►
From YouTube: Keeping Data Science Broad: Alternative Avenues for Development of Data Science Education Capacity
Description
This webinar will highlight efforts that build data science education capacity outside of the context of tradition curricular program development. This may include integration of data science into courses and curricula outside of the traditional CS/math/stats context (i.e. Arts and Humanities), expansion of capacity by integration of third party or shared resources (i.e. MOOCs and open source educational resources) into curricula, and additional educational options outside of traditional courses (i.e. Faculty training, Data Science for Social Good Programs, and Bootcamps).
Share Your Input on this Topic at: https://goo.gl/forms/8DO4Fu9PV4ZfGi7O2
Resources:
View the Final Keeping Data Science Broad Report, Learn more about the Workshop and View all Webinars in the Series from the links below.
Report: http://bit.ly/KeepingDataScienceBroad_Report
Workshop: http://bit.ly/KeepingDataScienceBroad_Workshop
Webinars: http://bit.ly/KeepingDataScienceBroad_Webinars
For more information on the South Hub Data Science Education and Workforce Working Group visit: http://bit.ly/SouthHub_EducationWG
A
All
right,
I
think
we
are
live
now,
so
welcome
everyone.
The
keeping
data
science,
broad
series,
I'm,
dr.
Renato,
Rawlings
gasps,
the
co-executive
director
for
the
South
big
data
hub
and
the
South
hub-
is
one
of
four
helps
for
the
nation
for
innovation,
around
big
data
that
includes
industry,
academia
and
government
partners
for
bettering
society,
there's
the
West,
the
Northeast
and
the
Midwest,
and
so
the
south
of
16
states
Delaware
to
Texas,
including
DC.
This
series
keeping
data
science
broad.
A
This
webinar
is
the
second
webinar
in
this
series,
where
it's
really
focused
on
the
engagement
of
primarily
teaching
institutions,
which
can
be
for
your
liberal
arts
colleges,
community
colleges,
minority
serving
institutions,
as
well
as
alternative
places
of
education
around
data
science,
including
museums,
training
programs,
as
well
as
mentorship.
So
today
is
exciting.
We
have
some
fantastic
panelists.
The
first
series
was
on
traditional
context,
which
you
can
go
back
to
our
YouTube
channel
and
watch,
and
this
one
is
on
alternative
mechanisms
of
data
science,
education
and
how
we
can
really
keep
data
science.
A
So
first
up,
we
have
actually
I'll
do
all
of
the
panelists
and
then
first
up
we're
gonna,
have
your
shimmer,
who
is
a
associate
professor
at
Lawrence,
Technical
University,
and
he
has
been
working
and
doing
a
lot
of
exciting
work
with
data
science
and
teaching
it
outside
of
STEM
disciplines.
We're
gonna
have
Sarah
stone,
who
is
the
co-executive
director
for
the
e
Science
Institute
at
the
University
of
Washington,
and
also
working
with
the
data
science
for
social
good
program?
A
We'll
have
al
Harris,
who
is
the
associate
they're,
an
assistant
director
for
the
IT
group
at
the
IT
department
at
the
breakfast
group,
which
is
a
mentoring
institution
and
a
mentoring
group
for
african-american
males
going
from
high
school
into
college
years.
Tracy
teal,
who
is
the
executive
director
of
data
carpentry,
which
provides
two-day
trainings
and
boot
camps
all
over
the
world
in
data
science
and
even
news
o
who's.
The
chief
scientist
at
the
New
York
Hall
of
Science.
A
B
Right
Thank,
You
Renetta
can
I
start.
Please
do
yeah!
Thank
you,
so
I'm
Leo
I'm,
like
you,
said,
associate
professor
of
computer
science.
Learns
technological
university,
also
assistant
dean
and
I've,
been
with
the
big
data
hub
since
day,
one
actually
and
I'm
still
I'm
still
around
and
in
the
context
of
education,
I'm,
mostly
recognized
with
the
teaching
through
research
using
research
for
teaching
at
all
levels,
whether
it's
undergraduate
level
graduate
level
non
stem
courses
and
now
going
into
the
k-12
level
to
use
research
as
a
tool
for
teaching.
B
So
I
hope
you
can
see
the
slide.
So
basically,
when
you,
when
we
think
about
research
experience
for
students,
what
comes
to
mind
is
a
student
working
with
the
professor
helping
him
or
her
to
do
the
work
in
their
lab
and
that's
not
exactly
what
I'm
developing
what
we
develop
in
here.
What
we
are
doing,
we
are
making
it
scalable
to
a
large
number
of
students,
so
many
students
can
participate
as
part
of
the
curriculum.
B
So
it's
not
a
curricular
activities,
it's
part
of
the
curriculum
and
it's
accessible
to
all
to
all
students,
and
there
are
a
lot
of
literature
about
how
about
the
impact
of
research
experience.
It's
one
of
the
most
one
of
the
strongest
interventions
that
we
have
right
now
that
it
that
we
know
about
right
now
in
higher
education.
B
These
are
some
reports
came
from
the
White
House
back
in
2012,
where
they
identify
that
the
you
reach
better
results
and
surely
a
better
engagement
when
the
courses
are
designed
as
research
courses
and
not
just
repeating
what
the
textbook
says
and
the
results
are
in
the
textbook,
but
actually
doing
research
in
the
in
the
classroom.
So
basically,
the
idea
is
to
deliver
the
same
materials
that
we
that
we
learn
in
a
regular
course,
but
doing
that
through
research,
not
through
the
textbook,
but
through
research.
Experience.
B
There
are
also
reasons
I
won't
get
to
it
right
now,
but
actually
the
main
reason
for
starting
this
project
in
the
first
place
was
to
reach
out
to
underrepresented
populations
in
stem,
and
these
are
so
that
leads
us
to
qui
of
a
classroom
based
research
experience
doing
the
research
inside
the
classroom
as
part
of
a
regular
course.
These
are
the
five
elements
of
of
cui
one
of
the
important
things:
the
the
the
ownership,
the
ownership
of
the
of
the
a
of
the
research
for
the
students.
B
Don't
necessarily
do
research
led
by
or
led
by
another
someone
else
as
we
saw
sprung,
but
they
actually
own
the
research,
it's
it's
their
part
and
they
and
they
lead
that
research
through
the
we're
doing
it
for
senior
graduate
level
introductory
level
freshmen
sophomore
level
courses,
core
curriculum
courses,
did
not
data
science
courses,
I'll
get
to
that
and
now
working
on
the
k-12.
So
here's
the
first
example
I
would
like
to
show
that's
an
an
example
from
a
graduate
course
actually.
B
So
the
purpose
here
is
to
do
to
analyze
employee
compensate
compensation,
so
the
employee
compensation
analytics
the
employees
actually
are
soccer
players
well-known
soccer
players.
Why?
Because,
first
of
all,
it's
interesting
and
it's
engaging
it
engaged
the
students,
but
also
we
had
a
lot
of
data
data
are
available.
So
in
that,
through
that
research,
you
actually
cover
the
descriptive
analytics
like
same
testing.
How
how
the
compensation
responds
to
the
skills
of
the
soccer
player,
also
predictive
analytics.
We
can
predict
how
much
a
soccer
player
can
make
based
on
their
skills
and
also
with
further
analysis.
B
We
also
get
to
the
prospective
analytics
to
what
skills
the
soccer
player
need
to
to
improve
their
compensation
to
to
get
a
higher
wage
and
for
the
course
we
cover
all
that.
Actually,
the
debt
also
and
one
of
the
students
I
also
published
a
paper
about
them
to
became
just
recently.
It
became
pretty
it
made
it
to
the
the
news
in
Europe,
mainly
in
in
the
UK,
so
so
that
was
a
nice
example.
That
example
is
less
comprehensive.
B
It's
a
senior
undergrad
example
of
analyzing
astronomical
databases,
so,
basically
taking
a
very
large
databases
and
annotating
them,
classifying
them
understanding,
what's
what's
in
them,
so
that
example
is
not
as
comprehensive
as
des.
