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

College or Bust… or Both

View more insights in Professional - Education - Policy and Politics in Education

View more figures from UC Berkeley

View the page of Tolani Britton

Read the Study

· Tools Cite Embed

Tolani Britton discusses whether the Great Recession led to changes in two-year and four-year college enrollment patterns for students aged 18–24.

Image courtesy of interviewee. May 17, 2024

Log-in or Sign-up to Faculti
Currently viewing this subject insight as a guest. You have insight(s) remaining for this month.

Workspace

Faculti AI is thinking...

< >

Summary
Related Studies
Faculti AI
Key Points
Key Quotes
Translate
Transcript

Export as PDF

This video video discusses the impact of economic recessions, specifically the Great Recession, on college enrollment rates, particularly for Black and Latina students. It highlights that previous research indicated people are more likely to return to college during recessions due to higher unemployment, which decreases the opportunity cost of education.

The author cites significant increases in unemployment rates among young individuals during the Great Recession and questions whether the trends observed in previous recessions hold true in this context. The analysis focuses on the ability of families of color, especially Black and Latina families, to afford college given the economic hardships they faced—namely, job losses and loss of home assets.

The study uses data from the Current Population Survey to analyze college enrollment patterns between 2002 and 2012, employing a difference-in-difference-in-difference methodology. The findings reveal an overall increase in college enrollment in states with higher unemployment rates and particularly in two-year colleges. This video video suggests that unemployed individuals turned to education to upskill and enhance their employability.

For Black and Latina students, it was found that their enrollment increased by 2 to 4 percentage points during the recession, although structural barriers like credit constraints in high unemployment states may have affected their ability to enroll. The author concludes by emphasizing the importance of providing support and resources, including financial aid and wraparound services, to help students complete their degrees, particularly in light of challenges exacerbated by economic downturns. Future research is suggested to explore better support mechanisms for students facing economic and structural challenges within the educational system.

Language:

Export as PDF

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

00:00.305 --> 00:01.805
We know that there was clearly the

00:01.805 --> 00:02.965
great recession that occurred.

00:03.505 --> 00:07.885
Um, and part of my interest in the work was prior work

00:07.885 --> 00:10.885
that actually demonstrated that in times of recession,

00:10.885 --> 00:13.965
and this is people like Betson McFarland in 1995

00:13.985 --> 00:17.165
and Pennington McGinty and Williams in 2002.

00:17.225 --> 00:19.965
And part of what they demonstrated was that,

00:20.275 --> 00:23.245
when there were economic recessions,

00:23.245 --> 00:26.005
people were actually more likely to return to college,

00:26.185 --> 00:28.805
in part because unemployment rates were higher.

00:29.385 --> 00:31.405
And so, the opportunity cost

00:31.425 --> 00:34.325
or what they were giving up was less, right?

00:34.345 --> 00:35.405
So if they were more likely

00:35.405 --> 00:38.005
to be unemployed than they would oftentimes return

00:38.185 --> 00:41.325
and be more likely to return to college than they would,

00:41.325 --> 00:43.765
during times of, economic prosperity.

00:44.465 --> 00:47.565
And so that really sparked my interest in understanding,

00:47.905 --> 00:49.845
is this true more broadly?

00:50.105 --> 00:54.005
And is this true during a period, of,

00:54.765 --> 00:57.205
I would say, you know, sort of greater economic severity.

00:57.235 --> 00:58.965
This was the finding in past recessions.

00:58.965 --> 01:00.965
But we know part of what happened with the Great Recession,

01:00.965 --> 01:03.525
particularly with, you know, people, you know,

01:03.525 --> 01:06.125
the subprime mortgages, people losing their homes,

01:06.595 --> 01:08.445
that it was more severe in many ways.

01:08.505 --> 01:10.885
We had relatively high rates of unemployment.

01:11.505 --> 01:14.885
Um, you know, one study found that the unemployment rate

01:14.885 --> 01:16.245
for people under the age

01:16.245 --> 01:20.685
of 25 increased from 11.5% in the first quarter of 2008.

