June 29, 2025 • 47 mins
Welcome to episode 34 of the Designing with Love podcast! In this episode, I had the pleasure of interviewing Dr. Matthew Metzgar, an expert in the higher education field and the author of a new book titled The Overnight AI Educator: Transform Your Course in 24 Hours.
Dr. Matthew Metzger explores how AI is transforming education and shares his innovative five-step approach to help educators adapt their courses for this new reality. He discusses his journey from discovering students using ChatGPT on exams to developing a comprehensive framework for integrating AI ethically and effectively into higher education.
Some of the elements we discussed in the episode include:
• Identifying skills in demand using AI research to ensure coursework aligns with workplace requirements
• Creating realistic assignments that are more resistant to AI shortcuts by making them practical and concise
• Supporting learner development with AI-generated resources like podcasts, videos, and digital twins
• Providing efficient feedback using simplified rubrics and exploring AI-assisted grading
• Iterating gradually by implementing small changes each semester rather than overwhelming overhauls
• Shifting the educator's role from content delivery to evaluation and guidance
• Motivating students through realistic, workplace-relevant assignments
• Embracing diverse perspectives by allowing students to use various AI tools rather than instructor-limited chatbots
• Normalizing AI as an educational tool through open discussions with students
Please visit Dr. Metzger's website, The Overnight Educator, to find current AI tools for education and information about his book "The Overnight Educator: Transform Your Course in 24 Hours."
Use the link to join PodMatch, a place for hosts and guests to connect.
Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you.
💟Designing with Love + allows you to support the show by keeping the mic on and the ideas flowing. Click on the link above to provide your support.
☕Buy Me a Coffee is another way you can support the show, either as a one-time gift or through a monthly subscription.
🗣️Want to be a guest on Designing with Love? Send Jackie Pelegrin a message on PodMatch, here: Be a guest on the show
📝Check out the companion blog site here: Designing with Love Blog
📱Send a text to the show by clicking the Send Jackie a Text link above.
👍🏼Please make sure to like and share this episode with others. Here's to great learning!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Jackie Pelegrin (00:01):
Hello and welcome to the Designing with
Love podcast.
I am your host, Jackie Pelegrin, where my goal is to bring you
information, tips and tricks asan instructional designer.
Hello, GCU students, alumni andeducators, welcome to episode
34 of the Designing with Lovepodcast.
(00:21):
Today, I have the pleasure ofinterviewing Dr Matthew Metzger,
an expert in higher educationand the author of a new book
titled the Overnight Educator.
Dr. Matthew Metzgar (00:35):
Transform your Courses in 24 Hours.
Welcome, Dr Metzger Hi.
Jackie Pelegrin (00:37):
Jackie, thanks for having me.
Dr. Matthew Metzgar (00:41):
Thank you.
So would you mind telling us alittle bit about yourself?
Sure, I'm a clinical professorof economics at the Bell College
of Business at UNC, charlotte.
I've been there coming on maybe15 years, and I worked in the
private sector before that, withsome consulting and other jobs,
and most of my focus lately hasbeen on education and teaching
(01:03):
techniques and, of course, ai.
Jackie Pelegrin (01:05):
Great, yes, and we know AI is just.
It's in all sectors right, andespecially education.
So it's exciting, definitely.
Dr. Matthew Metzgar (01:15):
It's a seismic shift.
That's for sure it really ishow sudden, how much changes
it's causing and how adaptablepeople have to be.
It's just amazing, amazing, howmuch impact it's causing and
how adaptable people have to be.
It's just amazing, amazing howmuch impact it's having.
Jackie Pelegrin (01:28):
Absolutely yes.
So what inspired, you to writeyour new book, and why now?
Dr. Matthew Metzgar (01:36):
Well, I tell you I was a little late to
the AI party, so to speak,because I started hearing about
it and other people were sayingyou know, look into it, all that
, and I kind of just put it offto the side.
But then last year in myclasses, you know, my students
started using AI and then, youknow, magically, these test
(01:57):
averages started going up and upand up and the time to complete
the exams are going down.
And so, you know, I start toask the students what's going on
and everything like that.
I surveyed them and most ofthem were using ChatGPT.
And then I got a little moreserious about the tool myself
and I realized I really had tochange everything because the
(02:22):
assessments I was using were nolonger valid.
And then, once you change thoseassessments, everything else
has to change as well.
You know, as far as what you'reteaching for and the structure
and so on and so forth,transformed my course.
(02:47):
But other professors in thesame boat and a lot of people
were looking for answers and youknow asking me and you know
other faculty.
You know I presented at someconferences on this and said you
know lots of questions becauseit's so new and there isn't just
, you know, one answer out there, but I feel like I have a good
enough grasp and I've had somesuccess that I wanted to share
that with others and say, look,here's a way you can think about
(03:09):
it, here's a process you canuse that can hopefully allow you
to be successful.
Jackie Pelegrin (03:14):
Wow, that's great.
And so that way they're notshying away from it, but they're
embracing it.
And I know some of the facultyI work with it's hard because
they want to embrace it butthey're scared, they don't know
how to do that.
And if they open it up, what isthat going to cause, right?
So and I'm glad you mentionedabout the curriculum we can't
(03:34):
just go and change one thing andthen not have a ripple effect
on everything else.
So it's so important with meworking in curriculum design, I
always mention that to thefaculty that we work with when
we go and revise the curriculum.
If you want to take a topic out, well that's going to cause a
shift and it's going to causechanges.
(03:54):
So being able to have that inmind is always important, that
we can't just take something outand then not have something be
affected somewhere.
So it's important.
Dr. Matthew Metzgar (04:04):
Absolutely.
I mean with something like thisthat you know, again, it's
affecting all pieces of itbecause you know it can affect
the materials you can produce asa teacher.
You can use AI, you can use AIfor grading, your students can
use AI to help on assignments,and so it's affecting all the
different parts on the studentside and the teacher side, and
so if you change one thing, theother pieces have to change as
(04:27):
well.
There was I don't know if yousaw in the news I can't remember
what school it was but astudent wanted a refund on her
tuition because she said herprofessor didn't want them to
use AI.
But then they found out he wasusing AI to produce notes and
things like that oh my gosh.
(04:49):
So she asked for this tuitionback or something.
He was in the news last week.
Jackie Pelegrin (04:52):
Oh wow, that is interesting.
Dr. Matthew Metzgar (04:55):
If you're going to use that as a teacher,
you have to allow your studentsto use it.
Jackie Pelegrin (04:58):
Right.
It's almost like that sayinggoes if you don't allow them to
use it, they'll find a waysomehow to use it, and they'll
do backdoor channels and thingslike that, and so they'll find a
way somehow to utilize it andtry to be smart about it.
So it's better to at leastexpose them to it while they're
(05:19):
in school and while they'regetting their education, and
then by the time they get to theworkforce it's not going to be
foreign to them, and so I think,as educators, it's something we
can do to help prepare them forthat and learn how to do it in
an ethical way as well.
So that way they translate itto the workplace, that way.
Dr. Matthew Metzgar (05:39):
It really is.
You know we had you knowCharlotte's a pretty big city
and as far as business and allthat and we had some business
leaders come in and speak to usand so you know I asked are you
guys using AI in your companies?
And they all 100% said yes, youknow, so it's not theoretical
or this is something we're goingto do in a few years.
(05:59):
You know these are all majorcompanies.
You know that are all using AIevery day in multiple projects
and so you know if a student,graduates and all they've used
chat GPT for was the copy andpaste exam questions, like
multiple choice exam questions,you know you haven't done them.
You know any service.
You know you want to preparethem, like you said.
(06:20):
You know here's how to use thattool in an ethical and a
productive way.
You know it's supposed to makepeople more productive.
You know here's how to use thattool in an ethical and a
productive way.
You know it's supposed to makepeople more productive.
You know at work and in othersituations.
So you've got to show studentshere's how you can use it to
make yourself more productiveand to help whatever company or
organization you're going towork for.
Jackie Pelegrin (06:38):
Right, absolutely Great.
I love that.
So can you walk us through yourfive-step overnight cycle for
AI driven course design that yououtlined in your new book
coming up?
Dr. Matthew Metzgar (06:49):
Yeah, and you know the when you talk about
that fear part earlier, I meanthat definitely is something I
thought about, you know, whenwriting the book, because you
know if you look at surveys whyfaculty might be reluctant to
make a change.
You know one is time.
You know the majority offaculty are actually, you know,
(07:10):
part-time in some way and sothey may be working other jobs
and don't have a ton of time to,you know, put into a course,
might just be teaching one ortwo courses.
So time's a factor.
You know the way you've donethings is a factor.
If you've been teaching thisclass for 20 years a certain way
you might be reluctant tochange.
And then three, like you said,the fear part.
(07:31):
You know well.
You know what do I do with this.
You know it's always that fearthat the students are going to
be way more advanced than theteacher at it and so maybe just
forget about it and kind of putyour head in the sand.
So I was trying to make a realsimple process.
You know that faculty could gothrough in a fairly quick way
(07:53):
and I think if it's, the morestructured it is and you know if
it can be done in a shortperiod of time, the more likely
someone's going to do it.
Absolutely so if you say, here,we want you to integrate AI in
your course, it's going to takeyou six months, well, you know,
you might have some peopledisappear.
Jackie Pelegrin (08:12):
Right, that's so true, especially, as you said
, with most of them teachingpart-time, like myself, where
you know I have a full-time joband then I do this on the side
for extra income and also for anopportunity to be able to stay
current in the field as well.
So there's multiple factors why.
You know we have a lot ofadjunct faculty that teach
(08:33):
courses at universities.
Dr. Matthew Metzgar (08:35):
Yeah, absolutely yeah, yeah, and so
definitely a lot of thoughtabout designing that product, so
to speak, for the audience.
But yeah, as far as the fivesteps, it's a little bit of a
backwards design from GrantWiggins.
He called it Understanding byDesign, I don't know if you're
familiar with it probably someof you are but I always was a
(08:58):
fan of his.
He passed away in 2017 orsomething, but he had a lot of
writing and books and a lot ofgreat ideas, and so that
backwards design.
It always stuck with me throughthe years as a good way to
think about your curriculum oryour class is start with the end
(09:19):
goal in mind, and so that endgoal is like we talked about.
You know, students are going tograduate and go to work for an
organization or company and beexpected to use AI right away.
