Introducing Artificial Intelligence Literacy in Schools: A Review of Competence Areas, Pedagogical Approaches, Contexts and Formats

In this episode I unpack Olari, Tenório, and Romeike’s (2023) publication titled “Introducing artificial intelligence literacy in schools: A review of competence areas, pedagogical approaches, contexts and formats,” which is a review of literature exploring how researchers from 31 papers investigated AI-related literacies in schools.

Over the last several months Educators
have been wondering about the potential
for AI in education whether it be
teaching AI or using AI within classroom
context or just for learning outside of
classroom but some Educators might be
wondering well what are some of the best
practices for AI in schools one way you
can actually explore that is by doing a
review of literature now having done
reviews of literature this can take a
long time hundreds of hours depending on
the amount of literature you have to
reveal or you could read somebody else
who's done that work for you or in this
case listen to a podcast on that today's
episode of the csk8 podcast is going to
talk about a review of literature on how
schools are introducing AI within School
context we're going to do that by
looking at the paper which is titled
quote introducing artificial
intelligence literacy in schools colon a
review of competence areas pedagogical
approaches context and formats end
quilts it was written by Victoria allari
Camila tenorrillo and Ralph romica
apologies if I'm mispronounce any names
here's the abstract for this paper quote
introducing artificial intelligence AI
literacy to school students is
challenging as AI education is
constantly growing Educators can
struggle to decide which content is
relevant and how it can be taught
therefore examining which practices and
formats have already been evaluated with
students and are used repeatedly and
which are challenging or should be
explored further is necessary to
facilitate teaching Ai and encourage the
development of new activities in this
literature review We address this need
using a directed and conventional
content analysis we systematically
analyzed 31 cases of introducing AI
literacy in schools in terms of three
categories a competence areas B
pedagogical approaches and c contexts
and formats when analyzing the results
we identified underrepresented
competence areas and summarize common
pedagogical practices in recurrent
formats and context additionally we
investigated the approach to using data
to make abstract AI knowledge accessible
to novices end quote vital summarizes
particular publication into a single
sentence I'd say that this review of
literature explored how researchers from
literacies in schools we're going to be
talking about this particular paper in
relation to K-12 education but also
higher education as that's what we do in
these unpacked King scholarship episodes
of the CSK8 podcast this is episode 190
which means we've got a ton more content
on my website but if you go to
jaredoleary.com you can find the show
notes for this episode as well as
hundreds more and hundreds of more free
resources lesson plans etc for computer
science Educators as well as 1500 hours
of drumming content and a bunch of
gaming content because I create content
for a living and by the way if you're
curious about my professional background
you can find my CV on my website in case
you were curious about all the things
I've done working with evergrade
kindergarten through doctoral student in
a variety of contexts now in the
introduction of the paper the authors
talk about how AI literacy has been a
subject of research for many years it's
actually something we've talked about in
the podcast since episode 13 which was
titled AI for all curriculum development
and gender discourse with Sarah Judd so
if you haven't listened to that episode
I highly recommend taking a look at it
but with AI being such a common Topic in
education at least in recent months has
been more popular outside of computer
science to talk about it the author's
question will well which content is
useful for AI education and how might we
actually teach it or how is it being
taught or explored in the various
research Publications on AI literacy in
schools so in section 2 of the paper
which is titled artificial intelligence
literacy and school education the
authors kind of explore this a bit so
they summarize some Publications and
reviews the literature on this topic so
for example on PDF page two of this
particular publication the authors talk
about AI literacy and so they summarize
this as quote a set of competencies that
enable individuals to evaluate AI
Technologies critically communicate and
collaborate effectively with AI and use
AI as a tool online at home and in the
workplace end quote while other authors
look at AI as a potential way for
expanding computational thinking to do
what's called computational thinking 2.0
at least proposed by some of the authors
and other authors talk about how AI
could be introduced to students in terms
of focusing on perception representation
and reasoning learning natural
interaction and societal impact as it
relates to Ai and learning schooling Etc
there were a number of papers in here
that really pique my interest so I
recommend taking a looking at section
two which is basically the review of
literature to find some other
Publications that might interest you
I'll be doing some unpacking scholarship
episodes in the future on some of these
on PDF page 3 the authors mentioned two
research questions that kind of guide
this particular review of literature one
is which competence areas are
underrepresented in the current AI
literacy approaches and then the second
research question is what are the common
pedagogical practices recurrent formats
and context in AI education so on the
third section of this paper this is on
the method so the authors talk about how
they used an exploratory leader to
review using some various qualitative
