Background
This toolkit was originally developed over the course of 2022 to support Transforming Hawaiʻi’s Food Systems Together (THFST) initiative’s internal efforts to draft a statewide policy framework for Hawai’i. We were asked by THFST to attempt to answer two questions:
To answer these questions, we read, summarized, and analyzed 25 plans and charters, interviewed 20 food systems planners and academics, and conducted extensive desk research. Over the course of our research, we heard from food systems planners and groups across the country that there is a strong need for this kind of foundational and systematic research in their own work. Thus, starting in June of 2022, our team worked to translate original research into the publicly available resources that you see here.
Selection of Plans & Charters
Because our research was originally developed for a state-level planning initiative in Hawai’i, the majority of the plans and charters analyzed are at the state or inter-state level. These plans were all selected based on their inclusion in the 2021 report from Michigan State University’s Center for Regional Food Systems, “Participatory State and Regional Food System Plan and Charters in the U.S.: A Summary of Trends and National Directory.” State- and inter-state plans and charters were selected along the following criteria:
a. “are systems-based and cross-sector (covering the entire food system),
b. “propose visions beyond 2021 or that are currently being updated (e.g., the 2005 California plan proposed a vision of the food system into 2030), and
c. “were participatory and collaboratively developed (those that crowdsourced ideas and attempted to mobilize stakeholders throughout the state using numerous strategies such as summits, working groups, etc.).”1
In addition, three city- and two county-level plans and charters were identified via snowball sampling and included based on internal relevance to the original project. Given the small sample size, we recognize that this is not a comprehensive or representative list of city- and county-level planning efforts. As a result, these local plans are excluded in any summary statistics, but remain within the database for the valuable insights that they may offer to other food systems initiatives. The content and creation strategies within city- and county-level plans and charters can be filtered separately within database searches. Future iterations of this project may strive for more robust inclusion of major city and county plans and charters.
Finally, this toolkit includes any major plan updates as separate inputs from their original versions, on the assumption that their individual processes may still hold relevance to future initiatives. For example, Michigan’s 2022 update and Los Angeles’s 2017 update are included as separate entries from their original iterations.
Data Entry & Review
The two databases (plan catalog and recommendation database) were created in tandem with each other. The format and information of each of these databases were designed to answer the primary questions of our original research.
For initial data entry, plans and charters were read through by a primary researcher. The catalog of plans and charters was populated by distilling information available within the plan and supplemental plan documents, as well as from the websites of the planning bodies, funders, and related stakeholders. The recommendation database was populated by entering, verbatim, all recommendations contained within each plan or charter. After both databases were filled, each entry was reviewed for accuracy by a secondary reviewer. To ensure complete accuracy within the catalog of plans and charters, individual entries were sent via email to key plan authors and/or coordinators for review and confirmation. We were able to reach 80% of authors across all plan types (city, county, state, and inter-state) for final review. 90% of the entries for state-level plans/charters were reviewed and approved by their authors. 2
Data Analysis
Recommendation Analysis
In order to analyze content within the plan recommendations, a coding system of food system elements was developed using an iterative, three-step process of thematic analysis3.
We first created a list of basic elements within food systems, building upon the established systems maps and literature from the Food and Agriculture Organization and GAIN/JHU/FAO. We subsequently cross-referenced their articulated components of a food system across the literature (Mui et al. 2021; Karetny et al. 2022) that analyzes existing food systems plans and charters’ content and added new elements, if identified, to our initial list of codes. From there, we finalized our working coding system by adding codes for social components around equity and justice (Mui et al.) and for more specific subcategories of food system elements (Karetny et al.). Using this list as a reference, all food systems plan and charter recommendations were subsequently reviewed. Food systems elements consistent with the literature and initial coding system were identified, and new food systems elements found within each plan and charter were highlighted and considered for inclusion as new codes. Following this, a complete list of “Key Planning Elements” (KPEs) was refined, consisting of 16 key planning elements and 65 subcodes (a complete list of codes can be found below). Recommendations were then analyzed and cross-coded based on this finalized list of “Key Planning Elements” by a team of three independent reviewers. When there were disputes about how to code, the complete reviewer team would meet to form a consensus. Consistent with the interdisciplinary nature of food systems, many recommendations affect multiple elements of the food system and bear multiple codes as a result.
Once the coding was completed, the KPEs were translated into a visual map to highlight common points of connection between goals and recommendations within city, county, and state plans and charters.
Lastly, metadata on KPEs was generated by finding “frequency” data on each KPE’s use within state-level plans and charters. This included 1) information on the proportion of state-level plans/charters that made recommendations coded with each KPE at least once, and 2) information on the proportion of all recommendations coded with each KPE, both within each state plan/charter and across all state plans/charters. As our sample size for plans and charters from municipalities and counties was low (n = 5), they were not included in any summary statistics (see “Selection of plans & charters”).