You know most of these jobs areusing AI.
Now, if you look at, microsoftdid a survey last year and
(09:41):
showed, you know, the majorityof workers I don't know 75, 80%
are using AI, and I'm sure it'sgone up in a year since.
You know I think the new reportwill come out soon this year so
they're entering thatenvironment where AI is part of
it, and so you've got to trainthem for that.
So that first step is looking atthe skills in demand and,
(10:05):
coincidentally, you can use AI,chat, gpt, whatever tool you
want to help you do that, and soif you're teaching like in my
case, economics or differentcourse, biology, chemistry,
whatever.
You can just go into chat GPTand say you know, chat GPT has
that deep research capability,depending on your plan or
whatever.
(10:25):
And you can say you know,survey current job ads and tell
me what are the main skills indemand, not just AI, but just
you know all the main skills andso you know you can do that
fairly quickly.
And what you'll find often isthat you know, yes, ai
(10:47):
technology is in there, but asAI has become able to do some of
these quantitative things,those other skills about you
know presenting, communicating,you know working with others on
teams, kind of those likeevergreen type skills have
become even more important.
And so you know, if you pull upa job ad, you know today, for
(11:09):
whatever this is a biologist,there'll be all the biology part
in there, but it's also a lotabout communicating.
You know presenting, usingpresenting using Microsoft tools
and things like that to present, and so on and so forth.
Jackie Pelegrin (11:21):
Right.
Dr. Matthew Metzgar (11:21):
So that's step one is figure out.
You know your graduates aregoing to finish and go into what
type of jobs.
You can also try to speculate abit and say, all right, well,
that's what they may need in2025.
But what about these studentsthat are freshmen or graduated
in four years?
And you can try to have alittle bit more of a future
(11:43):
focus.
That involves a little morerisk and speculation, because no
one knows for sure what the jobmarket will look like in four
years.
Jackie Pelegrin (11:54):
Yeah, it can be hard to predict the future job
market will look like in fouryears.
Dr. Matthew Metzgar (11:56):
Yeah, it's going to be hard to predict the
future, exactly, exactly.
But I mean again, I would say,still those at least core
elements about communication,teamwork, so on and so forth,
self-directed learning, aregoing to be there forever.
Jackie Pelegrin (12:09):
Right those key soft skills, right that they
need Definitely, definitely,yeah.
Dr. Matthew Metzgar (12:16):
And so that's step one um again.
So then you come up with the,the skills in demand, current
and future um, and now you'relooking at designing the tasks.
Right, that will, you know,mimic or be similar to those um.
And this is where it's reallychanged, because the class I
(12:36):
teach it's technically,historically, been a lot of like
quantitative work, you know, solike find this optimal price or
calculate this cost, and sostudents would do that.
Well, now I can do all thatvery quickly, and so a lot of
that quantitative work has beenkind of like outsourced, so to
speak.
Oh, wow, quantitative work hasbeen kind of like outsourced, so
to speak.
And so now you have to look at,okay, you become almost the
(13:07):
evaluator of the AI results.
You're making sure it's true.
So you could say, all right,calculate this elasticity, and
it gives you this number, but itmay not be right.
Ai can make mistakes.
So you don't have tonecessarily do the calculation,
but you have to know whetherit's correct or not and then, if
it's incorrect, obviously comeup with the right number.
Jackie Pelegrin (13:23):
Oh, so helping students to disseminate the
content that's coming out of AIand whether it's correct or not,
wow.
Dr. Matthew Metzgar (13:30):
Definitely and it's definitely not always
correct.
I mean, I've done some things,you know, like sample
calculations.
I'll put in there.
You know calculate this,calculate that, don't give me a
wrong number, and then you knowI'll write back.
You know this is not correct,and then it'll write back, you
know think and be like, oh yeah,you are right.
So hold on, let me recalculate.
Jackie Pelegrin (13:48):
Yes, I've done that.
Oh, I'm sorry, let me try thatagain.
It's almost like you're havinga conversation with the tool.
It's pretty interesting.
Dr. Matthew Metzgar (14:04):
Well, as you say, at least it's polite
right, Right?
Jackie Pelegrin (14:06):
That's true.
I wonder if we were rude to it,if it would be rude back.
I just wonder that sometimes,because I'm always polite Say
can you do this please?
And it'll have that tone, it'llmimic that tone.
You know that's where thatprompt engineering comes into
place really well with that.
So they say what you input willget what you get in the output.
Dr. Matthew Metzgar (14:29):
So right, Absolutely, absolutely and so,
yeah, so designing those tasks,the other key pieces I think
about is one I try to make it arealistic task as good as I can,
because I'm a little afraid.
You know you can go in AI andsay you know, generate this
scenario about you know, companyX, you know, sells widgets at
(14:52):
$2 a widget or something.
You know sells widgets at $2 awidget or something.
But I think if it's a fake orartificial problem, a lot of
students are then going to justrun it through chat GPT because
they're going to say, like Idon't care about this, it's not
real, it's just something he orsomeone made up.
Right, you know they're notgoing to be really motivated to
(15:13):
do it, and so I've found thatthe more realistic problem I can
make, the more likely studentsare to actually try it and not
just totally outsource it insome way, like I just gave a
task when the semester ended andI ran it through Chet GPT
myself to see what the answer,what it would give.
(15:34):
But I didn't have any studentgroups submit that answer and
that's because I think theythought they perceived that test
to be valuable, saying, hey,this might be something that I
would actually do and work atsome point, so it'd be a good
idea for me to try it now, youknow, because it was a real task
.
But if it was just fake, youknow, make these widgets or
something like that I thinkyou're task, but if it was just
(15:58):
fake make these widgets orsomething like that I think the
motivation is going to fall andthe odds of cheating, so to
speak, go up.
Jackie Pelegrin (16:03):
Wow, that's amazing.
Yeah, and you're taking itbeyond a textbook and putting it
in real life for them andsomething that they can imagine
themselves having to face right,a real problem or real
situation where they have tocome up with a solution in a
company.
Yeah, that's great.
Dr. Matthew Metzgar (16:22):
Yeah, absolutely.
I mean, you know it's like, ifyou want to, you know whatever,
learn how to ride a bike orsomething you know.
And if someone tells you justyou know, pedal on a stationary
bike, you know then you mightthat's not going to help me,
right?
You know I on a stationary bike, you know, then you might, uh,
that's not going to help me,right?
You know I wouldn't learn howto ride a bike out on the street
(16:42):
or on the mountains or whatever.
So you know you want thepractice to be as similar to the
real as you can get.
So you know I pull stuff, youknow, from the news.
You know, the week I'm makingthe assignment, you know.
So it might be something thathappened two days ago and I'll
pull it and make a task out ofit, and that way there is no
history that someone did this.
You know, someone solved thisproblem 10 years ago or
(17:04):
something you know.
It's brand new.
It's right out there now.
Jackie Pelegrin (17:07):
Wow, I love that.
That's great.
Dr. Matthew Metzgar (17:09):
So one with the task, I think, one about
making them realistic and havingthem in a good context, and
then two, giving them at leastthe option to work with others.
Because, again, if you look atwork situations, most work now
involves other people, whetherit's in person or virtual.
And so I, when the semesterstarts, the first task I assign
(17:34):
the groups of five.
And then you know, becausepeople may not know each other
and all that but then tasksafter that, then I open it up
and say you can kind of what ineconomics we call vote with your
feet, but you can move.
You can, you know, leave agroup and start a new group, or
you can go to a smaller group,or you can work by yourself or
(17:54):
whatever.
But at least give that initialchance for everyone to be in a
group.
Maybe it works, and so on andso forth.
And then you know, if peopleare unhappy I give them the
option to work in different ways.
But you know, now that I'msaying this, I should probably
calculate what percent of peoplestayed in the groups, but I'd
(18:15):
say it's pretty high, definitelyover half.
I'd probably say two-thirds ofpeople choose to stay with the
group, even though they don'thave to.
Obviously it can be easierbecause you can spread the
workload.
But two, I think you can end upwith a better project or
product output by working withothers than just yourself
(18:37):
usually, and so, again, that'sgiven a more of a realistic
scenario that if you're at workand you're trying to accomplish
X, it's probably going to payoff to get some people to help
you with that Right, becauseotherwise it's a long slog on
your own.
Jackie Pelegrin (18:55):
Yes and so.
Dr. Matthew Metzgar (18:56):
I'm trying to get them to think about that
as well, right, becauseotherwise it's a long slog on
your own.
Jackie Pelegrin (18:59):
Yes, so I'm trying to get them to think
about that as well, right, sothat they're not feeling like
they have to do all of this inisolation when they get to the
workplace, because so much isabout collaboration and it's
hard to accomplish anything bigon your own in the workplace,
right?
It's always about delegating or, like you said, bringing
together a team of people thatcan come up with a solution
(19:20):
together and work on that tasktogether, as well, absolutely,
absolutely, yeah, great.
Dr. Matthew Metzgar (19:27):
And then let's see.
The third step is supportingthe learner's development, and
that's where kind of the typicalteaching, so to speak, would
come into play, where you knowwhatever classroom setup you
have or an online setup throughan LMS and all that.
But I'd say the other piece,the biggest change to that with
(19:49):
AI, is that AI can help youdevelop all of this, all of
these resources now to helplearners on their journey.
If you have notes on a topicthat you yourself, the
instructor, wrote, you can haveAI turn that into a podcast.
There's tools out there to dothat.
(20:11):
Or you can have it turned intoa video.
There's digital twinning nowmaybe call it something
different, but it's where youonly have to record a couple
minutes of yourself on video andthen you can give it a Word
document and it'll now create avideo of you saying these things
(20:33):
that you've never said.
Jackie Pelegrin (20:35):
Oh, wow, yeah, it's called a tool to link out.
Dr. Matthew Metzgar (20:39):
there is a HeyGen H-E-Y-G-E-Ncom.
Jackie Pelegrin (20:43):
Oh, I've heard of that, I've heard of.
Heygen, I haven't used it yet,but I've heard of it.
Dr. Matthew Metzgar (20:47):
Yeah, it just creates a digital twin and
so you could have yourselftalking about whatever physics,
biology, you know something youdon't know nothing about.
That wouldn't be good, but I'mjust saying it's a.
It's a that type of tool, youknow.
There's all sorts of tools outthere like that I said.
