Popular in Professional

Culture and Society

The City as a Growth Machine: Toward a Political Economy of Place

New York University Harvey Moltoch

Education

The Role of Education in Reducing Racial Inequality

University of Southern California Pedro A. Noguera

Law

Diversity and Social Mobility in the UK legal profession

Law Society I. Stephanie Boyce

Internet Law

The Law of AI for Good

University of San Diego Orly Lobel

Pharmacy

Cytotoxic Activity and Anti-Migratory Effect of Berberine-Phytantriol Liquid Crystalline on Non-Small-Cell Lung Cancer

University of Technology Sydney Kamal Dua

Human Rights Law

Transforming Restorative Justice for Transitional Settings

University of Nottingham Kerry Clamp

Education

Teaching Writing in a Digital and Global Age

View more insights in Professional - Education - Literacy, Language and Communication

View more figures from University of California Berkeley

View the page of Sarah Freedman

Read the Study

Sarah Freedman discusses theories and research studies that shed light on writing in a digital and global age

Image courtesy of interviewee. May 9, 2024

Generating Workspace

Export as PDF

The transcript discusses the speaker's extensive experience in studying the teaching and learning of writing, particularly in the context of access and opportunity disparities among students. Since collaborating on a literature review for the American Educational Research Association published in 2016, the speaker notes significant changes in the landscape of writing education, especially with the advent of artificial intelligence (AI) tools like ChatGPT and Gemini.

Key points include:

1. **Inequality in Access**: The speaker emphasizes the unequal distribution of opportunities for young people to engage creatively and practically in writing. Access to educational resources and digital tools is uneven, both globally—where poorer countries lag behind wealthier ones—and locally, within wealthier nations, where socioeconomic status influences educational outcomes.

2. **Literacy Rates**: While global literacy rates have improved overall, discrepancies remain, particularly in low-income countries, with literacy rates lower than in wealthier nations. Gender disparities also exist, though they are narrowing.

3. **The Role of Policy and Teacher Support**: In affluent countries, disparities in writing education are influenced by social policies, funding, and teacher training. The quality of teaching and the reading/writing opportunities students receive are often determined by socioeconomic factors. The speaker highlights the need for policy reforms to better support teachers and ensure equitable learning conditions.

4. **The Social Nature of Writing**: Drawing from theorists like Vygotsky, the speaker discusses the importance of social interaction in writing development, arguing that writing is not a solitary activity but one enriched through feedback and collaboration with diverse voices.

5. **Embracing Generative AI**: The speaker calls for a nuanced understanding of generative AI's role in writing education, arguing that rather than fearing these tools, educators should explore how they can enhance writing skills. Effective use of AI requires sophisticated engagement and should be integrated into learning without assuming students will misuse it.

6. **Future Directions**: The speaker advocates for research into how generative AI can support writing instruction, emphasizing the balance between AI's quick feedback and human educators’ nuanced understanding of the writing process.

In conclusion, the lecture highlights the complexity of teaching writing in an evolving educational landscape, where equity, access, and the integration of new technologies play critical roles in shaping students' writing experiences.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.245 --> 00:01.815
I've been studying the teaching

00:01.915 --> 00:05.215
and learning of writing since the early 1970s,

00:05.395 --> 00:08.735
and I've always examined issues

00:08.875 --> 00:10.695
of opportunity for students.

00:11.275 --> 00:15.615
Um, in a few years ago,

00:15.895 --> 00:20.335
I was invited to write a chapter, a review chapter for

00:20.435 --> 00:23.535
by the American Educational Research Association

00:24.355 --> 00:27.935
for their upcoming handbook of Research on teaching.

00:28.555 --> 00:31.775
Um, that article was published in 2016,

00:31.995 --> 00:35.495
and I worked with my colleague at Berkeley,

00:35.505 --> 00:38.015
who's also an expert in writing Glenda Hall

00:38.675 --> 00:43.055
and two of our then students to write a comprehensive

00:43.735 --> 00:45.615
literature review of the teaching

00:45.615 --> 00:49.295
and learning of writing in this digital and global age.

00:50.395 --> 00:54.375
Um, I'm gonna focus today on what I've learned about

00:55.175 --> 00:57.655
teaching writing, both from my research

00:57.955 --> 01:01.175
and what my colleagues and I learned in writing that review.

01:02.585 --> 01:06.295
Since, since writing that review though, I wanna say that,

01:06.995 --> 01:08.895
much has changed.

01:09.435 --> 01:13.175
And so I'm going to also discuss,

01:14.565 --> 01:16.415
this current moment when

01:17.455 --> 01:19.735
artificial intelligence powered writing

01:19.905 --> 01:21.495
assistance like chat.

01:21.775 --> 01:26.575
GPT and Gemini have literally introduced revolutionary

01:26.875 --> 01:29.895
and publicly accessible new tools for writing,

01:30.265 --> 01:33.535
which have the potential to profoundly changed the nature

01:33.555 --> 01:38.295
of the writing process for those with access to those tools

01:40.085 --> 01:41.935
from all these sources,

01:41.935 --> 01:45.775
including the new generative AI affordances, I wanna begin

01:45.875 --> 01:48.135
by stressing that it is consequential

01:48.445 --> 01:52.175
that young people do not have equal opportunities

01:52.955 --> 01:55.775
to engage expressively creatively

01:56.475 --> 01:58.935
and practically in the practice of writing.

01:59.845 --> 02:03.415
When I study whether young people have equal opportunities,

02:04.695 --> 02:07.295
I question whether they have equal access

02:07.515 --> 02:08.935
to learning opportunities

02:09.675 --> 02:13.015
and whether they have equal access to tools

02:14.355 --> 02:17.325
like computers, earlier on

02:17.825 --> 02:21.205
and, word processing programs.