01:20.685 --> 01:24.605
So at the beginning of the Great Recession to 18.3%

01:25.225 --> 01:27.165
by the fourth quarter of 2010.

01:27.225 --> 01:28.885
So essentially over a two year period,

01:28.905 --> 01:32.405
we see this increase in the unemployment rate, particularly

01:32.745 --> 01:35.325
for, persons under the age of 25.

01:35.825 --> 01:40.525
And so my questions were, well, in the past we saw

01:40.525 --> 01:42.645
that people were more likely to return to college.

01:44.055 --> 01:46.075
Is this true now?

01:46.935 --> 01:49.835
And in particular, given that we know

01:49.835 --> 01:53.075
that the Great Recession hit families of color.

01:53.135 --> 01:54.515
And when I say families of color,

01:54.655 --> 01:58.595
I'm really thinking in this particular study of, black

01:58.595 --> 02:01.755
and Latina families, on the one hand,

02:01.935 --> 02:04.275
you have these higher unemployment rates.

02:04.775 --> 02:06.915
On the other hand, you have families losing

02:06.985 --> 02:08.915
what in the United States generally tends

02:08.915 --> 02:11.675
to be the primary asset, which is the home.

02:12.405 --> 02:15.395
Given this reality, were students even able

02:15.495 --> 02:16.835
to return to college?

02:16.975 --> 02:19.155
And why, why do we care about whether

02:19.155 --> 02:20.515
or not they were returning to college?

02:20.515 --> 02:21.555
When we think about college,

02:22.305 --> 02:24.755
there's clearly this whole debate in the US about college.

02:24.855 --> 02:27.035
Is it worth it? Students are taking on debt.

02:28.535 --> 02:31.475
And I think those things are true, right?

02:31.515 --> 02:35.795
I I think it'd be, careless not to name those realities.

02:35.795 --> 02:38.555
And at the same time, in the absence of college,

02:38.795 --> 02:41.915
particularly given some of the realities of discrimination

02:41.935 --> 02:45.715
for black Latina young adults, what are the other options?

02:46.225 --> 02:47.995
What are the ways that you can build skills?

02:48.545 --> 02:50.235
What are the ways that you can increase your

02:50.235 --> 02:51.395
likelihood of employment?

02:51.875 --> 02:53.835
Absolutely. So there is an extensive amount

02:53.835 --> 02:54.875
of work that's been done.

02:55.075 --> 02:57.755
I mean, I will certainly leave some people out.

02:58.095 --> 02:59.565
Um, but I, you know, I think

02:59.565 --> 03:02.045
that there's this broad literature,

03:02.045 --> 03:04.365
and some of the literature is very descriptive, right?

03:04.425 --> 03:06.005
So we think about people like Snyder

03:06.005 --> 03:09.605
and Dillo, who, who, named it in 2007, black

03:09.605 --> 03:12.645
and Latina young adults between 18 to 24 had lower rates

03:12.645 --> 03:16.285
of college enrollment, in the year following graduation,

03:16.585 --> 03:20.765
high school graduation, then their white counterparts.

03:21.265 --> 03:24.125
Um, and so if we think about, for example, black enrollment,

03:24.265 --> 03:27.125
the enrollment of black young adults, it was about 56%.

03:27.145 --> 03:31.245
And this is nationally, about 64% for Latin a students,

03:31.305 --> 03:32.765
and about 70%, right?

03:32.765 --> 03:34.965
This is the year following high school graduation,

03:35.385 --> 03:36.565
for white students.

03:36.705 --> 03:37.765
And so part of

03:37.765 --> 03:40.245
what we understand is these are not simply individual

03:40.605 --> 03:43.245
destruct, individual decisions.

03:43.625 --> 03:46.085
Um, they're about the structures in which they exist.

03:46.585 --> 03:48.765
We know that students are not attending the

03:48.765 --> 03:49.885
same types of high school.