content analyzes methods so they use
snowball approaches and keyword searches
to try and find some papers that would
kind of like match some broader criteria
of artificial intelligence in relation
to schooling they specifically were
looking in the Publications within ACM
ultimately they ended up identifying 31
studies and these were from 2010 through
passages within the text like taking a
condensed paper and turning into outline
format and kind of using those like big
headers as codes to kind of summarize
the main sections in this and then you
take those codes and you compare it with
other Publications to see okay well what
are people actually talking about across
all these papers how are they similar
different Etc all right but let's get
into the results so 4.1 is category 1
which is on the competence areas all
right so in this section they talked
about how to use the Doug stove triangle
which is quote a framework for
describing the phenomenon of
digitization that should be included in
education this approach involved
examining each study from three
perspectives a technological does the
study Foster technological competencies
do the students learn how AI Works B
social cultural does the study promote
social cultural competencies did
students learn what the impact of
technology is and see user oriented does
the study promote a user-oriented
perspective did the student learn how to
use AI end quote it's from page four so
from the technological perspective so
this was kind of exploring well what
were students learning about AI in
particular so like how did it work how
would you actually create your own AI or
engage in deep learning techniques Etc
the social cultural perspective was kind
of exploring papers that might have
looked at okay what are some of the
biases that are found when using AI then
the third perspective might include
something like okay well how is it that
we're able to use AI with our own
learning or creating Etc now in their
findings they found that a very small
number of Publications actually explored
all three of those areas simultaneously
or within the same study most of the
studies focused on the technological and
user-oriented perspectives so in other
words quote most studies concern
students ability to know how AI systems
work and how to use them but not what
their effects are end quote so in other
words the researchers were mainly
focusing on the content area of AI or
how to use AI but not necessarily be
ethical impacts of AI in whatever
context is being applied in going back
to some of the relatively recent
episodes that we've talked about with
ethics this presents a potential area
for educators to explore not only
artificial intelligence but how it
connects with ethics if you want to
learn more about that make sure you
check out some of the podcast episodes
that I will link in the show notes at
o'leary.com so now that we've learned
that researchers have mainly explored
the technological and user-oriented
perspectives when it comes to artificial
intelligence and education but let's
talk about well what were the
pedagogical approaches that were used so
this is category two so they found that
many of the research Publications were
talking about how it's important to use
data that is relevant to students rather
than abstract disconnected data that
doesn't really have any kind of impact
or relationship to the students who are
learning it so one of the things that
was frequently brought up was that
students could actually go and collect
their own data and use that to kind of
feed into the machine learning processes
for AI this could lead into the that
third area that wasn't really discussed
very much in terms of biases so if you
are an educator who is interested in
kind of exploring machine learning and
having students collecting and feeding
in their own data into AI when two
different students have two different
results for their AI on whatever the
context may be this presents an
opportunity to kind of discuss well how
might bias in the data that we provided
to have an impact on the result so for
example you might use like a facial
Wrecking thing and talk about how hey
this student with a lighter skin
complexion collected photos from their
family who had similar complexions and
when compared to another student who had
darker complexions the results of the AI
kind of was very different in terms of
making it so that it was easier for the
lighter complexion to detect lighter
complexion and harder to detect a darker
complexion and the opposite might be
true for the other scenario this could
open up a discussion on okay well
depending on what kind of data we
provide it's going to make it so that
the AI is good in one area in terms of
detecting a certain type of complexion
and bad when detecting a different type
of complexion and if we want to make it
so that it's able to equally detect it
or across a range of complexions perhaps
we should feed it a variety of different
photos from different complexions and
whatnot that's just one example but
another example might be like a text bot
or something having students create
something based off of text that you
provided if one student were to feed in
text from like a very technical book
versus another student who's going to
feed in text from like a Twitter feed or
something like that that you might get
very different results in terms of the
kind of responses that you'll get from
that Tech spot and again that presents
an opportunity to discuss well why is
that you don't necessarily have to go
into those projects with the front
loading of hey I'm going to teach you
about bias but can instead anticipate
that bias is likely going to show up in
some way that is going to be easily seen
by the students so you can know in
advance hey at some point down the road
we're going to have a conversation on AI
and bias or Ai and ethics Etc so you can
prepare for that without necessarily
even having to let students know that
hey this is going to happen at some
point if for some reason it does not
