Creative Data Literacy: A Constructionist
Approach to Teaching Information Visualization Catherine D'IgnazioEmerson Collegecatherine_dignazio@emerson.edu
Catherine D’Ignazio is a scholar, artist/designer and hacker mama who
focuses on feminist technology and data justice. She has run reproductive justice hackathons, designed global news
recommendation systems, created talking and tweeting water quality
sculptures, and led walking data visualizations to envision the future
of sea level rise. With Rahul Bhargava, she built the platform Databasic.io, a suite of
tools and activities to introduce newcomers to data science. Her 2020
book from MIT Press, Data Feminism, co-authored with Lauren Klein,
charts a course for more ethical and empowering data science practices.
Since 2019, she has co-organized Data Against
Feminicide, a participatory action-research-design project,
with Silvana Fumega and Helena Suárez Val. D'Ignazio's research at the
intersection of technology, design and social justice has been published
in Patterns, the Journal of Community Informatics, and the proceedings
of Human Factors in Computing Systems (ACM SIGCHI). Her art and design
projects have won awards from the Tanne Foundation, Turbulence.org and
the Knight Foundation and exhibited at the Venice Biennial and the ICA
Boston. D’Ignazio is an Associate Professor of Urban Science and
Planning in the Department of Urban Studies and Planning at MIT. She is
also Director of the Data + Feminism Lab which uses data and computational methods
to work towards gender and racial justice, particularly in relation to
space and place.
Rahul BhargavaMassachusetts Institute of Technologyrahulb@media.mit.edu
Rahul Bhargava is an educator, researcher, designer, and facilitator who
builds collaborative projects to interrogate our datafied society with a
focus on rethinking participation and power in data processes. He has
created big data research tools to investigate media attention, built
hands-on interactive museum exhibits that delight learners of all ages,
and run over 100 workshops to build data culture in newsrooms,
non-profits, and libraries. With Catherine D’Ignazio, he built Databasic.io, a suite of
tools and activities that introduce learners from various domains to
working with data. Rahul has collaborated with a wide range of groups,
from the state of Minas Gerais in Brazil to the St. Paul library system
and the World Food Program. His academic work on data literacy,
technology, and civic media has been published in journals such as the
International Journal of Communication, the Journal of Community
Informatics, and been presented at conferences such as IEEE Vis and
ICWSM. His museum installations have appeared at the Boston Museum of
Science, Eyebeam in New York City, and the Tech Interactive in San Jose.
Rahul is an Assistant Professor in Journalism and Art + Design at
Northeastern University, where he directs the Data Culture
Group.
Alliance of Digital Humanities OrganizationsAssociation for Computers and the Humanities00040301244 February 2019article
This is the source
DHQ classification scheme; full list available at http://www.digitalhumanities.org/dhq/taxonomy.xmlKeywords supplied by author; no controlled vocabularydata literacyconstructionismconstructivismbig datapedagogydata visualizationThe version history for this file can be found on GitHub
Data visualization has rapidly become a standard approach to interrogating and
understanding the world around us in domains that extend beyond the technical
and scientific to arts, communications and services. In business settings the
Data Scientist has become a recognized and valued role . Journalism has re-oriented itself around
data-driven storytelling as a potential saviour for an industry in peril . Governments are moving to more data-driven decision
making, publishing open data portals and pondering visualization as an
opportunity for citizen participation . This
journal itself has numerous examples that use visualization tools and techniques
within the digital humanities as a tool for exploration .
This boom in attention has led large new populations of learners into the field.
Formal educational settings have rushed to create new approaches and
introductions to this content, but often they fall back on traditional
approaches to things such as scientific charting and graphing . Many view data visualization as a new technology,
which runs the risks of replicating old approaches without acknowledging the
unique affordances and domains that data visualization relies upon. Data
visualization is not simply another technology to integrate into education. It
is visual argument and persuasion, far more closely associated with rhetoric and
writing than spreadsheets .
In this paper we present novel approaches to learning technologies and
activities, focused on novice learners entering the field of data driven
storytelling. We begin with a deeper dive into the problems we see with
introducing new learners into a field characterized by inequality, continue with
a discussion of approaches for introducing technologies to education, and
summarize the inspirational pedagogies we build on. We then offer some design
principles and three activities as examples of the concept of creative data
literacy. We assert that creative approaches grounded in constructionist
educational theories are necessary to empower non-technical learners to be able
to tell stories and argue for change with data.
What is creative data literacy?
Introduction
Data visualization has rapidly become a standard approach to interrogating and
understanding the world around us in domains that extend beyond the technical
and scientific to arts, communications and services. In business settings, the
Data Scientist has become a recognized and valued role . Journalism has re-oriented itself around
data-driven storytelling as a potential savior for an industry in peril . Governments are moving to data-driven decision
making, publishing open data portals and pondering visualization as an
opportunity for citizen participation . This
journal itself has numerous examples that use visualization tools and techniques
within the digital humanities as a tool for exploration .
This boom in attention has led new populations of learners into the field. Formal
educational settings have rushed to create new approaches and introductions to
this content, but often they fall back on traditional approaches to things such
as scientific charting and graphing . Many view data visualization as a new technology,
which runs the risks of replicating old approaches without acknowledging the
unique affordances and domains that data visualization relies upon. Data
visualization is not simply another technology to integrate into education. It
is visual argument and persuasion, more closely associated with rhetoric and
writing than software code .
In this paper we present novel approaches to learning technologies and
activities, focused on novice learners entering the field of data driven
storytelling. We begin with a deeper dive into the problems we see with
introducing new learners into a field characterized by inequality, continue with
a discussion of approaches for introducing technologies to education, and
summarize the inspirational pedagogies we build on. We then offer some design
principles and three activities as examples of the concept of creative
data literacy. We assert that creative approaches grounded in
constructionist educational theories are necessary to empower non-technical
learners to be able to tell stories and argue for change with data.
Motivations
A key challenge within the rise of data has been the unequal distribution of data
literacy. When those in positions of power use a new discourse (i.e. data) to
engage in their cultural practices, those that are influenced and don't speak
this discourse are actively excluded from collaborative construction. This
creates data inequality, between those that speak data and
those that do not , which we outline in more detail
in this section.