03:49.885 --> 03:53.165
They're not receiving the same types of preparation access

03:53.185 --> 03:56.525
to, high quality coursework,

03:56.865 --> 03:59.445
things like advanced courses, AP courses.

04:00.265 --> 04:02.165
Um, and what's, you know, part of

04:02.165 --> 04:05.485
what has really sparked my interest is that this recognition

04:05.485 --> 04:08.205
that college enrollment rates certainly differ by race

04:08.225 --> 04:09.725
and ethnicity in the United States,

04:10.425 --> 04:12.445
but work, for example, done by Cameron

04:12.585 --> 04:15.485
and Heckman in 2001, found that

04:15.485 --> 04:18.005
after you controlled for family background, so right,

04:18.005 --> 04:20.125
really thinking about income, black

04:20.125 --> 04:23.125
and Latina young adults were actually more likely to enroll.

04:23.265 --> 04:25.805
So when we just look at the statistics, we say,

04:25.915 --> 04:27.205
they're less likely to enroll.

04:27.205 --> 04:30.725
But then when we start controlling, for some

04:30.725 --> 04:33.085
of the factors related to enrollment,

04:33.085 --> 04:35.925
things like family income, we see that black

04:35.925 --> 04:38.285
and Latina students are actually more likely to enroll.

04:38.905 --> 04:41.805
Um, we, there's sort of an extensive body of work that,

04:41.905 --> 04:43.805
you know, including some work that I've done looking at

04:43.805 --> 04:46.565
expectations in both middle school and high school.

04:46.905 --> 04:50.365
And for black and Latina students, they're as likely,

04:50.585 --> 04:53.805
if not more likely, than their peers to say they want to go

04:53.805 --> 04:55.445
to college, they intend to go to college.

04:56.145 --> 04:57.405
And then we have to think about

04:57.555 --> 04:59.645
what are the circumstances in which they're living

04:59.665 --> 05:02.525
and learning, and how does that either enable

05:02.705 --> 05:04.405
or constrain access to college.

05:05.225 --> 05:08.165
Um, finally, we know that there's certainly the role

05:08.165 --> 05:11.045
of federal and state financial aid, particularly

05:11.225 --> 05:13.405
for students who have, you know, sort

05:13.405 --> 05:15.005
of limited access to college.

05:15.545 --> 05:18.365
And we, you know, there, once again, BARR

05:18.365 --> 05:21.005
and Turner in 2013, looked at the role

05:21.005 --> 05:25.205
of financial aid in, mitigating sort of individual

05:25.305 --> 05:27.085
and familial credit constraints.

05:27.745 --> 05:30.405
Um, and demonstrated that if students had more access

05:30.425 --> 05:32.165
to federal and state aid

05:32.165 --> 05:34.645
and particularly grants, they were more likely

05:34.645 --> 05:36.365
to enroll in colleges compared to students

05:36.365 --> 05:37.925
who had lesser access to these grants.

05:38.585 --> 05:41.485
Um, and we also know a work by, Musu Gillet,

05:41.665 --> 05:43.645
for example, in 2016, Santiago

05:43.645 --> 05:47.365
and Cunningham in 2005, that black students are actually,

05:47.705 --> 05:50.045
more likely to receive access to grants

05:50.065 --> 05:52.925
and loans, than students from other groups.

05:53.105 --> 05:57.405
And so the thought is, if they have access to aid,

05:58.325 --> 06:01.015
then they may be able to, in some ways,

06:01.095 --> 06:02.295
I don't wanna say overcome,

06:02.555 --> 06:04.935
but it may mitigate in some ways some

06:04.935 --> 06:08.255
of the structural barriers around lower levels of employment

06:08.315 --> 06:10.535
as well as the loss of familial assets.

06:10.835 --> 06:14.695
So the dataset was the, current population survey.

06:15.275 --> 06:17.895
And in particular, I used the education supplements.

06:18.095 --> 06:22.095
I did it over a 10 year period from 2002 to, 2012.

06:22.205 --> 06:25.975
Part of the thought is being able to establish pretre,

06:25.975 --> 06:28.695
which is important given the methodologies that I use,

06:28.695 --> 06:30.575
which was a difference in difference in differences.