02:21.385 --> 02:23.925
And now these new AI programs,

02:25.105 --> 02:29.045
access is not equally distributed, either in the form of

02:29.605 --> 02:32.485
learning activities or an access to digital tools.

02:32.665 --> 02:34.165
And this is true across the

02:34.165 --> 02:35.685
curriculum, not just for writing.

02:36.745 --> 02:39.685
So where are the inequalities that are consequential

02:39.705 --> 02:41.965
for learning to write first?

02:42.275 --> 02:44.445
I've done a lot of international research,

02:44.705 --> 02:48.325
and globally there are huge differences between access

02:48.545 --> 02:52.405
to even basic literacy across countries with those

02:52.465 --> 02:54.885
who live in, who live in poorer countries,

02:54.885 --> 02:58.005
having less access to robust teaching and learning,

02:58.265 --> 03:01.045
and to tools than those living in wealthy countries.

03:01.825 --> 03:06.325
The result is that even the most basic adult literacy rates

03:06.825 --> 03:09.525
are lower in poor countries than in rich countries.

03:10.105 --> 03:14.605
Um, the overall literacy rate globally has been rising,

03:14.785 --> 03:19.045
and the statistics show that in 1976,

03:19.275 --> 03:23.445
only 66% of adults in the world,

03:24.195 --> 03:26.405
were had basic literacy skills.

03:26.745 --> 03:30.565
And by 20 22, 80 7%

03:30.665 --> 03:33.085
of adults in the world are literate.

03:34.515 --> 03:37.695
But the rates for low income countries are shocking.

03:38.355 --> 03:40.095
Um, in 2022,

03:40.675 --> 03:44.735
the literacy rates in those countries were at 62%,

03:45.065 --> 03:49.775
which is even lower than the global rate in 1976.

03:50.635 --> 03:53.535
Um, these countries are mostly in Sub-Saharan Africa

03:53.715 --> 03:55.135
and some in Asia.

03:56.115 --> 03:59.055
Um, globally, there are also gaps between male

03:59.155 --> 04:01.695
and female literacy rates,

04:01.845 --> 04:05.135
with men generally having more access to learning

04:05.155 --> 04:06.495
to read and write than women.

04:07.475 --> 04:12.295
The good news is that the gender gap has decreased from 1976

04:12.475 --> 04:16.655
to 2022, but the bad news is that it's still there.

04:17.755 --> 04:20.455
Um, these statistics refer only to,

04:21.055 --> 04:24.735
I don't wanna stress only to very basic kinds of literacy,

04:24.915 --> 04:27.815
not to the kinds of advanced uses of reading

04:27.835 --> 04:32.015
and writing that, help people participate

04:32.555 --> 04:35.255
in the global e economic world

04:35.635 --> 04:39.735
or in, in academics.

04:40.675 --> 04:44.615
Um, there is not much information about these higher rates

04:44.755 --> 04:46.215
of literacy globally.

04:47.195 --> 04:49.775
Now, locally, within the wealthy countries

04:50.145 --> 04:51.815
where there's more access.

04:53.395 --> 04:55.695
My own research and the research of others show

04:55.695 --> 04:58.015
that opportunities to learn to write

04:58.595 --> 05:02.055
are also not equally distributed within those countries.

05:02.355 --> 05:05.815
And so where are the inequalities within the rich countries

05:05.815 --> 05:09.055
that have, have,

05:10.035 --> 05:13.175
better literacy, better access to literacy,

05:13.875 --> 05:15.175
not unexpectedly.

05:15.175 --> 05:17.255
Those with lower socio come

05:17.315 --> 05:20.495
who come from lower socioeconomic status groups

05:21.125 --> 05:24.095
have poor learning opportunities than those

05:24.095 --> 05:26.135
with higher socioeconomic status.

05:27.195 --> 05:29.815
Um, their schools expect less

05:29.955 --> 05:32.695
and their teachers expect less,

05:33.075 --> 05:35.495
and their teachers have less training in the teaching

05:35.495 --> 05:38.855
of writing in the us there are also differences across

05:38.995 --> 05:41.695
ethnic and racial groups, which are associated

05:41.695 --> 05:44.255
with socioeconomic status with whites

05:44.255 --> 05:46.615
and Asian Americans having more opportunity

05:46.635 --> 05:49.935
to reach high levels of literacy than those from,

05:49.985 --> 05:51.095
other ethnic groups.

05:52.235 --> 05:53.935
And there's evidence

05:53.955 --> 05:56.095
for these differences on standardized tests

05:56.435 --> 05:59.345
and on, kind

05:59.345 --> 06:02.265
of national exams in the us like the national Assessment

06:02.325 --> 06:03.705
of educational progress.

06:04.615 --> 06:07.065
It's also the case that those in wealthier countries

06:07.205 --> 06:10.825
and the more wealthy within all countries have more access

06:10.925 --> 06:12.945
to literacy learning and to digital tools

06:12.975 --> 06:14.825
that support writing development.

06:16.085 --> 06:17.385
Um, computers

06:17.405 --> 06:21.385
and access to the internet is also unequally distributed in

06:21.525 --> 06:26.185
in the same ways as, as, opportunities to learn.

06:26.845 --> 06:30.425
So for poorer countries where most citizens live in poverty,

06:31.055 --> 06:34.945
they need better access to schooling for their populations

06:35.325 --> 06:37.465
and attention to gaps in wealth.

06:38.125 --> 06:40.305
And this seems to be an ongoing

06:40.705 --> 06:41.945
struggle for these countries.

06:42.405 --> 06:45.945
And only then will even basic literacy rates

06:46.135 --> 06:47.825
have a chance of rising.

06:48.085 --> 06:50.585
So the big focus in those countries is just

06:50.585 --> 06:52.025
getting kids to school.