The
graduate
course
also
there's
a
lot
of
like
real-world
problems,
looking
working
with
a
noisy
data,
because
these
databases
are
not
as
clean
they're,
not
they're,
not
really
close
to
textbook
and
databases,
and
we
also
do
some
things
that
the
introductory
level
courses
and
here's
one
example.
B
That's
freshman
level
course
analyzing
lyrics
of
the
Beatles
using
data
science
tools,
so
basically
the
students
can
make
create
their
own
database.
They
go
I
get
the
lyrics
of
certain
song.
They
annotated
the
lyrics,
it's
fairly
simple
level.
That's
a
freshman
level
course
I
remind
you,
then
organizing
the
the
data
putting
the
lyrics
in
files
files
were
split
correspond
to
the
spreadsheets
that
they
have.
B
Then
they
analyze
the
lyrics
quantitatively,
with
some
tools
that
we
developed
so
add
those
tools
extracts
all
kind
of
descriptor,
the
descriptors
of
the
lyrics
and
then
get
to
the
to
the
analytic
spot
and
finding
the
statistical
features.
Looking
how
the
lyric
style
change
between
different
author,
different
song,
writers,
Lennon
McCartney,
how
we
changed
over
time,
what
changed
over
time
doing
the
statistical
analysis
of
that.
B
So
that
covers
a
lot
of
data
science
tools,
but
in
a
fairly
a
simple,
straightforward
way
and
again,
while
doing
research,
because
that
way
of
analyzing
lyrics
either
the
Beatles
or
any
other
song
writer
that
they
choose.
It's
a
new
way
of
doing
that,
and
also
that's
actually
what
we
did.
First,
we
started
with
curriculum
courses,
courses
that
are
not
necessarily
span
courses
like
humanities,
I'm
sharing
one
example.
For
me:
many
disco,
that's
an
art
history
course,
and
how
do
we
do
that'
science
in
art
history?
So
what
do
we
do
here?
B
The
students
come
up
with
some
hypotheses
that
they
have
about
art.
They
need
to
come
up
to
select
their
favorite
artist
or
favorite
artistic
style
and
come
up
with
a
hypothesis
about
it,
and
then
they
test
it
using
computational
tools.
So
we
have
some
computational
tools
that
we
developed
for
that
and
something
that
we
use
from
other
places
and
we
combine
those
into
a
process
that
the
students
collect
the
the
art
sample.
So
here's
example
of
several
simple
hypothesis,
so
they
collect
some.
B
Let's
say
if
you
look
at
the
middle
graph,
these
are
all
produced
by
a
computer,
so
you
see
that
the
scene
collected
about
50
paintings
from
Indian
art,
50,
paintings
of
Chinese
out
50,
Italian,
art
and
50
French
art,
and
then
they
let
the
computer
analyze
the
automatically
and
see
if
they're
there.
But
this
is
they
initially
came
up
with,
is
really
supported
by
the
analysis
of
the
of
the
computer
and
that's
an
example
of
actually
the
class,
the
entire
class
we
combined.
B
They
combined
all
the
the
students
work
to
let
the
computer
analyze
that
to
create
that
you
have,
and
then
they
discuss
it
in
the
class
but
they're
using
quantitative
tool.
There
isn't
that
we
developed
to
do
that,
and
here
again
that's
these
are
other
experiments
on
the
Left
I'd
like
to
see
how
out
changed
over
the
years
can
did
either
a
progression
of
art
is
a
limb
of
progression.
The
left,
the
graph
until
you
see
on
the
left,
clearly
show
the
computer
can
identify
that
progression
that
that
hypothesis
happens
to
be
true.
B
So
that's
a
new
way
and
there
in
the
in
the
links
that
I
provided
with
it
you'll
see.
There
are
some
videos
taken
in
the
class,
so
you
can
see
in
the
classroom
how
it's
done.
It's
pretty
interesting
to
see.
I
think
there
are
some
assessment
results
for
that.
That's
the
question
is:
do
you
enjoy
a
class
assignment?
What
did
you
enjoy
course
assignments,
and
we
can
see
that
there's
a
statistic,
significant
and
difference
when
we
use
that
over
the
traditional
art
history
courses
that
we
don't
use,
data
analytics.
Another
example
that's
from
psychology.
B
So
in
this
case
we
we
change
in
a
freshman
level
introductory
psychology
course
they
collect
the
students,
collect
facebook
post.
So,
first
of
all,
they
need
to
create
a
database
using
spreadsheet,
so
they
collect
a
Facebook
post.
They
create
a
spreadsheet,
they
annotate
the
posts
and
they
create
their
database
like
this,
and
then
they
end
up
with
a
database.
Then
we
combine
all
the
students
we
work
is
combined
to
one
day,
one
big
dataset,
and
then
we
analyze
the
data
set.
B
We
using
some
text
analytic
tools,
we
can
analyze
the
post
correlate
the
post
with
the
annotations
starting
to
see
if
we
can
quantify
differences
between
different
post,
different
topics,
the
age
of
the
person
writing
the
post,
all
kind
of
questions
like
that
and
different
the
divided
into
groups.
Each
group
come
up
with
their
hypothesis
and
they
test
the
hypothesis
using
the
data
and
using
data
science
tools.
B
We
have
some
preliminary
findings
and
the
purpose
is
not
necessarily
to
discover
something.
It's
basically,
we
used
research
for
a
student
engagement,
but
it
can
also
lead
to
discovery.
We
actually
discovered
several
things
while
using
those
courses
and
the
several
papers
were
published
about
it.
So
that's
also
a
nice
byproduct
to
have,
and
also
some
assessment
for
that.
The
question
I
prefer
lecture
to
hands-on
experience
in
the
class,
and
we
can
see
that
the
the
treatment
group
is.
B
A
A
So
once
we
go
through
and
have
people
speak,
we'll
be
looking
at
those
questions
and
I'll
ask
them
at
the
end
so
that
we
can
go
through,
but
I,
really
like
the
idea
of
students
being
able
to
make
their
own
question
to
kind
of
form
their
own
hypothesis
and
test
it
in
the
classroom.
I
think
that's
a
fantastic
way
of
getting
them
really
involved
and
engaged
in
the
process
and
it
not
being
a
data
science
class.
C
Thank
You
Renaud
and
thank
you
for
inviting
me
to
be
part
of
this
panel.
I'm
really
excited
to
be
here
and,
and
so
as
we're
not
I
mentioned
I'm.
The
co-executive
director
of
the
East
science
Institute,
which
is
on
the
UW
campus
I'm,
also
the
deputy
director,
one
of
the
deputy
directors
for
the
West
big
data,
innovation
hub
and
so
Renata
and
I
have
been
working
closely
on
a
variety
of
activities
and
and
so
that's
that's
been
a
lot
of
fun
and
so
I'll
go
ahead
and
share
my
slides.
C
C
Ok,
so
are
those
showing
up?
Yes,
ok,
great!
So
let
me
move
through
this.
So
today,
I'm
gonna
be
focused
primarily
on
our
data
science
for
social
good
program,
but
I
wanted
to
start
by
providing
some
framing
about
the
e
Science
Institute
and
our
mission,
because
we
do
a
lot
of
different
data
science,
education,
related
activities,
in
addition
to
data
science
for
social
good.
So
these
science
Institute's
our
mission
is
to
accelerate
and
facilitate
data
intensive
discovery
across
all
fields.
C
So
we're
really
a
neutral
entity
on
the
u-dub
campus
and
we're
one
of
three
data
science,
environment
institutes
and
that
were
generously
funded
by
the
more
in
Sloan
foundations,
the
others
being
UC,
Berkeley
and
NYU,
and
our
Institute
to
some
extent
is
structured
around
working
groups
and
one
of
our
working
groups
is
data.
