Sentiment Analysis for Qualitative Research in Applied Linguistics

Extending Research Horizons in Applied Linguistics - Between Interdisciplinarity and Methodological Diversity - Hadrian Aleksander Lankiewicz

Karolina Rudnicka [+-]
University of Gdańsk
Karolina Rudnicka. PhD in linguistics obtained from the University of Freiburg, Germany, in 2019, MA in applied linguistics obtained in 2009 from the University of Warsaw, four years of professional experience as a translator/interpreter/lecturer (2009-2013). She currently works as an assistant professor at the Institute of Applied Linguistics and Translation Studies at the University of Gdańsk, Poland. Her scientific interests include i) the application of corpus linguistics in research and teaching, ii) the study of different genres of English and German with an aim of detecting larger cross-linguistic trends and new phenomena, iii) the use of quantitative methods to complement qualitative studies.


In the times when more people than ever before are able to publicly share their knowledge, express their opinions, make commentaries and share pieces of information on every subject possible, researchers in the field of linguistics interested in this kind of output might feel they are faced with too much data to be able to work with. At the same time, the diversity and size of data might still be tempting to be dealt with, as it offers hope for larger trends or important results being hidden somewhere in there. One of novel possibilities is brought by sentiment analysis, also termed as opinion-mining, a notion which refers to “the mining of opinions of individuals, their appraisals, and feelings in the direction of certain objects, facts and their attributes” (Pawar et al. 2016). In an exemplary application of sentiment analysis, answers to open-ended questions are processed with the help of purpose-oriented language ontologies. As a result, information about, e.g. attitude of employees towards their company is successfully inferred. There are a few highly-specialised companies which offer “intelligent” high-quality sentiment analysis to business and governmental customers (e.g. Chenope or Clarabridge). The chapter presents methodological input of sentiment analysis to the field of applied linguistics and discusses problems pertaining to its applications ranging from doing research (also by people without a degree in computational science) to training of future linguists. In particular, the language users (L1 and L2) taking part in the experiment are asked to assess numerous text samples with regard to different criteria pertaining to their textual complexity (e.g. Štajner & Mitkov 2012). The online survey being filled has form of open-ended questions – to enable the participants to give any answer they reckon is appropriate. The answers are processed with the use of a free online tool for sentiment analysis. The obtained results are discussed and compared with a control sample – the surveys which have been assessed by the Author in a purely qualitative, manual way only.

Notify A Colleague


Rudnicka, Karolina. Sentiment Analysis for Qualitative Research in Applied Linguistics. Extending Research Horizons in Applied Linguistics - Between Interdisciplinarity and Methodological Diversity. Equinox eBooks Publishing, United Kingdom. Feb 2023. ISBN 9780000000000. Date accessed: 01 Dec 2021 doi: 10.1558/equinox.41516. Feb 2023

Dublin Core Metadata