LoLo students held their 3rd annual conference on the 16 June 2016, held at UCL, welcoming over 60 participants from across the UK to explore the different methods used in energy demand research.
The day kicked off with a presentation from Despoina Christantoni (University College Dublin) whose work uses an EnergyPlus simulation in combination with machine learning algorithms to predict the demand response potential of a commercial building. Her presentation was followed by a talk by Caroline Hughes (University of Bath) on her work using temperature sensors and questionnaires to investigate the impact of indoor heating patterns on the health of the elderly.
The conference then split into parallel sessions to enable a more in depth discussion of particular research methods. Michelle Shipworth (UCL Energy Institute) chaired a session on quantitative methods where presentations covered the use of quantitative methods as part of a mixed methods approach, the use of randomised-control-trials and the difficulties of measurement in a field where there is a lack of consensus on definitions – one person’s demand temperature is another person’s thermal comfort. Rokia Raslan (UCL Institute of Environmental Design & Engineering) chaired a session on the opportunities and challenges of using modelling to investigate energy demand, presentations highlighted issues such as uncertainty and the need for sensitivity analyses as well as the need to validate assumptions using empirical data.
In the afternoon two LoLo students who have finished their research, Faye Wade (UCL Energy Institute) and Argyris Oraiopoulos (Loughborough University), reflected on the challenges of the methods they’d used. Faye conducted an ethnography of heating engineers and discussed issues such as access to participants, the ethics of reporting what people say, and the challenges in justifying your “hunches” in qualitative research, but also the inherent value of an approach which allows such a detailed understanding to emerge. Argyris’ presentation entitled “4 funerals and a wedding” looked at the challenges of dealing with a huge set of data – in his case hourly recorded temperatures from 228 homes in Leicester – and how he had to try many different statistical techniques over a number of years before finally finding one which enabled him to forecast summer time temperatures in homes, showing that persistence pays off in the end!
In the final methods sessions of the day David Allinson (Loughborough University) chaired a session on measurement with presentations reporting work using both unoccupied and occupied test buildings and ranging in scale from modelling various parameters in a whole building to the moisture content of an area of wall. But despite the differences in focus and scales, all projects highlighted the need for careful planning, calibration, high costs and the difficulties of factoring people into the equation. Victoria Haines (Loughborough University) chaired a parallel session on qualitative methods where presenters used methods including focus groups and document analysis. The presentations led to an extended discussion about the distinction between descriptive and inferential research.
The chairs of all sessions then convened for a lively panel debate discussing the value of mixed methods and the dangers of it stretching researcher’s time and skills, the need both in qualitative and quantitative research to document the cleaning and coding process and recognise it’s interpretive nature, and the need for both quantitative and qualitative research in energy research.
Katy Janda (University of Oxford) then gave the keynote lecture entitled “Telling Tales” and discussing the predominance of “hero stories” that are overly preoccupied with success and failure and how we need to learn to tell a better story. “Learning stories” reflect the real world and help us move forward rather than “hero stories” which inevitably give a fictionalised account of reality. She advocates a “systems stories” approach where learning, caring and other stories are selected depending on the context.
Bob Lowe (Director UCL Energy Insitute & LoLo CDT Director (UCL)) closed the day with a reminder that failure and negative results contribute to the advancement of knowledge and learning and that all metrics, whether quantitative or qualitative, are theoretically constructed and evolve in symbiosis with theories.
Thanks from the LoLo CDT Management team go to the conference steering committee – Pamela Fennell, Ben Roberts, George Papachristou, Catherine Willan, & Zareen Sethna
Photo credits: Ben Roberts & George Papachristou (LoLo) & Mika Laiho (Durham University)