Science,
education
and
this
group
has
been
thinking
deeply,
both
about
the
development
of
more
formal
education
curriculum,
which
was
kind
of
the
focus
of
last.
C
The
department's
there's
a
lot
of
flexibility
on
these
options
and
so
there's
kind
of
a
template,
but
then
departments
can
figure
out
what
courses
make
the
most
sense
for
their
particular
programs,
and
so
it's
a
bit
of
a
diff
model
in
that
those
options
that
they
are
transcript
abut.
The
student
is
still
getting
a
degree
in
their
primary
discipline.
We've
also
been
involved
in
creating
a
data
science,
professional
master's
program
and
again
these.
C
At
this
point,
we
have
nine
different
departments
or
units
on
campus
that
are
participating
in
these
programs,
and
this
number
is
increasing
every
year
and
a
lot
of
these
courses
are
open
to
students
from
across
campus.
So,
even
if
a
student
isn't
able
to
get
the
transcript
double
option,
they
can
participate
in
these
courses
and
many
of
the
courses
that
are
coming
online.
Don't
have
any
prerequisites.
C
We
also,
we
also,
which
is
kind
of
where
the
focus
of
today's
webinar
do
a
lot
of
what
we
call
kind
of
ad
hoc
data
science,
education
programs-
and
these
include
and
quarter
long
programs
like
the
data
science
for
social
good
program.
That
will
be
more
of
the
focus
today.
We
also
run
an
incubator
program,
which
is
a
similar
format,
but
it's
during
the
school
year,
and
then
we
sponsor
mini
hack
weeks.
C
We
actually
just
wrapped
up
Astro
hack
week,
geo
hack
week
and
neuro
hack
week
in
the
past
few
weeks
and
other
kinds
of
short
tutorials,
and
we
are
an
active
partner
with
software
carpentry,
which
Traci
teal
will
talk
about
later,
so
I
won't
belabor
now,
and
then
we
offer
also
do
things
like
office
hours,
drop-in
programs.
We
really
think
that
these
ad
hoc
education
programs
help
to
fill
the
gaps
and
data
science
education.
C
We
know
that
many
folks,
you
know
they
may
be
later
in
their
careers
or
are
not
able,
within
the
context
of
their
education,
to
participate
more
fully
in
these
kinds
of
data
science
curriculum
options,
but
being
able
to
participate
in
a
short
term
program,
be
it
two
days
or
ten
weeks
allow
them
to
gain
an
important
skill
set.
That
is
valuable,
as
we
know
across
all
disciplines,
so
diving
more
deeply
into
our
data
science
for
social
good
program.
C
This
is
run
again
out
of
the
East
Science
Institute.
The
goal
of
this
program
is
really
twofold:
it
brings
together
students
from
across
the
country
and
provides
an
educational
opportunity
and
I
really
like
Leo's
emphasis
on
this
sort
of.
It
provides
a
hands-on
training
for
students
with
real-world
data,
which
we
all
know
is
messy
and
complex,
and
so
it
provides
that
sort
of
hands-on
training
working
with
real-world
data
and
then
at
the
same
time,
helps
to
advance.
C
You
know
research
projects
in
this
kind
of
social
good
space,
which
we
think
of
very
broadly
and
hopefully
that
research
will
then
help
to
facilitate
responsible
data-driven
decision-making
in
our
program
was
modeled
after
the
first
DSS
G
program,
which
you
guys
may
be
familiar
with
at
the
University
of
Chicago.
Georgia
Tech
also
has
a
program
that
I
know.
Renaud
is
deeply
involved
in,
and
we
are
also
modeled
this.
C
C
C
C
C
We
have
had
a
strong
focus
and
transportation,
and
that's
partly
due
to
having
some
really
strong
transportation
oriented
groups
on
the
UW
campus,
including
the
tasks
our
Center
for
accessible
technology,
who's
been
involved
in
many
of
our
projects
over
the
last
few
years,
but
we've
also
had
interesting
collaborations
and
such
as
looking
at
permanent
housing
for
homeless
families,
with
with
the
Gates
Foundation,
which
has
led
to
a
longer-term
collaboration
and
we've
also.
You
know
more
recently
been
involved
with
thinking
more
about
Neighborhood
Development
and
in
Seattle
and
equitable
development.
C
C
We
also
get
them
from
the
government,
primarily
from
the
city
of
Seattle
and
King
County
nonprofit
organizations,
industry
and
so
we're
very
broad
and
thinking
about
you
know
what
kinds
of
you
know
who
can
submit
proposals
we're
looking
for,
obviously
like,
like
you,
can
imagine
strong
science
and
strong
methods,
the
person
who
does
the
proposal-
and
this
is
a
very
lightweight
proposal-
I-
should
say
you
know.
This
is
something
on
the
order
of
two
or
three
pages,
but
the
person
who
submits
that
proposal
then
becomes
what
we
call.
C
The
project
lead
and
being
the
project
lead
requires
participation
like
participation
in
studio
with
us
16
hours
per
week
for
the
duration
of
the
program,
which
is
a
10-week
program,
and
so
that
is
a
very
high
level
of
commitment.
But
we've
found
that
that's
really
critical
because
we
rely
on
the
project
lead
not
just
to
manage
the
student.
C
And
so
that
leads
me
to
emphasize
that
we
find.
You
know
that
having
established
stakeholder
relationships,
that
the
project
lead
already
has
in
place
are
really
really
valuable,
not
just
for
the
success
of
the
project,
but
also
from
for
the
value
of
this.
You
know
the
student
teams,
in
particular
understanding
the
importance
of
stakeholder
relationships.
C
We
also
look
for
you
know,
measureable
outcomes,
and
many
of
these
projects
have
the
capacity
to
continue
to
be
have
sustained
engagement
and
for
us
to
be
involved
with
them.
You
know
for
many
years
to
come,
so
I've
mentioned
the
project
team
several
times.
This
is
what
a
project
team
looks
like,
so
each
team
has
one
to
two
project
leads:
that's
what
I've
kind
of
been
talking
about.
This
is
the
person
who
submits
the
proposal.
They
really
help
drive
the
goals
of
the
project
and
the
timeline
and
lead
the
the
team.
C
We
also
then
have
for
data
science
for
social
good
student
fellows.
These
include
graduate
students
and
advanced
undergraduates
I'll
just
to
this,
so
in
in
terms
of
selecting
these
students,
we're
very
fortunate
in
that
we
get.
You
know
typically
over
200
student
applicants
for
the
16
positions,
and
you
know-
and
that's
with
not
working
very
hard
at
all.
So
what
I
think
one
of
the
things
we've
learned
is:
there's
huge
interest
among
students
and
working
in
this
social
good
space
and
many
of
the
students
that
we
get
are
not
from
sort
of
methods
disciplines.
C
Necessarily
they
come
from
a
diverse
array
of
backgrounds.
You
know
from
the
Social
Sciences
the
physical
Natural,
Sciences
humanities,
but
they're
all
interested
in
working
in
this
space,
even
if
it's
just
for
the
10-week
period
of
the
summer
and
so
I
think
that's
something
to
keep
in
mind
and
running
these
programs
that
there
is
really
a
tremendous
interest.
C
C
C
We
also
do
require
some
level
of
technical
or
analytical
experience,
but
we've
been
very
slow
to
really
make
you
know
exact
requirements
here,
we're
looking
for
a
quantitative
experience,
but
that
can
look
quite
different
from
student
to
student.
We
also
look
for
a
demonstrated
interest
in
social
good
and
this
can
be
through
their
primary
research
or
through
paid
or
volunteer
experiences,
and
then
you
know
ideally
they've
had
some
experience
and
collaborative
projects.
C
Most
students
have
had
this
in
some
capacity
and
we
look
for
some
students
who
bring
a
human
centered
design
experience
just
because
so
many
of
these
teams
end
up
developing
apps
or
visualizations
that
really
it
really
helps
to
have
that
skill
set.
So
we're
often
looking
to
have
at
least
one
student
with
some
design
Ariane's
on
each
team.
