Research Protocol Diagrams as Didactic Tools to Act Critically in Dataset Design Processes

Abstract

In this paper we present the design of research protocol diagrams as a didactic tool in communication design education. The goal is to teach critical aspects of dataset design processes to design students. The context is a five-month course of Data Visualization addressed to master students in Communication Design. For the last five years, the course has been structured in three phases that gradually introduce students to the demanding issues deriving from communication with data in complex situations. In the second phase, students learn how to map and present -through an interactive report -, a controversial issue based on data coming from online sources. The process is question-driven: each group starts with a set of research questions, defines a protocol for data collection and analysis, and produces research findings using data visualizations. In this phase, students are asked to critically reflect on data collection as a design activity and to visually communicate the dataset design process creating protocol diagrams.Dataset design is a critical aspect of the entire research process, raising technical and ethical considerations. From our point of view, designing a dataset means dealing with data as an artifact, as the result of a series of steps leading to the construction of the dataset that will be encoded through data visualization. Even if the protocol diagram represents all the steps of the research, from the question to the final visualization, the most important part is the visual representation of the dataset design process.Visual representation of processes by diagrams is an established practice dating back more than one hundred years. While research protocol diagrams have been widely used in other scientific disciplines to represent processes, they can be repurposed for the field of Communication Design for educational purposes.In this context, protocol diagrams have two functions at different times: before the final delivery, when the diagram is continually updated as the research goes on, is a negotiation tool which helps students in previewing the possible path they should follow to complete the analysis, in disentangling and correlating individual actions, in examining, doubting and legitimizing the arbitrary choices they are led to make in the process of dataset design and finally is a tool for discussing within the group work and with teachers during the weekly reviews.Once students complete their research, they submit a final report that includes the protocols, visualizations, and main findings. Hence, the protocol diagram becomes a dissemination tool addressed both to researchers or future students that want to replicate the same process. Students are free to choose the shape of the diagram they prefer, and they are not provided with international standards to follow. Usually, results are hybrids forms of process charts and activity diagrams attributable to the visual archetype of the flow-chart. The protocol diagram teaches students about the non-objective, situated and interpretative natu...

Staging and storing data sprint-based research results: a communication design approach

“…Telling the process -Tracking and presenting the research process carried on during a data sprint is task-driven usually by designers (Mauri et al, 2019). According to Mauri et al, designing visual protocols is a common procedure for summarizing processes, making them accessible, transparent and repeatable, both during data sprints (Mauri et al, 2019) and as a didactic activity (Mauri et al, 2020). Drawing inspiration from this tradition, the format allows users to self-generate a diagram working as a visual summary and index to quickly access the diverse steps of the research process.…”

What are data sprints for?

“…Third, the methodology followed is described: the data and methods used to answer the research questions. Visual research protocols (see Mauri et al, 2020) help to inform the platform(s) under investigation, dataset building, tools and analyses. 4.…”

El potencial de las redes basadas en la API Google Vision para el estudio de imágenes digitales nativas

“…The protocol diagram (Figure 3) exposes what it takes to build and interpret a computer vision network, considering its technical content that reflects situated online activity, thick layers of technical mediation, and researchers' interventions. The protocol is meant to be a self-explanatory and self-sustainable visual tool highlighting both the dataset design and the visualization process (Mauri et al, 2020;Niederer & Colombo, 2019), also providing a dissemination artifact that allows replicating the same method for building other networks upon the features of different computer vision APIs. We call attention (see Figure 4) to the role of the researcher(s) whom, as we argue, should account for the content of the technical mediums, while engaging with technical practices.…”