The one other cool thing aboutdigital twinning is it can also
(21:09):
translate what you say into adifferent language.
Jackie Pelegrin (21:12):
Oh, I've heard about that.
Dr. Matthew Metzgar (21:14):
Yeah, and so if you have, you're teaching
a class, you know overseas orsomething and you don't speak
that language, you could, youknow, have it, create a video of
you saying your normal notes inthat language for those
learners, and so you can justcreate a whole bunch of
resources quickly that you, youknow, couldn't do before.
You know, with AI, you know youcan create podcasts, notes,
(21:36):
visuals, again videos, digitaltwins, all sorts of stuff out
there to help the learners getthe material.
Jackie Pelegrin (21:45):
Wow.
So that opens up institutionsthat are doing online education,
like Grand Canyon University,to be able to offer the courses
to students that are outside ofthe United States, too.
More often.
I've only had, I think, twostudents that have been outside
the United States becausethey're in the military or they
(22:06):
work on a military base orsomething like that, so it's
been rare, but it would be niceto be able to tap into more of
the international studentsabroad and get to learn from
them and their culture andthings like that and how they
perceive higher education wherethey are too.
So it would be neat for sure.
Dr. Matthew Metzgar (22:27):
Absolutely.
I think there's tremendouspotential for what you said.
That can really open up a lotof doors and make it accessible
to a lot more people thanpreviously.
Jackie Pelegrin (22:38):
Right.
Break down those barriers Right.
Dr. Matthew Metzgar (22:41):
Yeah, absolutely.
And then the other piece I'd saywith the supporting learners is
honestly, even if you give yourstudents all these different
resources, they may choose ontheir own to just go with a
conversational AI agent and Iknow a lot of my students do
that and I build in on thewebsite.
(23:01):
We have Google Gemini on campus,a co-pilot.
I kind of build that into pagesand with a textbook and all that
, just that some students aregoing to be more comfortable
learning about that topic.
So if you say I want to learnabout, you know, price
discrimination or something,well they can just put it in to
(23:22):
their agents, they tell me aboutit and go back and forth.
And so I think the interactivecontent or setups like that
obviously are going to beprobably more ups like that
obviously are going to beprobably more palatable to
students than a lot of the oneway stuff.
You know, I think these thedays of just sitting here
listening to this two hour videoon the instructor are probably
(23:47):
gone or near gone.
That you know students aregoing to be looking to get
information quicker and more ina back and forth situation with
the agents, and so obviouslythat's led to, you know,
instructors building chatbotsfor their class and maybe just
training those chatbots on theirmaterials, only putting it out
there.
But again, even if you don'tmake a chatbot for your class,
(24:10):
students can use any AI platformto get at that same information
.
So that's good to know,definitely, I would say one
thing on that.
That kind of a question comesup when, cause we talked about,
you know, faculty trying to newfaculty that are maybe hesitant
to do this and sometimes theymight feel like, well, I have to
(24:31):
make a chat bot for my classand I don't know how to do that.
I would say that, again, that'snot a requirement.
You know, if you aren't techsavvy and you don't want to do
it or whatever, I would say,just you know, skip it and let
them use existing chatbots thatare out there, because you can
create a chatbot to focus onlyon your.
(24:53):
You know the materials that youhave, that you've written or
whatever.
But in my mind that's a littlebit of what I'd call like in
economics, like creating amonopoly.
Like in other words I don't wantthe students to only be
dependent on just the way Ipersonally see things.
You know, I might explain itthis way, I might think of it
(25:17):
this way, but someone else mayexplain it this way, and in a
chatbot they're getting adifferent explanation.
And so, like for me, I did nottrain my chatbot on just my
stuff.
I just said I just made ageneral.
You know, it's based off acopilot and I would get
different type of answers thissemester than I would previously
in a good way semester than Iwould previously.
(25:38):
In a good way, because beforeit's like they were, I was
almost teaching them to thinklike me, and so the answers
would be answers like I wouldproduce.
But you know, you want a littlemore diversity of thought, I
think, and so you want diversityof sources, and so sometimes
they give me an answer thatreally surprised me and it's
like, oh, that's cool, it's adifferent way of thinking of it.
(26:00):
It's not wrong, it's justdifferent.
And so, you know, sometimes Iwould be like, oh, you know,
that's cool, you know, I'velearned something.
So I think it pays to have moresources than just your own to
the students Because, again,each person's limited.
You can only know so much, andit's better for them to get a
diversity of opinions onsomething than just what the
(26:22):
instructor says.
Jackie Pelegrin (26:24):
Oh, I love that .
That's a great way to look atit too when it comes to these
types of tools, and that it'syeah, you don't want to shoehorn
them into one thing, but youwant them to have that diverse
way of thinking.
I like that.
That's great.
Dr. Matthew Metzgar (26:39):
Yeah, yeah.
So definitely a factor.
And let's see the next stepprovide efficient feedback.
And this is something where alot of faculty are going to
start to worry.
Yes, because if you move away ina simplistic term, if you're
moving away from like multiplechoice questions or like numeric
(26:59):
response questions to thesekind of tasks I'm calling them,
could be many projects, whateveryou want to label them
open-ended type assignments.
Now I was in a conference andsomeone raised their hand and
said this well, how do I gradeall this?
You know, I can't give mymultiple choice exams because
they can use ChatGPT and get,you know, 95%.
(27:20):
I've got to give them all thesedifferent types of assignments.
How do I grade it all?
Well, if you have, you know, 15or 20 students, you can grade
it.
But if you have big numbers,then it gets to be more
problematic, and so if aninstructor is making this shift,
they have to really thinkbefore the course starts shift.
(27:42):
They have to really thinkbefore the course starts.
You know, how much feedback canI realistically offer?
Because you can offer feedbackon like drafts of, you know, the
assignments and then, of course, they have to be graded.
So there has to be feedbackthere.
But you have to think aboutthere's only so many hours in
the day and you know there's anopportunity cost to grading, to
giving feedback.
It takes your time as aninstructor and so there's no
(28:05):
right or wrong answer.
But it's just something youhave to be aware of because
otherwise you can get suckedinto just continuous drafts and
emails and everything else andyou may not be able to get
anything else done.
So what I found out?
I teach big classes.
This last semester my class was230 students.
Jackie Pelegrin (28:23):
Oh wow, that's a big class.
Dr. Matthew Metzgar (28:25):
Very big and so you know, first I get
them into the groups, like wetalked about, and so that cuts
it down from 230 to maybe.
I think maybe we'd end up withmaybe 80 or 90 assignments after
the group's resettled for thoseother tasks, but still that's a
lot.
So the one thing, a couple ofthings I learned was one to keep
(28:50):
the assignments shorter, and sowhen I assign these tasks
there's kind of steps to it, butI say the complete thing.
You submit a PDF, you know it.
Either we cap it at either oneor two pages, because if you
don't, someone's just going togo on there and make a 20-page
chat GPT output and give it toyou.
(29:11):
You're going to be stuck lookingat it when you know it's not
even real.
The students need to do itRight.
And also, like I was sayingearlier, think about that
student motivation.
If you tell them you have tosubmit this 20 page paper, their
motivation is going to be lowand more likely to use ChatGD2.
If it's one or two pages,they're going to.
You know again they're going tothink, okay, well, that's not
(29:32):
too bad, I can do it, you know,or at least you know I'm willing
to try it, rather than a 20page paper.
I don't want to write I'm noteven going to try that.
I'm willing to try it ratherthan a 20 page paper.
I don't want to write, I'm noteven going to try that.
Jackie Pelegrin (29:41):
Oh, that's so important.
Yes, especially with curriculumdesign, being able to keep that
in mind with their motivation.
Absolutely, definitely.
Dr. Matthew Metzgar (29:50):
Definitely, and the other thing, too, I'd
say is that it's not beyond that.
It's also realistic in thesense that a lot of times in a
work environment you may producea long paper or whatever, but
most of the time it might besomething like an executive
summary or the PowerPoint notesthat actually most people, 90%
(30:12):
of people, are going to look at,and so that's really what I'm
trying to get the students toproduce is something, you know,
a smaller document or somethinglike that, that would be similar
to what they present in theworkplace.
Jackie Pelegrin (30:29):
Right.
Dr. Matthew Metzgar (30:30):
And so, like I think to myself, like you
know where I work at theuniversity, no one's ever sent,
you know, a 50-page paper foreveryone to read, because, one
they know no one's going to readit, but two, it's just not
effective, right?
So instead you get memos andnotes or PowerPoints and things
like that.
So, again, that's what I'mtrying to think about them in
(30:50):
the workforce, you mayoccasionally have to write a
long white paper, but most ofyour communication is going to
be shorter for general audience,in something like, again, one
to two page in this situation.
Oh, that's great.
And so the other piece with thefeedback then comes into, since
it's an open-ended task, youknow grading it with some type
(31:13):
of rubric, and we started withthat, you know, in January.
You know you can use AI tocreate rubrics easily.
You know just a couple seconds.
But what we found with thatfirst task was it was still
getting too long of a process tograde myself and the teaching
(31:43):
assistants on the actual grade.
A paper should get you knowbecause, again, some of those
rubrics do allow forsubjectivity, so to speak.
You know, is this a three or afour in this category?
And so what we then went to forthe other.
For the rest of the task, therest of the semester was much
simpler rubrics where you knowyou might have some items, but
they were all essentially likebinary, like yes, no, type items
(32:07):
, and so you know we'd just belike you know, was this piece
calculated correctly?
You know it's yes or no, was agraph of a demand curve
presented, yes or no?
And we found that one that spedup the grading.
But two I think it reallyhelped the students.
(32:28):
It was more structured andallowed them to kind of see what
the end product should looklike.
And some of the first, when wedid the very first thing, the
rubric, we had all sorts ofdifferent formats and looks and
everything like that.
But as we simplified the rubricand cleaned it up, we were able
(32:52):
to get a lot more good-lookingpapers out of it, and so it's
kind of just like puttingconstraints on it is the way I
would think of it.
Jackie Pelegrin (33:02):
That's a good example I love that how you can
really utilize that AI to getthose outputs that you're
looking for.
Dr. Matthew Metzgar (33:11):
Oh yeah, yeah, I mean, you know it can
create rubrics in a snap.
And then, of course, the otherpiece is you know we messed
around with a little bit, wedidn't go all out, but you know
you can also use AI to give youa base grade on a paper, you
know.