C
C
So,
what's
really
exciting
is
to
see
that
this
data
science
skill
set,
as
many
of
you
know,
really
allows
you
then
to
work
across
many
disciplines
and
that
can
be
exciting
for
all
involved
in
these
teams.
Oh
yep.
So
this
is
our
data
science
staff
and
again
we're
so
fortunate.
To
have
such
an
awesome
group
of
people
participate
in
our
program,
you'll
see
down
on
the
in
the
bottom
right
corner.
Here
we
also
have
a
director
of
data
science
ethnography.
C
C
Recognizing
that
we
have
a
broad
suite
of
expertise
in
the
room
and
often
you
know
often,
there
can
be
sort
of
cross
talk
between
the
different
teams
and
cross.
You
know
solutioning
across
the
teams
and
we
also
have
a
lot
of
social
events
integrated
like
weekly
happy
hours.
We
encourage
the
project,
leads
to
arrange
field
trips
to
visit
stakeholders.
You
know
as
appropriate,
given
the
project
and
what
you
can
see
here,
we're
very
focused
lucky
to
have
data
science
studio
and
wear
itches
houses,
our
Institute,
but
also
houses.
A
A
A
No
so,
and
so
al
the
night.
Our
next
presenter
is
al
Harun,
who
is
like
we
said
at
the
beginning,
the
associate
director
for
the
assistant
director
for
IT
department,
the
breakfast
group
which
mentors
african-american
men
who
are
in
high
school
and
on
their
way
to
college
in
stem
and
other
fields.
So
al
well.
D
E
Right
excellent,
what
I
want
to
do
is
first
say
Sarah,
I'm,
not
sure
why
it
is.
We
have
met
okay,
but
we're
gonna
have
to
change
that
guys
very
quickly
because
a
lot
of
what
you
said
that
it's
directly
applicable
to
what
we're
trying
to
accomplish
breakfast
group,
so
I
should
expect
to
hear
from
you
now.
E
So
what
you
should
be
able
to
see
is
the
breakfast
group
website.
It's
at
the
the
breakfast
group,
org
yeah
and
I-
encourage
you
to
go
to
that
website.
If
you
choose
to
while
I'm
having
this
conversation
and
just
go
through
it
and
see
what
it
is
that
we
did
and
what
we
were
about,
the
the
breakfast
group
has
been
around
since
1976
and
we're
dedicated
to
education
and
economic
development
and
what
what
I
want
to
do
in
this
conversation
is
to
talk
about,
because
we
do
mentoring,
we've
been
doing
it
for
quite
some
time.
E
We
have
a
few
signature
programs
and
in
those
signature
programs
we
have.
There
are
a
lot
of
volunteers.
It's
an
all-volunteer
organization.
If
we
made
quite
a
bit
of
progress
over
the
years,
so
we're
very
good
at
what
we've
been
doing
and
we're
in
five
high
schools
now
and
the
work
goes
on.
But
one
of
the
things
that
you
have
to
deal
with
is
how
how
you
and
legacy
organization
like
the
record
spirit,
which
is
still
relevant
today
and
bring
them
up
to
speed
on
the
explosion
of
requirements
that
are
connected
with
data.
E
It's
data
data
data,
everybody
wants
performance
data
and
they
want
to
see
how
it
applies
to
any
funding
that
they
might
provide
you.
So
it
takes
a
little
bit
of
effort
to
bring
everybody
up
to
speed
in
the
organization
and
to
do
things
differently
and
in
an
improved
fashion
than
you
have
in
the
past.
Alright.
E
So
a
few
years
ago,
the
breakfast
group
did
come
to
realize
that
the
growing
growing
need
to
focus
more
on
internal
research
or
internal
resources
are
collecting
this
data
across
the
board
for
its
signature
program
project
mister,
the
IT
department
was
formed,
brought
in
a
couple
of
individuals
myself
and
another
gentleman
Chris
Webb
2-bit,
to
form
the
IT
department
and
then
build
the
appropriate
infrastructure
around
that
IT
department
to
support
the
entire
organization
to
support
the
programs
that
we,
our
signature
programs,
and
to
support
our
members
as
they
continued
you.
There
good
work,
yeah.
E
The
IT
department
was
tasked
with
making
changes
to
those
to
those
infrastructures
and
services
to
allow
for
systematic
program
reporting,
technical
support
and
data
management
needs
assessment
services
for
our
signature
programs.
What
I
want
to
discuss
are
the
steps
taken
in
in
accomplishing
this
task
and
learn
more
about
the
political
whims
and
community
priorities
and
funding
influences
that
drove
the
need
for
much
greater
prioritization
of
the
collection
of
data
as
a
function
of
organizational
success.
E
A
little
bit
a
little
bit
more
of
an
introduction.
The
group
was
the
oldest
nonprofit
non
fraternities
service
organization
of
African,
American
business
and
professional
men
in
the
northwest,
the
United
States.
The
group
has
been
active
in
Seattle
african-american
community
for
more
than
39
years,
a
team
into
some
fairly
significant
success
along
those
lines.
We
we
value
a
consistent
and
competent
and
reliable
presence
in
the
lives
of
youth
easily
serve,
so
we
have
to
model
for
those
youth.
E
What
it
is
that
we
hope
that
they
would,
you
know,
strive
for
in
their
post
high
school
lives.
Our
organization
of
power
is
african-american
and
other
low-income
at-risk
male
youth
of
color
by
imparting
the
values,
responsibility,
leadership,
accountability.
Our
mission
is
to
provide
effective
leadership
in
American
and
other
at-risk
male
youth
of
color
towards
the
achievement
of
academic
excellence.
E
So
outcome
measurements,
when
you're
driving
a
program
become
significantly
important
once
you
start
to
get
into
certain
kinds
of
funding,
one
of
the
things
that
the
record
script
has
been
quite
able
quite
capable
I
should
say
quite
successful
in
doing
over
the
years
is
working
with
different
organizations
of
corporate
entities
to
raise
funds
to
support
project
mister.
We
are
all
achievers
program,
our
tie,
one
on
luncheon,
which
is
the
celebration
of
the
progress
and
success
of
the
students
that
we
work
with
and
very
successful
in
doing
that
in
recent
years.
E
In
looking
for
additional
funding
to
expand
what
the
organization
does
does
well,
we've
had
we've
had
we
realize
that
in
working
with
entities
like
the
city
and
the
county
and
the
state
that
the
data
requirement
performance
data
requirement
in
order
to
both
apply
for
funding
and
to
support
the
the
maintenance
of
that
funding,
that
was
growing.
So
our
next
step
was
to
realize
that
we
had
to
make
clear
that
the
programs
that
we
promote
and
support
our
evidence-based
programs.
We
had
to
have
evidence
informed
curriculum
development
which
we,
which
we
have
been
growing.
E
There
need
to
be
some
consistency
and
expansion
of
the
cohort
size,
and
that
means
that,
rather
than
a
few
students
at
a
time,
we
need
to
reach
out
and
seek
to
grow
that
to
10
or
15
students
per
quart,
and
we
need
to
collect
all
the
data
that
that's
required
for
reporting
on
those
cohorts.
You
know
in
a
systematic
and
reliable
and
consistent
way.
E
It
is
it's
true
that
the
breakfast
group
has
been
doing
this
for
years,
but
as
as
in
years
of
past,
it's
become
less
about
hardcopy
data
collection
and
more
about
electronic
data
collection
and
then
making
sure
that
that
data
is
properly
stored,
partly
collected,
stored,
analyzed
and
reported,
increased
need
for
organization
data,
security
and
policy
development
became
a
part
of
it
and
an
increased
need
for
training
members
and
program
instructors
on
on
that
same
security,
well
about
having
neglected
all
that
data
and
properly
presented.
So
that
was
that
was
our
challenge.
E
E
Once
we
got
once
we
get
to
this
point
beyond
the
data,
we
would
normally
collect
on
success,
programming
for
our
life
skills,
urban
community
awareness,
community
engagement
and
indirect
classroom
performance.
