exploring reasons underlying students' needs and interests and their language-learning activities as well as gathering their views on the language provision available.
The survey was designed and distributed using Jisc Online Surveys between July and August 2017. It was sent to administrators and academics at 152 HEIs across the UK, with a request for the survey to be circulated among their PGRs. With its 32 questions, the survey was quite ambitious and required a significant commitment of time on the part of the respondents. Despite this, as described below, a larger than expected number of respondents completed it. Most questions had a 5-point Likert scale ranging from strongly agree to strongly disagree. References to the survey questions in this paper are indicated with a 'Q' followed by the relevant number (e.g. Q3). 'Survey data' under relevant excerpts refers to data collected through the online survey. The survey was followed by semi-structured interviews with a self-identified sample of the respondents to gain further insights. A total of 19 individual interviews lasting an average of 93 minutes each were held in September 2017. They were audio recorded and transcribed. Excerpts from the interviews are identified here as 'Interview data'.
The article uses descriptive statistics to report on the survey data via a quantification of the grading scales and binary questions but also from a quantification of the open comments, when appropriate, after a qualitative thematic analysis. Interview data was analysed thematically and a good level of comparability with the survey data was observed.

Participants
A sample of 565 self-selected PGRs took part in the study, which represents 0.5% out of a total of 112,520 officially registered as such in the UK in 2016-2017 (HESA data https://www.hesa.ac.uk/ data-and-analysis/students/whos-in-he) from 61 participating UK HEIs (Q3). These students came in various proportions from each of the 18 existing disciplinary categories listed by the Higher Education Statistic Agency (HESA). 1 According to HESA data from 2016 to 2017, there was roughly 1 postgraduate research student for every 10 postgraduate taught students in the UK (HESA data https://www. hesa.ac.uk/data-and-analysis/students/whos-in-he).
395 of the respondents (70%) were female and 164 (29%) were male while HESA data (https:// www.hesa.ac.uk/data-and-analysis/students/whos-in-he) reports that the entire postgraduate population in 2016/2017 consisted of 48% females and 52% males. This suggests that females were more willing to engage in research of this type than males for reasons that are beyond the scope of this study. Most participants (70%) were aged 23-37 (Q2) at the time, and this mirrors average student ages in official HESA statistics (https://www.hesa.ac.uk/data-and-analysis/students/whos-in-he). Figure 1 shows that although respondents were at different stages of their postgraduate studies, the majority were in their second and third years (Q5).
For almost two thirds of students (63%), English was their first language, followed by a wide range of languages none of which represented more than 5% of the entire sample (Q12). The great majority of the respondents (91%) declared knowledge of other languages (Q15). There is no official data to compare this with, so it is not possible to ascertain whether this level of multilingualism is representative of the entire population under study. Instead, it can be assumed that PGRs with an interest in languages were more likely to respond to the request to take part in this type of research. Therefore, the views and activities reported may be representative of those with a multilingual profile.
In terms of mode of study, the data collected (76% full-time and 24% part-time, Q6) is in line with national figures provided by HESA (https://www.hesa.ac.uk/data-and-analysis/students/whos-in-he). Finally, most PGRs (61%) declared to be funded the UK government, only 9% were non-UK funded, and 27% were self-funded; 6% were sponsored privately or by a corporation and only 1% had requested a loan (Q8).

Ethics
Ethical approval from the researching institution was granted. Some universities were unable to include their postgraduates in this study because their institutional policy did not allow them to distribute the survey link. Three universities requested for the survey to be submitted for scrutiny to their own internal ethics committee, and all three submissions received approvals.

Discussion of findings
The interview and survey data sets are fairly comparable. Findings from the study are divided into two main interrelated dimensions of PGRs' language learning experiences: (1) language learning needs and interests, and (2) type of language learning. Needs and interests are presented with respect to their research but also more widely. The study also explored changes in language learning needs as well as alternative approaches to developing language skills, the extent to which PGRs engaged in formal, non-formal and informal language learning and the types of language learning activities they did. Uncovering these learning experiences and students' views revealed a range of barriers and opportunities that impacted upon their language development.

Language learning needs and interests
Most PGRs expressed an interest in learning a new language (66%) but even more were keen to improve languages they already had (77%) (Q20.3/4). Participants' responses were classified between learning languages to support their research and learning for personal reasons.