And again, of course, you havedata collection issues and
privacy and so on and so forth.
Right, but you know there'ssolutions out there, like Timely
(33:35):
Grader, where it can, you know,run through the essays.
I think they're trying toconnect to Canvas and so
there'll be other companies soonthat'll do that.
But it's like the AI will dolike a first pass grading and
then you, as the human, can comein and agree with that or not
change it, make your commentsand then go through.
(33:55):
But to me, that's definitelygoing to be the future in some
way is that AI will do a firstpass grade and then you have the
human judgment that kind ofconfirms it and gives the
comments and, you know, providesthat value to the students.
Jackie Pelegrin (34:13):
Right, I love that.
So it's giving that firstanalysis of the paper or the
assignment for you as aninstructor.
I like that.
Dr. Matthew Metzgar (34:23):
Yeah, because a lot of times it can
save time.
You know if this assignment'syou know way off or something
like that, you know it mighttake you a while to figure out.
You know what's even going on.
But you know AI can kind ofquickly look at this and say, oh
, this thing is, you know, maybethe student didn't understand
the assignment or something likethat.
It's way off, and so that firstpass can definitely help.
(34:45):
Again, it's not perfect.
We were, because when we'd runsome papers through AI to get
what it would say for the rubric, again it would never exactly
line up with the way we werethinking.
You know as people, and so it'sa good start, but again, you
need that human element in there.
Jackie Pelegrin (35:08):
Yes, absolutely , that's so true.
I can see how that would be aconcern for faculty.
Wondering, you know, if I'mputting this in here, you know,
is it taking away some of what Ido?
But it sounds to me like whatyou've discovered is that it
enhances the experience as aneducator and also for the
students as well.
Dr. Matthew Metzgar (35:29):
Yeah, I mean to me I think also we talk
about in the book is AI has kindof shifted the role of the
educator, whereas before it wasmore about the educator was the
expert in the educator, whereasbefore it was more about, you
know, the educator was theexpert in the content.
And I'm going to, you know,talk about the content and
explain the content and so onand so forth.
(35:49):
And then maybe the grading wasdone again via multiple choice
questions and scan trot sheetsand all that.
I think it's really shifting it.
The other, almost having themswap places, where a lot of the
content and the delivery isgoing to be done online or with
AI, and so on and so forth.
And the job of the facultymoves more to an evaluate.
(36:12):
Okay, you know, because if it'sa realistic piece, you need a
person to tell you is this goodor bad?
You can't just have you knowit's not going to be enough,
right?
So I think it really isflipping the role to be more of
an evaluator.
Actually, I was interviewing anart professor about this and she
(36:35):
was saying you know, there's somuch content now for art school
students.
You can go on YouTube and finda video on anything you know.
Here's how to do paint orlandscape or sculpt or whatever
right, it's unlimited amount ofcontent on how to.
But she was saying you know,the reason then, to go to art
(36:56):
school is to have professorsthat you know, have experience
and know what they're doing,evaluate your work and tell you
is it good, bad, what can beimproved, and so on and so forth
.
She was saying that's our mainfunction.
Now, it's not really the how-to, because it's a universe of
how-to videos.
It's more of just okay, youwatch the video and you produce
(37:18):
this piece.
Now, is it good?
Is it going to meet the needsof the company, the consumers,
whoever?
If not, how can it be improvedto do that?
Jackie Pelegrin (37:28):
So definitely a shift in my view.
Yeah, absolutely.
Dr. Matthew Metzgar (37:34):
And the last piece of the cycle is just
iterate and improve.
And you know the tools arechanging so rapidly with AI and
all that.
That.
You know each semester there'sgoing to be a new opportunity
for learning and, you know,putting in new things, seeing
what worked with students, whatdidn't, so just kind of a normal
(37:54):
feedback process.
But the only thing I'd say withthat that like as far as again I
don't want to scare offinstructors from trying this out
is that you can have a bit oflike I'll call it tool fatigue
out there because there's somany new tools, so many things
(38:16):
coming out, so many changes, butyou don't have to use all these
tools, you don't have to doeverything.
You might just say, okay, thissemester I'm going to add in one
new AI tool or try one new AIthing I've never done before.
You don't have to try 10 things, or you don't even have to try
anything new, but at some pointyou're probably going to need to
(38:38):
update.
So it can be done in a in smallways.
And so when we're sayingiterate, that doesn't mean every
semester I've got to massivelychange this course and do all
these things different.
You know, once you've got tomake a shift and get AI in these
pieces.
The next semester should berelatively easy to kind of just,
you know, clean up and keep upwith the current changes, and
(39:01):
then, of course, the feedbackfrom the students.
Jackie Pelegrin (39:04):
Right, and then that way you can do that
iteration from each semesterright and see what needs to
improve, what works well, what,what doesn't, and what needs to
shift as well, from each Exactlyand that's what you know.
Dr. Matthew Metzgar (39:18):
I feel like this cycle is something you can
use.
You know before each term orcourse or whatever in a fairly
quick way.
You know because again you cango, you can run and now and say,
all right, spend six months,run some job ads and see is
there anything new or the skillsand this more or less, and so
on and so forth, so you can dothat.
Then you can kind of go throughyour task and see all right,
(39:41):
are they matching with what'scurrently in demand?
Do I need to update some of thetasks?
Are my resources?
You can check the stats and seeare students using like these
things more than this?
Do they like the podcast andnot the videos or vice versa?
And then you can think aboutthe feedback how much time did
(40:01):
it take?
Do I need to be more or less ormore efficient with it, or can
I give more feedback and thenagain think about small things
you can improve.
But ideally, again, I'm tryingto make an easily repeatable
process that someone could do ona semi-regular basis and that
keeps them up to date and withthe times and producing good
(40:25):
results for their students.
Jackie Pelegrin (40:27):
Right and I love how it's five steps, so
it's not complicated, and Ithink that's great, especially
for educators and, like you said, no matter if they've been
teaching for 15, 20 years or ifthey're just starting out.
Dr. Matthew Metzgar (40:47):
I think it's a really great process for
them to be able to use at anystage, and their teaching as
well.
Yeah, exactly that's what youknow.
When I started writing the book, it's like you know, we
initially had, you know, justdifferent chapters and chapter
on this, chapter on that, and Iwas just like you know, the
business instructor in me saidyou know, this has to be more
structured.
This has to be, you know, stepone, two, three, four, five, you
(41:08):
know, to really get someone tofollow through.
In other words, you know, a lotof times you might just read a
book and then put it down and itdoesn't do anything right.
But I'm trying to make a moreactionable book that's saying
okay, just here's this process,go through it and then you can
get you know again up to dateand get some good results.
Jackie Pelegrin (41:29):
That's great, Wonderful.
So a final question as we wrapup.
So what's one practical pieceof advice that you'd give to an
educator who's curious about AIbut not sure where to start?
Dr. Matthew Metzgar (41:45):
Well, I guess it would depend.
You know how often they'reusing AI themselves.
You know, like some people areheavy users of chat GPT, use it
every day or something like that.
I mean, I use it for a lot ofdifferent tasks.
So I would say, if they haven'tused any AI, to just start
(42:09):
going on there and trying to useit for personal use, seeing
what kind of value you can getfrom it.
You know, ask questions and soon and so forth.
I think probably most peoplehave, at this point, at least
experimented with it somewhat.
You know the other thing, asmall step, could just be even
asking their students.
(42:31):
You know how do you use ChatGPT, asking their students you know
how do you use ChatGPT.
And so trying to again get itmore out in the open that, okay,
this is a tool that probably75% of the population uses, at
(42:51):
least 80% of students use, andjust making it more kind of like
normalizing it.
Right, you know, okay, this isa tool, we know you use it, I
use it, I use it, everyone usesit.
And now start the conversationabout that.
Well, how can we maybeintegrate it in this class or
what can we do with it?
But I'd say that's.
This first thing is just maybesome awareness, some
transparency about it, justsaying, you know, hey, this is a
(43:12):
very powerful tool, right?
Right, that can help you asstudents, it can help me as a
teacher, it can help thebusinesses you work for, and so
let's just start talking aboutit, you know.
Let's see what we can do withit.
Jackie Pelegrin (43:27):
Yes, I love that.
Yeah, and keeping an open mindabout it and seeing where it'll
take us Absolutely, I love that.
Well, thank you so much forgoing over the steps.
That is very insightful andyour advice, because I think it
makes it more manageable andsomething that instructors can.
They feel like they can tacklethis and be able to integrate it
(43:51):
into their curriculum and workwith their students and not be
fearful of it Like so manyprobably still are.
So at least they can take that,some of those small steps and
do some things where they don't,like you said, they don't feel
like they have to learn everytool, but pick up the tools that
they think are going to beuseful and try it out.
And that's a great thing is wecan experiment and see what
(44:14):
works and what we need toimprove upon.
So that's a great thing aboutAI as well.
Dr. Matthew Metzgar (44:20):
It is and you know it's just if you look
historically, you know thingsthat seemed kind of odd at first
become, you know, very, verycommonplace.
I was talking with someone andI thought, you know, like the
idea of Uber and Lyft to thinkthat I'm just going to press
this button on the phone and astranger comes and picks me up
and takes me somewhere, you know, if you would have told someone
that before it happened, you'dthink that's crazy, that's not
(44:43):
stuck in real.
But of course it became real.
And so you know, ai is going tochange a lot of norms and you,
just like you said, you have tobe open-minded to it and be
willing to explore it and seewhere it can take you.
Jackie Pelegrin (44:55):
Yes, absolutely .
I remember as a kid watchingthe Jetsons and I thought that's
so futuristic.
We're never going to get tothat, but now, who knows, we
could have flying cars in thenext few years.
So I think AI is going toreally make a lot of things
possible and provide medicalbreakthroughs and all these
different things that they'retalking about.
(45:16):
So it's a pretty amazingadventure to see that where it's
going to take us.
So that's great.
Dr. Matthew Metzgar (45:23):
Absolutely.
Jackie Pelegrin (45:23):
Great, great Well, thank you so much for your
time, and I am so happy that mystudents are going to be able,
and others even around the world.
This podcast does have a globalreach, so I'm really glad that
those around the world will beable to get a lot of great
advice from what you'vementioned today.
So thank you so much for yourtime.
I appreciate it.