happen then maybe at the end of the unit
you can talk about okay now let's talk
about some examples of where this could
have gone wrong even though we were not
able to potentially detect some of the
biases that were happening because
there's always going to be some kind of
biases within the technology that we
create and the algorithms that we write
or the data that we provide to machine
learning Etc but we might not be able to
detect them but again check out some of
those discussions on ethics to kind of
get some ideas on how you could
potentially explore that in your classes
standpoint the authors found that
collaboration was a big focus and so was
active learning so because this was like
summarizing how to introduce AI into
school programs a lot of the
characterizations that could be used to
describe these Publications were
exploring like a low four barrier of
Entry to getting into Ai and education
making things Hands-On playful and
Project based so rather than doing
activities students were actually
creating or engaging in a longer term
project for example using some data for
some machine learning Concepts now in
these particular research Publications
that they're looking at they found that
this wasn't just for a particular age
group this could be done in lower age
groups as well as up through high school
college Etc so you can have younger
students like elementary students engage
in artificial intelligence like in data
collections for things that were like
relevant to them now in category three
this was on context and formats I'm
going to read a couple paragraphs from
page seven of the PDF quote the findings
indicate that the activities described
in the studies were align with the
students's context though they had an
artificially created purpose while the
studies reference context from students
is every day's life such as games
friends animals food
students performed artificially created
tasks example using particular software
to train a classification model with a
data set that is prepared by the
educator and did not transfer their
knowledge into other domains students
applying knowledge to new contexts and
working on projects they cared about
were rarely reported the formats used in
the unanalyze studies range from short
activities and one day workshops to
Summer Schools lasting several weeks
most activities reported were not part
of the regular computer science
curriculum and occurred in the context
of other subjects examples social
studies or outside regular school
classes the activities were based on the
assumption that students had no prior
knowledge of AI and were conducted by
researchers though the studies did not
specify whether School teachers were
also involved in the education process
only Heinz at all bergstein at all
Williams and Brazil casperson at all
reported involving teachers and
preparing them to the multipliers and
test AI materials end quote apologies
that probably mispronounced all those
names alright so this is important to to
think about so the majority of these
research studies that were explored from
again 2010 to 2021 were a short term so
maybe a single day or maybe for a summer
over the course of a few weeks so we
didn't really get a look at long-term
impacts here this was just introducing
AI into schools or school contexts and
these weren't necessarily for computer
science classes or Computer Science
Education that I imagine is going to
change I imagine there's going to be a
lot more discussion a lot more research
on AI and education in a variety of
contexts not only in computer science
context but also in education context
I.E classes that are not specific to
computer science this presents some
interesting things that I'm going to
bring up in some of the lingering
questions and thoughts at the end of
this particular podcast episode however
I want to now discuss the discussion
section which starts on page seven at
the bottom right so here's the first
main finding quote most studies were
concerned with developing students's
ability to know how AI systems work and
how to operate them but not what their
effects are end quote that makes sense
especially considering that the studies
were not necessarily taking place within
a context of Computer Science Education
so again it was like taking place in
like the social studies class or
something else the second main finding
was that quote students were actively
engaged in the learning process however
they were frequently reported to learn
about AI in restrictive context moreover
they did not apply their knowledge to
new domains end quote now why is this
important well transfer is important
because if you're learning something in
one context you want to be able to know
whether or not it actually benefits in
other areas
and that right there is my co-host
Minnie who is my dog now the authors
talk about how this is similar to the
idea of a playpen versus like a
playground in a playpen you are in the
little small enclosed area it's very
safe there's only things in there that
you're supposed to play with and there
are limited number of things that you
can do within that smaller space but in
a playground there might be more things
and more areas to explore maybe even
some areas that aren't necessarily
designed for the thing that you're doing
like having fun same thing we can kind
of apply in education talked about this
in the interview with Gail lovely which
is from episode 11 is titled exploring
Cs and CT in pre-k with Gail lovely so
from an educational standpoint the way
that this relates to this is that if
we're in a playpen you might have a very
nicely crafted educational experience
within this box but there might be
limited opportunities for thinking
creatively Etc within that small space
in a playground area there might be many
more opportunities to explore and to be
creative and run around Etc but it might
not be as tightly crafted of as an
educational experience where you lie on
on that Continuum between like a
playground and a playpen is kind of