Research by communications scholars and social scientists supports the idea that
despite the grand hype around Big Data and the knowledge
revolution it will create , there is
profound inequality in who is benefitting from the storage, collection and
analysis of data and who is not . Collectively, these authors argue that data has
become a currency of power. Decisions of public import, ranging from which
products to market, to which prisoners to parole, to which city buildings to
inspect, are increasingly being made by automated systems sifting through large
amounts of data . As a result, knowing how to
collect, find, analyze, and communicate with data is of increasing importance in
society. And yet ownership of data is largely centralized, mostly collected and
stored by corporations and governments . Critically,
the technical knowledge of how to work effectively with data is in the hands of
a small class of specialists. People are far more likely to be discriminated
against with data or surveilled with data than they are to use data for their
own civic ends . This has implications on how people
do social science , practice law , produce
policy , govern the city and create the news , to cite just some of the proliferating work in
this space.
While the scholarship of Critical Data Studies has
focused on algorithmic transparency, data discrimination and privacy concerns,
there has been comparatively less effort on issues of equity in terms of who has
access to the computing power and know-how to be able to make sense of data and
how they come to acquire and deploy that knowledge. Mark Andrejevic has termed
this the Big Data Divide and boyd and Crawford have referred to data-haves and have-nots . Crawford has written eloquently on Artificial Intelligence's White
Guy Problem. Certainly, the fact that there are equity and inclusion issues in data
science is not surprising given the persistence of digital inequality and the lack of women and minorities in STEM fields
. Cultivating data literacy in a more diverse
population is therefore clearly part of any solution or mitigating strategy for
data inequality.
However, teaching data literacy to computer scientists and statisticians is a
different proposition than teaching data literacy to non-technical, adult
newcomers such as humanities scholars, journalists, educators, artists and
non-profit staff. Both authors are educators that introduce data visualization
and storytelling to people that desire to learn new skills but do not come from
technical backgrounds. These include students in the humanities in graduate and
undergraduate settings, non-profit staff trying to become more data-centric in
their work, government officials striving to make data-driven decisions, and
communicators such as journalists, artists and educators. What makes these
audiences similar is that they self-identify more with words than with numbers,
they are often wary of technologies, and feel a lack of confidence when working
with data and looking at visualizations . For these
reasons, we assert that teaching information analysis and design to learners
without technical backgrounds requires a set of alternate approaches.
Creative Data Literacy
In prior work, we drew from prior work in numeracy, statistical literacy and
information literacy to define data literacy as the ability to read,
work with, analyze and argue with data as part of a broader process of
inquiry into the world. Popular press has argued for broad data literacy education . Workshops for nonprofits and activists throughout
the world are introducing tools and practices that can help learners use data to
advocate for social change . However, there is a lack of consistent and
appropriate approaches for helping novices learn to speak data. Some approach the topic from a math- and statistics-centric point of
view . Some build custom tools to support
intentionally designed activities based on strong pedagogical imperatives . Still others have brought together diverse
communities of interested parties to build documentation, trainings, and other
shared resources in an effort to grow the open data movement. Where many of these efforts fall short is on the relegation of data
literacy to a set of technical skills such as reading charts and making graphs
rather than connecting those skills to broader concepts of critical engagement,
social transformation and empowerment.
Reflecting on the background, contexts, and settings of the learners that we work
with drives us to offer more creative and engaging introductions related to data
storytelling and data visualization. This work can draw from long histories of
more creative and empowering approaches to learning. Here we trace a short
lineage of the constructionist approach to education pioneered by Seymour
Papert, inflected by the popular education of Paulo Freire. A constructionist
approach helps ground the introduction of new technologies and concepts into
learning settings without displacing all of the focus onto the technology
itself.
Pedagogical Underpinnings
There are a number of concrete pedagogical theories that educators interested in
creative data literacy can draw from for inspiration. Each suggests a set of
principles to follow, based on an epistemological theory. What follows is a very
short introduction to the key elements we draw from a variety of now-classic
approaches to learning.
The Progressivist approach offers education as a pathway for individuals to
become engaged in the social construction of a society. In the U.S., John Dewey
was a lead figure in the establishment of this approach to education. Their goal
was to give a learner command of himself. Dewey argues against the learner as an empty vessel; focusing on the
child as an active participant in their educational process. He viewed school as
the primary mechanism for social progress - the only sure method of social
reconstruction.
Jean Piaget fleshes out the process by which this type of learning occurs, via
his concepts of assimilation
and accommodation. This work describes how new information and experiences are either
assimilated into a learner’s existing theories, or how the new information
causes the learner to change her theories to accommodate the new information.
This approach values the student as a collaborative learner in educational
settings, helping respect their individual experiences and context.
Lev Vygotsky, coming from the field of psychology, put this type of learning
within a larger context with his discussion of the social construction of
knowledge. He argues that speech is intricately tied to learning through the
development of awareness. This thinking out loud is in fact
how we learn; and requires a social context to occur. His Zone of Proximal Development (ZPD) is his
suggestion for the optimal context within which such speech and cognitive
development occurs . One creates the ZPD by
connecting a learner with a well-informed expert to closely work together;
thereby allowing a learner that is about to master a concept to move beyond
their level of development. The emphasis on interaction in the learning setting
and learning as a social process is a key contribution of Vygotsky's work for
us. This emphasis on social practices for learning is also a central component
to new literacy studies , increasingly used as
multimodal model for literacy in the digital humanities.
Paulo Freire's approach to popular
education rethinks these approaches from a frame of learning as
liberation . He argues that education has
historically been used as a tool of oppression, and we need to change it into a
tool of re-humanization. In parallel, for the educator he offers a path of
re-examination and self-questioning. Freire builds on Dewey's criticism of the
idea of learners as empty vessels, leading to a concept of critical pedagogy.
This frames learners as active agents in their learning contexts, empowering
them to rebuild themselves through education. For Freire, a primary path to
social change is the building up of critical strategies.
bell hooks builds on Freire's critical pedagogy and draws from feminist theory to
articulate an approach to teaching that actively fosters learners' ability to
challenge an unjust status quo. For hooks as well as Freire, education may
either be used to reproduce the current status quo or to empower people to work
together for a more equitable, socially just world. hooks advocates that
education should be the practice of
freedom by which she means the ability to transgress and challenge
existing conditions. hooks introduces dimensions of race, gender and class into
critical pedagogy and argues that teachers should engender critical awareness
and engagement in their students .
