►
From YouTube: Education & Workforce WG
Description
Fostering data science pathways from associate’s to bachelor’s degrees
Presenter: Nicholas Horton
Amherst College
A
Well
again,
thank
you
very
much,
renate
and
everyone
and
john
a
really
good
talk
that
I
think
dovetails
really
well,
as
we
continue
up
the
age
spectrum.
You
know
we
know,
there's
a
tremendous
number
of
high
school
students
who
are
dual
enrolled
in
community
colleges,
and
I
think
that
need
to
be
thinking
about
data
science,
development
early
and
often
is
really
critical.
I
want
to
acknowledge
that
much
of
the
work
I'm
talking
about
today
is
with
my
collaborator
ben
baumer
from
smith
college,
who
couldn't
be
here
today.
A
A
Three
community
colleges,
five
four-year
colleges
and
a
university
doing
our
assessment.
The
project
has
two
goals:
one
is
to
foster
experiential
learning
with
community
organizations,
creating
opportunities
to
scale
up
data
science
for
social
good,
and
the
second
goal
is
what
really?
What
I
want
to
talk
about
today
is
trying
to
foster
data
science
at
community
colleges.
A
There's
really
I'm
not
going
to
talk
much
but
I've,
given
the
links
to
some
papers,
we've
written
about
our
work
of
how
to
develop
data
acumen
for
undergraduate
students,
doing
these
research
projects
using
scrum
and
agile
development
techniques
for
data
science.
Again
those
papers
are
available
and
the
links
are
there
and
that's
really
not
what
I'm
going
to
be
talking
about
today.
That's
pretty
well
done
that
part
of
the
project.
A
We've
learned
a
lot
and
what
I
really
want
to
talk
about
is
how
we
can
be
building
data
science
programs
at
community
colleges,
because
this
is
critical
for
a
number
of
reasons
that
are
outlined.
So
our
goal
is
to
really
kind
of
think
about
motivating
the
importance
of
data
science.
This
is
a
hard
thing
at
present,
given
that
community
colleges
have
been
slammed
by
the
pandemic.
If
you
look
nationally,
you
see
enrollments
down
10
to
15
percent.
A
The
people
who
are
kind
of
you
know,
students
at
community
colleges
have
had
the
worst
time
of
any.
I
think
in
terms
of
dealing
with
the
pandemic,
and
it's
really
sabotaged
these
community
colleges
efforts
to
be
able
to
kind
of
bolster
new
programs
they're
struggling
just
to
stay
alive
at
this
point,
so
we
have
to
start
by
motivating
the
importance
of
data
science
and
familiarizing
leadership,
with
some
of
the
options
that
are
being
explored
nationally.
A
Our
goal
is
to
help
prepare
faculty
to
teach
data
science
and
by
doing
so
learning
what
the
barriers
and
impediments
are
there
and
we're
seeing
those
are
considerable.
In
summary,
they
don't
have
time
and
they
don't
have
money,
but
again,
there's
a
lot
of
initiative.
There's
a
lot
of
motivation,
really
talented
and
engaged
people,
and
a
lot
of
the
same
kind
of
things
I'm
describing
are
the
same
initiatives.
A
I
think
we're
seeing
at
some
of
our
minority
minority
serving
institutions
and
other
institutions
that
have
been
historically
under
supported
in
our
system,
so
really
what
we
want
to
focus,
though,
on
the
student
experience,
what
are
the
barriers
for
students
to
be
able
to
pursue
data
science
programs?
They
need
to
hear
about
this
early.
They
need
to
be
able
to
understand
what
their
options
are
and
they
need
to
have
multiple,
flexible
pathways
to
be
successful
with
it.
A
I
think
they're
absolutely
critical.
We
know
that
there's
a
substantial
fraction
of
our
undergraduate
students
or
community
colleges.
There
are
more
people
taking
introductory
statistics
at
community
colleges
and
at
four-year
institutions.
It's
the
only
affordable
game
in
town.
There
are
some
exceptions
to
that
in
various
places,
but
when
we
think
about
affordable
education,
they're
really
structured
in
a
way
to
really
you
know,
foster
that
for
people
and
they're
really
critical
to
develop
the
educated
workforce
that
we
need.
A
I
think
one
of
the
points
that
came
out
really
directly
in
the
keep
data
science
broad
is
that
these
institutions
tend
to
be
far
more
representative
of
the
population,
the
most
four-year
institutions,
the
number
of
first-generation
students,
low-income
students
of
color,
are
dramatically
higher
at
these
institutions
and
when
we
talk
about
the
need
to
expand
opportunities
for
everyone,
not
just
those
who
are
kind
of
privileged
who've
gone
to,
you
know
well-resourced,
school
districts.