06:53.125 --> 06:57.745
Um, and oftentimes children are used as labor,

06:58.445 --> 07:00.225
and are needed at home.

07:00.565 --> 07:02.625
And so there's a real push and pull.

07:02.655 --> 07:06.105
It's a very diff very, very difficult and, problem.

07:06.885 --> 07:10.435
Um, for wealthy countries, the key

07:10.435 --> 07:12.315
to closing the opportunity gaps,

07:13.745 --> 07:16.675
wealthy countries think a little differently,

07:16.705 --> 07:19.675
because there's mandatory schooling,

07:19.935 --> 07:22.115
and kids do go to school,

07:22.735 --> 07:25.235
but then the question becomes what kind of schools

07:25.575 --> 07:30.475
and much of closing the opportunity gaps in,

07:30.735 --> 07:33.395
in these in wealthier countries lies in the hands

07:33.495 --> 07:37.475
of social policy makers who need funding formulas

07:37.475 --> 07:41.315
that don't, don't leave the underserved students behind.

07:42.255 --> 07:45.075
For example, there, there are not enough policies in place

07:45.075 --> 07:47.595
to support these schools generally,

07:47.775 --> 07:51.475
or teachers of writing in particular in the us for example.

07:52.175 --> 07:55.955
Um, continuing education programs for teachers

07:55.955 --> 07:57.755
of writing are losing funding.

07:58.615 --> 08:01.595
Um, there needs to be a better incentive structure

08:01.695 --> 08:05.515
to get good teachers into these schools in,

08:05.615 --> 08:09.195
in neighborhoods serving children who live in poverty.

08:10.135 --> 08:12.515
Um, and there needs to be motivation

08:12.655 --> 08:14.195
for outstanding teachers

08:14.295 --> 08:16.515
to teach in these, in these settings.

08:17.255 --> 08:20.235
Um, cross national studies show that countries

08:20.235 --> 08:22.435
that see teachers as professionals

08:22.575 --> 08:27.475
and support them, well produce more highly literate young

08:27.695 --> 08:32.515
people without the kinds of equity gaps that, that we see,

08:32.735 --> 08:34.315
in countries

08:34.315 --> 08:36.915
that don't provide sufficient support for teachers.

08:37.615 --> 08:41.275
Um, so for example, in Finland,

08:41.665 --> 08:46.605
there's a big stress on, on

08:47.235 --> 08:48.445
educating teachers

08:48.665 --> 08:51.685
and teachers being professionals, paying them well.

08:52.305 --> 08:56.485
Um, and so the

08:57.125 --> 09:00.525
literacy rates are, they're the highest in the world.

09:01.025 --> 09:03.565
So there, there's a kind of correlation there.

09:04.345 --> 09:07.405
Um, the other thing about teaching, writing

09:07.505 --> 09:09.445
and closing the opportunities gaps is

09:09.445 --> 09:11.645
that teaching writing is very labor intensive,

09:11.905 --> 09:16.005
and it requires teachers who first can write well themselves

09:17.025 --> 09:20.765
and who are able to analyze student work in ways

09:21.275 --> 09:24.685
that help them respond productively to student writing.

09:25.905 --> 09:30.165
And so teachers need to be able to read lots of papers,

09:30.345 --> 09:33.005
and that is, that takes a lot of time,

09:33.505 --> 09:35.485
and they need to be able to talk to their,

09:35.775 --> 09:37.045
their, pupils.

09:37.105 --> 09:38.645
And that takes a lot of time.

09:39.385 --> 09:43.285
Um, and they need to be able to engage their students and,

09:43.425 --> 09:46.845
and give them useful response that they can carry forward,

09:47.665 --> 09:49.285
with lowers.

09:49.545 --> 09:52.485
And the, and the teaching loads often are too high

09:52.585 --> 09:55.365
for teachers to be able to give the kind of attention that,

09:55.515 --> 09:56.805
that the students need.

09:56.905 --> 09:58.525
So the National Council of Teachers

09:58.525 --> 10:02.885
of English in the United States for, since for many years,

10:02.895 --> 10:06.805
since the 1960s, I think has, has had a policy statement

10:06.805 --> 10:08.405
that says that teachers

10:08.465 --> 10:11.165
of English should have no more than a hundred students

10:11.945 --> 10:15.045
in four classes of 25 students each.

10:15.225 --> 10:16.765
And that should be the highest.

10:17.675 --> 10:18.725
That should be the highest,

10:19.105 --> 10:22.605
and that is just not followed at all.

10:22.945 --> 10:27.685
Um, so, it's hard

10:27.705 --> 10:30.245
for teachers to get to know their individual students

10:30.945 --> 10:33.525
in countries like in, in a number

10:33.525 --> 10:36.005
of European countries in England as well.

10:36.035 --> 10:38.165
I've done a cross-cultural study of

10:38.685 --> 10:40.525
teaching writing in England and the United States

10:41.245 --> 10:44.285
teachers keep the same students for multiple years,

10:44.665 --> 10:46.925
and they are able to get to know them that way.

10:47.385 --> 10:50.125
In the United States, teachers may keep students

10:50.345 --> 10:51.485
for a semester.

10:52.025 --> 10:56.005
Um, and so it's really, it's really hard to do the kind

10:56.005 --> 10:58.405
of teaching that that needs to happen.

10:59.385 --> 11:02.405
And I mean, this is true for all students.

11:02.905 --> 11:06.965
Um, but if you can't, if you don't have way,

11:07.185 --> 11:10.805
if you don't have, if policies don't give you structures in

11:10.805 --> 11:14.325
which you can actually teach writing, well then it's hard

11:14.325 --> 11:15.885
to close the opportunity gaps

11:15.885 --> 11:18.365
because the wealthier students get help elsewhere.

11:18.985 --> 11:23.565
Um, and the poor students don't, don't have ways to do that.