06:31.155 --> 06:34.135
And the main threats of validity in the difference in

06:34.135 --> 06:35.655
difference, which is the two differences,

06:35.655 --> 06:37.295
and I'll get to the three differences in a minute,

06:37.875 --> 06:39.615
is really thinking about pretre.

06:39.875 --> 06:42.295
Um, if you have two different groups that you're looking

06:42.315 --> 06:44.975
to compare, we're looking to compare trends,

06:45.115 --> 06:47.815
not necessarily the absolute numbers.

06:48.035 --> 06:50.615
Um, and so the trends actually need to be parallel.

06:51.155 --> 06:53.735
Um, and so part of the reason I have to establish some time

06:53.755 --> 06:55.735
before is to have the parallel trends.

06:56.275 --> 06:59.575
Um, in terms of the actual dataset, it is,

06:59.645 --> 07:01.735
collected on a monthly basis.

07:02.635 --> 07:04.735
Um, and it's a random sample

07:04.735 --> 07:06.335
of the population of the United States.

07:06.475 --> 07:07.655
At the time that I was collecting

07:07.675 --> 07:09.095
or using this collected data,

07:09.095 --> 07:11.215
it was about 60,000 family units.

07:11.915 --> 07:13.375
Um, and part of what they're trying

07:13.375 --> 07:17.135
to understand are measures like employment, earnings,

07:17.155 --> 07:18.655
and then sort of, sort of other general

07:18.655 --> 07:20.175
characteristics of household.

07:20.795 --> 07:24.775
Um, it does exclude people who are, institutionalized.

07:24.775 --> 07:26.295
And in the United States, that's both people

07:26.295 --> 07:28.895
who might be institutionalized for, health reasons

07:28.955 --> 07:30.615
as well as incarcerated.

07:31.155 --> 07:34.055
Um, they usually interview a particular household about four

07:34.055 --> 07:37.335
times, then they have eight, months where they don't,

07:37.335 --> 07:39.535
and then they resurvey it for another four months.

07:40.355 --> 07:42.855
Um, and the education supplement in particular

07:42.915 --> 07:44.135
is only collected in October.

07:44.195 --> 07:45.895
So it is an annual supplement,

07:45.915 --> 07:48.095
and it looks at things like school enrollment,

07:48.685 --> 07:49.815
educational attainment.

07:50.555 --> 07:53.495
Um, you know, for me, part of what it does is also look at,

07:53.515 --> 07:55.375
intensity of enrollment.

07:55.375 --> 07:56.975
Are students going part-time, full-time?

07:57.035 --> 07:58.975
Are they going to two year institutions,

07:58.975 --> 08:00.055
four year institutions?

08:00.595 --> 08:02.455
Um, you know, some of the other factors

08:02.455 --> 08:05.935
that I looked at were things like the racial, racial,

08:06.195 --> 08:08.895
affiliation of, of individuals in the household,

08:08.915 --> 08:10.695
family composition,

08:10.695 --> 08:12.655
the financial characteristics of household.

08:13.315 --> 08:16.015
Um, and part of the reason I did that was it allowed me

08:16.035 --> 08:18.735
to include a whole set of covariates such as, you know,

08:18.735 --> 08:20.615
race, ethnicity, gender, whether

08:20.635 --> 08:23.135
or not families owned their home employment status

08:23.155 --> 08:25.215
of the heads of households, given the sort

08:25.215 --> 08:26.415
of analysis that I was doing.

08:27.235 --> 08:30.335
Um, I also accounted for geographic locale

08:30.335 --> 08:32.175
because part of the approach, which I'll get

08:32.175 --> 08:34.215
to once again a little bit later when I start talking about

08:34.215 --> 08:35.655
the difference and difference in differences,

08:36.275 --> 08:38.855
was thinking about, what is the role

08:38.855 --> 08:40.295
of the severity of the reception?