Building on Piaget and Vygotsky's work, Seymour Papert's concept of constructionism offered that
constructing out loud with
objects or ideas lets us think about the active embodiments of our learning . His approach revolves around the idea that optimal
learning occurs when people are designing and making theories and objects that
are meaningful to them or their peers — you learn by doing the task with your
peers. This creates opportunities to build knowledge out loud. Like others, this
pedagogy values the learner as a rich individual full of experiences, knowledge,
and ideas that can be engaged to introduce them to new material. Papert’s
intellectual descendants have created a variety of approaches to draw from,
including design principles for building software for learners and guidelines for creating activities to introduce
those tools .
A constructionist approach offers us support for multiple ways of learning. The
constructionist acknowledges that there are many paths for many learners, and
supports the learner in their path-finding endeavors. Support is found for this
idea by connecting constructionism to feminist theories of knowledge, arguing
against things like the superiority of scientific abstract thought http://web.media.mit.edu/~ascii/papers/turkle_papert_1990.pdf.
Deploying Technologies for Learning
How are new technologies, methods and tools adapted into existing learning
situations? It is worth reflecting on questions of adoption - how new
technologies either design for a current learning paradigm or seek to circumvent
or disrupt it. Technologies reflect the social and cultural context of their
creation. When a technology enters an educational situation, such as a school,
it is reshaped and reformed in the image of that setting. This can lead to modes
of use that the creators of that technology never intended, and perhaps even in
opposition to the modes of use they did intend. Technology developers can
respond to this schooling of technology in a variety of
ways. Three approaches that merit more discussion include: Conformity: designing for the dominant paradigm Institutionalism: embedding pedagogy into the technology Insurrectionism: attempting to disrupt the dominant paradigm
The easiest approach is conformity; namely designing for the dominant paradigm.
This approach acknowledges and respects the power of educators and
administrators in deciding what role technology should be allowed to play in
their school setting. It pushes the role of the technology developer to the
side, suggesting that their interests and motivations are secondary to that of
the educational institution. This acquiescence also lends itself to the school
recreating existing power structures and pedagogical paradigms within the
technology, as it seeks to perpetuate itself and its systems (as any existing
organization does). In software, this is known as Conway's law; the idea
that organizations which design
systems... are constrained to produce designs which are copies of the
communication structures of these organizations. For example: If the school uses worksheets to document infractions, the
educational tool ends up using worksheets to introduce students to its features.
We offer Hour of Code as an illustrative current example. This movement,
and the code.org website that supports it, uses traditional models of education
in an effort to introduce millions of children across America to software
coding. Students are brought in groups to computer labs and libraries,
introduced to the code.org website, and offered an hour to go through activities
on the computer step by step, introducing them to the basics of coding . This introduces new concepts within the mechanisms
and norms of the existing school settings.
The insurrectionist approach is more radical; using educational technology to
disrupt the existing paradigms of the intended
destination. This can create an adversarial nature between the developers of
said technology and those in control of the education setting where they intend
their tools to be used. This can be described as a question of working within
the system versus working around the system. An illustrative example here is
the One Laptop Per Child project (OLPC). They attempted to engineer around the
schooling of an educational technology by deploying it
without a standard roll-out, based on iterative early design to gauge student
engagement . Ultimately this approach was written
off by most as a failure, based on metrics of logistical success http://www.bu.edu/writingprogram/journal/past-issues/issue-3/shah/
or standardized testing outcomes http://foreignpolicy.com/2009/09/09/why-did-one-laptop-per-child-fail/.
That debate on the merit of those metrics continues in both academic and
development circles, because the program leaders argue that the pedagogical
impacts of the project have been profoundly felt and impact educational
technologies and technologists to this day.
Conformity, institutionalism, insurrectionism can all serve as models for any
educator looking to innovate on pedagogy for the technology-driven topic of data
visualization. One must pick a path, eyes open to the challenges and
opportunities of each.
Principles for a Constructionist Approach to Data Visualization
Melding these pedagogies, and deciding on an approach to introducing technologies
to support learning data visualization, leads to foundational approaches we
take. In general, we take an institutionalist approach to operationalizing
constructionist theory for the purposes of teaching data visualization. We take
this approach based on our current roles within the world of higher education,
our belief in the power of individual teachers to transform institutional
learning environments, and the opportunity provided by existing educational
structures to get in front of large numbers of learners. The three principles we
draw from the constructionist education are Project-based learning, Hands-on
learning and Peer learning. In this section, we define, explain and give
examples of each principle in turn as well as relate the principle more
specifically to the domain of data literacy and visualization.
Project-based learning (tied to the learner interests)
Project-based learning is a model that organizes learning around projects,
which can be defined as complex tasks, based on
challenging questions or problems, that involve students in design,
problem-solving, decision making, or investigative activities; give
students the opportunity to work relatively autonomously over
extended periods of time; and culminate in realistic products or
presentations. Project-based learning builds on the constructivist and
constructionist ideas that learning is something that is actively built by
learners rather than transmitted linearly from teacher to student - what
called the banking model of education. In
project-based learning, projects help students encounter the central
problems and questions of a discipline, involve students in a constructive
investigation, and are student-driven and realistic .
What does a project-based learning approach look like for data analysis and
information visualization? First, project-based learning is inquiry driven
so the first step of project-based pedagogy with data is helping students to
understand what kinds of questions can be answered (or at least partially
answered) with data and whether the data one needs has been collected in the
first place. Determining which institutions may have collected such data
(and why) involves creative political thinking as well as savvy information
retrieval skills. Once learners understand what kinds of questions may be
answered with data, a project-based learning approach grounded in Freire's
model of popular education guides them to ask questions that have meaning
for themselves and their communities. While many tutorials teach with
datasets such as car performance metrics or server logs, these are data for
which most new learners do not have a rich context of lived experience. If
learners can bring their own context and lived experience to the table, this
will enable them to ask better questions, draw a richer picture of the
data's limitations and perhaps even challenge the data collection practices.