In
fact,
that
meant
that
we
also
must
prepare
to
add
direct
testing
measures
as
a
part
of
the
stem
ravishment
programming
and
proper
collect,
store
analyze
report
and
protect
this
data
as
well.
E
So
this
was
the
journey
that
we
had
for
the
script
over
the
last
few
years,
we're
very
much
getting
to
where
we
need
to
be,
but
we're
also
seeking
to
continue
to
collaborate
with
organizations
to
improve
who,
how
we
do
that
and
I
I
see
that
as
something
that
we're
gonna
be
very
successful
with.
So
is
my
short
presentation,
an.
E
A
You
so
much
yes,
so
the
breakfast
group
is
really
a
long-standing
institution
and
how
data
is
now
wrapped
into
your
students
learning,
but
also
into
your
organization
or
process,
so
I'm
really
interested
as
people
think
about
you
know
if
you're
working
with
nonprofits
or
any
business
really
now
needs
this
data
hone
it
and
what
types
of
questions
you
might
have
for
al,
please
people
that
are
watching
and
put
them
in
the
chat
we're
asking
of
the
panel
later
and
then
we'll
go
back
now.
I
think
you're
still
screen
sharing.
So
you
can.
A
F
F
So,
as
Renata
said,
my
name
is
Traci
teal
and
I'm.
The
executive
director
of
data
carpentry
and
I'm
gonna
be
talking
today
about
software
and
data
carpentry,
and
how,
like
a
lot
of
these
other
programs,
we're
looking
at
doing
training
sort
of
outside
some
of
these
other
more
traditional
pathways,
and
so
I
really
want
to
start.
F
And
so
the
question
is:
how
do
we
scale
data
literacy
and
data
skills,
along
with
this
really
big
data
production
and
so
right?
This
is
really
the
theme
here
of
this
webinar,
where
we
talk
about
training
in
the
gaps,
and
so
what
I'm
going
to
focus
on
is
short
intensive,
trainings,
so
workshops
short
courses
and
the
reason
for
that
is
that
they
can
provide
more
training
opportunities,
so
semester.
Long
courses
are
really
great
and
integrated
into
university
programs.
So
that's
terrific,
but
in
terms
of
the
scaling
it's
often
they
can't
get
through.
F
Maybe
as
many
students,
someone
might
build
a
great
program
at
one
University,
but
not
have
it
available
at
other
universities.
So
these
shorter
workshops
are
something
that
can
be
deployed
broadly
and
made
available
to
more
people
and
so
data
in
software
carpentry
workshops
and
that's
what
we
do
and
we
provide
what
we
teach
in
our
workshops.
So
we
teach
the
foundational
skills
for
you
to
work
effectively
with
data
and
do
software
development.
F
That
people
know
is
that
people
walk
away
with
an
awareness
that
these
types
of
approaches
exist
and
that
they
have
the
confidence
to
continue
to
use
them
so
I
also
really
liked.
You
know
how
they
are
talked
about,
how
they
showed
how
you
could
use
these
skills
in
a
particular
context,
and
because
that's
really,
what
get
you
excited
to
learn
more
when
you
see
what's
possible?
Often,
if
you
haven't,
had
an
exposure
to
these
approaches,
you
don't
even
know
that
you
can
do
these
kinds
of
things
or
ask
these
kinds
of
questions.
F
So
how
is
it
that
we
teach
skills
and
build
confidence
in
two
days
and
so
what
we
do?
Their
two-day
hands-on
workshops,
and
so
all
I
think
all
of
us
here
have
emphasized
the
importance
of
the
hands-on
component
securely,
get
experience
with
the
tools,
so
all
of
our
workshops
are
hands-on.
We
follow
an
AI,
Wii
U
model
where
an
instructor
talks
about
a
skill
and
presents
it.
F
We
do
it
together
with
the
instructor
in
the
students
and
then
the
students
try
it
on
their
own,
so
it
gives
opportunities
to
try
and
get
feedback
as
they're
learning.
We
also
warily
work
to
create
a
friendly
learning
environment,
especially
around
computational
skills.
A
lot
of
times
the
people
have
had
sort
of
disempowering
experiences
trying
to
learn
these
things.
Either
they've
gone
to
a
course
that
you
know
wasn't
really
the
right
match
for
where
they
were
what
they
wanted
to
learn
or
had
some
experience
that
made
them
think
that
computational
skills
weren't
for
them.
F
So
we
really
emphasize
within
our
workshops
creating
that
friendly
learning
environment
and
meeting
people
where
they
are
all
of
the
lessons
that
we
develop.
So
we
have
lessons
on
all
these
topics
that
we
talked
about.
They're
all
collaboratively,
developed
and
openly
licensed
so
they're
all
online
available
for
use
and
reuse
and
adaptation.
F
Anything
anyone
wants
to
do
with
them
and
they're
collaboratively
developed,
so
they
have
in
some
cases
hundreds
of
contributors
and
the
great
thing
about
that
is
that
we're
constantly
teaching
these
materials
getting
feedback
from
people
on
how
it
went
and
how
we
can
improve
the
lessons.
So
it's
not
the
case
that
you
develop
a
course
for
for
your
semester.
You
teach
it,
then
you
come
back
to
it.
F
We've
run
hundreds
of
workshops
all
over
the
world
on
30
countries,
and
these
are
some
of
our
long-term
survey
results.
So,
looking
at
people
after
six
months
or
more
after
they've
taken
a
workshop,
and
so
we
see
two
important
things.
One
is
that
we
see
that
people
are
using
the
skills
that
we
taught
so
this
chart
on
the
left
with
the
blue
and
red
bars.
People
are
talking
about
how
often
they
use
programming
skills
before
the
workshop
and
after
the
workshop,
so
we
see
increased
usage
of
these
skills
in
people's
works.
F
The
other
really
important
thing
is
that
we
see
a
vast
improvement
in
confidence
that
persists
even
after
the
workshop.
So
you
could
imagine
that,
right
after
a
workshop,
someone
feels
like
yes,
I,
really
know
this.
They
feel
really
excited,
but
then,
as
time
goes
on,
their
confidence
decreases
or
they
don't
have
opportunities
to
use
it
or
find
it's
harder
than
they
thought.
F
But
in
fact,
this
improved
confidence
continues
to
persist
six
months
or
more
after
the
workshop,
and
that's
really
the
foundation
and
dome
college
is
also
doing
a
really
nice
program
for
undergraduates
in
training,
and
they
really
emphasized
in
their
surveys
that
self-efficacy
and
confidence
are
the
most
important.
They
never
see
a
skills
increase
if
they
don't
see
a
confidence
or
a
self-efficacy
increase.
F
So
I
mentioned
that
we'd
run
hundreds
of
workshops
over
all
over
the
world,
so
that
obviously
wouldn't
work
very
well
unless
we
had
a
lot
of
people
to
go
and
do
that,
so
we
have
an
instructor
training
program
and
that
instructor
training
program
has
run
actually
on
next
month.
We'll
have
run
100
instructor
trainings
and
that
instructor
training
program
teaches
people
how
to
teach.
So
it's
based
on
educational
pedagogy.
So
it
also
is
a
two-day
workshop,
so
teaching
people
how
to
teach-
and
it's
teaching
people
how
to
teach
computational
skills.
F
This
way
with
this
really
active
learning
component
to
it
and
I
kind
of
liken
this.
This
train
the
traders
model
for
scaling
data
training
as
a
first
aid
approach
rather
than
the
fire
fighter.
So
right
now
in
data
science,
they
have
kind
of
this
firefighter
model
where
we
have
these
people
who
are
really
well
trained.
You
know
maybe
nano
machine,
learning
and
Hadoop,
and
we
sort
of
think
that
we
need
this
fire
fighter
to
train
and
deploy
on
all
of
our
data
science
problems.
