Our Voices: using innovative techniques to collect, analyse and amplify the lived experiences of young people in Aotearoa
Assoc Prof Kane Meissel1, Assoc Prof Polly Atatoa-Carr2, Prof Chris Cunningham3, Prof Yun Sing Koh4, Dr Emma Marks5,
Prof Susan Morton6, Dr Georgia Rudd5, Assoc Prof Elizabeth Peterson7, Ash Smith5, Dr Caroline Walker5, Amelia Willems5
1. Faculty of Education and Social Work, University of Auckland, 2. Te Ngira Institute for Population Research, University of Waikato, 3. Research Centre for Hauora and Health, Massey University, 4. Centre of Machine Learning for Social Good, University of Auckland, 5. Social and Community Health, University of Auckland, 6. Foundation Director, Growing Up in New Zealand, 7. Psychology, University of Auckland,
Image source: https://ourvoices.auckland.ac.nz/
Background
Our Generation, Our Voices, All Our Futures (‘Our Voices’ for short) is a research programme funded by MBIE that aims to better understand the diverse and complex journey our young people experience growing up in Aotearoa so we can get a fuller picture of what shapes their wellbeing.
Currently there is a need to address the growing concern of a wellbeing gap among young people. Tackling this issue is challenging due to the complexity and resources required for comprehensive data collection and longitudinal analysis. It also requires a focus on underrepresented groups who may face disadvantages from biases or gaps in service access. By working closely with young people, Our Voices aims to provide rich, child-centric data and analyses to inform future policy decisions, services and support programmes for the benefit of current and future generations.
Data Collection
Typically, it is adults who decide what questions should be asked of children and young people to understand their experiences. ‘Our Voices’ worked directly with young people to co-design what questions should be asked and in what format, to help understand what wellbeing means to them. In partnership with technology and design experts, the young people created concepts for a digital interface that could ask the questions that matter to them in a secure and fun way.
These concepts were developed into an interactive web app called ‘Tō Mātou Rerenga – Our Journey’, which features friendly avatars, mini games and epic journeys of discovery to help users of the app feel safe, curious and engaged. Questions based on the young people’s suggestions were designed to enable multi-modal responses via free text, image, video and audio, inviting young participants to share their experiences in various themes such as identity, hobbies, school, friends and family.
Using technology for data collection from young people in this way required some innovative solutions. For example, this was the first time that Amazon Web Services were utilised in this way by researchers at Waipapa Taumata Rau. Given the sensitivity and complexity of the data being collected, a bespoke administrative portal was developed to enable efficient moderation and screening of the incoming data in accordance with child protection policy principles.
The resulting data was a set of over 60,000 lines of rich qualitative multi-modal data from a diverse group of young people in Aotearoa.
Analysis Methodologies
As the Our Voices data was collected from participants in the Growing Up in New Zealand longitudinal study, researchers had access to a significant amount of contextual data from the lives of these young people from birth to their current age of 14 years.
Given the volume and complexity of the data, the Our Voices team have been working with the Centre of Machine Learning for Social Good to explore the utility of advanced machine learning techniques to facilitate timely analysis. Using the new data collected and the existing Growing Up in New Zealand longitudinal data, machine learning has also enabled us to look at wellbeing trajectories over time.
Novel machine learning techniques are being used to handle both structured and unstructured data, including the multi-modal data of free text, images, audio and video. Leveraging existing state-of-the-art techniques by using pretrained models and fine-tuning them for a New Zealand context means we can analyse complex data to uncover deeper information about youth experiences and wellbeing in a timely manner. These techniques offer a more rapid solution to provide evidence to inform policy decisions and future research that can have a more immediate impact for our young people.
Rather than replacing traditional analyses however, the aim is to supplement results by providing a different lens to pre-existing qualitative methods and to assess the effectiveness of these new methods in the New Zealand context. By leveraging machine learning algorithms, new insights can be extracted from temporally imbalanced data, helping to mitigate biases and aid in expert analysis. Additionally, researchers can triangulate missing data points using historical and current data patterns.
By analysing the model’s outputs, researchers can validate model accuracy while also retrieving valuable insights from this approach. The cross-faculty and multi-institution partnerships of the Our Voices programme demonstrate a transdisciplinary approach towards a shared goal of improving youth wellbeing in Aotearoa. In combining innovative techniques with traditional qualitative methods, the team aims to address the challenges of collecting and analysing longitudinal data, particularly in capturing the voices of underrepresented youth.
Through this approach, Our Voices aims to ensure that no information is overlooked, valuing the experiences and input of all young participants, even those who have not completed every component of the data collection.
Amplifying the voices of young people
Young people are rarely given the opportunity to have their say on important matters that impact their lives and wellbeing, and seldom do their voices reach those in a position of power to make impactful changes.
In addition to more traditional dissemination avenues such as reports and policy briefs, the Our Voices team have partnered with Auckland Museum to create an interactive public engagement experience to share findings and amplify the voices of young participants.
By connecting our research to the public through an exhibit that is participatory in nature, we can ensure the voices of our young people are being heard within a space that is trusted by and accessible to our communities.
The voices of our young people can help make a difference, both for our current generation and future generations.
For more information head to https://ourvoices.auckland.ac.nz/
Assoc Prof Kane Meissel and his research group have been using the University’s storage, Dropbox for collaboration and research virtual machines for several projects. The data storage and computing platform at the University of Auckland are managed and maintained by the Centre for eResearch.