08:40.395 --> 08:44.615
For example, in states where we had, sort

08:44.615 --> 08:47.255
of greater increases in, in unemployment rates,

08:48.005 --> 08:52.655
were we likely to see, more, all right,

08:52.655 --> 08:53.655
college enrollment.

08:53.655 --> 08:55.485
So were students more likely to enroll

08:55.755 --> 08:58.525
because right, the opportunities

08:58.705 --> 09:01.765
or the access to jobs was lesser or less likely, right?

09:01.765 --> 09:04.725
Because if your family members aren't working in theory,

09:04.785 --> 09:05.965
you have time to go to college,

09:06.065 --> 09:08.885
but you may need to try to find, you know, a part-time gig

09:08.985 --> 09:12.805
or, you know, some other kind of, unemploy,

09:12.835 --> 09:15.005
some other kind of employment in order to contribute

09:15.005 --> 09:16.085
to your household, which is one

09:16.085 --> 09:19.045
of the things we've actually seen in this most recent,

09:20.025 --> 09:21.525
you know, covid period where students

09:21.945 --> 09:23.965
and young people were taking jobs in order

09:23.965 --> 09:25.005
to help their families.

09:25.785 --> 09:28.485
Um, the, the final, sort of component that I

09:29.045 --> 09:32.325
included was, some of the, funding dispersed

09:32.365 --> 09:33.925
by the federal government,

09:33.925 --> 09:37.245
for education related purposes in 2009.

09:37.945 --> 09:41.885
Um, so the, the empirical approach is the difference

09:41.885 --> 09:43.445
and difference and differences model.

09:43.475 --> 09:45.445
This is a quasi experimental approach.

09:45.545 --> 09:48.965
So, you know, one of the, realities

09:48.965 --> 09:51.405
of education is we often can't run experiments,

09:51.425 --> 09:53.685
and particularly during, a recession

09:54.225 --> 09:55.365
and the great recession,

09:55.365 --> 09:58.445
we can't necessarily assign families to treatment

09:58.445 --> 09:59.485
and control conditions.

10:00.025 --> 10:01.645
So part of what we do is we think about

10:01.645 --> 10:03.565
what are naturally occurring experiments,

10:03.585 --> 10:04.805
and that's what we think when,

10:04.865 --> 10:06.005
so that's the approach, right?

10:06.005 --> 10:07.885
The quasi experimental approach is essentially,

10:08.105 --> 10:10.605
how do I take this natural experiment,

10:11.105 --> 10:15.255
and use a methodo methodological approach that uses this,

10:15.515 --> 10:17.695
natural variation that's occurring.

10:18.115 --> 10:21.175
Um, and so for the triple difference model,

10:21.235 --> 10:23.015
the first difference is really to see

10:23.015 --> 10:25.615
how enrollment changed over the time period in question.

10:25.835 --> 10:28.495
So really thinking pre great recession.

10:28.495 --> 10:30.255
And then once we're in the great recession,

10:30.635 --> 10:32.015
how has enrollment changed?

10:32.715 --> 10:36.535
The second is the geographic variation intensity.

10:37.435 --> 10:41.415
Um, and that's looking at state's annual unemployment rate,

10:41.915 --> 10:43.415
prior to the recession.

10:43.435 --> 10:47.855
So looking at 2006 and then looking at 2009.

10:47.915 --> 10:49.975
So from 2006 to 2009,

10:50.325 --> 10:53.335
what was the change in state's annual unemployment rate?

10:53.555 --> 10:55.775
And once again, this is the, to understand,

10:55.915 --> 10:58.215
to measure the severity of the recession.

10:58.835 --> 11:01.945
Um, and then finally, I am thinking about,

11:02.325 --> 11:04.345
and my third difference is,

11:04.445 --> 11:06.945
and how might this be different for,

11:07.265 --> 11:09.065
separately model separate models for black

11:09.585 --> 11:12.305
students when compared to other students?

11:12.725 --> 11:14.345
Um, and how is this different

11:14.345 --> 11:17.065
for Latina students when compared to other students?