This
is
an
area.
These
schools
are
already
serving
these
folks
and
there's
really
a
need
for
flexible
and
innovative
articulation.
A
A
So
a
couple
of
things
I
want
to
just
mention,
if
you
haven't
seen
them,
the
national
academy's
report
on
data
science
for
undergraduates,
came
out.
It
defined
data,
acumen
components
thereof
and
really
talked
about
the
importance
of
community
colleges
and
the
need
for
cooperation
with
other
institutions.
A
The
link
to
the
free
report
is
given
below
it's
also
in
the
etherpad.
As
I
mentioned,
the
south
big
data
hub
and
renata
and
crew
brought
us
together
to
talk
about
ways
of
ensuring
that
data
science
is
broadly
inclusive,
that
we
don't
repeat
the
issues
that
have
led
to
the
dramatic
disparities
we're
still
seeing
racial
gender
disparities.
A
We
can't
let
that
happen.
We
have
to
make
sure
as
we
design
this
new
field,
it
remains,
including
equitable
and
there's
a
lot
of
work
for
us
to
be
doing.
I
think
we're
making
some
baby
steps,
but
we
need
to
really
double
down
in
those
efforts
that
report,
I
think,
helped
highlight
the
importance
and
the
two-year
college
data
science
summit
was
funded
by
the
national
science
foundation.
A
It
brought
together
a
group
of
people
to
try
to
identify
what
was
different
at
community
colleges,
how
we
can
support
faculty
development,
curriculum
initiatives
and
also
built
on
some
of
the
you
know,
results
and
findings
from
the
keeping
data
science
abroad.
So
those
are
great
background.
If
you
haven't
read
them,
I
think
they're
really
useful.
I
want
to
spend
just
a
couple
of
minutes
talking
about
what
are
some
of
the
barriers
and
opportunities
for
transfer
pathways.
Again,
I
said
we
need
a
student-centered
approach
to
foster
success.
A
We
need
to
kind
of
be
thinking
about
this
in
mind
of
how
students
are
able
to
kind
of
proceed,
what
we
need
them
to
be
doing
and
how
to
structure
and
support
them
for
success.
They
said
they're
motivated,
they're
engaged,
but
again
we
need
there's
a
lot
of
barriers
in
their
place.
There
needs
to
be
more
assort
of
these
associates
to
workforce
programs,
we're
starting
to
see
those
the
office
of
personnel
management
has
put
together
kind
of
what
jobs
look
like
from
the
associates
level.
A
I
think
that's
going
to
be
happening
over
the
next
three
to
five
years
in
larger
numbers.
My
focus
today
is
to
talk
about
associates,
to
transfer
to
think
about
a
two-year
degree
that
leads
directly
directly
and
seamlessly
to
a
four-year
degree.
So
what
are
the
points
of
friction?
We
need
to
smooth
or
eliminate.
I
could
talk
for
a
long
time
about
each
one
of
the
five
that
we've
identified
here.
Let
me
just
briefly
introduce
those
the
first
one
is
duh.
We
need
a
first
course
to
data
science
to
be
taught
at
these
schools.
A
We,
you
know,
in
order
for
any
of
these
pathways
to
work,
students
must
have
an
opportunity
to
take
a
first
course
in
data
science.
They
can't
wait
till
their
juniors
to
do
that
if
they
want
to
develop
sufficient
data
fluency
and
data
acumen.
Later
on
and
again,
this
is
ties
in
directly
with
the
nasam
recommendation
that
academic
institutions
should
encourage
to
encourage
the
development
of
a
basic
understanding
of
data
science
in
all
undergraduates.
A
This
could
be
involve
a
joy
of
data
course,
as
well
as
a
more
technical
one
like
what
we've
seen
the
berkeley
data
eight,
but
there
needs
to
be
a
basic
data.
Science
course
this
course
could
be
valuable
for
students
in
psychology
or
nursing
or
biology
as
well,
to
give
them
kind
of
basic
introduction
to
their
data
literacy.
But
we
need
to
make
sure
that's
there.
Most
schools
are
still
not
offering
such
a
course.
A
A
One
of
the
challenges
here
is
that
when
you
look
at
four-year
institutions,
their
second
courses
in
data
science
are
hugely
heterogeneous
and
we
need
to
kind
of
make
sure
that
there's
kind
of
some
clear
understandings
of
what's
transferable
in
there
as
a
second
course
again.