F
So
you
look
at
the
impact,
that's
something
like
the
first
Red
Cross
first-aid
training
program.
The
number
of
people
who've
gone
through
there,
that
are
the
people
on
the
ground
and
something
happens
that
can
do
the
CPR
that
scales
much
better
than
these
firefighters
and
they're
different
sets
of
problems
that
people
we
need
to
approach.
F
Obviously
we
still
need
firefighters,
but
if
we
think
about
things
a
little
differently
in
terms
of
the
level
of
training
and
how
we
reach
people,
it
thinks
make
kind
of
shifts
how
we
think
about
scaling
training
opportunities,
so
I
mentioned
with
our
instructor
training
program
and
that
it
teaches
how
to
teach.
But
the
other
really
important
thing
there
is
that
it
builds
the
local
capacity
for
training
in
building
communities.
F
When
you
train
instructors
to
teach
at
an
institution,
they
can
then
teach
themselves
and
they
can
teach
other
people's
in
their
community
and
it
helps
to
build
a
community
of
practice,
a
set
of
people
that
they
can
look
to.
Who
are
using
these
skills.
They
can
continue
to
work
together
and
changing
that
kind
of
the
culture
around
data
science
is
almost
as
important
as
giving
people
the
tools.
So
how
can
you
get
involved?
So
right?
F
Sara
mentioned
a
lot
about
running
the
software
carpentry
workshops
at
University
of
Washington
right
and
that's
a
great
example
about
building
local
capacity
right,
they've,
really
taken
the
software
and
data
carpentry
and
built
this
program
around
running
continued
programs,
instructor
trainings
and
workshops
so
then
becoming
a
member
of
organization
is
what
really
facilitates
building
the
local
training
capacity.
You
can
also
request
individual
workshops,
though
you
can
become
an
instructor
as
a
part
of
this
program
contributed
to
our
collaborative
lesson
development.
A
F
A
So
we're
gonna
do
it
at
the
end.
I
did
give
a
heads
up
about.
You
know
how,
like
the
members,
are
the
breadth
of
people
in
all
of
your
programs
that
are
kind
of
being
impacted
and
what
kind
of
the
demographics
of
those
are,
whether
it's
the
more
diversity
you
see,
I
mean
alternative
programs
and
sort
of
a
similar
breakdown,
so
like
a
traditional
computer
science
type
discipline
or
where
you
see
dude
is
kind
of
being
different.
So
any
other
questions
to
from
the
audience.
G
Everybody
can
see
this
I
guess
yeah,
okay,
so
I'm
here
from
the
New
York
Hall
of
Science,
as
Renata
mentioned,
and
we
are
actually
a
member
of
the
Northeast
big
data.
Innovation
hub
and
we've
been
working
on
a
big
data
literacy
project
to
help
bridge
the
gap
between
data-driven,
Science
and
Technology
practice
and
public
understanding
of
big
data
and
big
science.
G
We're
calling
big
data
for
little
kids,
which
is
a
series
of
workshops
to
study,
data,
literacy
skills
and
very
young
children.
We
also
held,
which
is
possibly
the
first
public
festival
and
big
data
back
in
2015,
so
I
guess
I.
Let
me
start
by
just
reviewing
where
this
gap
came
from
and
how
we
got
into
this
predicament
of
of
science
moving
in
one
way
and
education
moving
in
another.
G
As
you're
all
aware,
we
live
in
an
age
in
which
traditional
science,
in
fact
every
sector,
every
aspect
of
human
culture,
is
flooded
with
data,
so
believe
it
or
not.
This
transformation
was
preceded
by
a
period
in
which
the
future
of
science
was
in
doubt
at
the
six
International
Congress
on
the
history
of
science,
dirt,
the
so
low
price
predicted.
The
exponential
growth
of
science
would
make
research
so
complex
and
costly.
We
would
soon
exhaust
our
ability
to
support
science
as
an
enterprise.
G
The
when
gathering
data
was
hard
also
expensive,
but
Moore's
law
took
care
of
that
problem
and
by
the
turn
of
the
millennium,
our
ability
to
acquire
process
and
store
data
accelerated
rapidly,
obviously
due
to
advances
in
computer
power,
but
also
cheap
and
automated
data
capture
systems
for
everything,
from
DNA
sequencing
to
remote
sensing.
We're
now
gathering
data
at
such
a
fantastic
rate,
we're
drowning
in
it.
G
Science
engineering
have
been
transformed
in
an
hour
we're
now
facing
a
kind
of
Moore's
law
for
data-driven
science,
which
is
an
exponential
growth
in
what
Jim,
Gray
called
the
fourth
paradigm
of
science,
research
and
here
in
which
computer
users
worldwide
generate
enough
digital
data.
Every
15
minutes
to
fill
the
u.s.
Library
of
Congress.
Our
ability
to
gather
data
has
vastly
outstripped
our
ability
to
analyze
it
all.
The
effect
of
this
trend
is
reflected
in
the
transformation
of
the
work
place
throughout
domains.
G
The
kinds
of
skills
demanded
by
everything
from
startups
to
Fortune
500
companies
to
research
labs
have
dramatically
shifted
away
when
we'll
focus
on
individual
compartmentalize
skills
in
hierarchies
to
nimble,
highly
creative,
collaborative
interdisciplinary
and
analytical
skills
kinds
of
questions
we
ask
of
ourselves
the
degree
of
complexity
of
nature
at
all
scales
and
the
complexity
of
the
problems.
Policymakers,
industry
and
citizens
are
called
upon
to
address
our
increasingly
interdisciplinary
and
complex.
The
routine
manual
and
repetitive
tasks
are
either
completely
automated
or
no
longer
relevant
and
there's
no
indication.
G
Yet
the
demands
on
our
education
system
are
the
same
as
they
were
back
in
1950s
data
sciences
are
not
taught
at
any
depth,
except
in
small
specialized
programs.
They're
increasingly
emphasized
in
standards,
but
uptake
has
been
slow
in
this
lack
of
penetration
threatens
that
pace
of
technological,
scientific
and
engineering
progress.
G
We
have
the
opportunity
to
transform
human
knowledge,
along
with
big
data,
to
bring
the
public
along
with
us,
which
is
not
only
the
most
official
thing
to
do.
It's
the
only
way
this
process
can
continue
at
a
rapid
pace
without
a
big
make
data
knowledgeable
public
policy
and
education
sectors,
we're
building
ourselves
a
house
of
cards
that
will
fall
purely
from
the
lack
of
skills
and
intellectual
prowess.
G
G
We
then
brought
them
together
with
other
data
scientists
and
learning
experts
in
a
workshop
in
the
spring,
to
validate
conclusions
and
look
at
what
we
would
take
to
fill
the
gap
between
assessment,
dis
assessment
and
Big
Data
practice
across
all
settings
and
domains
of
work.
The
outcome
of
this
process
helped
validate
the
need
for
and
to
identify
an
emerging
community
of
practice
around
big
data,
and
it
also
revealed
the
following:
an
advanced
Big
Data
literacy
is
essential
in
a
society
and
there's
a
willingness
among
practitioners
to
envision
a
way
forward.
An
inclusive,
collaborative
multi
sector.
G
Multidisciplinary
approach
is
needed
to
develop
a
bias
aware
and
unified
framework
and
sustainable
infrastructure
that
will
allow
for
big
data,
literacy,
project
implementation
and
oversight.
While
there's
a
need
to
specifically
target
the
northeast
there's,
clearly,
a
global
need
for
big
data
literacy
and
scalability
and
generalizability
are
important
goals
and,
finally,
the
breadth
of
expertise
and
interest
in
equity
ethics
and
in
bridging
sectors
and
educational
opportunity.
Areas
represented
by
workshop.
G
So
the
next
phase
of
this
work
is
to
develop
a
strawman
which
can
be
used
to
draft
a
framework
for
big
data
literacy.
This
process
will
be
based
on
a
successful
process.