Rather than focusing the majority of attention on the
right and wrong visual tactics
to create polished data visualizations, taking a project-based learning
approach focuses more attention on the exploring and analysis stages of the
process. This involves helping learners ask better questions about the data,
spot missing or bad data, and understand the limitations of what they may
and may not be able to conclude from any given data set. Finally, a
project-based learning approach to information visualization scaffolds
learners through the whole data analysis and visualization pipeline rather
than using different datasets and subjects at each stage. In this way, the
process becomes iterative, rather than idealized and procedural, and the
learners may return to prior stages as they encounter hurdles or identify
better questions to ask.
Hands-on learning (embodied)
Hands-on learning emphasizes the experiential nature of learning over the
transmission (or banking) model of education where an
instructor talks to a room of students who sit quietly and listen. Doing and
making with the active participation of one's hands and body is central to
this approach, articulated well by Progressive Era educators . For Dewey, learning was not about the
passive reception of knowledge which exists a priori out in the world but
about active learning that engages the learner's hands
and five senses in the construction of knowledge. Building off of Vygotsky's
thinking out loud approach described above, Papert shifts the
focus from verbalizing thoughts to creating objects to think with. In particular, Papert showed how learning about computational
thinking - the seemingly abstract digital processes of personal computers
and robots - could be grounded in making concrete representations and
tangible objects. Externalizing the learning process through embodied
creation aids in learners' constructing their own knowledge about the world
.
Creating situations for hands-on learning may seem to be a challenge for
information design, whose outputs one most often sees in two dimensions on
screens. Increasingly, however, educators and artists are experimenting
with ways to visceralize,
physicalize and otherwise make data tangible, embodied and
felt. While creative information displays such as these may not be the end
goal of a particular pedagogical process, showing these examples can expand
learners' thinking about what constitutes information design and what
possible outputs of a data analysis process may look like. There are also a
number of ways that hands-on learning can be incorporated into the process
(rather than the product) of data analysis. Warm-up exercises, introduced at
the beginning of the learning process, might include making data sculptures
from a simple data set or doing a group critique of a large printed
infographic. Hands-on learning at this stage of the process helps to build
confidence for non-technical learners and create a low barrier to entry for
technical subject matter, essential for later learning . In later parts of the data analysis process,
collaborative sketching and low-fidelity paper prototyping can be extremely
important methods to focus learners on the story they want to tell with data
rather than on the mechanics of operating a particular software program or
writing working code . The sculptures, sketches,
and paper prototypes constitute objects to think with
that help learners simultaneously see and reflect on their data analysis
process, as well as test it out with others, in order to take the next
step.
Peer learning
Constructivist approaches to education believe that learning is grounded in
experience and human experience, according to
psychologists such as Vygotsky, is rooted in a social context with sustained
relationships. This points towards incorporating peer learning into
educational situations so that learners may work together on meaningful
problem-solving activities, negotiating their learning through language.
Peer learning can be defined as the acquisition of knowledge through active
helping and supporting among status equals or matched companions . Because this type of learning requires peers to
verbalize concepts, a form of simultaneously teaching oneself and another,
this helps embody and crystallize thought into language . Peer learning connects back to both
project-based learning and hands-on learning because the basic premise is
that people learn better through collaborative, meaningful problem-solving
activities than through listening to an expert or working through exercises
alone.
Our Pedagogy in Practice
We have put this pedagogy into practice in a number of ways, in a variety of
contexts, and with diverse learners. This section describes three specific
activities and discusses how they are motivated by the pedagogy and approach to
using technology in educational settings as described above.
Just-in-Time Experts
Working with data involves piecing together a patchwork quilt of tools. The
sheer number of tools to choose from has made it impossible to introduce
even the most important of them all to students. The co-authors of this
paper have, in separate work, logged more than 500 free or freemium tools
designed for non-technical users to collect and analyze data or create data
visualizations. More critically, there isn't well-established criteria for
helping students understand how and when to piece together this quilt. To
address this challenge we look to peer learning approaches.
In our undergraduate and graduate courses at MIT and Emerson College,
respectively, we have students write reviews of tools for each other as a
way to distribute expertise in the classroom. We created a website called NetStorieswww.netstories.org/tools with reviews of online tools for
working with data, where the reviews are crowdsourced as assignments to the
participants in our classes. In parallel, we maintain an informal list of
any tool for working with data that may or may not have yet been reviewed.
In class, we have learners pick the top 3 tools they are interested to
learn, and then assign each person one tool to learn and write a review
about. Often these are tools that we, as the instructors, have never used.
The audience for each review is other learners, including other teachers.
We have each student present the tool they reviewed for 5 minutes in class,
allowing for broader awareness across the class. For the duration of the
course, students are then considered the class expert
in that tool and the instructor refers other students to them if they need
to use it in their final projects.
Another example involves having students teach tools to each other to
establish criteria for assessing when and how to use them in appropriate
ways. As an example, Bhargava uses this technique to introduce his
undergraduate and graduate students to useful tools for mapping. He
assigned half the class a tutorial that introduced them to CARTO, and
assigned the other half a tutorial for mapping in Tableau Public.https://ocw.mit.edu/courses/comparative-media-studies-writing/cms-631-data-storytelling-studio-climate-change-spring-2017/assignments/
Each group was assigned the task of analyzing the same
dataset geographically using the tool specified. In the subsequent class
the groups were paired across tools; each pair comprised of one person that
learned CARTO and one that learned Tableau. They were then given 15 minutes
each to introduce the other to what they made, how they had done it, and why
that tool might be useful. The activity concluded with reflections about
the comparative advantages, affordances, and limitations of each tool.
Whereas the Just-in-Time experts examples, above, were for undergraduate and
graduate college students, we typically run the WordCounter Sketch a Story activityhttps://databasic.io/en/wordcounter/wordcounter-activity-guide.pdf
in more ad-hoc, one-time workshops for digital humanists, journalists,
nonprofit and government workers. The activity introduces learners to basic
concepts of quantitative text analysis. Learners often approach text as
qualitative data, unaware of the potential for quantitative analysis of text
that computational approaches provide. Counting words is a fundamental
strategy of most computational text processing, including more complex
techniques such as machine learning and topic modeling.
