Skip to main content

Research aim and objectives

This Rapid Evidence Assessment (REA) aims to assist the Welsh Government (WG) in providing guidance to practitioners, trainers and policy makers on how to deliver effective immersion education for learners aged 3 to 16 in Wales.

In this study, ‘immersion education’ is defined as learners’ experience of receiving education through the medium of the language of the school, with that language being different from the learners’ home language. 

The current direction of Welsh Government policy for increasing Welsh-medium education and education in two languages (bilingual) is reflected in the objectives of the Cymraeg 2050 strategy and through the Curriculum for Wales. The Government intends to incorporate immersion models into the mainstream work of the education system and expand the late immersion programme for pupils. 

The four objectives of the project were:

  • to identify, appraise and summarise the research evidence available in Wales and internationally; 
  • to draw conclusions that will inform guidance on providing effective immersion education for 3 to 16-year-olds in Wales; 
  • to make recommendations about areas and research questions that may need to be explored through further research;   
  • create a practical and accessible summary that provides guidance to practitioners, trainers and training providers on how to deliver effective immersion education based on the findings of the rapid evidence assessment.

The main research question used to inform this study was:

  • What does the literature tell us about effective approaches and methods in immersion education and about effective pedagogy in immersion education?

The secondary research questions were:

  • What does the evidence tell us about the effectiveness of introducing and using a language/languages other than the target language at different stages of the learner's education and linguistic journey?   
  • What does the evidence tell us about the effectiveness of delivering and maintaining immersion education for learners of different ages i.e. what are the characteristics of effective provision in relation to early, mid and late immersion education?
  • What does the evidence tell us about approaches to deal with a range of linguistic backgrounds and previous linguistic experiences amongst cohorts or groups of learners?
  • Are there specific considerations that merit attention in relation to additional learning needs and immersion education provision?

There is an additional output to this study, namely an accessible summary of the study findings for practitioners, trainers and training providers, which presents the key findings on immersion approaches and methods.


The research questions were used to draw up terms and search criteria to ensure the REA was as feasible and fit for purpose as possible. 

In accordance with the protocol drawn up for the study, the research was conducted in four stages. An initial database of 4,342 items was created (Stage 1). 41 literature items were included in the final synthesis of evidence (Stage 4).

Stage 1: Establish the scope and boundaries of the review and conduct the literature searches

Following the initial scoping work, nine databases were selected, including open access sources (Met Search, the National Library of Wales and Gwerddon) in order to access Welsh language items that would possibly not be found in larger international databases. The search terms were determined by considering the research questions for the REA. The search terms were confirmed, and refined by drawing on the research team’s expertise and understanding of literature in the field, and in consultation with practitioners, trainers, training providers and system leaders working in Welsh medium immersion education, in a virtual focus group meeting. Search terms were trialled (in both Welsh and English) to ensure that the search terms, databases and inclusion and exclusion criteria used were suitable for the purpose of the REA. Based on the pilot process it was decided to include items published between 2000 and 2023.  

Three internationally recognised experts in immersion education were invited to draw the research team’s attention to the research publications relating to effective approaches and methods in immersion education that would be relevant to the study, from their experience, and which met the screening criteria. They were also asked for examples of current and forthcoming research. Only one of the three was able to contribute to this stage of the research within the time available. 44 items were received from the one contributing individual. The same process was followed as was used for the items found through the literature searches when screening and extracting information about these items.

In accordance with the criteria for Stage 1 a total of 4,342 items were collected for screening at Stage 2.

Stage 2: Screening information

In order to include them in the information extraction stage (Stage 3), the 4,342 literature items screened at Stage 2 had to meet the following criteria:

A total of 180 items that met the criteria were obtained at Stage 2.

Stage 3: Information extraction and quality assessment and weight of evidence

The information extraction, quality assessment and weight of evidence work was carried out based on reading the full text of items that had met the screening criteria at Stage 2. A Data Extraction Form containing 53 questions was modified to suit the purpose of the REA. This form was used to collect key information from each item.[footnote 1] Of the 180 items from which information was extracted, 115 were excluded as they had not obtained a high weighting score. There were 65 items that met the criteria, and these were the items that were used to develop the synthesis in Stage 4.

Stage 4: Synthesis development

Following the searches, data extraction and weighing of the evidence, the synthesis of the research was drafted. Owing to the complex nature of some of the definitions of ‘immersion education’ and ‘minority language’, the process of producing the synthesis was iterative. An additional quality assurance stage was introduced in the production of the synthesis, to ensure that the research questions were addressed appropriately by the literature items that were obtained. 24 additional items were excluded at this quality assurance stage. A total of 41 items were included in the final synthesis.

The evidence submitted in the synthesis includes 23 studies specifically covering early immersion, 10 discussing mid immersion and 10 discussing late immersion (these were categorised based on the age of learners).[footnote 2] There were 12 studies relating to partial immersion, with seven discussing intensive immersion. Most of the evidence comes from the United States of America (n=14), with a wide range of other countries included.

Producing a summary for practitioners

Based on the findings of the REA, a summary was produced on how to provide effective immersion education for practitioners, trainers and training providers. Stakeholders who attended the first Focus Group session were invited to attend the second Focus Group session, to offer their views and comments on a draft version of this summary.