Building
on
the
basic
data,
visualization
and
and
data
wrangling
that
might
one
might
see
in
the
first
course
we
see
later
on,
there's
a
need
for
a
course
that
covers
topics
in
scientific
computing,
data,
science,
workflow
and
reproducible
computing.
A
We
know
that
reproducibility
and
responsible
workflow
is
a
critically
important
part
of
structuring
and
organizing
data
science.
We
need
people
not
to
have
to
go
on
the
streets
to
do
that.
We,
you
know
it
used
to
be.
People
were
saying:
oh
learn,
github
we're
assuming
it
again.
We
found
that
leads
to
more
disparities.
We
need
to
prepare
students
for
the
workforce
for
kind
of
the
the
big
picture
on
under
taking
on
data
analysis,
and
this
is
these
are
skills
that
really
need
to
be
incorporated
in
the
curriculum.
A
Our
fourth
point
relates
to
lab
sciences.
Many
and
all
in
massachusetts
of
stem
transfer
pathways
from
our
community
colleges
require
a
lab
science,
and
this
is,
you
know,
not
a
bad
thing.
It
would
be
even
better,
however,
if
such
a
course
could
build
their
opportunities
to
make
meaning
of
data
in
these
subs
in
these
areas
of
physics,
biology
or
chemistry,
and
we
could
also
imagine
that
a
future
data
science
infused
lab
course
could
be
a
really
great
way
to
build
this
in
context
in
a
domain
area.
A
Last
but
not
least,
communication
ethics
and
application
domains
are
critical,
and
so
bachelor's
programs,
you
know,
include
training
in
explicit
training
in
communication
and
ethics.
This
is
a
big
part
of
what
was
incorporated
in
data
acumen,
definitions
from
the
national
academies.
My
colleague
ben
has
a
great
paper
on
how
to
structure
a
data
ethics
kind
of
program
throughout
it
throughout
a
major,
but
we
really
need
institutions
to
think
carefully
and
holistically
about
how
requirements
for
communication,
ethics
and
domain
applications
can
be
used
to
build
credits
that
can
be.
A
You
know,
successfully
transferred
in
a
flexible
way
closing
thoughts
of
the
you
know
as
part
of
the
national
academy's
roundtable
in
post-secondary
data
science,
education,
which
was
focused
on
community
colleges,
two-year
colleges,
erie
treisman,
described
data
science
programs
as
powerful
resources
for
students
seeking
upward
mobility.
I
think
we
owe
it
to
that
those
students
to
help
create
these
opportunities
in
a
way
that
really
can
allow
them
to
be
successful,
so
ben
and
I
have
been
working,
and
we
believe
that
there's
ways
to
make
progress
on
these
points
of
friction.
They
exist.
A
There
are
challenges
and
barriers,
but
I
think
we
can
come
up
with
solutions
to
make
a
lot
of
them
more
feasible.
I
put
a
link
to
a
white
paper
that
we've
been
developing,
that
has
more
details
and
resources
and
really
hope.
You'll
engage
with
this
work
and
again
I
thank,
and
I
look
forward
to
the
conversations.
A
Is
there
a
project
there
are
there?
Are
some
projects,
I'd
be
glad
to
kind
of
you
know,
start
with
that
personal
contact
and
put
my
email
in
the
in
there
and
it's
also
in
the
ether
pad,
but
there
are
some
groups
that
are
that
are
working
more
generally
on
some
of
these
these
issues,
I
think
one
of
the
challenges
of
renata
is
that
we're
fragmented.
A
You
know
I'm
seeing
that
on
that
on
the
k-12
data
science
site,
there's
a
whole
bunch
of
people,
doing
really
cool
stuff
in
the
learning
sciences,
same
thing
on
the
math
education
side,
the
science,
education
side
and
the
computing
education
side,
and-
and
they
don't
generally
talk
to
each
other-
there's
not
really
a
good
mechanism
to
bring
those
folks
together,
and
I
think
that's
something
that
we
need
to
be
doing.
There
is
a
pretty
active
acm
community
college
group.
A
We've
been
in
touch
with
there's
an
active
group
with
the
association
of
two-year
colleges,
math
teachers,
the
amatic
group,
but
this
is
a
place
where
I
think
we
need
to
kind
of
come
up
with
some
other
ways
to
connect
and
support
these
folks,
and
particularly
when
we
think
about
k-12
when
we
think
about
community
college.
We
know
the
time
you
know
is
not
available
for
them.
My
job,
I'm
incredibly
lucky
with
my
job.