Dr. Matthew Metzgar (45:42):
Definitely, definitely, and I list some of
the current tools on my website,overnighteducation.
I try to do kind of a roundupof the current tools but, as we
talked about, things arechanging so quick that I'll
probably do, on a somewhatregular basis, try to update
what tools are out there andgive people some things to work
with that can give them value.
Jackie Pelegrin (46:06):
Great, that's wonderful.
I'll make sure to link yourwebsite in the show notes so
that way all of my listeners cango there and check that out.
And then I did notice on yourwebsite that you have
information on your book, soonce it's out, then everybody
can check it out and order acopy.
Dr. Matthew Metzgar (46:24):
Yeah, excellent, definitely.
Jackie Pelegrin (46:26):
Great.
Thank you Appreciate it.
Thank you for taking some timeto listen to this podcast
episode today.
Your support means the world tome.
If you'd like to help keep thepodcast going, you can share it
with a friend or colleague,leave a heartfelt review or
offer a monetary contribution.
Every act of support, big orsmall, makes a difference and
(46:48):
I'm truly thankful for you.
Hello and welcome to the Designing with
Love podcast.
I am your host, Jackie Pelegrin, where my goal is to bring you
information, tips and tricks asan instructional designer.
Hello, GCU students, alumni andeducators, welcome to episode
34 of the Designing with Lovepodcast.
(00:21):
Today, I have the pleasure ofinterviewing Dr Matthew Metzger,
an expert in higher educationand the author of a new book
titled the Overnight Educator.
Dr. Matthew Metzgar (00:35):
Transform your Courses in 24 Hours.
Welcome, Dr Metzger Hi.
Jackie Pelegrin (00:37):
Jackie, thanks for having me.
Dr. Matthew Metzgar (00:41):
Thank you.
So would you mind telling us alittle bit about yourself?
Sure, I'm a clinical professorof economics at the Bell College
of Business at UNC, charlotte.
I've been there coming on maybe15 years, and I worked in the
private sector before that, withsome consulting and other jobs,
and most of my focus lately hasbeen on education and teaching
(01:03):
techniques and, of course, ai.
Jackie Pelegrin (01:05):
Great, yes, and we know AI is just.
It's in all sectors right, andespecially education.
So it's exciting, definitely.
Dr. Matthew Metzgar (01:15):
It's a seismic shift.
That's for sure it really ishow sudden, how much changes
it's causing and how adaptablepeople have to be.
It's just amazing, amazing, howmuch impact it's causing and
how adaptable people have to be.
It's just amazing, amazing howmuch impact it's having.
Jackie Pelegrin (01:28):
Absolutely yes.
So what inspired, you to writeyour new book, and why now?
Dr. Matthew Metzgar (01:36):
Well, I tell you I was a little late to
the AI party, so to speak,because I started hearing about
it and other people were sayingyou know, look into it, all that
, and I kind of just put it offto the side.
But then last year in myclasses, you know, my students
started using AI and then, youknow, magically, these test
(01:57):
averages started going up and upand up and the time to complete
the exams are going down.
And so, you know, I start toask the students what's going on
and everything like that.
I surveyed them and most ofthem were using ChatGPT.
And then I got a little moreserious about the tool myself
and I realized I really had tochange everything because the
(02:22):
assessments I was using were nolonger valid.
And then, once you change thoseassessments, everything else
has to change as well.
You know, as far as what you'reteaching for and the structure
and so on and so forth,transformed my course.
(02:47):
But other professors in thesame boat and a lot of people
were looking for answers and youknow asking me and you know
other faculty.
You know I presented at someconferences on this and said you
know lots of questions becauseit's so new and there isn't just
, you know, one answer out there, but I feel like I have a good
enough grasp and I've had somesuccess that I wanted to share
that with others and say, look,here's a way you can think about
(03:09):
it, here's a process you canuse that can hopefully allow you
to be successful.
Jackie Pelegrin (03:14):
Wow, that's great.
And so that way they're notshying away from it, but they're
embracing it.
And I know some of the facultyI work with it's hard because
they want to embrace it butthey're scared, they don't know
how to do that.
And if they open it up, what isthat going to cause, right?
So and I'm glad you mentionedabout the curriculum we can't
(03:34):
just go and change one thing andthen not have a ripple effect
on everything else.
So it's so important with meworking in curriculum design, I
always mention that to thefaculty that we work with when
we go and revise the curriculum.
If you want to take a topic out, well that's going to cause a
shift and it's going to causechanges.
(03:54):
So being able to have that inmind is always important, that
we can't just take something outand then not have something be
affected somewhere.
So it's important.
Dr. Matthew Metzgar (04:04):
Absolutely.
I mean with something like thisthat you know, again, it's
affecting all pieces of itbecause you know it can affect
the materials you can produce asa teacher.
You can use AI, you can use AIfor grading, your students can
use AI to help on assignments,and so it's affecting all the
different parts on the studentside and the teacher side, and
so if you change one thing, theother pieces have to change as
(04:27):
well.
There was I don't know if yousaw in the news I can't remember
what school it was but astudent wanted a refund on her
tuition because she said herprofessor didn't want them to
use AI.
But then they found out he wasusing AI to produce notes and
things like that oh my gosh.
(04:49):
So she asked for this tuitionback or something.
He was in the news last week.
Jackie Pelegrin (04:52):
Oh wow, that is interesting.
Dr. Matthew Metzgar (04:55):
If you're going to use that as a teacher,
you have to allow your studentsto use it.
Jackie Pelegrin (04:58):
Right.
It's almost like that sayinggoes if you don't allow them to
use it, they'll find a waysomehow to use it, and they'll
do backdoor channels and thingslike that, and so they'll find a
way somehow to utilize it andtry to be smart about it.
So it's better to at leastexpose them to it while they're
(05:19):
in school and while they'regetting their education, and
then by the time they get to theworkforce it's not going to be
foreign to them, and so I think,as educators, it's something we
can do to help prepare them forthat and learn how to do it in
an ethical way as well.
So that way they translate itto the workplace, that way.
Dr. Matthew Metzgar (05:39):
It really is.
You know we had you knowCharlotte's a pretty big city
and as far as business and allthat and we had some business
leaders come in and speak to usand so you know I asked are you
guys using AI in your companies?
And they all 100% said yes, youknow, so it's not theoretical
or this is something we're goingto do in a few years.
(05:59):
You know these are all majorcompanies.
You know that are all using AIevery day in multiple projects
and so you know if a student,graduates and all they've used
chat GPT for was the copy andpaste exam questions, like
multiple choice exam questions,you know you haven't done them.
You know any service.
You know you want to preparethem, like you said.
(06:20):
You know here's how to use thattool in an ethical and a
productive way.
You know it's supposed to makepeople more productive.
You know here's how to use thattool in an ethical and a
productive way.
You know it's supposed to makepeople more productive.
You know at work and in othersituations.
So you've got to show studentshere's how you can use it to
make yourself more productiveand to help whatever company or
organization you're going towork for.
Jackie Pelegrin (06:38):
Right, absolutely Great.
I love that.
So can you walk us through yourfive-step overnight cycle for
AI driven course design that yououtlined in your new book
coming up?
Dr. Matthew Metzgar (06:49):
Yeah, and you know the when you talk about
that fear part earlier, I meanthat definitely is something I
thought about, you know, whenwriting the book, because you
know if you look at surveys whyfaculty might be reluctant to
make a change.
You know one is time.
You know the majority offaculty are actually, you know,
(07:10):
part-time in some way and sothey may be working other jobs
and don't have a ton of time to,you know, put into a course,
might just be teaching one ortwo courses.
So time's a factor.
You know the way you've donethings is a factor.
If you've been teaching thisclass for 20 years a certain way
you might be reluctant tochange.
And then three, like you said,the fear part.
(07:31):
You know well.
You know what do I do with this.
You know it's always that fearthat the students are going to
be way more advanced than theteacher at it and so maybe just
forget about it and kind of putyour head in the sand.
So I was trying to make a realsimple process.
You know that faculty could gothrough in a fairly quick way
(07:53):
and I think if it's, the morestructured it is and you know if
it can be done in a shortperiod of time, the more likely
someone's going to do it.
Absolutely so if you say, here,we want you to integrate AI in
your course, it's going to takeyou six months, well, you know,
you might have some peopledisappear.
Jackie Pelegrin (08:12):
Right, that's so true, especially, as you said
, with most of them teachingpart-time, like myself, where
you know I have a full-time joband then I do this on the side
for extra income and also for anopportunity to be able to stay
current in the field as well.
So there's multiple factors why.
You know we have a lot ofadjunct faculty that teach
(08:33):
courses at universities.
Dr. Matthew Metzgar (08:35):
Yeah, absolutely yeah, yeah, and so
definitely a lot of thoughtabout designing that product, so
to speak, for the audience.
But yeah, as far as the fivesteps, it's a little bit of a
backwards design from GrantWiggins.
He called it Understanding byDesign, I don't know if you're
familiar with it probably someof you are but I always was a
(08:58):
fan of his.
He passed away in 2017 orsomething, but he had a lot of
writing and books and a lot ofgreat ideas, and so that
backwards design.
It always stuck with me throughthe years as a good way to
think about your curriculum oryour class is start with the end
(09:19):
goal in mind, and so that endgoal is like we talked about.
You know, students are going tograduate and go to work for an
organization or company and beexpected to use AI right away.
You know most of these jobs areusing AI.
Now, if you look at, microsoftdid a survey last year and
(09:41):
showed, you know, the majorityof workers I don't know 75, 80%
are using AI, and I'm sure it'sgone up in a year since.
You know I think the new reportwill come out soon this year so
they're entering thatenvironment where AI is part of
it, and so you've got to trainthem for that.
So that first step is looking atthe skills in demand and,
(10:05):
coincidentally, you can use AI,chat, gpt, whatever tool you
want to help you do that, and soif you're teaching like in my
case, economics or differentcourse, biology, chemistry,
whatever.
You can just go into chat GPTand say you know, chat GPT has
that deep research capability,depending on your plan or
whatever.
(10:25):
And you can say you know,survey current job ads and tell
me what are the main skills indemand, not just AI, but just
you know all the main skills andso you know you can do that
fairly quickly.
And what you'll find often isthat you know, yes, ai
(10:47):
technology is in there, but asAI has become able to do some of
these quantitative things,those other skills about you
know presenting, communicating,you know working with others on
teams, kind of those likeevergreen type skills have
become even more important.