Summary of the evidence about the effectiveness of introducing and using a language/languages other than the target language at different stages of the learner's education and linguistic journey

  • Although some authors found that mainly using the target language was beneficial in immersion contexts, there is evidence that occasional use of the home language can also be beneficial in making comparisons between different languages. 
  • Structuring the use of various languages (in specific sessions or lessons for each language) as well as allocating time for translanguaging, are effective.
  • Interventions to develop literacy skills have a positive effect on language acquisition, for example progress in reading skills and metalinguistic awareness. However, this depends on a number of factors and the type of intervention in question.
  • There is some evidence of the benefits of using technology to develop phonological awareness and vocabulary acquisition when learning a language, but it is difficult to identify, based on the evidence available in the study, which elements of the technology are most effective.

Summary of the evidence on the effectiveness of delivering and maintaining immersion education for learners of different ages

Early immersion

  • Using bilingual books and code switching in reading material can support learners’ linguistic development in early immersion education in the target language. In addition, 1:1 reading time between adults and the young learners (for example, between children and their parents) is beneficial. 
  • The evidence notes that early immersion learners should be taught in small groups. When teaching in small groups, the specific needs of learners may be identified, and teachers can differentiate on the basis of those needs. 
  • Teacher talk affects learners’ oral ability, and using the target language effectively (by repeating, confirming, detailing and elaborating) predicts learners’ linguistic outcomes. 
  • Technology programmes can support reading skills development in immersion education. This can assist teachers where resources are limited.
  • Using the target language on its own in immersion education is not enough; the teachers’ use of the target language must also be meaningful. That is, teachers should use language which gives learners the opportunity to extend and elaborate when speaking in the target language, rather than just giving instructions in the target language. 

Late immersion

  • Extensive evidence was found that methods targeting reading skills support language acquisition in late immersion contexts. Specifically, teaching learners to read with phonics can help them to link phoneme and graphemes and help with phoneme segmentation. 
  • There is a strong argument in the evidence examined for this study that introducing two languages simultaneously in reading material is beneficial. 
  • The evidence notes the importance of recognising learners’ linguistic context when considering linguistic development in late immersion contexts. This extends to considering learners’ individual needs.
  • The evidence reports that Content and Language Integrated Learning (CLIL) is an effective tool to develop learner confidence, improve knowledge about subjects other than the target language and to develop linguistic skills. However, the evidence does not provide much detail about the exact skills. 
  • Communication methods help learners to acquire the target language. There is evidence that small groups encourage learners to use the target language more often. Also, by using an inductive and explicit approach (learners finding out grammatical structure rules for themselves) learners are equipped to work with peers to better develop and remember grammatical rules. 
  • There is evidence to suggest that using games or technology to assist with learning a language or languages can be effective. Virtual reality technology, specifically, can create a sense of presence which reinforces the immersion experience.

Summary of the evidence about approaches to deal with a range of linguistic backgrounds and previous linguistic experiences among cohorts or groups of learners

  • Evidence about using various languages in immersion classes where learners come from different language backgrounds show that metalinguistic awareness does not lead to higher ability in the target language. 
  • There is strong evidence that using the learner’s home language was important to show the value of those languages. The evidence suggests that intercultural education is required to promote social inclusion and that moving immigrants to the mainstream early on could help with this. However, the link between seeing value in languages and the effective acquisition of linguistic competence is unclear. 
  • Mixing monolingual and multilingual groups can have a positive effect on the linguistic ability of all learners, not just migrants. 
  • Pair work can help develop reading skills in the target language among mixed ability groups.

Summary of the evidence about specific considerations in relation to additional learning needs and immersion education provision

There is scope to support learners with ALN in immersion classes, and small group teaching and encouraging collaboration between peers are effective ways of improving linguistic skills, bearing in mind the learner’s specific ALN.

Translanguaging can assist in identifying and assessing multilingual learners with ALN.

Phonics intervention can address the lack of phonics awareness among learners with ALN.

  • The synthesis presents a wide range of literature on international immersion education methods and approaches. The languages taught in the studies reviewed all represent an education context where the situation is relevant to the Welsh language (i.e. education in a language that is considered a minority language in a bilingual and multilingual context).
  • The evidence offers guidance to policy makers and educators on offering effective immersion education in Wales. However, it should be borne in mind that effective provision could be different in Wales, in comparison with international contexts. When planning provision and evaluating its effectiveness, requirements and circumstances specific to Wales should be a central consideration.

Further research considerations

This report includes evidence of international immersion education approaches and methods. Further research could be carried out into immersion methodology in Wales, looking at the approaches and methods used in early and late immersion contexts, with immigrants, with learners with ALN, and those who use a language other than the target language at home. Research specifically focussing on the effectiveness of the methods and approaches would provide valuable information about immersion education in Wales.


[1] The EPPI Centre (Evidence for Policy & Practice Information Centre) Data Extraction Form was modified. See EPPI-Centre (2007) Review Guidelines for Extracting Data and Quality Assessing Primary Studies in Educational Research. Version 2.0 London: EPPI-Centre, Social Science Research Unit. 

[2] These numbers do not correspond exactly to the number of items, as some include more than one geographical location and immersion context.

Contact details

Report authors: Kathryn Jones and Iolo Jones (IAITH: Welsh centre for language planning), Mirain Rhys, Katharine Young and Adam Pierce (Cardiff Metropolitan University)

Views expressed in this report are those of the researchers and not necessarily those of the Welsh Government.

For further information please contact:
Dr Eleri Jones
Welsh Language Research Branch

Social research number: 37/2024
Digital ISBN 978-1-83577-988-0

GSR logo