A
It
expects
me
to
do
things
and
it
provides
me
a
teaching
load
and
other
kinds
of
supports
travel
money
to
make
that
possible.
I
know
there
are
many
people
without
that
and
the
like
in
the
world
of
community
colleges,
there's
not
the
same
kind
of
release,
time
and
other
other
support,
so
we
need
to
kind
of
figure
out
ways
of
both
bringing
people
in
and
also
kind
of
making
sure
their
time
is
valued
and
supported.
C
A
Patty,
I
I
agree
completely
with
with
that
and
a
lot
of
the
work
that
you've
been
doing.
You
would
put
in
the
link
to
the
you
know:
expanding
participation
and
computing
esp
project,
which
I
think
is
has
a
lot
of
similarities
to
this
they're,
both
questions
about
what
we
teach
but,
more
importantly,
how
we
teach
them
and
to
do
so
in
a
way,
that's
really
inclusive.
A
A
One
of
them
is
housed
in
a
computer
science
department
and
another
one
is
an
is
in
in
an
ikea,
is
department,
and
so
again
we're
also
seeing
business
programs
and,
like
so
there's
a
kind
of
a
tower
of
babel,
that's
developed
here,
but
I
think
the
principles
we
know
what's
can
be
effective
for
students
in
terms
of
active
learning,
fostering
a
growth
mindset
and
there's
really
kind
of
work.
We
need
to
do
as
instructors
and
as
curriculum
developers
to
really
make
that
happen.
C
Yeah,
I
think
it's
great
what
you're
doing
it's
so
necessary.
I
just
want
to
emphasize
and
encourage
you
to
keep
doing
that,
because
the
honestly
one,
I
think,
one
of
the
hardest
things
I
had
in
puerto
rico,
when
I
was
when
I,
since
I've
been
working
here
with
convincing
my
fellow
faculty
members
that
you
know
lecture
based,
training
isn't
the
best
way
and
so
like.
C
If
they
see
it
happening
like
coming
from
a
peer
there's,
no
respect
for
that
right,
they're,
just
like
oh
who
does
she
think
she
is,
but
coming
from
a
collaboration,
it
makes
a
whole
different
like
it
changes
everything
and
so
yeah.
I
just
wanted
to
to
emphasize
how
important
this
is.
So
thank
you.
D
A
You
know
we
can't
afford
to
kind
of
wait
until
people
get
their
master's
program
to
be
able
to
develop
these
skills.
It's
really
kind
of
ensuring
that
they're
getting
them
in
high
school
and
their
associates
and
then
at
the
bachelor's
level
is
going
to
be
really
critical
and
I
think
we
need
to
kind
of
be
really
creative
about
how
we
make
that
all
work.
One
thing
I
think
is
pretty
exciting
was
the
approach
that
berkeley
took
with
their
data
eight
course
and
then
designing
a
set
of
connector
courses.
A
There
were
25
or
30
of
them
at
that
you
know,
and
it's
at
its
peak.
I
think
that
tied
in
directly
to
the
learning
outcomes
from
that
introductory
course
in
linguistics
and
psychology
and
history
and
other
disciplines,
I
think
we
could
be
thinking
about
courses
that
integrate
computation
and
data
and
to
kind
of
be
building
on
some
of
those
kind
of
common
things.
This
requires
departments
and
programs,
never
good
at
talking
with
each
other
to
work
with
each
other,
and
we
need
to
then
get
the
kind
of
cross
institutional
things
going
on.
A
There
are
some
examples:
california
is
a
great
one,
where
pretty
much
every
course
in
the
public
school
system
has
a
mapping
on
the
spreadsheet
from
here
to
here,
and
you
can
tell
exactly
where
it
transfers
and
how
it
comes
together.
These
things
take
a
lot
of
time,
effort
and
both
bottom-up
and
top-level
top-down
things.
I
think
it's
particularly
challenging
because
data
science
still
is
evolving,
and
so
exactly
what
these
things
are
are
not
quite
as
well
set
as
they
are
in
biology.
But
again
I
think
that's.
D
E
Yeah
this
is
mark,
so
during
the
academic
year
faculty
and
teacher,
they
don't
have
a
lot
of
time
to
get
training,
but
a
new
brand.
Do
you
offer
a
program
for
teachers
and
and
faculty
so
they
can
learn
and
work
on
projects,
maybe
small
research
project
where
they
can
learn
new
skill
and
then
incorporate
what
they're
learning
in
the
in
in
the
teaching
during
the
academic
year.