11:24.305 --> 11:27.235
Um, so the task is huge,

11:27.495 --> 11:31.475
and closing achievement gaps really requires well-trained

11:31.475 --> 11:34.355
and dedicated teachers who have support to do their work.

11:34.705 --> 11:38.955
Writing development happens across people's lifespans

11:40.135 --> 11:41.755
and in,

11:41.975 --> 11:45.115
and the context outside of schools matter to

11:45.115 --> 11:47.195
that development as well as the context

11:47.825 --> 11:50.195
that are available inside schools.

11:51.015 --> 11:54.365
And to understand the process, I rely on two theorists,

11:54.905 --> 11:59.405
Lev Vygotsky and Mikhail Bach team who connect the larger

12:00.005 --> 12:01.205
societal issues, some

12:01.345 --> 12:05.765
of which I've been talking about already, to the everyday

12:06.285 --> 12:08.405
practices that support writing development.

12:09.505 --> 12:11.725
My main takeaway from the writings

12:11.725 --> 12:16.125
of Vygotsky is his definition of development as opposed to

12:16.645 --> 12:20.565
learning and his understanding of how social interaction

12:21.085 --> 12:22.445
stimulates that development.

12:22.985 --> 12:25.565
So the key words here are development

12:26.065 --> 12:27.525
and social interaction.

12:28.785 --> 12:31.965
So too often we think in units

12:32.075 --> 12:33.805
that are too small.

12:34.585 --> 12:39.205
So for Vygotsky learning, learning is the small bits

12:39.205 --> 12:40.605
that we, that we get.

12:40.865 --> 12:44.765
Um, and development is the more profound changes,

12:45.985 --> 12:48.525
and we need bits of learning

12:48.585 --> 12:51.165
to feed into development, but development.

12:51.625 --> 12:55.325
Um, and too often we focus on the small bits like learning

12:55.345 --> 12:59.245
to punctuate or, form complete written sentences.

12:59.745 --> 13:02.005
Um, and we need small units of learning.

13:02.105 --> 13:05.045
But our goal is to see development,

13:05.045 --> 13:08.525
to see young people make these big strides in writing.

13:08.705 --> 13:12.125
And the big strides would include consistent ability

13:12.225 --> 13:15.125
to argue a point and provide compelling evidence,

13:15.785 --> 13:20.725
or to learn to grab the reader and tell a good story

13:21.385 --> 13:24.125
or, to engage a reader.

13:25.825 --> 13:28.125
Of course, the kinds of arguments

13:28.145 --> 13:30.485
and the ways of engaging readers continue

13:30.505 --> 13:34.445
to develop across time, and their ups

13:34.445 --> 13:37.485
and downs, depending on the complexity of one's ideas.

13:38.065 --> 13:42.085
But development involves moving in these ways,

13:42.185 --> 13:43.405
in these bigger ways.

13:44.345 --> 13:45.845
And for underserved students,

13:46.065 --> 13:48.965
too often the focus is on the learning bits.

13:50.345 --> 13:55.005
Um, but they need to, they need to develop in,

13:55.305 --> 13:56.805
in much deeper ways,

13:57.585 --> 14:00.085
and they're disadvantaged by being denied help

14:00.115 --> 14:02.365
with the most important kinds of learning.

14:03.935 --> 14:07.155
And my second big takeaway from Vygotsky is the importance

14:07.155 --> 14:08.355
of social interaction.

14:08.535 --> 14:10.795
So too often when people think about writing,

14:10.865 --> 14:14.755
they think about it as a solo act, the stereotype

14:14.755 --> 14:19.275
of the wor writer in a tower someplace working alone at his

14:19.335 --> 14:22.835
or her desk, and coming out with a finished manuscript

14:22.935 --> 14:26.155
or a finished product with no social interaction

14:26.155 --> 14:28.195
with anyone else during this lonely,

14:28.255 --> 14:29.675
lonely composing process.

14:30.575 --> 14:33.315
And Vygotsky turns the stereotype on its head,

14:33.535 --> 14:37.715
and he portrays development as a writer, as a profoundly

14:38.055 --> 14:42.475
and, and as a writer at, to develop as a writer, it has

14:42.475 --> 14:45.755
to be a profoundly social activity in which, yes,

14:45.755 --> 14:49.195
writers might spend some time alone, but they learn

14:49.335 --> 14:51.525
and develop as they interact with others,

14:52.465 --> 14:55.205
as they generate ideas for what they're going to say.

14:55.465 --> 14:58.525
And they depend on the feedback of others to be sure

14:58.555 --> 15:00.965
that they're communicating what they intend

15:02.185 --> 15:05.005
and to help them elaborate and clarify,

15:05.425 --> 15:08.565
and sometimes even change what they're going to say.

15:09.665 --> 15:14.505
And this interaction is in,

15:14.685 --> 15:19.625
in, as, as kids grow up is, has to be with people,

15:19.845 --> 15:22.985
but also includes interacting with what you read

15:23.285 --> 15:26.105
and the having the words of others come, come

15:26.105 --> 15:27.625
to you in varied ways.

15:28.525 --> 15:30.865
And it's through these crucial interactions

15:30.865 --> 15:33.465
that writers develop their ability to write.

15:33.965 --> 15:37.385
And most of my research on writing has focused on

15:37.455 --> 15:39.665
understanding these interactions

15:39.845 --> 15:41.505
and what makes them work best.

15:42.805 --> 15:47.745
Now, my main takeaway from the writings of Bach team is that

15:47.965 --> 15:52.225
to develop as a writer in these social inter interactions,

15:52.445 --> 15:56.345
we benefit from a diversity of other voices

15:57.085 --> 15:58.785
who challenge our thinking.

15:59.285 --> 16:02.345
Um, and that is part of what's really crucial

16:02.445 --> 16:04.385
to stimulating our development.

