Anonymized interview transcripts for Towards recognizing the heterogeneity of English medium instruction (EMI) teacher backgrounds

Open data API in a single place

Provided by Stockholms universitet

Get early access to Anonymized interview transcripts for Towards recognizing the heterogeneity of English medium instruction (EMI) teacher backgrounds API!

Let us know and we will figure it out for you.

Dataset information

Country of origin
Updated
2026.01.12 00:00
Created
2026.01.12
Available languages
English
Keywords
Applied linguistics and educational linguistics
Quality scoring

Dataset description

As the number of English medium instruction (EMI) courses around the world continues to increase, so too does the number of teachers operating in EMI. Previous research on EMI teachers has investigated teacher identity construction, teacher development, perspectives on the role of language in teaching and learning, and classroom practices. However, much existing literature on EMI teachers tends to treat them as a homogeneous group and overlooks the heterogeneity that may offer additional and more nuanced insights into teachers working on the frontlines of EMI. To address this gap, the present study emphasizes the heterogeneity of EMI teachers by investigating two groups of teachers working in the Swedish EMI context: Swedish EMI teachers and expatriate EMI teachers in Sweden. Employing qualitative semi-structured interviews analyzed through deductive content analysis, it examines between-group variables such as whether teachers share a first language with students and the amount and location(s) of their EMI teaching experience. The interviews covered the following topics: teacher background and development; views on student group characteristics; and perspectives on student English proficiency. Comparisons between the Swedish teachers and the expatriate teachers in Sweden show several overlaps but also divergences, including the impact of having EMI experience from multiple contexts and the option of translanguaging.
European data infrastructure with broad catalog discovery, free evaluation access and production-grade API options.
190K+
indexed dataset pages
32
countries and EU institutions
2019
API-first since
Free API quota
for evaluation and prototypes
SLA
history and push on production APIs
FAQ

Questions before production use

Practical answers on evaluation, licensing, freshness, versioning and support.

api.store is built and operated by Apitalks s.r.o. Company details and a direct contact path are linked in the footer for vendor checks and procurement review.
Yes. Selected APIs include a free API quota, so your team can validate coverage, freshness, response shape and workflow fit before asking for a production plan.
Often yes, but usage rights depend on the source license and dataset. We surface source, license and update metadata where available, and can help review terms before a production integration.
Maintained APIs include update metadata where available. For production integrations, we can add history, monitoring and push updates so changes are easier to detect and act on.
Production APIs can add SLA, stable identifiers, versioning support, history, push updates and direct support around the data your product or AI workflow depends on.

Didn't find the API you need?

Let us know and we will figure it out for you.

European data discovery with free evaluation access and production-grade API options.

Copyright © 2026. Made by Apitalks