11:17.365 --> 11:20.355
And the final thing that I did, I'm just recognizing

11:20.355 --> 11:22.995
that the difference in difference in differences model,

11:23.585 --> 11:25.875
like all methodological approaches,

11:25.875 --> 11:27.835
like all empirical approaches, has its limitation.

11:27.955 --> 11:30.675
I also did, I fit a set of maximum likelihood,

11:30.825 --> 11:32.395
multinomial logistic models.

11:33.015 --> 11:36.115
And the reason that I did this is I wanted to look at,

11:36.255 --> 11:39.275
or compare average probability of college enrollment in two

11:39.335 --> 11:42.955
and four year institutions, as opposed to what I did

11:42.955 --> 11:44.115
with the difference and difference

11:44.115 --> 11:46.635
and differences model, which was really looking at,

11:46.775 --> 11:48.875
are you enrolled or not enrolled?

11:49.135 --> 11:52.755
The major finding was that, or ma, the major findings were

11:52.755 --> 11:54.125
That college enrollment increased

11:54.145 --> 11:57.605
by about 2.4 percentage points in states

11:57.635 --> 11:59.405
with higher rates of unemployment.

11:59.425 --> 12:01.445
So this was overall, right?

12:01.465 --> 12:03.125
So when overall in states

12:03.215 --> 12:07.525
where the recession in some ways was more severe as measured

12:07.545 --> 12:11.645
by, state unemployment rates, we actually saw

12:11.645 --> 12:15.165
that people were more likely to enroll in college.

12:16.005 --> 12:17.305
What's important here is that we,

12:17.445 --> 12:20.865
we primarily saw the increase in two year colleges.

12:20.865 --> 12:22.865
And what that says to us,

12:22.965 --> 12:25.425
or what that at least suggests is that one

12:25.425 --> 12:28.065
of the mechanisms is that people who were unemployed were,

12:28.605 --> 12:31.905
returning to shorter programs in order

12:32.045 --> 12:33.305
to upskill, right?

12:33.325 --> 12:34.985
So they were going back,

12:35.725 --> 12:39.665
and saying, okay, how do we for some shorter period of time,

12:40.125 --> 12:42.345
pot potentially taking a few courses

12:42.885 --> 12:45.265
or taking, a certificate program,

12:45.315 --> 12:47.945
which would be shorter than the degree program they were

12:48.225 --> 12:50.705
going back to, update their skills,

12:50.735 --> 12:52.665
potentially receive new skills.

12:53.325 --> 12:58.205
Um, and I think that, you know, part of the context here is

12:58.235 --> 13:00.325
that, with the Great Recession,

13:00.325 --> 13:02.725
there were certainly places where we saw,

13:02.745 --> 13:04.405
higher rates of unemployment.

13:04.545 --> 13:09.005
And so if we think about, the construction industry,

13:09.065 --> 13:13.005
for example, we saw, you know, huge reductions in the number

13:13.005 --> 13:15.405
of homes being built, particularly be

13:15.435 --> 13:18.605
because so many people lost their homes,

13:18.605 --> 13:20.525
because of the, the mortgage crisis.

13:21.425 --> 13:24.685
And so, it could be the case that,

13:24.705 --> 13:27.365
in the places we saw higher rates of unemployment,

13:27.365 --> 13:30.165
they were also states like Nevada that,

13:30.945 --> 13:34.805
had large construction industries that were no longer,

13:35.745 --> 13:37.685
you know, sort of vibrant industries.

13:37.705 --> 13:38.765
And so it could have been the case

13:38.765 --> 13:42.325
that people were returning to, receive new skills

13:42.545 --> 13:44.285
to learn new skills potentially.

13:44.865 --> 13:47.285
Um, and, you know, there's not great evidence

13:47.345 --> 13:48.525
to change industries.

13:49.105 --> 13:52.605
Um, but when we think about two year colleges, we know

13:52.605 --> 13:55.565
that the kind of courses that are offered are often,

13:55.985 --> 13:58.885
you know, the, the two year institutions are,

13:58.885 --> 14:00.805
very diverse institutions, right?