16:05.525 --> 16:09.225
So to the social interactions that Vygotsky discusses,

16:09.615 --> 16:13.865
bine adds diversity, he claims that what we think

16:14.005 --> 16:16.305
and how we think is not our own.

16:17.015 --> 16:18.985
It's, we don't have our own thoughts,

16:18.985 --> 16:20.065
they don't really exist.

16:20.325 --> 16:22.545
Our thoughts include within them

16:24.665 --> 16:27.675
what other people say and think in our world.

16:28.615 --> 16:33.035
And often unconsciously, we incorporate the voices

16:33.375 --> 16:34.515
and the words of others.

16:35.375 --> 16:38.875
Um, and so people get all upset about plagiarism

16:38.935 --> 16:40.435
and helping people write,

16:40.695 --> 16:45.675
and that's just off course in, in terms of these theorists,

16:45.675 --> 16:48.435
there's not a clear line between our thinking

16:48.495 --> 16:49.795
and the thinking of other people.

16:50.775 --> 16:54.395
Um, BAK team shows the great value of inclusivity

16:54.615 --> 16:56.395
of all kinds of different people

16:56.655 --> 16:58.515
during the teaching learning process.

16:59.655 --> 17:04.435
Um, the more challenging the words

17:04.575 --> 17:07.795
are, we shouldn't be running away from that.

17:08.015 --> 17:11.985
And I guess one of the dangers that we've seen

17:12.845 --> 17:14.065
in action today is

17:14.065 --> 17:16.065
that people are living in their little bubbles.

17:16.605 --> 17:18.665
And in the United States, in particularly,

17:19.085 --> 17:21.865
our societies becoming more and more divided,

17:22.565 --> 17:26.265
and we, we, with social media,

17:26.525 --> 17:28.985
we get ourselves into an echo chamber

17:29.395 --> 17:31.665
where we only hear like voices.

17:32.685 --> 17:35.985
And that doesn't allow us to understand the thoughts

17:35.985 --> 17:39.505
of others to live together with others,

17:40.205 --> 17:44.985
and to think and write, outside the box.

17:45.445 --> 17:49.885
Um, so it's actually dangerous

17:50.505 --> 17:53.765
in a way to not be inclusive.

17:54.265 --> 17:57.085
Um, and Bine shows

17:57.465 --> 18:01.445
how such echo chambers depress learning and development.

18:02.925 --> 18:06.115
Um, I think that,

18:07.425 --> 18:11.015
these theories provide a path for research in the future.

18:11.315 --> 18:12.455
In the area of writing

18:12.595 --> 18:14.975
and schools, it becomes clear

18:15.085 --> 18:18.215
that diverse schools not segregated schools

18:18.475 --> 18:22.735
and diverse classrooms, not those tracked by narrow notions

18:22.755 --> 18:27.135
of ability can lead to better learning for everybody.

18:27.925 --> 18:31.015
It's also the case that students need to develop strategies

18:31.015 --> 18:32.495
for writing as well

18:32.595 --> 18:35.415
as build from their sociocultural context.

18:36.475 --> 18:39.055
Um, researchers have played explicit attention

18:39.195 --> 18:41.895
to underachieving students,

18:43.255 --> 18:46.675
and students benefit by hearing

18:48.125 --> 18:49.865
not just the opinions of others

18:50.085 --> 18:51.465
and the experiences of others,

18:52.165 --> 18:54.945
but by incorporating those experiences

18:55.165 --> 18:56.905
and making it part of their own.

18:57.205 --> 18:59.225
Now, I, I conducted a research study

18:59.225 --> 19:02.665
with a teacher in a Berkeley Middle school, Verdel,

19:02.965 --> 19:06.105
and we wrote an article entitled Teaching English in

19:06.175 --> 19:07.465
Untracked Classrooms.

19:08.045 --> 19:12.505
And we described how, how productive,

19:12.725 --> 19:15.905
how you can help students use one another's ideas

19:16.205 --> 19:18.665
and experiences in the process of learning

19:18.725 --> 19:21.385
to think more deeply about reading

19:21.485 --> 19:26.145
and writing, in responding to their,

19:29.995 --> 19:32.575
in responding to their reading and writing.

19:32.875 --> 19:37.185
Um, she was masterful at being able to push students

19:37.285 --> 19:40.265
to achieve more than they could without her questioning

19:40.445 --> 19:42.065
and without, and,

19:42.205 --> 19:44.505
and she was also masterful at

19:45.015 --> 19:47.945
helping them keep notes about other people's thoughts

19:48.525 --> 19:51.505
and figuring out what of those thoughts

19:52.015 --> 19:53.785
they found useful for themselves.

19:54.535 --> 19:59.025
Yeah, I think that with respect to future research, this is

19:59.025 --> 20:00.185
where the field needs to go.

20:00.285 --> 20:02.985
We need to understand the affordances

20:03.365 --> 20:06.985
of the new writing supports provided by generative ai.

20:07.725 --> 20:12.425
Um, so I think we can use the, the same theories,

20:13.085 --> 20:14.705
and,

20:15.565 --> 20:20.025
and question how generative AI can best support student

20:20.025 --> 20:22.905
writers as the machine responds to

20:23.045 --> 20:24.945
and helps all of us produce text.

20:25.485 --> 20:27.945
And I think that most people who have access,

20:28.055 --> 20:31.945
including my fellow academics, have tried and used

20:32.045 --> 20:34.145
and are using generative ai.

20:34.805 --> 20:36.905
And in the schools there's been,

20:37.605 --> 20:40.705
at least in the United States, there's been this, fear

20:41.335 --> 20:44.945
that no longer will kids have to learn to write

20:44.965 --> 20:47.025
or know how to write, but you really have

20:47.125 --> 20:49.125
to be quite sophisticated to be able

20:49.125 --> 20:52.085
to use generative AI in ways that are productive

20:52.145 --> 20:54.365
and it can help you learn to write better.