And so you know, if you pull upa job ad, you know today, for
(11:09):
whatever this is a biologist,there'll be all the biology part
in there, but it's also a lotabout communicating.
You know presenting, usingpresenting using Microsoft tools
and things like that to present, and so on and so forth.
Jackie Pelegrin (11:21):
Right.
Dr. Matthew Metzgar (11:21):
So that's step one is figure out.
You know your graduates aregoing to finish and go into what
type of jobs.
You can also try to speculate abit and say, all right, well,
that's what they may need in2025.
But what about these studentsthat are freshmen or graduated
in four years?
And you can try to have alittle bit more of a future
(11:43):
focus.
That involves a little morerisk and speculation, because no
one knows for sure what the jobmarket will look like in four
years.
Jackie Pelegrin (11:54):
Yeah, it can be hard to predict the future job
market will look like in fouryears.
Dr. Matthew Metzgar (11:56):
Yeah, it's going to be hard to predict the
future, exactly, exactly.
But I mean again, I would say,still those at least core
elements about communication,teamwork, so on and so forth,
self-directed learning, aregoing to be there forever.
Jackie Pelegrin (12:09):
Right those key soft skills, right that they
need Definitely, definitely,yeah.
Dr. Matthew Metzgar (12:16):
And so that's step one um again.
So then you come up with the,the skills in demand, current
and future um, and now you'relooking at designing the tasks.
Right, that will, you know,mimic or be similar to those um.
And this is where it's reallychanged, because the class I
(12:36):
teach it's technically,historically, been a lot of like
quantitative work, you know, solike find this optimal price or
calculate this cost, and sostudents would do that.
Well, now I can do all thatvery quickly, and so a lot of
that quantitative work has beenkind of like outsourced, so to
speak.
Oh, wow, quantitative work hasbeen kind of like outsourced, so
to speak.
And so now you have to look at,okay, you become almost the
(13:07):
evaluator of the AI results.
You're making sure it's true.
So you could say, all right,calculate this elasticity, and
it gives you this number, but itmay not be right.
Ai can make mistakes.
So you don't have tonecessarily do the calculation,
but you have to know whetherit's correct or not and then, if
it's incorrect, obviously comeup with the right number.
Jackie Pelegrin (13:23):
Oh, so helping students to disseminate the
content that's coming out of AIand whether it's correct or not,
wow.
Dr. Matthew Metzgar (13:30):
Definitely and it's definitely not always
correct.
I mean, I've done some things,you know, like sample
calculations.
I'll put in there.
You know calculate this,calculate that, don't give me a
wrong number, and then you knowI'll write back.
You know this is not correct,and then it'll write back, you
know think and be like, oh yeah,you are right.
So hold on, let me recalculate.
Jackie Pelegrin (13:48):
Yes, I've done that.
Oh, I'm sorry, let me try thatagain.
It's almost like you're havinga conversation with the tool.
It's pretty interesting.
Dr. Matthew Metzgar (14:04):
Well, as you say, at least it's polite
right, Right?
Jackie Pelegrin (14:06):
That's true.
I wonder if we were rude to it,if it would be rude back.
I just wonder that sometimes,because I'm always polite Say
can you do this please?
And it'll have that tone, it'llmimic that tone.
You know that's where thatprompt engineering comes into
place really well with that.
So they say what you input willget what you get in the output.
Dr. Matthew Metzgar (14:29):
So right, Absolutely, absolutely and so,
yeah, so designing those tasks,the other key pieces I think
about is one I try to make it arealistic task as good as I can,
because I'm a little afraid.
You know you can go in AI andsay you know, generate this
scenario about you know, companyX, you know, sells widgets at
(14:52):
$2 a widget or something.
You know sells widgets at $2 awidget or something.
But I think if it's a fake orartificial problem, a lot of
students are then going to justrun it through chat GPT because
they're going to say, like Idon't care about this, it's not
real, it's just something he orsomeone made up.
Right, you know they're notgoing to be really motivated to
(15:13):
do it, and so I've found thatthe more realistic problem I can
make, the more likely studentsare to actually try it and not
just totally outsource it insome way, like I just gave a
task when the semester ended andI ran it through Chet GPT
myself to see what the answer,what it would give.
(15:34):
But I didn't have any studentgroups submit that answer and
that's because I think theythought they perceived that test
to be valuable, saying, hey,this might be something that I
would actually do and work atsome point, so it'd be a good
idea for me to try it now, youknow, because it was a real task
.
But if it was just fake, youknow, make these widgets or
something like that I thinkyou're task, but if it was just
(15:58):
fake make these widgets orsomething like that I think the
motivation is going to fall andthe odds of cheating, so to
speak, go up.
Jackie Pelegrin (16:03):
Wow, that's amazing.
Yeah, and you're taking itbeyond a textbook and putting it
in real life for them andsomething that they can imagine
themselves having to face right,a real problem or real
situation where they have tocome up with a solution in a
company.
Yeah, that's great.
Dr. Matthew Metzgar (16:22):
Yeah, absolutely.
I mean, you know it's like, ifyou want to, you know whatever,
learn how to ride a bike orsomething you know.
And if someone tells you justyou know, pedal on a stationary
bike, you know then you mightthat's not going to help me,
right?
You know I on a stationary bike, you know, then you might, uh,
that's not going to help me,right?
You know I wouldn't learn howto ride a bike out on the street
(16:42):
or on the mountains or whatever.
So you know you want thepractice to be as similar to the
real as you can get.
So you know I pull stuff, youknow, from the news.
You know, the week I'm makingthe assignment, you know.
So it might be something thathappened two days ago and I'll
pull it and make a task out ofit, and that way there is no
history that someone did this.
You know, someone solved thisproblem 10 years ago or
(17:04):
something you know.
It's brand new.
It's right out there now.
Jackie Pelegrin (17:07):
Wow, I love that.
That's great.
Dr. Matthew Metzgar (17:09):
So one with the task, I think, one about
making them realistic and havingthem in a good context, and
then two, giving them at leastthe option to work with others.
Because, again, if you look atwork situations, most work now
involves other people, whetherit's in person or virtual.
And so I, when the semesterstarts, the first task I assign
(17:34):
the groups of five.
And then you know, becausepeople may not know each other
and all that but then tasksafter that, then I open it up
and say you can kind of what ineconomics we call vote with your
feet, but you can move.
You can, you know, leave agroup and start a new group, or
you can go to a smaller group,or you can work by yourself or
(17:54):
whatever.
But at least give that initialchance for everyone to be in a
group.
Maybe it works, and so on andso forth.
And then you know, if peopleare unhappy I give them the
option to work in different ways.
But you know, now that I'msaying this, I should probably
calculate what percent of peoplestayed in the groups, but I'd
(18:15):
say it's pretty high, definitelyover half.
I'd probably say two-thirds ofpeople choose to stay with the
group, even though they don'thave to.
Obviously it can be easierbecause you can spread the
workload.
But two, I think you can end upwith a better project or
product output by working withothers than just yourself
(18:37):
usually, and so, again, that'sgiven a more of a realistic
scenario that if you're at workand you're trying to accomplish
X, it's probably going to payoff to get some people to help
you with that Right, becauseotherwise it's a long slog on
your own.
Jackie Pelegrin (18:55):
Yes and so.
Dr. Matthew Metzgar (18:56):
I'm trying to get them to think about that
as well, right, becauseotherwise it's a long slog on
your own.
Jackie Pelegrin (18:59):
Yes, so I'm trying to get them to think
about that as well, right, sothat they're not feeling like
they have to do all of this inisolation when they get to the
workplace, because so much isabout collaboration and it's
hard to accomplish anything bigon your own in the workplace,
right?
It's always about delegating or, like you said, bringing
together a team of people thatcan come up with a solution
(19:20):
together and work on that tasktogether, as well, absolutely,
absolutely, yeah, great.
Dr. Matthew Metzgar (19:27):
And then let's see.
The third step is supportingthe learner's development, and
that's where kind of the typicalteaching, so to speak, would
come into play, where you knowwhatever classroom setup you
have or an online setup throughan LMS and all that.
But I'd say the other piece,the biggest change to that with
(19:49):
AI, is that AI can help youdevelop all of this, all of
these resources now to helplearners on their journey.
If you have notes on a topicthat you yourself, the
instructor, wrote, you can haveAI turn that into a podcast.
There's tools out there to dothat.
(20:11):
Or you can have it turned intoa video.
There's digital twinning nowmaybe call it something
different, but it's where youonly have to record a couple
minutes of yourself on video andthen you can give it a Word
document and it'll now create avideo of you saying these things
(20:33):
that you've never said.
Jackie Pelegrin (20:35):
Oh, wow, yeah, it's called a tool to link out.
Dr. Matthew Metzgar (20:39):
there is a HeyGen H-E-Y-G-E-Ncom.
Jackie Pelegrin (20:43):
Oh, I've heard of that, I've heard of.
Heygen, I haven't used it yet,but I've heard of it.
Dr. Matthew Metzgar (20:47):
Yeah, it just creates a digital twin and
so you could have yourselftalking about whatever physics,
biology, you know something youdon't know nothing about.
That wouldn't be good, but I'mjust saying it's a.
It's a that type of tool, youknow.
There's all sorts of tools outthere like that I said.
The one other cool thing aboutdigital twinning is it can also
(21:09):
translate what you say into adifferent language.
Jackie Pelegrin (21:12):
Oh, I've heard about that.
Dr. Matthew Metzgar (21:14):
Yeah, and so if you have, you're teaching
a class, you know overseas orsomething and you don't speak
that language, you could, youknow, have it, create a video of
you saying your normal notes inthat language for those
learners, and so you can justcreate a whole bunch of
resources quickly that you, youknow, couldn't do before.
You know, with AI, you know youcan create podcasts, notes,
(21:36):
visuals, again videos, digitaltwins, all sorts of stuff out
there to help the learners getthe material.
Jackie Pelegrin (21:45):
Wow.
So that opens up institutionsthat are doing online education,
like Grand Canyon University,to be able to offer the courses
to students that are outside ofthe United States, too.
More often.
I've only had, I think, twostudents that have been outside
the United States becausethey're in the military or they
(22:06):
work on a military base orsomething like that, so it's
been rare, but it would be niceto be able to tap into more of
the international studentsabroad and get to learn from
them and their culture andthings like that and how they
perceive higher education wherethey are too.