14:01.065 --> 14:04.645
Um, so, you know, in many cases they,

14:04.955 --> 14:06.925
they provide GED courses, right?

14:06.925 --> 14:08.245
General education diplomas.

14:08.245 --> 14:11.545
So people who haven't gotten their, their,

14:11.545 --> 14:14.025
their high school diplomas can get their diplomas at these

14:14.025 --> 14:15.105
community colleges.

14:15.485 --> 14:17.465
Um, they have vocational training.

14:18.085 --> 14:20.425
So for example, if you wanted to become,

14:22.575 --> 14:23.575
a welder, many

14:23.635 --> 14:26.295
of the two year colleges have programs in welding

14:26.395 --> 14:27.575
or to become an electrician.

14:27.595 --> 14:29.855
So, they're just, you know, in addition,

14:29.855 --> 14:31.975
they also have programs in sociology, things

14:31.975 --> 14:34.055
that we generally think about for the four year degree.

14:34.475 --> 14:37.415
Uh, so there is certainly evidence that once again,

14:37.415 --> 14:38.695
college enrollment increased.

14:39.435 --> 14:41.375
Um, the second part, so that was sort of the first,

14:41.635 --> 14:42.975
major finding.

14:43.835 --> 14:46.175
The second part was, and does this hold

14:46.175 --> 14:47.575
for Black and Latine students?

14:48.075 --> 14:49.415
And what I found was that black

14:49.415 --> 14:50.895
and Latine students were between two

14:50.895 --> 14:53.845
and four percentage points more likely to enroll in college

14:53.895 --> 14:55.765
after the onset of the great Recession.

14:56.585 --> 15:00.885
Um, however, this was not necessarily this the case

15:01.425 --> 15:03.685
in states with higher rates of unemployment.

15:03.705 --> 15:05.805
So overall they were more likely to enroll in college.

15:06.625 --> 15:11.285
Um, however, part of what this suggests is that, some

15:11.285 --> 15:13.565
of the ways in which families might have been credit

15:13.675 --> 15:16.325
constrained in states with higher rates

15:16.325 --> 15:19.725
of unemployment served as a structural barrier for students.

15:20.345 --> 15:23.085
Um, and so I think earlier in the conversation,

15:23.245 --> 15:26.965
I mentioned, well, I use the word conversation lightly.

15:27.625 --> 15:30.205
Um,

15:30.525 --> 15:34.645
I mentioned earlier in this monologue, I mentioned that,

15:35.185 --> 15:39.445
you know, it could be the case that, black

15:39.445 --> 15:42.125
and Latina students were more likely to take on jobs.

15:42.505 --> 15:44.045
Um, and we know that actually black

15:44.045 --> 15:45.765
and Latina students are more likely

15:45.765 --> 15:48.085
to be both enrolled in college and working.

15:48.465 --> 15:49.725
And this might be one of the reasons

15:49.725 --> 15:51.845
that we didn't necessarily see,

15:52.055 --> 15:54.725
while we saw overall increases in the likelihood

15:54.725 --> 15:57.485
of college enrollment, as I said, on the order of two

15:57.485 --> 15:59.005
to four percentage points for black

15:59.005 --> 16:01.405
and Latin students, it wasn't necessarily in places

16:01.515 --> 16:03.885
that had higher rates of unemployment.

16:04.705 --> 16:09.125
Um, the final thing that I'll say is that, in part

16:09.125 --> 16:12.365
because of methodology, I interpret,

16:12.365 --> 16:15.005
this is a quasi experimental approach, but in part

16:15.005 --> 16:18.605
because, particularly for Latin A students, the,

16:19.025 --> 16:22.485
the prior trends were not, as parallel

16:22.485 --> 16:24.565
that I interpret LA the results

16:24.725 --> 16:26.085
for Latin a students with caution.

16:26.185 --> 16:27.485
So I always add that caveat.