20:55.105 --> 20:56.845
Um, it's a kind of interaction

20:57.235 --> 20:59.925
that happens when you use generative ai.

21:00.465 --> 21:04.005
And, we need to find out

21:04.145 --> 21:06.725
and we need to know how

21:06.745 --> 21:09.565
to support students in using it.

21:09.745 --> 21:12.885
And we, it, it's not something that needs to be banned.

21:13.025 --> 21:16.365
So I think that there's, there's lots of room for research

21:16.365 --> 21:18.245
because these tools are changing.

21:18.345 --> 21:21.845
And as the tools are changing, the then writing is

21:21.955 --> 21:23.925
what writing is is changing.

21:24.665 --> 21:29.125
And so researchers need to be taking a very careful look

21:29.745 --> 21:32.605
at what about AI is useful?

21:32.995 --> 21:34.565
What about AI is harmful?

21:34.945 --> 21:38.525
How do humans control how they use ai?

21:38.525 --> 21:42.955
Because, I think that kids need support.

21:43.335 --> 21:46.835
Um, and the AI warns you,

21:46.835 --> 21:48.395
everything I say is not right,

21:49.215 --> 21:52.675
but how do teach, how do kids learn to

21:53.415 --> 21:56.475
figure out how to sort through

21:57.225 --> 22:00.675
what AI produces versus what they produce versus

22:00.745 --> 22:03.075
what they intend to say, what they want to say?

22:04.135 --> 22:08.655
Um, I think the worst response is to assume

22:08.655 --> 22:11.415
that teach kids are gonna use these tools to cheat.

22:12.075 --> 22:15.415
And that's really was the immediate response in,

22:15.635 --> 22:20.175
in many parts of the, educational world in the US with,

22:20.635 --> 22:23.975
schools banning it and that kind of thing.

22:24.155 --> 22:26.295
But it, it's, that's not gonna work.

22:26.715 --> 22:31.055
Um, we, we need to know what humans do best

22:31.235 --> 22:33.535
and what machines do best at this point,

22:34.315 --> 22:36.895
and we need to,

22:37.035 --> 22:39.855
figure out what they can do.

22:40.195 --> 22:43.495
So, and when I was thinking about this, I decided

22:43.595 --> 22:45.335
to ask chat, GP, t

22:45.555 --> 22:49.495
and Gemini to provide their answer to the question of

22:49.725 --> 22:52.975
what do machines do best and what do kids do best?

22:53.155 --> 22:57.975
And I can read you what chat GPT said, just as an example.

22:58.155 --> 23:02.695
So chat GBT said, generative AI has unique strengths

23:02.715 --> 23:04.135
as a teaching tool for writers.

23:04.715 --> 23:08.775
It offers instant feedback, is accessible around the clock,

23:09.035 --> 23:12.455
and can process and generate large amounts of text quickly,

23:12.865 --> 23:15.055
which is helpful for brainstorming

23:15.315 --> 23:18.335
and providing examples of varied writing styles.

23:18.995 --> 23:21.735
It can also simulate different audiences

23:21.795 --> 23:25.975
for a writer's work, helping to refine, targeting and voice.

23:26.945 --> 23:28.525
Humans, on the other hand,

23:29.335 --> 23:31.565
excel in providing nuanced feedback,

23:32.115 --> 23:34.965
understanding complex emotional tones,

23:35.385 --> 23:37.365
and fostering a more personalized

23:37.465 --> 23:39.645
and motivational learning environment.

23:40.855 --> 23:44.285
Human instructors can also guide on ethical writing

23:44.725 --> 23:47.645
Practices and offer deep contextual critiques

23:47.995 --> 23:50.245
that AI may not fully grasp.

23:50.815 --> 23:52.445
These aspects are crucial

23:52.545 --> 23:55.885
for developing a RI writer's unique voice and style,

23:56.225 --> 23:58.845
and for navigating subtler aspects of writing

23:58.955 --> 24:00.645
that AI might overlook.

24:01.585 --> 24:03.525
So this is what AI thinks,

24:04.145 --> 24:07.725
but humans need to weigh in as well.

00:00 - I've been studying the teaching and learning of writing since the early 1970s, and I've always examined issues of opportunity for students. in a few years ago, I was invited to write a chapter, a review chapter for

00:20 - by the American Educational Research Association for their upcoming handbook of Research on teaching. that article was published in 2016, and I worked with my colleague at Berkeley, who's also an expert in writing Glenda Hall and two of our then students to write a comprehensive literature review of the teaching and learning of writing in this digital and global age. I'm gonna focus today on what I've learned about teaching writing, both from my research

00:57 - and what my colleagues and I learned in writing that review. Since, since writing that review though, I wanna say that, much has changed. And so I'm going to also discuss, this current moment when artificial intelligence powered writing assistance like chat. GPT and Gemini have literally introduced revolutionary and publicly accessible new tools for writing, which have the potential to profoundly changed the nature

01:33 - of the writing process for those with access to those tools from all these sources, including the new generative AI affordances, I wanna begin by stressing that it is consequential that young people do not have equal opportunities to engage expressively creatively and practically in the practice of writing. When I study whether young people have equal opportunities, I question whether they have equal access to learning opportunities

02:09 - and whether they have equal access to tools like computers, earlier on and, word processing programs. And now these new AI programs, access is not equally distributed, either in the form of learning activities or an access to digital tools. And this is true across the curriculum, not just for writing. So where are the inequalities that are consequential for learning to write first?