So it would be neat for sure.
Dr. Matthew Metzgar (22:27):
Absolutely.
I think there's tremendouspotential for what you said.
That can really open up a lotof doors and make it accessible
to a lot more people thanpreviously.
Jackie Pelegrin (22:38):
Right.
Break down those barriers Right.
Dr. Matthew Metzgar (22:41):
Yeah, absolutely.
And then the other piece I'd saywith the supporting learners is
honestly, even if you give yourstudents all these different
resources, they may choose ontheir own to just go with a
conversational AI agent and Iknow a lot of my students do
that and I build in on thewebsite.
(23:01):
We have Google Gemini on campus,a co-pilot.
I kind of build that into pagesand with a textbook and all that
, just that some students aregoing to be more comfortable
learning about that topic.
So if you say I want to learnabout, you know, price
discrimination or something,well they can just put it in to
(23:22):
their agents, they tell me aboutit and go back and forth.
And so I think the interactivecontent or setups like that
obviously are going to beprobably more ups like that
obviously are going to beprobably more palatable to
students than a lot of the oneway stuff.
You know, I think these thedays of just sitting here
listening to this two hour videoon the instructor are probably
(23:47):
gone or near gone.
That you know students aregoing to be looking to get
information quicker and more ina back and forth situation with
the agents, and so obviouslythat's led to, you know,
instructors building chatbotsfor their class and maybe just
training those chatbots on theirmaterials, only putting it out
there.
But again, even if you don'tmake a chatbot for your class,
(24:10):
students can use any AI platformto get at that same information
.
So that's good to know,definitely, I would say one
thing on that.
That kind of a question comesup when, cause we talked about,
you know, faculty trying to newfaculty that are maybe hesitant
to do this and sometimes theymight feel like, well, I have to
(24:31):
make a chat bot for my classand I don't know how to do that.
I would say that, again, that'snot a requirement.
You know, if you aren't techsavvy and you don't want to do
it or whatever, I would say,just you know, skip it and let
them use existing chatbots thatare out there, because you can
create a chatbot to focus onlyon your.
(24:53):
You know the materials that youhave, that you've written or
whatever.
But in my mind that's a littlebit of what I'd call like in
economics, like creating amonopoly.
Like in other words I don't wantthe students to only be
dependent on just the way Ipersonally see things.
You know, I might explain itthis way, I might think of it
(25:17):
this way, but someone else mayexplain it this way, and in a
chatbot they're getting adifferent explanation.
And so, like for me, I did nottrain my chatbot on just my
stuff.
I just said I just made ageneral.
You know, it's based off acopilot and I would get
different type of answers thissemester than I would previously
in a good way semester than Iwould previously.
(25:38):
In a good way, because beforeit's like they were, I was
almost teaching them to thinklike me, and so the answers
would be answers like I wouldproduce.
But you know, you want a littlemore diversity of thought, I
think, and so you want diversityof sources, and so sometimes
they give me an answer thatreally surprised me and it's
like, oh, that's cool, it's adifferent way of thinking of it.
(26:00):
It's not wrong, it's justdifferent.
And so, you know, sometimes Iwould be like, oh, you know,
that's cool, you know, I'velearned something.
So I think it pays to have moresources than just your own to
the students Because, again,each person's limited.
You can only know so much, andit's better for them to get a
diversity of opinions onsomething than just what the
(26:22):
instructor says.
Jackie Pelegrin (26:24):
Oh, I love that .
That's a great way to look atit too when it comes to these
types of tools, and that it'syeah, you don't want to shoehorn
them into one thing, but youwant them to have that diverse
way of thinking.
I like that.
That's great.
Dr. Matthew Metzgar (26:39):
Yeah, yeah.
So definitely a factor.
And let's see the next stepprovide efficient feedback.
And this is something where alot of faculty are going to
start to worry.
Yes, because if you move away ina simplistic term, if you're
moving away from like multiplechoice questions or like numeric
(26:59):
response questions to thesekind of tasks I'm calling them,
could be many projects, whateveryou want to label them
open-ended type assignments.
Now I was in a conference andsomeone raised their hand and
said this well, how do I gradeall this?
You know, I can't give mymultiple choice exams because
they can use ChatGPT and get,you know, 95%.
(27:20):
I've got to give them all thesedifferent types of assignments.
How do I grade it all?
Well, if you have, you know, 15or 20 students, you can grade
it.
But if you have big numbers,then it gets to be more
problematic, and so if aninstructor is making this shift,
they have to really thinkbefore the course starts shift.
(27:42):
They have to really thinkbefore the course starts.
You know, how much feedback canI realistically offer?
Because you can offer feedbackon like drafts of, you know, the
assignments and then, of course, they have to be graded.
So there has to be feedbackthere.
But you have to think aboutthere's only so many hours in
the day and you know there's anopportunity cost to grading, to
giving feedback.
It takes your time as aninstructor and so there's no
(28:05):
right or wrong answer.
But it's just something youhave to be aware of because
otherwise you can get suckedinto just continuous drafts and
emails and everything else andyou may not be able to get
anything else done.
So what I found out?
I teach big classes.
This last semester my class was230 students.
Jackie Pelegrin (28:23):
Oh wow, that's a big class.
Dr. Matthew Metzgar (28:25):
Very big and so you know, first I get
them into the groups, like wetalked about, and so that cuts
it down from 230 to maybe.
I think maybe we'd end up withmaybe 80 or 90 assignments after
the group's resettled for thoseother tasks, but still that's a
lot.
So the one thing, a couple ofthings I learned was one to keep
(28:50):
the assignments shorter, and sowhen I assign these tasks
there's kind of steps to it, butI say the complete thing.
You submit a PDF, you know it.
Either we cap it at either oneor two pages, because if you
don't, someone's just going togo on there and make a 20-page
chat GPT output and give it toyou.
(29:11):
You're going to be stuck lookingat it when you know it's not
even real.
The students need to do itRight.
And also, like I was sayingearlier, think about that
student motivation.
If you tell them you have tosubmit this 20 page paper, their
motivation is going to be lowand more likely to use ChatGD2.
If it's one or two pages,they're going to.
You know again they're going tothink, okay, well, that's not
(29:32):
too bad, I can do it, you know,or at least you know I'm willing
to try it, rather than a 20page paper.
I don't want to write I'm noteven going to try that.
I'm willing to try it ratherthan a 20 page paper.
I don't want to write, I'm noteven going to try that.
Jackie Pelegrin (29:41):
Oh, that's so important.
Yes, especially with curriculumdesign, being able to keep that
in mind with their motivation.
Absolutely, definitely.
Dr. Matthew Metzgar (29:50):
Definitely, and the other thing, too, I'd
say is that it's not beyond that.
It's also realistic in thesense that a lot of times in a
work environment you may producea long paper or whatever, but
most of the time it might besomething like an executive
summary or the PowerPoint notesthat actually most people, 90%
(30:12):
of people, are going to look at,and so that's really what I'm
trying to get the students toproduce is something, you know,
a smaller document or somethinglike that, that would be similar
to what they present in theworkplace.
Jackie Pelegrin (30:29):
Right.
Dr. Matthew Metzgar (30:30):
And so, like I think to myself, like you
know where I work at theuniversity, no one's ever sent,
you know, a 50-page paper foreveryone to read, because, one
they know no one's going to readit, but two, it's just not
effective, right?
So instead you get memos andnotes or PowerPoints and things
like that.
So, again, that's what I'mtrying to think about them in
(30:50):
the workforce, you mayoccasionally have to write a
long white paper, but most ofyour communication is going to
be shorter for general audience,in something like, again, one
to two page in this situation.
Oh, that's great.
And so the other piece with thefeedback then comes into, since
it's an open-ended task, youknow grading it with some type
(31:13):
of rubric, and we started withthat, you know, in January.
You know you can use AI tocreate rubrics easily.
You know just a couple seconds.
But what we found with thatfirst task was it was still
getting too long of a process tograde myself and the teaching
(31:43):
assistants on the actual grade.
A paper should get you knowbecause, again, some of those
rubrics do allow forsubjectivity, so to speak.
You know, is this a three or afour in this category?
And so what we then went to forthe other.
For the rest of the task, therest of the semester was much
simpler rubrics where you knowyou might have some items, but
they were all essentially likebinary, like yes, no, type items
(32:07):
, and so you know we'd just belike you know, was this piece
calculated correctly?
You know it's yes or no, was agraph of a demand curve
presented, yes or no?
And we found that one that spedup the grading.
But two I think it reallyhelped the students.
(32:28):
It was more structured andallowed them to kind of see what
the end product should looklike.
And some of the first, when wedid the very first thing, the
rubric, we had all sorts ofdifferent formats and looks and
everything like that.
But as we simplified the rubricand cleaned it up, we were able
(32:52):
to get a lot more good-lookingpapers out of it, and so it's
kind of just like puttingconstraints on it is the way I
would think of it.
Jackie Pelegrin (33:02):
That's a good example I love that how you can
really utilize that AI to getthose outputs that you're
looking for.
Dr. Matthew Metzgar (33:11):
Oh yeah, yeah, I mean, you know it can
create rubrics in a snap.
And then, of course, the otherpiece is you know we messed
around with a little bit, wedidn't go all out, but you know
you can also use AI to give youa base grade on a paper, you
know.
And again, of course, you havedata collection issues and
privacy and so on and so forth.
Right, but you know there'ssolutions out there, like Timely
(33:35):
Grader, where it can, you know,run through the essays.
I think they're trying toconnect to Canvas and so
there'll be other companies soonthat'll do that.
But it's like the AI will dolike a first pass grading and
then you, as the human, can comein and agree with that or not
change it, make your commentsand then go through.
(33:55):
But to me, that's definitelygoing to be the future in some
way is that AI will do a firstpass grade and then you have the
human judgment that kind ofconfirms it and gives the
comments and, you know, providesthat value to the students.
Jackie Pelegrin (34:13):
Right, I love that.
So it's giving that firstanalysis of the paper or the
assignment for you as aninstructor.
I like that.
Dr. Matthew Metzgar (34:23):
Yeah, because a lot of times it can
save time.
You know if this assignment'syou know way off or something
like that, you know it mighttake you a while to figure out.
You know what's even going on.