A
If,
if
I'm
understanding
that
correctly
mark-
and
I'm
I'm
pretty
impressed
by
from
what
I've
heard
about
some
of
the
programs
recently
in
the
nasam
technology
and
undergraduate
stem
discipline,
there
were
some
really
neat
projects
that
were
happening
with
research
experiences
for
students
from
university
of
the
virgin
islands
and
really
innovative
approaches
that
have
been
brought
into
the
curriculum
to
give
students
that
practice.
What
I
think
I
heard
from
you
was
a
sense
of
how
do
we
provide
such
opportunities
for
faculty?
Many.
E
E
E
A
And
again,
I
think
I
think
we
need
to
be
doing
that
in
all
sorts
of
ways.
We've
seen
ways
that
that
faculty
development
has
scaled
beyond
the
typical
model
of
a
small
number
of
people
in
a
room
for
a
week
or
so
in
the
summer,
and
we
know
that's.
You
know
that
workshop
kind
of
approach
is
going
to
be
a
lot
of
shopping,
not
so
much
work.
A
It's
not
clear
what
the
outcomes
are,
but
I
think
we've
seen
models
for
how
the
ap
computer
science
principles
course
developed
a
cadre
of
teachers
that
aren't
your
usual
suspects
that
are
in
much
more
diverse
communities
and
a
lot
of
more
under-resourced
communities,
I'm
most
familiar
with.
What's
going
on
in
new
england,
where
microsoft
funded
200
schools
worth
of
kind
of
faculty
development
and
brought
them
together.
A
I
think
we've
learned
a
lot
of
lessons
from
the
pandemic
of
what
you
know,
what
some
combination
of
in-person
and
then
follow-up
hybrid,
can
do
to
allow
people
without
kind
of
generating
a
lot
of
carbon
and
other
things
and
to
be
away
from
their
families
and
things
to
do
remote
instruction.
And
so
I'm
I'm
hopeful.
We
can
think
about
scaling
such
faculty
development
things.
A
This
would
be
a
great
nsf
grant
to
kind
of
explore
or
to
kind
of
be
bring
together,
because
I
think
the
world
needs
it
and
I
think,
there's
a
kind
of
a
benefit
for
how
we
can
be
thinking
about
scaling.
These
things
up.
That's
going
to
require
a
team
that
involves
many
of
the
disciplines
that
are
represented
here
today
and
I'd
be
game
to
talk
about
some
of
those
things.
If
there's
an
interest,
because
I
know,
there's
a
need.
B
Yeah,
so
we're
definitely
interested
in
the
faculty
development
piece.
You
know-
and
I
think
most
of
you
know
about
our
data
up
program,
which
was
aimed
at
faculty
to
do
that
in
used
to
be
in
person
now
virtual
faculty
development
in
like
an
actual,
hands-on
workshop,
but
then
also
a
pedagogical
workshop
that
teaches
about
teaching
data
science
and
best
practices
and
teaching,
but
scaling
that
even
virtually,
I
think
will
require
collaborators
as
as
we're
thinking
about
the
program.
B
You
know
you
can
still
only
have
a
certain
number
of
people
in
a
virtual
room
that
you
can
actually
have
enough
assistance
to
really
answer
everyone's
questions
and
do
what
needs
to
be
done
to
get
people
ready
to
be
able
to
teach
and
not
just
a
lecture
style.
So
I
really
would
be
interested
in
how
people
are
thinking
through
this,
and
if
there
is
a,
if
there
is
a
larger
grant
that
we
can,
you
know
think
about
how
some
of
these
programs,
the
best
practices
have
worked,
would
have
worked
for
us.
B
So
that
was
one
question
or
I
I
thought
process
around
this
limited
capacity
for
new
credits,
and
you
know
the
limited
capacity
for
new
courses
in
k12
and
other
things
of
specifically
gearing
towards
towards
disciplines.
So,
and
we
have
one
minute
so
patty,
let's
say
your
question
and
then
nick
I'll,
let
you
close
us
out
with
any
of
your
thoughts
so.
C
A
A
That
that
would
be
my
closing
thought.
Renata
is
exactly
what
paddy
was
saying.
The
data
carpentry's
approaches,
I
think,
can
help
us
kind
of
bring
people
up
one
level
and
I
think
the
aptly
named
data
up
program
that
you
did
really
kind
of
helped
build
the
communities
locally
and
within
within
the
region,
and
I
think
those
have
those
have
really
kind
of
continued
to
pay
off
in
terms
of
what
we
need
to
do
to
bring
the
next
generation
of
instructors
into
that.
So
thank
you.