02:42 - I've done a lot of international research, and globally there are huge differences between access to even basic literacy across countries with those who live in, who live in poorer countries, having less access to robust teaching and learning, and to tools than those living in wealthy countries. The result is that even the most basic adult literacy rates are lower in poor countries than in rich countries. the overall literacy rate globally has been rising, and the statistics show that in 1976,

03:19 - only 66% of adults in the world, were had basic literacy skills. And by 20 22, 80 7% of adults in the world are literate. But the rates for low income countries are shocking. in 2022, the literacy rates in those countries were at 62%, which is even lower than the global rate in 1976. these countries are mostly in Sub-Saharan Africa and some in Asia.

03:56 - globally, there are also gaps between male and female literacy rates, with men generally having more access to learning to read and write than women. The good news is that the gender gap has decreased from 1976 to 2022, but the bad news is that it's still there. these statistics refer only to, I don't wanna stress only to very basic kinds of literacy, not to the kinds of advanced uses of reading and writing that, help people participate

04:32 - in the global e economic world or in, in academics. there is not much information about these higher rates of literacy globally. Now, locally, within the wealthy countries where there's more access. My own research and the research of others show that opportunities to learn to write are also not equally distributed within those countries. And so where are the inequalities within the rich countries

05:05 - that have, have, better literacy, better access to literacy, not unexpectedly. Those with lower socio come who come from lower socioeconomic status groups have poor learning opportunities than those with higher socioeconomic status. their schools expect less and their teachers expect less, and their teachers have less training in the teaching

05:35 - of writing in the us there are also differences across ethnic and racial groups, which are associated with socioeconomic status with whites and Asian Americans having more opportunity to reach high levels of literacy than those from, other ethnic groups. And there's evidence for these differences on standardized tests and on, kind of national exams in the us like the national Assessment

06:02 - of educational progress. It's also the case that those in wealthier countries and the more wealthy within all countries have more access to literacy learning and to digital tools that support writing development. computers and access to the internet is also unequally distributed in in the same ways as, as, opportunities to learn. So for poorer countries where most citizens live in poverty, they need better access to schooling for their populations

06:35 - and attention to gaps in wealth. And this seems to be an ongoing struggle for these countries. And only then will even basic literacy rates have a chance of rising. So the big focus in those countries is just getting kids to school. and oftentimes children are used as labor, and are needed at home. And so there's a real push and pull.

07:02 - It's a very diff very, very difficult and, problem. for wealthy countries, the key to closing the opportunity gaps, wealthy countries think a little differently, because there's mandatory schooling, and kids do go to school, but then the question becomes what kind of schools and much of closing the opportunity gaps in, in these in wealthier countries lies in the hands of social policy makers who need funding formulas

07:37 - that don't, don't leave the underserved students behind. For example, there, there are not enough policies in place to support these schools generally, or teachers of writing in particular in the us for example. continuing education programs for teachers of writing are losing funding. there needs to be a better incentive structure to get good teachers into these schools in, in neighborhoods serving children who live in poverty. and there needs to be motivation

08:12 - for outstanding teachers to teach in these, in these settings. cross national studies show that countries that see teachers as professionals and support them, well produce more highly literate young people without the kinds of equity gaps that, that we see, in countries that don't provide sufficient support for teachers. so for example, in Finland, there's a big stress on, on

08:47 - educating teachers and teachers being professionals, paying them well. and so the literacy rates are, they're the highest in the world. So there, there's a kind of correlation there. the other thing about teaching, writing and closing the opportunities gaps is that teaching writing is very labor intensive, and it requires teachers who first can write well themselves and who are able to analyze student work in ways

09:21 - that help them respond productively to student writing. And so teachers need to be able to read lots of papers, and that is, that takes a lot of time, and they need to be able to talk to their, their, pupils. And that takes a lot of time. and they need to be able to engage their students and, and give them useful response that they can carry forward, with lowers. And the, and the teaching loads often are too high

09:52 - for teachers to be able to give the kind of attention that, that the students need. So the National Council of Teachers of English in the United States for, since for many years, since the 1960s, I think has, has had a policy statement that says that teachers of English should have no more than a hundred students in four classes of 25 students each. And that should be the highest. That should be the highest,

10:19 - and that is just not followed at all. so, it's hard for teachers to get to know their individual students in countries like in, in a number of European countries in England as well. I've done a cross-cultural study of teaching writing in England and the United States teachers keep the same students for multiple years, and they are able to get to know them that way. In the United States, teachers may keep students

10:50 - for a semester. and so it's really, it's really hard to do the kind of teaching that that needs to happen. And I mean, this is true for all students. but if you can't, if you don't have way, if you don't have, if policies don't give you structures in which you can actually teach writing, well then it's hard to close the opportunity gaps because the wealthier students get help elsewhere. and the poor students don't, don't have ways to do that.

11:24 - so the task is huge, and closing achievement gaps really requires well-trained and dedicated teachers who have support to do their work. Writing development happens across people's lifespans and in, and the context outside of schools matter to that development as well as the context that are available inside schools. And to understand the process, I rely on two theorists, Lev Vygotsky and Mikhail Bach team who connect the larger

12:00 - societal issues, some of which I've been talking about already, to the everyday practices that support writing development. My main takeaway from the writings of Vygotsky is his definition of development as opposed to learning and his understanding of how social interaction stimulates that development. So the key words here are development and social interaction. So too often we think in units

12:32 - that are too small. So for Vygotsky learning, learning is the small bits that we, that we get. and development is the more profound changes, and we need bits of learning to feed into development, but development. and too often we focus on the small bits like learning to punctuate or, form complete written sentences. and we need small units of learning. But our goal is to see development,

13:05 - to see young people make these big strides in writing. And the big strides would include consistent ability to argue a point and provide compelling evidence, or to learn to grab the reader and tell a good story or, to engage a reader. Of course, the kinds of arguments and the ways of engaging readers continue to develop across time, and their ups and downs, depending on the complexity of one's ideas. But development involves moving in these ways,