determined by the goals that you have
for the students that you're working
with and you like the context that
you're in and like even administrator
pressures stresses Etc but I personally
generally lean much more towards the
playground rather than towards the
playpen so the authors point out that
hey most of these studies we're focusing
on the playpen experience but we should
really explore playgrounds if we were
trying to focus on creativity the play
pen may be a good stepping point but not
necessarily a great destination to end
at now the third main discussion area is
quote AI education is still a marginal
Topic in schools however if the goal is
to spread AI literacy widely it should
be integrated into regular school
lessons more formats should be available
for advanced students and teachers
should be more involved end quote and
the authors point out this is important
to consider because if we're going to
have ai education in schools and in
these research studies we really need to
involve teachers in them and we also
need to help train them which I'll talk
about more in a moment the last thing
that they mention as a main discussion
point on page 8 is quote AI education
appears to be inextricably linked to
data literacy however a solid
theoretical Foundation that explores the
relationship between Ai and data
literacy is lacking end quote in episode
data analysis and implications of Cs as
a skill with and Gunn we talk about data
analysis a little bit so I highly
recommend checking out that episode but
this is a really interesting point that
data goes hand in hand with a lot of
stuff that's going on with AI so if you
want to kind of get a twofer and be able
to explore AI Concepts and practices but
also do some data analysis and
collection Etc you might be able to do
it by combining the two and that right
there is kind of like a summary of the
findings and discussion of this
particular paper it leads to some
lingering questions and thoughts that I
have so one of them is when teaching AI
what balance do you strive for between
learning about Ai and using AI as a tool
for learning or creating in other words
do you focus more on teaching how AI
works or more on how you can use various
AI Tools in your everyday life when do
you focus on one over the other do you
strive for a balance between the two or
do you mainly just focus on I'm only
going to focus on AI as a tool or I'm
only going to focus on how it actually
works it reminds me a lot of like the
discussions on do we focus on Theory or
practical application within a classroom
given that I have like a drum set behind
me do we focus on music theory or do we
focus on composing do we focus on how we
might describe music could we focus on
how to actually write to our own music
there's not necessarily a right or wrong
answer to this but it's something that
you really should consider if you're
going to explore AI in education when
you think of AI literacy are you
focusing on AI as a tool or are you
actually going to encourage kids to
learn how it works or both I'm joined
Again by my doggo mini another lingering
thought that I have is AI and education
is an underexplored area of potential
professional development for educators
having previously designed professional
development and curriculum that's used
by hundreds of thousands of students
around the world this is something I've
spent a lot of time thinking about over
the last several years whether we're
using AI as a tool or learning the
concepts and practices related to Ai and
how it works I imagine that this is
something that individual educators are
going to want to know whether it's
computer science Educators or Educators
at large I also Imagine That individual
consultants and PD providers will be
able to develop content to help teachers
to not only teach how to use Ai and what
it is but they can also use it in their
own work as an educator so for example
using AI to draft emails to parents or
to document student Behavior or progress
or to generate ideas for lessons Etc so
if you are a consultant like some of the
guests that I've had on the show or you
are a professional development or
curriculum provider these are some areas
that you could potentially explore if
you're already exploring these areas and
you'd like to come on to talk about this
in an upcoming episode as an interview
feel free to let me know there's a
contact me button on my website
jaredelary.com now this review of
literature is not necessarily an
exhaustive review of literature but it
is interesting to explore how are people
exploring AI or at least how have they
done it from 2010 to 2021 I'm really
curious how this is going to change over
the next decade like how are teachers
going to be using AI in their class what
about students how we can use it for
teaching how we're going to use it for
learning or for creating what kind of
research is going to come out of this
I'm very skeptical of adding yet another
thing that General classroom teachers
are going to be required to teach in
their classroom but I also think AI is
going to be one of those things that
people are going to use whether they
want to or not just like technology was
but next week I'm going to do a little
bit of a different episode related to AI
so stay tuned for that a little teaser
for you and then we're going to explore
some other areas related to professional
development so stay tuned for that if
you are using AI in your classroom or
you are teaching AI in your classroom
let me know how you're doing that you
can reply in the comments on YouTube or
on social media if you enjoyed this
episode please consider sharing it with
somebody else it just helps more people
find the free podcast episodes and
resources on my website jaredalary.com
or if you know a drummer or a gamer
who'd be interested in the drumming and
gaming content you can share that as
well it's all neatly organized stay
tuned next week for another episode
until then I hope y'all staying safe and
are having a wonderful week