We've
used
in
the
past
the
game,
consensus
around
science
disciplines
or
areas
of
literacy
concern
that
can
be
infused
into
learning,
standards,
inform
curriculum
development
and
instruction
and
ultimately,
help
result
in
a
more
literate
population.
G
We're
continue
to
work
to
work
on
getting
the
support.
We
need
to
create
and
disseminate
such
a
framework
for
big
data
literacy
and
we're
seeking
partners
to
join
this
effort,
of
course,
we're
very
interested
in
the
work
of
all
the
hubs
and
making
this
an
inclusive
process.
So
I'd
like
to
invite
not
only
the
people
on
this
panel,
but
those
listening
and
those
in
the
South
hub
and
other
hubs
to
really
think
about
joining
this
effort
and
contacting
us
so
that
we
can
begin
this
important
work
together.
G
I'd
also
like
to
acknowledge
the
support
of
this
project
under
National
Science,
Foundation,
sais
research
area
for
division
of
information,
intelligent
systems
and
also
additional
support
from
the
North
East
big
data.
Innovation
hub
I'd
also
like
to
thank
the
southern
hub
and
renata
for
inviting
me
to
present
today
and
all
the
members
of
the
North
East
sub
that
participate
in
this
work.
Thank
you.
A
Thank
You
Steven,
yes,
so
I
mean
the
work
that
you're
doing
here
is
key
and
kind
of
core
to
the
mission
here.
I'm
looking
at
what
are
the
skills
that
are
really
needed
in
data
science,
and
we
found
that
to
be
very
broad
and
very
industry
specific,
so
having
core
skills
is
something
that
I
think
we
all
in.
This
call
are
pretty
interested
in
and
I
know
that
keeping
this
discussion-
and
this
particular
group
going
is
something
that
we're
taking
up
as
the
South
hub.
A
A
In
the
description
of
this
particular
one,
you
can
click
to
the
link
to
the
community
input
and
share
it
broadly
we'd
like
to
put
together
a
report
as
well
in
connection
with
several
different
education
initiatives
around
data
science,
to
look
at
this
broadly
and
to
get
as
much
feedback
as
possible
from
people
doing
it
and
using
it
and
interested
in
data
science
and
seeing
it
I
have
a
colleague
that
I
think
I'll.
You
know
that
says
that
everybody,
every
company
now
and
most
nonprofits,
no
matter
what
your
core
mission
is.
E
A
So
I
mean
some
of
this
I'd
like
to
open
up
to
some
of
the
questions
that
we
talked
about,
but
really
this
key
issue
of
data
literacy
and
then
also
the
workforce
component
of
how
to
broadly
train
the
workforce
needed
to
keep
innovation
going
in
this
country
and
around
the
world.
With
that.
So
one
question
that
we
did
get
was
specifically
around
data
carpentry
and
how
you
do
assessment,
so
how
you
assess
your
students
to
ensure
that
they're
competent
and
whether
they
get
any
credit
or
a
continuing
education
or
badges.
F
Great
I
can
hear
me:
okay,
yes,
okay
thanks.
Those
are
great
questions
and
so
I'll
take
the
assessment
question
first,
so
we
do
have
someone
on
our
team
on
assessment.
I'm
Carrie
Jordan,
who
was
gonna,
also
be
on
this
call
today,
but
is
in
fact
teaching,
and
so
we
do
two
types
of
assessments.
So
we
do
one
pre
and
post
workshop
survey
and
so
that's
assessing
feelings
about
the
workshop,
as
well
as
self
reporting
skills.
F
Before
and
after
the
workshops,
and
then
the
other
one
is
the
long-term
survey
and
that's
the
one
that
I
showed
the
data
from
which
six
months
our
mover
out.
So
actually,
if
you
go
to
the
data,
carpentry
org,
slash
assessment,
you'll
see
all
of
our
assessment
materials
there.
So
our
surveys
and
our
short-term
survey
report
and
then
I'm
not
sure
if
the
long
term
but
I,
think
the
long
term
survey
reports
also
linked
from
there.
F
So
we
have
primarily
been
relying
on
self-reporting
and
doing
non
paired
assessments,
but
we've
now
just
developed
a
new
instrument
that
will
allow
us
to
do
paired
reporting
and
has
some
actual
skills
based
questions
that
are
not
self
reporting.
But
actually
are
checking
those
skills,
so
we'll
have
a
little
bit
more
information
about
skills
going
forward.
I,
don't
know.
F
Questions
so
if
there
are
put
them
in
the
chat
and
then
I,
the
second
question
was
about
Oh,
badges
and
certificates.
So
actually
no,
we
we
don't
really
provide
anything
around
learners.
Having
taken
the
workshops,
we
do
sometimes
give
completion
certificates
or
letters
that
people
have
taken
those
because
some
people
do
require
them
for
continuing
education,
but
we
don't.
F
We
don't
have
a
program
where
we
say
we
certify
that
you've
learned
these
skills
and
the
reason
for
that
is
one.
We
don't
want
to
create
a
test,
because
that
often
does
not
make
you
feel
like
you're,
leaving
having
learned
things
and
the
other.
Is
that
we're
not
really
sure
exactly
what
would
be
in
that
test
when
our
goals
are
not
just
around
skill
space,
but
about
preparing
you
for
continued
learning
and
learning
how
to
look
for
help
and
do
troubleshooting
so
probably
going
forward.
A
E
Yes,
this
is
al
Heron
I
immediately
went
to
the
website
for
the
instruction
and
I
can
tell
you
that,
based
on
what
I've
read
given
the
the
data,
literacy,
financial
literacy
and
stem
literacy
combination
that
we
are
bring
to
the
table
for
our
students
that
it
seems
to
me
and
I
need
to
learn
more.
It
seemed
to
me
that
this
would
be
very
well
placed
in
helping
bring
our
organization
and
our
students
further
into
the
world
of
data
literacy.
That's
my
immediate
impression.
A
B
A
lot
of
our
main
goals
is
basically
engagement
and
self-efficacy.
We
use
surveys,
we
use
a
exit,
entry
and
exit
interviews.
We
have
external
evaluators,
helping
us
doing
that
so,
and
the
approach
is
actually
bigger
than
just
data.
Science
I
talked
about
data
science
here,
but
we're
actually
changing
the
the
entire
college.
So
it's
a
it's
a
more
comprehensive
program
than
than
just
the
data
science
and
the
assessment
is
in
that
context.
So
it's
a
it's
a
comprehensive
assessment.
B
And
it's
it's
about
inclusion.
It's
about.
Like
I,
said
it's
engagement,
self-efficacy
identification
is
them
and
professionals
among
the
general
population
of
students,
but
also
under
represented
populations.
So
it's
a
the
project
is
just
like
you
said
it's
a
it's,
not
just
data
data
science,
but
not
the
goals
are
not
necessarily
just
to
train
data
scientists.
It's
we
have
a
bigger
goal
which
is
stem
in
general.
C
So
we,
similarly
that's
what's
been
talked
about.
We
do
assessments
both
pre,
so
this
is
particularly
for
the
DSS
to
program
and
we
do
pre
program
assessments
that
include
both
kind
of
technical
skills,
as
well
as
other
kinds
of
factors
like
things
to
do
with
the
domain
knowledge.
We
come
from
and
kind
of
their
career
aspirations,
and
then
we
follow
up
on
that
at
the
end
of
the
program.
C
So
we're
looking
both
at
program
success,
but
also
you
know
in
terms
of
teaching
the
technical
skills,
but
also
how
that
kind
of
changes,
the
framing
of
what
students
want
to
do
with
their
lives,
and
we
also
often
find
that
that
really
there's
really
a
shift
in
either
the
focus
of
the
type
of
research
they
want
to
do,
or
you
know
kind
of
where
they
want
to
spend
their
lives
and
academia
versus
going
more
into
the
social
sector,
because
they
have
experience
really
making
a
difference.
And
so
that's
that's
super
exciting
to
see
and
I.