To introduce the tool and the activity we discuss motivations for
quantitative text analysis and try to give domain-specific examples. In the
humanities, for example, scholars are looking to computational methods for
alternate kinds of insights into large bodies of text, which is often
referred to as distant reading. In journalism, news organizations are increasingly under pressure to
make sense of large document dumps like the Panama Papers or the WikiLeaks
diplomatic cables. This is text at a scale which would be impossible for a
single person to read and must be treated quantitatively for the sheer
reason of scale. And in government, cities are increasingly gathering
citizen ideas and comments in qualitative form and need a way to start to
draw out patterns. For the WordCounter activity, we introduce the idea of
analyzing music lyrics quantitatively and show several inspirational
examples of prior work that analyzes music lyrics, including the Rap
Research Lab and
Spotimap.
Once we have explained the motivations, we demo the tool for just five
minutes. WordCounter is a simple web application in which users can paste a
block of text or upload a plain text file. The back-end server uses open
source libraries to count the words, bigrams (two-word phrases) and trigrams
(three-word phrases) in the uploaded text. After submitting their corpus,
the user sees tables listing the most frequently used words, bigrams, and
trigrams along with an option to download a CSV file (Figure 1). WordCounter can operate on sample
data we have included, text the learner pastes in, a file the learner
uploads, or text from a website URL the learner pastes in. It includes two
advanced options - toggles for case sensitivity and the removal of common
words (stopwords).
Working in groups of three, learners have fifteen minutes to decide which
lyrics to run through WordCounter in order to find a
story to tell with the results. The sample data for
the tool includes lyrical corpora for a variety of popular musical artists
in English, Spanish and Portuguese (including Beyoncé, Elvis Presley, Maná,
Legião Urbana, etc.). Each team uses crayons and large pads of paper to
create a sketch of a visual presentation of their story to share with peers
for feedback. The visual outputs resemble those seen in . In our courses, the class then spends 10-15
minutes discussing the stories and using them as a jumping off point for
further discussion of text mining and text analysis concepts.
While the activity is short, the focus on finding a
story is a prompt for learners to use the tool,
interpret and filter the results using prior knowledge of musical artists,
and translate their organizing concept into visual language. The activity
constitutes a simple model for a project-based learning approach to data
visualization in a low-stakes environment with fun subject matter. Peer
groups of three learners work together to negotiate each stage of the
process, including selecting relevant data, suggesting focus ideas, and
developing visual representations with color, hand-drawn icons and pictures.
Translating the story into visual language is a key
part of the hands-on learning aspect of this activity. Learners often start
to make key visual design decisions, such as using color keys to represent
data about individual artists, using familiar icons to represent concepts
such as love, and using size and shape to denote differential
quantities. How novices use visual mappings and narratives in these types of
data-driven sketches is an active area of study in the information
visualization field . In the post-activity
shareback, facilitators work to highlight these emerging design decisions
and point out how they lead to more intentional and systematic strategies
for visual information design. Simple, short activities can embody all three
principles of project-based learning, hands-on learning and peer
learning.
Creating a Narrative Storybook
Our second example is related to constructing strong narratives in
explanatory visualizations. Our hands-on activity guides students through
telling their data-driven story in storybook form, to flesh out the
narrative structure and the key plot points.https://datatherapy.org/activities/activity-write-a-data-storybook/
While introducing the idea of telling stories with visualizations, educators
often break them into two buckets - exploratory or explanatory . Exploratory visualizations provide some
interface, visual or interactive, to manipulate the data being shown. They
allow the user to explore the data and find a story of their own.
Explanatory visualizations use data to tell a strong and clear story
through the use of visual mapping and graphical symbols. They rely on
narrative structure to convey the story of the data to the reader; built
through time, physical space, or both.
Explanatory visualizations that strive to tell a story need strong narrative
structure. We find students often critique this aspect of examples we
discuss. These critiques open lively discussions, but don't provide a space
for students to practice the art of stitching together a narrative. To
allow this, we offer the storybook activity as a playground where students
can explore their story's narrative, without worry about the details of the
visual mappings or symbolical glossary they will build on.
The activity works best once a group of learners has started to narrow in on
the main story they hope to tell. We often use this in classroom settings
after a group of students has picked a dataset they will be working with and
analyzed it to find a particular narrative that they find engaging. Learners
are asked to bring their current story in the form of a template that says
the data say _______, we want to tell that story because _______.
We then give each group a large piece of paper and a pair of scissors. We
then lead them through a classic technique of folding that paper, with one
small cut, into a small book with 3 two-page spreads, a front cover, and a
back cover. Once the paper is in book form, each group is instructed to
write once upon a time… on the cover and the end on the back.
The rest of the pages are available for them to sketch out their story in a
form similar to a children's storybook, using large graphics and story text.
We offer crayons or thick markers as the implements to write with, to both
suggest the playful approach and keep the visuals they choose to include at
low fidelity.
This activity lets participants focus in on the backbone of their story to
make sure it is strong and clear. It doesn't argue against nuance and
detailed story-telling, but rather provides the skeleton from which to hang
the details from. If their story can't be reduced to this simple form, then
their narrative needs further iteration. As with the WordCounter sketching
exercise, this activity gets learners off the screen and back into a space
they might be more familiar with - namely crayons and paper. The
story-time we host at the end, where each group
reads their story to the rest of the class, offers a context for the kind of
social construction of knowledge that Vygotsky discusses. The storybook
artifact itself is a classic thing to think with from a
Constructionist point of view; an embodiment of the learning process that
can be passed around, discussed, and iterated upon until a final polished
data visualization is created. In addition, it brings a simple element of
fun into learning settings in a way that makes sense. This isn't fun for
fun's sake, but rather fun that is had while doing the precise activity that
is being learned.
Other Creative Processes and Outputs
To truly rethink approaches to teaching data visualization, one must take a
step back and reconsider what visualization implies, who it includes,
and who it excludes. The term itself has become loaded - evoking mental
pictures of strong, well designed graphics that depict some data-driven
image of truth. This mythology of the all-knowing data visualization,
occasionally propagated by designers themselves , can be broken down by renaming it as an
information presentation. This recasting is both non-intimidating and
welcoming. Most of our audiences are intimidated by data, but feel
comfortable with information. Few work on visualization, but most give
presentations. We choose our words carefully in workshop and classroom
settings to be more inclusive.