But you know AI can kind ofquickly look at this and say, oh
, this thing is, you know, maybethe student didn't understand
the assignment or something likethat.
It's way off, and so that firstpass can definitely help.
(34:45):
Again, it's not perfect.
We were, because when we'd runsome papers through AI to get
what it would say for the rubric, again it would never exactly
line up with the way we werethinking.
You know as people, and so it'sa good start, but again, you
need that human element in there.
Jackie Pelegrin (35:08):
Yes, absolutely , that's so true.
I can see how that would be aconcern for faculty.
Wondering, you know, if I'mputting this in here, you know,
is it taking away some of what Ido?
But it sounds to me like whatyou've discovered is that it
enhances the experience as aneducator and also for the
students as well.
Dr. Matthew Metzgar (35:29):
Yeah, I mean to me I think also we talk
about in the book is AI has kindof shifted the role of the
educator, whereas before it wasmore about the educator was the
expert in the educator, whereasbefore it was more about, you
know, the educator was theexpert in the content.
And I'm going to, you know,talk about the content and
explain the content and so onand so forth.
(35:49):
And then maybe the grading wasdone again via multiple choice
questions and scan trot sheetsand all that.
I think it's really shifting it.
The other, almost having themswap places, where a lot of the
content and the delivery isgoing to be done online or with
AI, and so on and so forth.
And the job of the facultymoves more to an evaluate.
(36:12):
Okay, you know, because if it'sa realistic piece, you need a
person to tell you is this goodor bad?
You can't just have you knowit's not going to be enough,
right?
So I think it really isflipping the role to be more of
an evaluator.
Actually, I was interviewing anart professor about this and she
(36:35):
was saying you know, there's somuch content now for art school
students.
You can go on YouTube and finda video on anything you know.
Here's how to do paint orlandscape or sculpt or whatever
right, it's unlimited amount ofcontent on how to.
But she was saying you know,the reason then, to go to art
(36:56):
school is to have professorsthat you know, have experience
and know what they're doing,evaluate your work and tell you
is it good, bad, what can beimproved, and so on and so forth
.
She was saying that's our mainfunction.
Now, it's not really the how-to, because it's a universe of
how-to videos.
It's more of just okay, youwatch the video and you produce
(37:18):
this piece.
Now, is it good?
Is it going to meet the needsof the company, the consumers,
whoever?
If not, how can it be improvedto do that?
Jackie Pelegrin (37:28):
So definitely a shift in my view.
Yeah, absolutely.
Dr. Matthew Metzgar (37:34):
And the last piece of the cycle is just
iterate and improve.
And you know the tools arechanging so rapidly with AI and
all that.
That.
You know each semester there'sgoing to be a new opportunity
for learning and, you know,putting in new things, seeing
what worked with students, whatdidn't, so just kind of a normal
(37:54):
feedback process.
But the only thing I'd say withthat that like as far as again I
don't want to scare offinstructors from trying this out
is that you can have a bit oflike I'll call it tool fatigue
out there because there's somany new tools, so many things
(38:16):
coming out, so many changes, butyou don't have to use all these
tools, you don't have to doeverything.
You might just say, okay, thissemester I'm going to add in one
new AI tool or try one new AIthing I've never done before.
You don't have to try 10 things, or you don't even have to try
anything new, but at some pointyou're probably going to need to
(38:38):
update.
So it can be done in a in smallways.
And so when we're sayingiterate, that doesn't mean every
semester I've got to massivelychange this course and do all
these things different.
You know, once you've got tomake a shift and get AI in these
pieces.
The next semester should berelatively easy to kind of just,
you know, clean up and keep upwith the current changes, and
(39:01):
then, of course, the feedbackfrom the students.
Jackie Pelegrin (39:04):
Right, and then that way you can do that
iteration from each semesterright and see what needs to
improve, what works well, what,what doesn't, and what needs to
shift as well, from each Exactlyand that's what you know.
Dr. Matthew Metzgar (39:18):
I feel like this cycle is something you can
use.
You know before each term orcourse or whatever in a fairly
quick way.
You know because again you cango, you can run and now and say,
all right, spend six months,run some job ads and see is
there anything new or the skillsand this more or less, and so
on and so forth, so you can dothat.
Then you can kind of go throughyour task and see all right,
(39:41):
are they matching with what'scurrently in demand?
Do I need to update some of thetasks?
Are my resources?
You can check the stats and seeare students using like these
things more than this?
Do they like the podcast andnot the videos or vice versa?
And then you can think aboutthe feedback how much time did
(40:01):
it take?
Do I need to be more or less ormore efficient with it, or can
I give more feedback and thenagain think about small things
you can improve.
But ideally, again, I'm tryingto make an easily repeatable
process that someone could do ona semi-regular basis and that
keeps them up to date and withthe times and producing good
(40:25):
results for their students.
Jackie Pelegrin (40:27):
Right and I love how it's five steps, so
it's not complicated, and Ithink that's great, especially
for educators and, like you said, no matter if they've been
teaching for 15, 20 years or ifthey're just starting out.
Dr. Matthew Metzgar (40:47):
I think it's a really great process for
them to be able to use at anystage, and their teaching as
well.
Yeah, exactly that's what youknow.
When I started writing the book, it's like you know, we
initially had, you know, justdifferent chapters and chapter
on this, chapter on that, and Iwas just like you know, the
business instructor in me saidyou know, this has to be more
structured.
This has to be, you know, stepone, two, three, four, five, you
(41:08):
know, to really get someone tofollow through.
In other words, you know, a lotof times you might just read a
book and then put it down and itdoesn't do anything right.
But I'm trying to make a moreactionable book that's saying
okay, just here's this process,go through it and then you can
get you know again up to dateand get some good results.
Jackie Pelegrin (41:29):
That's great, Wonderful.
So a final question as we wrapup.
So what's one practical pieceof advice that you'd give to an
educator who's curious about AIbut not sure where to start?
Dr. Matthew Metzgar (41:45):
Well, I guess it would depend.
You know how often they'reusing AI themselves.
You know, like some people areheavy users of chat GPT, use it
every day or something like that.
I mean, I use it for a lot ofdifferent tasks.
So I would say, if they haven'tused any AI, to just start
(42:09):
going on there and trying to useit for personal use, seeing
what kind of value you can getfrom it.
You know, ask questions and soon and so forth.
I think probably most peoplehave, at this point, at least
experimented with it somewhat.
You know the other thing, asmall step, could just be even
asking their students.
(42:31):
You know how do you use ChatGPT, asking their students you know
how do you use ChatGPT.
And so trying to again get itmore out in the open that, okay,
this is a tool that probably75% of the population uses, at
(42:51):
least 80% of students use, andjust making it more kind of like
normalizing it.
Right, you know, okay, this isa tool, we know you use it, I
use it, I use it, everyone usesit.
And now start the conversationabout that.
Well, how can we maybeintegrate it in this class or
what can we do with it?
But I'd say that's.
This first thing is just maybesome awareness, some
transparency about it, justsaying, you know, hey, this is a
(43:12):
very powerful tool, right?
Right, that can help you asstudents, it can help me as a
teacher, it can help thebusinesses you work for, and so
let's just start talking aboutit, you know.
Let's see what we can do withit.
Jackie Pelegrin (43:27):
Yes, I love that.
Yeah, and keeping an open mindabout it and seeing where it'll
take us Absolutely, I love that.
Well, thank you so much forgoing over the steps.
That is very insightful andyour advice, because I think it
makes it more manageable andsomething that instructors can.
They feel like they can tacklethis and be able to integrate it
(43:51):
into their curriculum and workwith their students and not be
fearful of it Like so manyprobably still are.
So at least they can take that,some of those small steps and
do some things where they don't,like you said, they don't feel
like they have to learn everytool, but pick up the tools that
they think are going to beuseful and try it out.
And that's a great thing is wecan experiment and see what
(44:14):
works and what we need toimprove upon.
So that's a great thing aboutAI as well.
Dr. Matthew Metzgar (44:20):
It is and you know it's just if you look
historically, you know thingsthat seemed kind of odd at first
become, you know, very, verycommonplace.
I was talking with someone andI thought, you know, like the
idea of Uber and Lyft to thinkthat I'm just going to press
this button on the phone and astranger comes and picks me up
and takes me somewhere, you know, if you would have told someone
that before it happened, you'dthink that's crazy, that's not
(44:43):
stuck in real.
But of course it became real.
And so you know, ai is going tochange a lot of norms and you,
just like you said, you have tobe open-minded to it and be
willing to explore it and seewhere it can take you.
Jackie Pelegrin (44:55):
Yes, absolutely .
I remember as a kid watchingthe Jetsons and I thought that's
so futuristic.
We're never going to get tothat, but now, who knows, we
could have flying cars in thenext few years.
So I think AI is going toreally make a lot of things
possible and provide medicalbreakthroughs and all these
different things that they'retalking about.
(45:16):
So it's a pretty amazingadventure to see that where it's
going to take us.
So that's great.
Dr. Matthew Metzgar (45:23):
Absolutely.
Jackie Pelegrin (45:23):
Great, great Well, thank you so much for your
time, and I am so happy that mystudents are going to be able,
and others even around the world.
This podcast does have a globalreach, so I'm really glad that
those around the world will beable to get a lot of great
advice from what you'vementioned today.
So thank you so much for yourtime.
I appreciate it.
Dr. Matthew Metzgar (45:42):
Definitely, definitely, and I list some of
the current tools on my website,overnighteducation.
I try to do kind of a roundupof the current tools but, as we
talked about, things arechanging so quick that I'll
probably do, on a somewhatregular basis, try to update
what tools are out there andgive people some things to work
with that can give them value.
Jackie Pelegrin (46:06):
Great, that's wonderful.
I'll make sure to link yourwebsite in the show notes so
that way all of my listeners cango there and check that out.
And then I did notice on yourwebsite that you have
information on your book, soonce it's out, then everybody
can check it out and order acopy.
Dr. Matthew Metzgar (46:24):
Yeah, excellent, definitely.
Jackie Pelegrin (46:26):
Great.
Thank you Appreciate it.
Thank you for taking some timeto listen to this podcast
episode today.
Your support means the world tome.
If you'd like to help keep thepodcast going, you can share it
with a friend or colleague,leave a heartfelt review or
offer a monetary contribution.
Every act of support, big orsmall, makes a difference and
(46:48):
I'm truly thankful for you.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com