C
We
give
kind
of
a
title
which
is
you
know,
for
the
students
that
participate
there,
science
for
social
good
fellow
and
you
know
that
is
a
title-
that
they
can
then
use
on
their
resumes
and
transcripts.
And
there
isn't
you
know
in
terms
of
participating,
we
don't
have
a
badge
per
se
or
you
know
a
continuing
education,
kind
of
credit
program
and.
G
G
We
do
pull
out
with
with
artifacts
the
sort
of
narrative
you
know
analysing
narrative,
we
do
observations,
we
do
pre
and
post
surveying
and
interviews,
and
we
also
do
what
it
will
call
think
Aloud's
in
which
we
actually,
while
the
the
the
subject
is,
is
engaged
in
the
activity
they're,
actually
talking
about
what
they're
thinking.
So
that's
one
whole
area
that
we're
dealing
with
in
terms
of
doing
a
value
to
study,
and
that's
primarily
done
for
our
sake.
The
the
subject
doesn't
necessarily
get
a
get
a
grade
or
any
kind
of
evaluation
out
of
it.
G
So
some
of
our
programs
are
actual
programs.
We
run
with
schools
and
with
students,
and
they
get
a
certificate
and
and
they're.
You
know
they
they
were
they.
We
we
generally
get
pretty
much
unheard
percent
participation.
Typically,
a
student
that
isn't
doing
well
will
just
drop
out,
because
these
are
very
intensive
kinds
of
things
that
we
do
you.
The
term
was
used
before
boot
camp.
We
call
them
that,
and
you
know
in
general,
we
we
don't
give
badges
or
that
kind
of
thing,
but
we
do
give
certificates
which
kids
appreciate
enemies.
G
A
Right
so
I
do
then
we
have
the
five
minutes
left
for
the
panel
and
I
have
a
question
for
each
of
you
so
aspiring
you.
It
was
five
minutes,
but
I
wad
question
about
envisioning
the
future
for
data
science,
so
for
data
science
to
thrive
and
to
be
the
data
literacy
that
we
see,
we
need,
as
a
society,
to
push
forward
innovation
to
push
forward
work.
A
How
do
you
see
or
envision
a
future
where
some
of
these
alt,
what
could
be
considered
alternative
mechanisms,
whereas
outside
of
Education,
can
connect
to
really
broaden
and
change
the
educational
sphere
around
a
data
literacy
or
data
science
and
how
it
can
remain
inclusive?
So
a
vision
of
you
know
how
we
could
work
together
with
traditional
programs
as
well
as
initiatives
and
museums,
so
I'll
kind
of
start
fine
or
let
someone
pick
but
really
kind
of
your.
G
B
I'd
like
to
add
one
of
the
things,
the
data
sense,
just
like
many
other
field.
There's
this
much
you
can
learn
from
the
textbook.
You
do
need
the
hands-on
experience
when
you
go
out,
you
do
need
and
there's
a
lot
of
just
like
we
heard
today
from
almost
all
speakers,
there's
a
lot
of
emphasis
on
the
hands-on
experience,
the
things
that
it
doesn't
necessarily
need
to
be
outside
of
the
traditional
programs.
B
The
traditional
programs
can
change
and
can
incorporate
those
things
in
the
classroom
as
part
of
the
of
the
program
it
started
to
happen
and
higher
education
is
starting
to
take
that
path.
Yeah.
It
happens
slowly
higher
education,
that
conservative
field,
so
it
happens
slowly
and
slowly
it
changes
at
first.
You
see
like
anecdotal
changes.
Then
you
see
a
larger,
bigger
programs,
and
now
you
see
that
trend,
but
it
is
happening
in
some
other
fields,
not
necessarily
data.
A
C
Well,
I
was
just
gonna,
say,
I
really
agree,
I
think
you
know
we
get
the
question
a
lot.
You
know
once
these
programs
are
in
the
university
curriculums
College
curriculums,
is
there
really
going
to
be
a
need
for
these
types
of
you
know
less
traditional
programs
and
I
think
there
always
is
because,
as
you
know,
the
wealth
of
panels
on
this
webinar
have
noted
these
programs
fill
gaps.
That
will
continue
to
be
there
and
because
not
everybody
has
access
to
these.
C
You
know
academic
courses,
and
it
also
these
other
programs,
and
especially
integration
across
these
programs,
allows
us
to
work
across.
So
many
different
sectors
across
different
populations
tackle
questions
of
getting
this
sort
of
education
to
diverse
populations,
and
you
know
so
I
think
there's
these
programs
really
serve
an
important
role
and
that's
likely
to
change
over
time,
as
we
have
more
sort
of
academic
programs
in
place,
but
I
don't
think
it's
gonna
go
away.
F
So
I
love
this
question
because
our
vision
statement
is
building
communities,
teaching
Universal
data
literacy,
so
we
do
have
a
vision
of
the
future
that
includes
everyone
knowing
about
data,
and
you
know
how
we
get
there
I
think
there's
all
these
programs
plus
many
more,
but
the
community
aspect
is
key
and
building
up
people
from
all
different
backgrounds
and
experiences
to
have
these
skills
and
support
each
other
and
be
a
part
of
this
broader
data
community,
so
about
a
lot
of
different
ways
to
get
there
and
that's
why
I
love
panels
like
this?
Oh.
E
Yes,
so
quite
a
thing
I've
been
trying
to
communicate
is
that
we've
had
to
begin
with,
educating
our
actual
members
and
our
board
and
for
the
students.
Our
enrichment
programs
allows
allow
us
to
data
science
education
into
their
experience
as
a
part
of
our
program.
The
fact
that
our
instructors
actually
teach
in
the
schools
allow
us
to
bring
that
into
the
school
and
we
are
beginning
to
work
with
the
starting
with
the
principals
of
those
of
those
schools
and
then
with
the
math
instructors
and
and
others
to
bring
that
more
into
their
normal
education
experience.
E
A
I
mean
I'd
like
to
thank
all
of
you.
This
has
been
a
really
rich
panel,
I
think
for
listening
to
all
the
things
that
are
going
on
and
happening
in
this
space
as
we
move
for
as
I
said,
this
is
the
second
webinar
and
they
keep
the
data
science
broad.
Also,
the
website
for
the
series
is
in
the
links
for
this
Perdition
of
this
particular
youtube.
A
Video
and
we'll
post
more
webinars,
as
we
are
partnering
with
the
National
Academy
of
Science
as
they're
doing
a
study
on
him
on
this
topic
of
envisioning,
the
future
of
data
science,
and
so
there's
gonna,
be
a
series
of
webinars.
There's
already
been
some
workshops,
and
really
my
goal
is
to
make
sure
that
this
discussion
of
data
science
even
stayed
as
broad
as
as
this
panel
and
and
other
things
that
is
there
going
on
so
that
it
doesn't,
you
know,
become
pigeon-holed
in
a
certain
type
or
a
certain
particular
type
of
mechanism
for
going
forward.
A
So
that's
another
benefits
to
filling
out
the
input
form,
but
also
to
coming
back
and
viewing
some
of
the
other
webinars
that
we'll
run
through
the
National
cutting
of
science,
which
will
be
posted
on
the
website.
And
if
you
want
to
get
more
involved
in
the
hubs
you
can
visit
any
of
our
web
pages,
you
can
go
to
the
South
hub.
I
think
we
also
have
links
to
the
West
Midwest
that
northeast
hub
and
you
can
get
involved
in
the
education
and
workforce.
A
So
thank
you
very
much
for
watching
and
particularly
and
also
I,
think
Costa,
who
is
in
technical
support
for
this,
as
well
as
Beth
and
Donnellan
Costa,
from
no
innovation
and
and
been
battlin
from
no
innovation
in
who
do
fantastic
work,
as
well
as
that
Russell
who's
been
a
consultant
with
the
South
hub
for
some
of
our
education
programs.
She
is
also
fantastic,
and
so
thank
you.
Both
thank
you.