Similarly, the materiality and media used to create data visualizations can
be questioned to move beyond technological fetishism. Technological
expertise need not be a barrier to creating data visualizations, if one
calls them simple visual presentations of information. This opens a door to
using paper printouts of data tables and pie charts with a community group
to collaboratively design a mural, for example. Working together with
community groups, Bhargava has painted 10 such murals around the world . Similarly, one can look to classic low-tech
media from novel sources such as kindergarten to offer a data
sculpture activity that allows learners to sketch
their data-driven stories with pipe-cleaners, googly eyes, and other such
craft materials. These types of creative invitations align more closely
with the pedagogy we described earlier. They specifically follow a
constructionist approach to building out loud, and borrow Freire's
approach to starting with materials that are familiar to and owned by the
learner. While data literacy does not end with pompons and googly eyes,
these can function as a productive starting point to get more people in the
door (and to make the door a little larger).
Conclusion
In this paper we have outlined creative approaches to teaching data analysis and
visualization to adult learners from non-technical backgrounds, including
digital humanists, journalists, government officials, non-profit staff, and
artists. Data is a currency of power but the technical know-how to make meaning
with data is distributed unequally in contemporary Western society. Our
pedagogical strategies are grounded in learning theories out of the tradition of
constructivist education, which view the learner as an active participant in
constructing new knowledge and applying it in productive ways to transform their
social reality. We assert that data literacy for these audiences should be
taught with creative, social strategies over solo exercises and rote learning.
Appraches that emerge from the constructivist perspective include project-based
learning, hands-on learning and peer learning. We describe examples of how these
may be productively applied to learning about data analysis and visualization.
These are just a few examples of what we hope is an expanding repertoire of
constructionist approaches to building data literacy at scale.
That said, we consider this work preliminary. Much work remains to be done in
clarifying and standardizing the definition of data literacy, especially in
relation to a shifting field of technological developments and visual
communication practices. What should data literacy look like for non-technical
learners, who will not go on to be data scientists, but who will need to
communicate with data in their professional lives? Finally, there is room for
inquiry into how to best measure and evaluate the learning that is taking place
for non-technical adult learners. That said, this paper argues that the best way
forward is through engaging learners where they are with hands-on creative
activities that build their capacity. Without such invitations any efforts to
work with novices will fall into a techno-centric focus on software skills
acquisition, which has little chance of connecting learners to the opportunity
of data to help them achieve their goals. This is a critical research agenda for
those in the digital humanities space, who have a history and practice of
working on precisely this concern.
Ackermann, E.K., 2004.
Constructing Knowledge and Transforming the World. In M. Tokoro & L. Steels,
eds. A learning zone of one’s own: Sharing representations
and flow in collaborative learning environments. Amsterdam, Berlin,
Oxford, Tokyo, Washington, DC: IOS Press, pp. 15–37.Agalianos, A.,
Noss, R. & Whitty, G., 2001. vLogo in mainstream schools: the struggle over
the soul of an educational innovation. British Journal of
Sociology of Education, 22(4), pp. 479–500.Andrejevic, M., 2014. The big
data divide. International Journal of
Communication, 8, pp. 1673–1689. Available at: http://ijoc.org/index.php/ijoc/article/view/2161.Berret, C. &
Phillips, C., 2016. Teaching Data and Computational Journalism, Available at:
https://www.gitbook.com/book/columbiajournalism/teaching-data-computational-journalism/details.Bhargava, R., 2014. Speaking Data.
Data Therapy. Available at: https://datatherapy.org/2014/07/09/speaking-data/ [Accessed November
15, 2016].Bhargava, R. et al., 2015.
Beyond Data Literacy: Reinventing Community Engagement and Empowerment in the
Age of Data. Data-Pop Alliance, (September).
Available at: http://datapopalliance.org/wp-content/uploads/2015/11/Beyond-Data-Literacy-2015.pdf. Bhargava, R. &
D'Ignazio, C., 2015. Designing Tools and Activities for Data Literacy Learners.
I Data Literacy Workshop.Bhargava, R. et al., 2016. Data
{Murals}: {Using} the {Arts} to {Build} {Data} {Literacy}. Journal of Community Informatics, 12. Available at: http://ci-journal.net/index.php/ciej.Boyd, D.M. & Crawford,
K., 2012. Critical Questions for Big Data. Communication
& Society, 15(5), pp.662–679.Calzada, P.J. &
Marzal, M.Á., 2013. Incorporating Data Literacy into Information Literacy
Programs: Core Competencies and Contents. Libri,
63(2), pp.123–134. Available at: http://www.degruyter.com/view/j/libr.2013.63.issue-2/libri-2013-0010/libri-2013-0010.xml
[Accessed April 3, 2016].Cockburn, A. &
Williams, L., 2001. The Costs and Benefits of Pair Programming. Extreme Programming Examined, pp.223–243. Available
at: http://dl.acm.org/citation.cfm?id=377517.377531.Collective,
T.T., 2014. Visualizing Information for Advocacy
Second. The Netherlands: Tactical Technology Collective.Conway, M.E., 1968. How do committees
invent? Datamation, 14, pp. 28–31. Available at:
http://libproxy.mit.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&AuthType=cookie,sso,ip,uid&db=bpr&AN=518990226&site=eds-live.Crawford, K., 2016. Artificial
Intelligence’s White Guy Problem. The New York
Times. Available at: http://www.nytimes.com/2016/06/26/opinion/sunday/artificial-intelligences-white-guy-problem.html
[Accessed September 12, 2016]. Crawford, K., Miltner, K.
& Gray, M.L., 2014. Critiquing Big Data: Politics, Ethics, Epistemology.
International Journal of Communication, 8, pp.1
663–1672. Available at: http://ijoc.org/index.php/ijoc/article/view/2167/1164D’Ignazio, C. &
Bhargava, R., 2016. Databasic: Tools and Activities for Data Literacy Learners.
Journal of Community Informatics, 12. Available
at: http://ci-journal.net/index.php/ciej. Dalton, C.M.,
Taylor, L. & Thatcher, J., 2016. Critical Data Studies:
A Dialog on Data and Space, Rochester, NY. Available at: http://papers.ssrn.com/abstract=2761166 [Accessed September 12,
2016].Data Physicalization Wiki
contributors, 'Data Physicalization', Data Physicalization Wiki, 30 April 2021,
12:25 UTC, dataphys.org [Accessed 21
July 2022] (Note: citation added post publication by DHQ Editors)Davenport, T. &
Patil, D.J., 2012. Data Scientist: The Sexiest Job of the 21st Century.