Research-related language needs and interests
In the survey, PGRs were asked about the relevance of language learning in relation to their postgraduate research. Almost half of the respondents (44%) stated that their research involved language(s) other than English (Q11). The highest demand identified by students was related to French, German, Spanish, Latin, Italian, Portuguese and Arabic. In contrast, 37% of students reported that they had no language training needs directly related to their research (Q20.1). Some students whose research involved languages (38%) declared to have developed their language skills previously in formal, non-formal and informal contexts (Q20.2), while 7% of students (Q28) did not have the language skills they needed for their research and were learning the language from scratch, as this example illustrates: I'm starting to learn Swahili a little bit, to prepare for potential research in East Africa. (Survey data) It is remarkable that less than a quarter (21%) of students who needed languages for their research were pursuing studies in Modern Languages (Q4), as this is evidence that languages in research are not confined to those engaged in language-related postgraduate studies.
Research-related language learning was linked to a number of needs and interests, such as direct access to original texts, conducting fieldwork, writing grants, writing about their research, being able to communicate at conferences, accessing research networks and partners, establishing academic collaborations and improving their post-doctoral and employment opportunities (Q29), as the following quotes illustrate: Though not directly relevant to my degree, knowledge of other languages is important from a networking and collaborating perspective. It is also useful if I am applying for a job in another country. (Survey data) In science you're very much encouraged to do a post-doc abroad. You're not expected to stay in the same place, you're expected to move labs. So if you're wanting to stay in Europe … We have an Italian PI, a couple of Dutch PIs, we have PIs from all over. So you might get away with English in your lab, but to live in the country you'd need the language. (Interview data) The issue of accessing knowledge in other languages was also raised by participants, suggesting the identification of a fundamental need for languages at postgraduate level: So, I do sometimes wonder if I had more of an understanding and knowledge of French, would I be more able to engage with the philosophers, more than on a surface level at least. (Interview data) Being able to communicate with something in the original language almost gives you the ability to understand it in a way that you can't in translation. (Interview data) I think for this project, if I was using translations, I do think I'd lose something because I do a lot of close reading and textual analysis. You could do the same project but you'd need to refer to the original texts at least. (Interview data) This idea of 'losing something' when accessing texts in translation rather than in their original form was, in fact, a recurrent subtheme in the data, confirming Svensson and Wihlborg's (2010) view on the role of languages in the development of student knowledge: The point is that language plays an important role in the internationalisation of knowledge content in higher education, since it is a medium of expressing and communicating this knowledge. The question then becomes one about the use of language as a condition for students' development of knowledge. A critical aspect of this question concerns access. Both concerning in what language there is access to certain knowledge, and the issue of how students can access knowledge and develop personal knowledge depending on their access to or mastering of particular languages. (Svensson and Wihlborg 2010: 605-6) During an interview, another participant referred to the importance of this fact and quoted, 'as Wittgenstein said, "the limits of my language are the limits of my world"' (Interview data). This was also reflected in their answers to the survey where almost a third (32%) recognised that their research would have been enhanced or been different if they had had greater language skills (Q20.8), either because it would add legitimacy or advance knowledge further. A student wrote in this respect: Research-wise, my work is going to have much more credibility if I can read the actual original German. (Survey data) In addition, students also pointed out that their language knowledge was critical in their choice of research: In contrast, for a few students, the value of knowing other languages for their research was unclear. Some participants recognised having identified relevant research published in other languages but doubted that being able to read such publications would have made a difference. Comments by these participants also suggested that learning a new language is a lengthy process that may compete with other priorities during a research degree: A lot of research in my topic has been conducted in French. It would be useful to be able to have greater reading skills in this area, although it is not directly obvious that this is the case. (Survey data) Since my research is UK based and largely depends on interviews I have no direct need for languages although, for a variety of reasons, Gaelic languages might be helpful and there is some anthropology on Death Studies written in French that is not currently accessible to me, and which has never been translated into English, but none of this is crucial or central. (Survey data) I have been completing a literature review. Some papers were excluded from this because they were written in another language (mostly Spanish). I am not sure to what extent it would have made a difference if I could have understood these papers and I'm sceptical that I would have had the time to develop my language skills enough over the last two years in order to have been able to understand them. (Survey data) Some participants did not perceive learning a language as essential given the dominance of research publications in English, since most international researchers share their research in English: I'd like to learn German as they are the leaders in my field. However, nearly all the research is in English so it wouldn't really benefit my research. (Survey data) Some peer-reviewed literature is available in languages other than English, but the majority of major conferences and literature is in English. Whilst learning another language would be nice, I would not consider it essential. (Survey data) In parallel, one recurrent and explicit need widely acknowledged was learning academic English, echoing the preponderance of this theme in the international literature (e.g. Fenton-Smith and Humphreys 2015; Ryan et al. 2012).
Doing my PhD helps me learn to write formal English and that is sufficient at the moment. (Survey data) As an EdD student, my language needs are mostly to do with academic writing. (Survey data) Finally, 34% of students indicated that, despite their interest in learning and developing languages while doing their research, they were unable to do so (Q20.6). They highlighted some practical difficulties, related mostly to lack of access to suitable language provision and lack of time, particularly for part-time students who usually juggle work and family commitments with their studies (HEA n.d.).