Article

Olari, V., Tenório, K., & Romeike, R. (2023). Introducing Artificial Intelligence Literacy in Schools: A Review of Competence Areas, Pedagogical Approaches, Contexts and Formats.

Abstract

“Introducing artificial intelligence (AI) literacy to school students is challenging. As AI education is constantly growing, educators can struggle to decide which content is relevant and how it can be taught. Therefore, examining which practices and formats have already been evaluated with students and are used repeatedly and which are challenging or should be explored further is necessary to facilitate teaching AI and encourage the development of new activities. In this literature review, we address this need. Using a directed and conventional content analysis, we systematically analyzed 31 cases of introducing AI literacy in schools in terms of three categories: (a) competence areas, (b) pedagogical approaches, and (c) contexts and formats. When analyzing the results, we identified underrepresented competence areas and summarized common pedagogical practices and recurrent formats and contexts. Additionally, we investigated the approach to using data to make abstract AI knowledge accessible to novices.”

Author Keywords

Artificial intelligence literacy, AI education, data literacy

My One Sentence Summary

This review of literature explored how researchers from 31 papers investigated AI-related literacies in schools.

Some Of My Lingering Questions/Thoughts

When teaching AI, what balance do you strive for between learning about AI and using AI as a tool for learning/creating?
AI in education is an under explored area of potential professional development for educators

Resources/Links Relevant to This Episode

Other podcast episodes that were mentioned or are relevant to this episode
- AI4ALL, Curriculum Development, and Gender Discourse with Sarah Judd
  - In this interview with Sarah Judd, we discuss what Sarah learned both in the classroom and as a CS curriculum writer, the curriculum Sarah continues to develop for AI4ALL, advice and philosophies that can guide facilitating a class and designing curriculum, some of our concerns with discourse on gender in CS, my recommended approach to sustainable professional development, and much more.
- Exploring CS and CT in Pre-K with Gail Lovely
  - In this interview with Gail Lovely, we discuss navigating appropriate behavior with digital technologies, some considerations for early and pre readers, how to respond to concerns about screen time, metaphors of education as playpens and playgrounds, learning CS/coding through literacy, and much more.
- Integration through Data Analysis and Implications of CS as a Skill with Anne Gunn
  - In this interview with Anne Gunn, we discuss Anne’s background as both a CS professional and educator, thoughts on data analysis for classroom integration (e.g., sonification), implications of understanding CS as a skill rather than a topic, our experience helping develop the Wyoming Computer Science Standards, and other topics relevant to #CSK8 educators.
- More episodes related to AI
- More episodes related to ethics
- All other episodes
Find other CS educators and resources by using the #CSK8 hashtag on Twitter