Module 2. Analysis and Platform
This page gives a step by step guide to analysing the data from the Cornerstone Community workshops and Questionnaire responses.
This is work-in-progress documentation based on a prototype of the Cornerstone Indicators Platform.
Introduction
This Playbook offers a guide for performing thematic analysis on your community data using the Cornerstone Indicators Platform.
Thematic analysis is essentially a way of looking for patterns in data. While the manual process of coding data is available in our previous version of documentation, we have designed an automatic process using LLM ( Large Language Model ) algorithms to greatly simplify the process and tailor it to the needs of our framework.
The analysis tool is based on the open-source framework developed by Google called JigSaw . That has been used for large-scale deliberation, for example, by a community in Kentucky.
Aims and outputs
The aim is to identify areas and themes, factors, and metrics that represent what thriving means to your community. In the Cornerstone Communities Framework, we are using these terms to mean the following:
Areas and Themes: an overarching area of interest or an outcome that is considered important by the community; for example wellbeing, environmental sustainability, economic stability and community resilience;
Factors: things that contribute to the thematic outcomes; for example good health and green behaviours;
Metrics: existing data points that represent the dimensions; for example mortality rates and cycle route utilisation rates.
Overview of the platform structure
The Cornerstone Platform is organised into the following pages
Overview: information about your community.
Team: management of team members in your workspace.
Taxonomy: an analysis process that results in a taxonomy of relevant outcomes and measurements.
Indicators: an interface that helps to visually design final indicators.
Analysis
1. Registration and Organisation
After registering in The Cornerstone Platform , confirming via email ( Supabase Auth ), and logging in, you will see your workspace.
Create your workspace and give it a name, description and location. The latter is important as the LLM algorithm will tailor results to your specific location.
2. Team Members
You can invite your community members directly into your organisation and work on the indicator together by inviting them and choosing one of 2 permission types. Members will be able to see every page, but won't be able to edit them. Owners will have the same access as you and will be able to edit everything.
3. Taxonomy and Analysis
The Taxonomy page will guide you through the Data Analysis process step by step. You can come back to any stage and adjust it anytime.
1. Contextual Information
This screen will ask you to upload results from your community workshops into the platform. We designed templates that are suitable for the LLM analysis process. Inside the templates, there are two boxes:
Context of the workshop and the question that has been asked
Comments that LLM will analyse to assign them to Themes and generate Factors.
The Platform would take any textual data separated by a line break ( space ), but if you've used our Workshop Playbook from the previous module, you can use templates with the Context box already filled in.
It is important to ensure that there is a line break ( enter ) after each of the Comments from your workshop, as well as ensuring there isn't any in the middle of it. The algorithm will separate each comment when it recognises a line break.
4. Indicator Design
You can now move on to the Indicators page and create your first indicator. The platform serves as a tool to rank your indicator wheel and achieve a quick and easy process of design. For ideating indicators together with your community, we recommend using the Indicator Design Playbook from the next chapter, and come back to the platform when you have a final selection of the Indicators, or in between ideation rounds, to design the wheel and see how your ideas of proxy indicators affect your taxonomy of outcomes.
Last updated