Change of needs
Participants were also asked about the extent to which their needs had changed over the course of their research degree. As Figure 2 shows, over a quarter (27%) declared to have experienced changes (Q20.9).
The changes reported were varied but often related to future needs after graduating. Students mentioned the benefits of language learning for visiting or moving to other countries for research, attending international conferences, having better career prospects, giving talks, networking, and having teaching opportunities.

Alternative approaches to language learning for research
Some of the PGRs with language needs for their research but without the language skills felt that there were alternatives or shortcuts to meeting their language needs, namely, using, producing and requesting translations.
It would perhaps have been useful to be able to read certain texts in the original German or Spanish, but those texts weren't a major part of my research, so I felt happy using translations. (Survey data) The easiest way to actually work on him is to go to the German editions and work on my own translations and discuss how and why I'm translating in a certain way, or why often its easiest not to translate at all. (Interview data) Regarding the German that I used in my research -I knew enough about the structures and the syntax to know what sort of sentences I needed for the experiment, but I had a native German speaker produce the materials for me because I did not have the skills to do it myself. (Survey data) I would be interested in learning Japanese as I am involving Japanese art students in my research but as I am selffunding I cannot afford to do this. I have a friend who is Japanese and I have paid him a small amount to do any translation required. (Survey data) In addition to translation, another approach was to resort to technological solutions or traditional dictionaries to access the necessary research sources.
Most foreign language texts are accessed using machine translation (Google, etc.) or a traditional paper dictionary. (Survey data)

Personal language needs and interests
To some extent, PGRs' language needs and interests go beyond those connected to research. 14.6% of students who were studying languages during their research spoke of the value of languages beyond instrumental benefits for their postgraduate studies (Q30). They articulated a benefit for the expansion of knowledge and of the human experience of the world: Although language training is not part of my PhD research degree, and whether I learn a language or not will make no different on my PhD topic, I do think that learning a new language (hence a new culture) helps you have a different perspective on different things, including education and research. (Survey data) Learning a language is almost something that's helping to increase an understanding and ability to communicate with the world, almost. I enjoy it on an intellectual level, as a challenge, it's that thing of seeing something that you can't understand, it's like a puzzle, and there's the satisfaction of solving it. (Interview data) Interview answers, in particular, revealed that students engaged with language for socialising or enjoyment, to communicate with family and friends, and to spend time abroad, as the following examples illustrate: Learning a language is one of the activities that you can do at university to socialise and become part of the community. (Interview data) Figure 2. PGRs' responses to whether they experienced changes in language learning needs (n = 565).
Once I'd brushed up my French, also I look up to my older sister and my mother. All very intelligent bright women. Me the youngest. Got to catch up! (Interview data) I wanted, as well as the family thingthat's one part of itthe other was just doing something mentally engaging that wasn't my thesis. (Interview data) But also if I worked for a multinational company, then you could have opportunities to work in Germany. So that has been a thought, it's not just something I'm doing purely for entertainment or family reasons. It's a little bit with career in mind. So it's just multiple reasons really. (Interview data) Most of these personal needs and interests can be said to be related to the students' wellbeing, to balancing life and work, a point that is also stressed in the 2017 PRES report (HEA n.d.).