13:42 - in these bigger ways. And for underserved students, too often the focus is on the learning bits. but they need to, they need to develop in, in much deeper ways, and they're disadvantaged by being denied help with the most important kinds of learning. And my second big takeaway from Vygotsky is the importance of social interaction. So too often when people think about writing,

14:10 - they think about it as a solo act, the stereotype of the wor writer in a tower someplace working alone at his or her desk, and coming out with a finished manuscript or a finished product with no social interaction with anyone else during this lonely, lonely composing process. And Vygotsky turns the stereotype on its head, and he portrays development as a writer, as a profoundly and, and as a writer at, to develop as a writer, it has to be a profoundly social activity in which, yes,

14:45 - writers might spend some time alone, but they learn and develop as they interact with others, as they generate ideas for what they're going to say. And they depend on the feedback of others to be sure that they're communicating what they intend and to help them elaborate and clarify, and sometimes even change what they're going to say. And this interaction is in, in, as, as kids grow up is, has to be with people, but also includes interacting with what you read

15:23 - and the having the words of others come, come to you in varied ways. And it's through these crucial interactions that writers develop their ability to write. And most of my research on writing has focused on understanding these interactions and what makes them work best. Now, my main takeaway from the writings of Bach team is that to develop as a writer in these social inter interactions, we benefit from a diversity of other voices

15:57 - who challenge our thinking. and that is part of what's really crucial to stimulating our development. So to the social interactions that Vygotsky discusses, bine adds diversity, he claims that what we think and how we think is not our own. It's, we don't have our own thoughts, they don't really exist. Our thoughts include within them what other people say and think in our world.

16:28 - And often unconsciously, we incorporate the voices and the words of others. and so people get all upset about plagiarism and helping people write, and that's just off course in, in terms of these theorists, there's not a clear line between our thinking and the thinking of other people. BAK team shows the great value of inclusivity of all kinds of different people during the teaching learning process.

16:59 - the more challenging the words are, we shouldn't be running away from that. And I guess one of the dangers that we've seen in action today is that people are living in their little bubbles. And in the United States, in particularly, our societies becoming more and more divided, and we, we, with social media, we get ourselves into an echo chamber where we only hear like voices.

17:32 - And that doesn't allow us to understand the thoughts of others to live together with others, and to think and write, outside the box. so it's actually dangerous in a way to not be inclusive. and Bine shows how such echo chambers depress learning and development. I think that, these theories provide a path for research in the future. In the area of writing

18:12 - and schools, it becomes clear that diverse schools not segregated schools and diverse classrooms, not those tracked by narrow notions of ability can lead to better learning for everybody. It's also the case that students need to develop strategies for writing as well as build from their sociocultural context. researchers have played explicit attention to underachieving students, and students benefit by hearing

18:48 - not just the opinions of others and the experiences of others, but by incorporating those experiences and making it part of their own. Now, I, I conducted a research study with a teacher in a Berkeley Middle school, Verdel, and we wrote an article entitled Teaching English in Untracked Classrooms. And we described how, how productive, how you can help students use one another's ideas

19:16 - and experiences in the process of learning to think more deeply about reading and writing, in responding to their, in responding to their reading and writing. she was masterful at being able to push students to achieve more than they could without her questioning and without, and, and she was also masterful at helping them keep notes about other people's thoughts and figuring out what of those thoughts

19:52 - they found useful for themselves. Yeah, I think that with respect to future research, this is where the field needs to go. We need to understand the affordances of the new writing supports provided by generative ai. so I think we can use the, the same theories, and, and question how generative AI can best support student writers as the machine responds to and helps all of us produce text.

20:25 - And I think that most people who have access, including my fellow academics, have tried and used and are using generative ai. And in the schools there's been, at least in the United States, there's been this, fear that no longer will kids have to learn to write or know how to write, but you really have to be quite sophisticated to be able to use generative AI in ways that are productive and it can help you learn to write better.

20:55 - it's a kind of interaction that happens when you use generative ai. And, we need to find out and we need to know how to support students in using it. And we, it, it's not something that needs to be banned. So I think that there's, there's lots of room for research because these tools are changing. And as the tools are changing, the then writing is what writing is is changing.

21:24 - And so researchers need to be taking a very careful look at what about AI is useful? What about AI is harmful? How do humans control how they use ai? Because, I think that kids need support. and the AI warns you, everything I say is not right, but how do teach, how do kids learn to figure out how to sort through what AI produces versus what they produce versus

22:00 - what they intend to say, what they want to say? I think the worst response is to assume that teach kids are gonna use these tools to cheat. And that's really was the immediate response in, in many parts of the, educational world in the US with, schools banning it and that kind of thing. But it, it's, that's not gonna work. we, we need to know what humans do best and what machines do best at this point, and we need to,

22:37 - figure out what they can do. So, and when I was thinking about this, I decided to ask chat, GP, t and Gemini to provide their answer to the question of what do machines do best and what do kids do best? And I can read you what chat GPT said, just as an example. So chat GBT said, generative AI has unique strengths as a teaching tool for writers. It offers instant feedback, is accessible around the clock, and can process and generate large amounts of text quickly,

23:12 - which is helpful for brainstorming and providing examples of varied writing styles. It can also simulate different audiences for a writer's work, helping to refine, targeting and voice. Humans, on the other hand, excel in providing nuanced feedback, understanding complex emotional tones, and fostering a more personalized and motivational learning environment. Human instructors can also guide on ethical writing

23:44 - Practices and offer deep contextual critiques that AI may not fully grasp. These aspects are crucial for developing a RI writer's unique voice and style, and for navigating subtler aspects of writing that AI might overlook. So this is what AI thinks, but humans need to weigh in as well.

Faculti AI can make mistakes