As part of the COST Action CA21114 (CLIL Network for Languages in Education: Towards bi- and multilingual disciplinary literacies (CLILNetLE)), Working Group (WG) 4 hosted a 2-day online workshop on 6th and 7th of February 2024. The focus of the two days was on developing and enhancing a core set of research skills that are integral to the completion of our research deliverables as part of this COST action.

In WG4, two of our main deliverables include the creation, piloting and dissemination of two questionnaires that form part of our survey of:

1) The digital practices that CLIL learners from across Europe engage in outside of school and how these link to the development of their bi- and multilingual disciplinary literacies in school.

2) The digital practices and resources used in school-based CLIL teaching.

To date, these questionnaires have been piloted in Albania. The questionnaires are predominantly quantitative in nature and we hope to disseminate them in, at least, 6 countries across Europe in the hope that we will be able to gather insights into our two deliverables on a pan-European scale allowing us to draw comparisons between different educational contexts where CLIL is used.

Our focus of the WG4 workshops has been on the steps that we will take once the data has been collected in these different countries: namely, data cleaning and transformation, data coding and data analysis. There was also a focus on the implementation of FAIR data principles and ethical research, as this forms a fundamental part of our work in the COST Action. Our workshops were hosted by Dr Katharina Ghamarian-Krenn and Dr Craig Neville who are part of WG4.

Day 1: The focus of day one was to discuss not only the notions of FAIR data principles and Data Management Plans but, more specifically, how these should be adhered to when undertaking our work for WG4. For us, in order to ensure high levels of reproducibility, we discussed the importance of maintaining good records of data analysis in order to refine our analytical approaches as well as the roles played by coding dictionaries and data glossaries. Some discussion was also afforded to the importance of metadata and how our own work will feed into the creation of the metadata that will eventually accompany our deliverables when they are committed to a data repository at the University of Vienna. The second part of the day then focused on data cleaning and data coding. In this session, participants were able to look at the stages of dataset preparation from the moment that it is downloaded from the questionnaire dissemination tool (in this case, Qualtrics) to the removal of incomplete data. The second part of this session focused on the idea of coding the dataset and, in particular, how different variables can be combined and recoded to support further analysis later on.

Day 2: The second day of our training school focused on getting to grips with our data analysis software, JASP. The first part of the day focused on how JASP can be used to analyse data descriptively and to understand the types of conclusions that can be drawn from descriptive data including detecting normal distributions, measures of central tendency and measures of variability. Participants were then invited to undertake some of this analysis individually using the software to help familiarise themselves with its functionality. The aim is that when the full dataset has been created following the dissemination of the questionnaires in early 2024, participants will be able to interrogate the data through a data analysis lens. The second part of the day saw participants then using basic inferential statistics (such as T-tests and ANOVA analysis) to seek out correlations and similarities between groups.

Overall, these days of the training school offered by WG4 were positively received by participants and we now feel equipped to return with these newly refined skills to further contribute to the deliverables of WG4 and the wider COST Action. These sessions also provided less experienced researchers with the opportunity to engage in continuing professional development related to a very specific skillset required by many researchers, which may support them to go on and undertake other research in which they use many of the skills refined here.

‍

Feedback after 2^ndday (taken from the Zoom chat):

“I wish my stats profs had been so GOOD!!”

“Thank you for the training. We learned so many things!”

“Thank you very much, Katharina! Very practical and useful.”

“Thank you very much for this very useful training.”

“Thank you all for organizing these sessions, it was super clear and useful!”

‍