Formal and non-formal language learning
In this study, formal and non-formal language learning are 'institutionalized, intentional and planned' (UNESCO 2011a(UNESCO , 2011c. To simplify, the distinction was not made explicit in the survey questions.
PGRs were asked to refer to any language learning that took place with a teacher present, including private tuition.
When considering formal learning, students were asked about their awareness of the provision and support at their home institution, their views on the adequacy of this provision, the difficulties encountered accessing this training as well as the languages they studied.
Close to a quarter of respondents (21%) said that they had been engaged in formal (and nonformal) language learning activities (Q21) during the last academic year. From those, four students pointed out that they needed support from their funders (Q26) and around a quarter (23%) said that access to formal learning required the approval of their supervisors, who in such cases hold the key that opens language learning for PGRs or not.

Awareness of and access to formal and non-formal language provision
When the PGRs considered doing their degree, language provision at their institution of choice did not seem to influence their decision as 95% of the respondents said that it was not a factor in their choice to undertake their research there (Q19). In fact, when they were asked about the existence of a Language Centre at their institution, 35% did not know whether it existed (Q17), and 77% were unaware of whether there was institution-wide language provision (IWLP) or a Languages for All programme on offer (Q18). These results are surprising since almost half of the students (41%) had studied previous qualifications at the same institution (Q7), but they suggest that students may not be familiar with the current structures and terminology around language provision. Universities may need to do more to advertise the language learning opportunities on offer.
Among the students who were undertaking institutional language learning, the majority (62%) were doing courses that were not formally assessed (Q27). More than half (59% of them) were learning at their home institution, mostly at the Language Centre or IWLP unit (Q21a). These students acknowledged having access to funding support from their universities that could be used for language learning (Q25): I don't need to learn another language but if I needed to for fieldwork, I could apply for funding in the [research pooled] funding pot. (Survey data) Another great thing [name of institution] doesfor one year as a postgrad student, you can do a free year of language learning. (Interview data) However, the remaining students commented that their institutions seemed to consider language learning as an additional need, and, as a result, they had to access support somewhere else, fund it and organise it for themselves. These PGRs arranged non-formal language training through private tuition, specialist language institutes (e.g. Confucius Institute, Goethe Institute), summer schools, online tuition and community centres. Students specified a series of reasons why they engaged in non-formal language learning elsewhere (Q21ai): . Incompatibility with language training schedule on offer . Lack of language provision at home institution . Lack of language learning opportunities for postgraduates . Lack of institutional response to language learning requests . Lack of available spaces in language classrooms . Lack of language intensive courses . Limited choice of languages and unsuitable teaching content at home institution . Cost of language courses offered . Teaching level too low or too high . Travelling distance to lessons . Lack of arrangements at home institution with other providers This list reflects some of the factors negatively affecting PGRs' experiences of formal and nonformal language education (Q31). In their own words: In my university, the language classes which are offered for free for research students (as my budget is very limited) are extremely boring, and only focused on translating from a foreign language to English. This way you can only acquire a passive understanding of the language, which I often already have in the languages I'm interested in. My university offers a very limited number of languages. Portuguese, for instance, was not available. (Survey data) My institution does offer courses for modern languages, but these often begin before the start of the postgraduate academic year and seem to be more focused on speaking and daily interactions, rather than reading academic texts, which would be most useful for my research. (Survey data) I found my experience to be frustrating as I was in a level 1 class with people who had little or no prior experience of language learning and thought the pace of the lessons was extremely slow. I'd quite like for there to be more language learning provisions for linguists but understand that resources in languages are limited at the moment. (Survey data) I chose my university believing there was language learning provision suitable to my needs (for example, French for academic reading). It transpired this provision has been cut and no suitable replacement exists. This is very disappointing. The only things offered is a charge for continuing education class that happens at unsuitable times and for an unsuitable duration (full year commitment) for a semi-distance learner with multiple additional commitment including part of second term spent in Greece. (Survey data) In relation to this, 47% students suggested that greater provision of language learning would be beneficial to their research (Q20.5) and that a more flexible approach to language teaching with courses which did not require a full-semester or a full-year commitment was needed. In a fast-changing world, the significance of flexible training models for PGRs is key to meet societal needs and their own aspirations.
Significantly, only around half of these students (54%) were happy with their formal language learning experiences (Q31). The picture captured shows low levels of satisfaction and engagement with formal and non-formal language learning for postgraduates due to issues of access to adequate training and funding. As a consequence, as one student put it: A generation of scholars from the UK, who can only read English, are being produced each year due to these constraints! (Survey data)

Languages learnt formally and non-formally
The languages studied by PGRs are varied, although the majority were formally and non-formally learning French, German and Spanish (Q24). English was also among the most studied, which could reflect the considerable body of international students at postgraduate (research) level in the UK, one of the highest in the world at 42%, according to HESA data for 2016/2017 (https://www. hesa.ac.uk/data-and-analysis/students/whos-in-he) (Figure 3).
Among these, most students (68%) had prior knowledge of the language they were studying (Q28). At the same time, 16 of them declared to be studying a second language (German being the most-common additional language), with only one student studying three languages at the same time (one of which was English) (Q32).