Available at: https://hbr.org/2012/10/data-scientist-the-sexiest-job-of-the-21st-century
[Accessed May 14, 2015].Dewey, J., 1987. My Pedagogic Creed.
School Journal, 54, pp. 77–80. Available at:
http://dewey.pragmatism.org/creed.htm.Dewey, J., 1938. Experience and Education, Collier Books.Diakopoulos, N., 2015.
Algorithmic Accountability: Journalistic investigation of computational power
structures. Digital Journalism, 3(3), pp. 1–18.
Available at: http://www.tandfonline.com/doi/abs/10.1080/21670811.2014.976411
[Accessed September 12, 2016].DiMaggio, P., Hargittai, E.
& others, 2001. From the digital divide to
digital inequality: Studying Internet use as
penetration increases. Princeton: Center for Arts and
Cultural Policy Studies, Woodrow Wilson School, Princeton
University, 4(1), pp.2–4. Available at: http://www.princeton.edu/~artspol/workpap/WP15 -
DiMaggio+Hargittai.pdf [Accessed September 12, 2016].Emerson, J. &
Tactical Technology Collective, 2013. Visualizing
information for advocacy, Bangalore [India: Tactical Technology
Collective.Forlini, S.,
Hinrichs, U. & Moynihan, B., 2016. The Stuff of Science Fiction: An
Experiment in Literary History. Digital Humanities
Quarterly, 10(1).Freire, P., 2000. Pedagogy of the Oppressed, Bloomsbury Publishing. Goldsmith, S. &
Crawford, S., 2014. The responsive city: Engaging
communities through data-smart governance, John Wiley & Sons.
Available at: http://books.google.com/books?hl=en&lr=&id=ttRLBAAAQBAJ&oi=fnd&pg=PA203&dq=info:aSm-_3OsV-MJ:scholar.google.com&ots=TEz8OIQdgJ&sig=3FjAjGOlXvTZRuqiTmfQvoDsoqY
[Accessed September 12, 2016].Gordon, E. &
Mihailidis, P. eds., 2016. Civic Media : Technology ,
Design , Practice, MIT Press.Gray, J.J.,
Chambers, L. & Bounegru, L., 2012. The Data Journalism
Handbook 1 edition. Sebastopol, CA: O’Reilly Media, Inc.Gurstein, M.B., 2011. Open data:
Empowering the empowered or effective data use for everyone? First Monday, 16(2). Available at: http://firstmonday.org/ojs/index.php/fm/article/view/3316 [Accessed
July 16, 2014].Harel, I.R., 1988. Software design for
learning : children’s construction of meaning for fractions and LOGO
programming.Harris, J., 2012. Data Is Useless
Without the Skills to Analyze It. Harvard Business
Review. Available at: https://hbr.org/2012/09/data-is-useless-without-the-skills [Accessed
May 14, 2015].hooks, bell, 1994.
Teaching to transgress : education as the practice of
freedom, New York: Routledge.Hourcade, J.P. et al.,
2009. Early OLPC experiences in a rural uruguayan school. Mobile Technology for Children, pp. 225–243. Howard, A., 2014. The Art and Science
of Data-Driven Journalism. Tow Centre for Digital
Journalism - a Tow/Knight Report. Available at: http://towcenter.org/the-art-and-science-of-data-driven-journalism/.Hoyt, E., Ponto, K. &
Roy, C., 2014. Visualizing and Analyzing the Hollywood Screenplay with
ScripThreads. Digital Humanities Quarterly, 8(4). Huron, S. et al., 2014.
Constructive visualization. In ACM, pp. 433–442. Available at: http://dl.acm.org/citation.cfm?id=2598566 [Accessed May 12,
2015].Iliinksy, N. &
Steele, J., 2011. Designing Data Visualizations,
O’Reilly Media. Jacobs, J.M., Cairns, S.
and Strebel, I., 2016. Materialising vision: Performing a
high-rise view. In Visuality/Materiality (pp. 147-166). Routledge.Keating, J., 2009. Why did One
Laptop Per Child fail? Foreign Policy. Available at: https://foreignpolicy.com/2009/09/09/why-did-one-laptop-per-child-fail/.Kennedy, H. et al., 2016.
Engaging with (big) data visualizations: Factors that affect engagement and
resulting new definitions of effectiveness. First
Monday, 21(11). Available at: http://firstmonday.org/ojs/index.php/fm/article/view/6389. Kirchner Surya Mattu, J.L.L.J.A.,
2016. Machine Bias: There’s Software Used Across the Country to Predict Future
Criminals. And it’s Biased Against Blacks. ProPublica. Available at: https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing
[Accessed September 12, 2016].Letouzé, E. et al., 2015.
Beyond Data Literacy: Reinventing Community Engagement
and Empowerment in the Age of Data.Maycotte, H.O., 2014. Data
Literacy -- What It Is And Why None of Us Have It. Forbes. Available at: http://www.forbes.com/sites/homaycotte/2014/10/28/data-literacy-what-it-is-and-why-none-of-us-have-it/
[Accessed May 14, 2015].McCandless, D., The beauty of
data visualization | TED Talk | TED.com. 2010. Available at: http://www.ted.com/talks/david_mccandless_the_beauty_of_data_visualization.Montessori, M., 2013. The montessori method, Transaction publishers. Moretti, Franco. Distant reading. Verso Books, 2013.Murnane, J.S., 2010. And They Were
Thinking? Basic, Logo, Personality and Pedagogy. International Federation for Information Processing,
pp.112–123.Naylor, B., 2016. Data Brokers:
Companies Are Buying, Sharing Your Online Info. What You Can You Do About It?
All Tech Considered : NPR. Available at: http://www.npr.org/sections/alltechconsidered/2016/07/11/485571291/firms-are-buying-sharing-your-online-info-what-can-you-do-about-it.Neuhauser, A., 2015. 2015 STEM
Index Shows Gender, Racial Gaps Widen. US News & World
Report. Available at: https://www.usnews.com/news/stem-index/articles/2015/06/29/gender-racial-gaps-widen-in-stem-fields.O’Hear, B., 2016. Paper Prototyping.
