Corpus approaches to discourse: using topic modelling to understand widening participation discourses through the Access Corpus

In recent years, widening access to higher education has become an ever-increasing priority for English higher education institutions. This is evidenced by: an increase in the amount invested by the sector in this area of activity to £248 million annually (in 2020), an increased policy focus on the evaluation of widening participation initiatives, and an increased acknowledgement of the central widening participation problem: “You are still, however, 2.3 times less likely to enter higher education if you are from the least advantaged group relative to the most advantaged group and 5.5 times less likely in the higher tariff institutions” (OfS, 2018: 3).
This talk presents the findings of topic modelling and corpus linguistic techniques (e.g. collocation analysis) applied to a novel corpus – the Access Corpus. The Access Corpus consists of documents, produced each year by English higher education institutions, which explain how particular universities aim to widen participation. The corpus contains ~12 million words and represents documents from 109 English higher education institutions over 15 years (2006 – 2021). Analysis focuses on two specific junctures within the timeline of the corpus: a. 2012 at the tripling of tuition fees; b. 2018 at the introduction of the Office for Students. Further analysis explores discussions relating to student characteristics: specifically concerning ethnicity and student ‘aspirations’ for higher education.

All welcome to join in person.
If you wish to join online, pre-registration is required:
After you register, you will receive a confirmation email with joining instructions.