Informal language learning
Students were also asked about their informal language learning, meaning activities that did not take place with a teacher present, and 44% of them said that they had engaged in informal language learning (Q16). Table 1 shows the range of informal activities they carried out on their own.
Web-based language learning is the first preference for 40% of these students but further studies are needed to understand the nature of the students' web-based experiences. Among the top-ranked  options, some could be equally classed as web-based or internet-dependent (e.g. using an app, watching films and doing an online course) and it is worth stressing that independent and interdependent activities (i.e. conversing with speakers and studying on their own) share a similar percentage at around 16%. Finally, as stated earlier, Figure 4 below shows the wide-ranging variety of reasons PGRs have for doing languages.
Research reasons comes first (20%) but it is followed closely by pleasure (20%) and practice (19%). Overall, non-research-related reasons seem to drive most of the participants' choices for engaging with informal learning.

Conclusions
This study, the first of its kind in the UK, suggests that most research students need and are interested in improving their language skills in languages they already have knowledge of rather than picking up new languages. Against a background of lack of research and partial provision in this area, this article provides some evidence for policymakers that languages may play a key role in knowledge transfer and advancement in a research context across all disciplines (not just in Modern Languages) and thus further research and support may be needed. For instance, despite the dominance of English in academic environments and publications, a third of the survey participants said that their research could have been improved with greater language support. Against this, one in three students found it impossible to pursue their interest in language learning due to lack of suitable provision or time.
This calls for an approach to language provision for PGRs with greater scope and flexibility, perhaps following the model used in the Languages for London Arts and Humanities Partnership (LAHP) student scheme led by the Modern Language Centre at King's College London. Highquality classes are offered from ab-initio to advanced in the main languages (French, German and Spanish) and additional languages (Arabic, Italian and Russian) have been added more recently (LAHP website: https://www.kcl.ac.uk/modern-language-centre/collaborations/lahp.) Some of the defining features are longer semesters, assessment by attendance, weekly and weekend options and use of authentic materials with an emphasis on interdisciplinarity. This is an example of how Language Centres in universities within Doctoral Training Consortia can develop language modules to meet the needs of postgraduate students (including those studying part-time or with other commitments and scarce time available). One way forward could be to investigate where cross-consortia partnerships for language delivery could be envisaged. It would also be invaluable to explore collaborations with consortia outside the Humanities. This is relevant to a quarter of the participants who had been learning languages formally in the academic year when the study was conducted, but a sizeable 40% did so non-formally, following provision not provided by their institutions, and a third of PGRs where unaware of the existence of a Language Centre at their institution. Unsurprisingly, a great proportion of students engaged in informal language learning mainly through web-based or internet-dependent tools and by using the target language with others, although this activity was not reported as linked to their formal studies.
The most commonly-learnt languages were French, German, English and Spanish. The reasons for learning languages were split equally between research-related purposes and personal pursuits such as pleasure and leisure but more importantly three quarters of our respondents who declared a need for languages for research were not language specialists, which points to a potential institutional duty to evaluate language needs of PGRs more carefully lest we should 'lose something' fundamental in the development of new knowledge. This is further supported by responses from one third of the survey participants indicating that their research could have been enhanced or been different had they had greater language skills, including better academic English skills.
The results of this study present us with evidence that could inform future policy and provision: there is a significant need for language training for researchers, which must be carefully tailored and widely advertised. The need is not entirely research-based, as personal interest and engagement with languages can be seen as a way to aid in improving wellbeing, stress levels and life-work balance, aspects institutions are surely keen to invest in. As our sample is not entirely representative, there is also a need for systematic and up-to-date data pertaining to language learning at PGR level that could be gathered by including language-related questions in large-scale annual surveys such as PRES by Advance HE, the PhD surveys by Nature and some PGR-related questions in the specialist annual surveys by UCML and AULC.