Medium.com. Available at: https://blog.revelate.de/paper-prototyping-d53fafc71b78#.21ek6q4uh.O’Neil, C., 2016. Weapons of math destruction : how big data increases inequality and
threatens democracy, New York : Crown.Papert, S., 1994. The Children’s Machine: Rethinking School In The Age Of The
Computer Revised ed., Basic Books. Papert, S., 1980. Mindstorms: children, computers, and powerful ideas, New York:
Basic Books. Pasquale, F., 2015. The black box society : the secret algorithms that control
money and information.Patil, T.H.D.J., 2012. Data Scientist:
The Sexiest Job of the 21st Century. Available at: https://hbr.org/2012/10/data-scientist-the-sexiest-job-of-the-21st-century
[Accessed May 14, 2015].Perovich, L., 2014. Laura {Perovich}.
Big Bar Chart in Laura Perovich’s Portfolio.
Available at: http://www.lauraperovich.com/projects/bigbarchart.html [Accessed July
28, 2016]Piaget, J., 1952. The origins of intelligence in children., New York: International
Universities Press.Porter, L. et al., 2013.
Success in Introductory Programming: What Works? Communications of the ACM, 56(8), pp. 25–28. Available at: https://proxy.tamuc.edu:2048/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=89594060&site=eds-live.Resnick, M. &
Rosenbaum, E., 2013. Designing for Tinkerability. Design,
Make, Play: Growing the next generation of STEM innovators, pp.
163–181. Resnick, M. &
Silverman, B., 2005. Some Reflections on Designing Construction Kits for Kids.
Proceeding of the 2005 conference on Interaction design
and children (IDC ’05), pp. 117–122. Available at: http://dl.acm.org/citation.cfm?id=1109540.1109556.Roberts-Smith, J.
et al., 2013. Visualizing Theatrical Text: From Watching the Script to the
Simulated Environment for Theatre (SET). Digital Humanities
Quarterly, 7(3).Sandvig, C. et al., 2014.
Auditing algorithms: Research methods for detecting discrimination on internet
platforms. Data and Discrimination: Converting Critical
Concerns into Productive Inquiry. Available at: https://pdfs.semanticscholar.org/b722/7cbd34766655dea10d0437ab10df3a127396.pdf
[Accessed September 12, 2016].Shah, M., 2011. A Blurry Vision:
Reconsidering the Failure of the One Laptop Per Child Initiative. WR 150, no. 3. Available at: http://www.bu.edu/writingprogram/journal/past-issues/issue-3/shah/Mayer-Schönberger, V. & Cukier, K., 2013. Big Data: A
Revolution That Will Transform How We Live, Work, and Think,
Houghton Mifflin Harcourt.Slater, D. & Garcia, M.,
2016. The Methodologies of Data Literacy. School Of Data
Blog. Available at: http://schoolofdata.org/2016/01/14/research-results-part-2-the-methodologies-of-data-literacy/.Street, B., 2003. What’s
new in New Literacy Studies?: Critical approaches to
literacy in theory and practice. Current Issues in
Comparative Education, 5(2), pp. 77–91.Thomas, J.W., 2000. A Review of
Research on Project-Based Learning. The Autodesk
Foundation, pp. 1–45. Available at: http://www.bobpearlman.org/BestPractices/PBL_Research.pdf.Topping, K.J., 2005. Trends in Peer
Learning. Educational Psychology, 25(6), pp.
631–645.Turkle, S. & Papert,
S., 1992. Epistemological Pluralism and the Revaluation of the Concrete. Journal of Mathematical Behavior, 11(1), pp.
3-33.Tufekci, Z., 2014. Engineering the
public: Big data, surveillance and computational politics. First Monday, 19(7), pp. 1–38.Tygel, A. & Kirsch, R.,
2015. Contributions of Paulo Freire for a critical data literacy. Proceedings of Web Science 2015 Workshop on Data
Literacy. Available at: http://www.researchgate.net/profile/Alan_Tygel/publication/278524333_Contributions_of_Paulo_Freire_for_a_critical_data_literacy/links/5581469d08aea3d7096e6a31.pdf
[Accessed July 23, 2015].Voakes, P.S., Beam,
R.A. & Ogan, C., 2003. The Impact of Technological Change on Journalism
Education: A Survey of Faculty and Administrators. Journalism and Mass Communication Educator, 57(4), pp. 318–34. Vygotsky, L., 1980. Mind in society: The development of higher psychological
processes, Harvard University Press.Walny, J., Huron,
S. & Carpendale, S., 2015. An Exploratory Study of Data Sketching for Visual
Representation. Computer Graphics Forum, 34(3), pp.
231–240.Webber, H. et al., 2014. The
graph choice chart. The Science Teacher, 81(8), pp.
37–44.Welles, B.F., 2014. On minorities and
outliers: {The} case for making {Big} {Data} small. Big
Data & Society, 1(1), pp. 1–2. Available at: http://bds.sagepub.com/content/1/1/2053951714540613 [Accessed April
25, 2015]Willett, W. & Huron,
S., 2016. A Constructive Classroom Exercise for Teaching InfoVis. In IEEE VIS. Available at: http://vgl.cs.usfca.edu/pdvw/2016/abstracts/Willett.pdf.Williams, S. et al., 2015.
City Digits: Local Lotto: Developing Youth Data Literacy by Investigating the
Lottery | Journal of Digital and Media Literacy. Journal of
Digital and Media Literacy, 2(2). Available at: http://www.jodml.org/2014/12/15/city-digits-local-lotto-developing-youth-data-literacy-by-investigating-the-lottery/
[Accessed March 19, 2015]. Wilson, C. & Cameron,
2015. Hour of code---a record year for computer science. ACM Inroads, 6(1), pp. 22–22. Available at: http://dl.acm.org/citation.cfm?doid=2735450.2723168 [Accessed
November 14, 2016]. Wolff, Annika and Kortuem, Gerd (2015).
Visualising energy: teaching data literacy in
schools. In: Sencity 2, 08 Sep 2015, Osaka.Zer-Aviv, M., 2014. Disinformation
Visualization: How to lie with datavis. Visualizing
Information for Advocacy. Available at: https://visualisingadvocacy.org/node/643.html.Zimmerman, B. J., 2000.
Attaining self-regulation: A social cognitive perspective. In M. Boekaerts, P.
Pintrich & M. Zeidner (Eds.), Handbook of
self-regulation, pp. 13–39. San Diego, CA: Academic Press.