16:27.825 --> 16:30.805
So I would say that part of, and,

16:30.945 --> 16:35.205
and you know, one of, one of the lessons from covid

16:35.345 --> 16:38.645
and some of the current research that's being done,

16:38.785 --> 16:41.965
places like the Community College Research Center at

16:42.125 --> 16:45.845
Columbia University and many other places is trying

16:45.845 --> 16:47.885
to understand,

16:48.355 --> 16:50.405
what happens when government provides an

16:50.405 --> 16:51.525
extensive amount of aid.

16:51.625 --> 16:54.805
So part of what the, context, you,

16:54.825 --> 16:56.125
you may have to switch the order.

16:56.285 --> 16:57.365
'cause I think I've started off a little bit with

16:57.365 --> 16:58.605
implications, not conclusions.

16:58.945 --> 17:02.565
Um, but you know, part of, you know, part of

17:03.195 --> 17:05.605
what we saw during COVID-19 was

17:05.605 --> 17:07.845
that more money was provided.

17:08.105 --> 17:10.525
Um, and, and this is the order of billions of dollars

17:10.745 --> 17:13.005
to both colleges,

17:13.345 --> 17:15.485
and to students in the form of financial aid.

17:15.485 --> 17:17.005
And so some of the work that's actually being done

17:17.585 --> 17:22.085
is we have this, you know, this sort

17:22.085 --> 17:25.325
of severe shock to, the economy,

17:25.665 --> 17:27.405
to families, to people's health.

17:27.595 --> 17:29.005
What happens when government actually

17:29.005 --> 17:30.165
provides additional aid?

17:30.515 --> 17:32.645
Does this mitigate some of those effects?

17:33.225 --> 17:36.525
Um, and in particular, does it actually help keep people

17:37.045 --> 17:39.765
enrolled in college and able to complete?

17:40.505 --> 17:44.165
Um, and so the answer to that question is we don't know.

17:44.225 --> 17:47.205
So I would absolutely say that that is one future direction

17:47.205 --> 17:48.565
for research is

17:48.565 --> 17:50.765
what happens when the government actually does step in

17:50.765 --> 17:51.925
and provide additional aid.

17:51.985 --> 17:53.725
Is it enough? Um,

17:53.725 --> 17:55.405
and one of the reasons that this, these

17:55.925 --> 17:58.245
questions are complex is it's not just money.

17:58.715 --> 18:01.285
It's actually how money is spent, right?

18:01.465 --> 18:03.085
It, you know, and some of the money, for example,

18:03.145 --> 18:05.605
during covid was spent on computers.

18:06.385 --> 18:10.685
Um, it was spent on ensuring students had access to, to wifi

18:11.025 --> 18:13.045
and access to networks.

18:13.185 --> 18:15.805
Um, you know, one of the things that I found as,

18:16.485 --> 18:18.725
a professor at this time was that there were students

18:18.745 --> 18:21.085
who didn't have places particularly, you know, given

18:21.085 --> 18:22.085
that students had gone home,

18:22.405 --> 18:24.365
students didn't have place quiet places

18:24.415 --> 18:26.965
where they could do my three hour class.

18:27.745 --> 18:30.565
Um, and so there were all of these sort of conditions

18:31.185 --> 18:33.485
and right, there were some pretty incredible things

18:33.485 --> 18:36.445
that happened, like setting up wifi centers at some

18:36.445 --> 18:39.245
of the California community college parking lots so

18:39.245 --> 18:41.325
that students could just come and sit in that parking lot

18:41.385 --> 18:43.445
and right, and then do their class.

18:43.665 --> 18:45.685
But at the same time, if a student had a job,

18:45.935 --> 18:47.685
maybe they weren't able to sit and do that class.

18:48.225 --> 18:51.635
So I think that, you know, one of the things

18:51.635 --> 18:54.235
that we've seen, during the Great recession,

18:54.235 --> 18:56.595
we saw increases in the likelihood of college enrollment

18:56.615 --> 18:58.555
for black and Latina students and for all students.

18:59.255 --> 19:03.595
Um, however, we saw declines in this most recent COVID-19,

19:03.595 --> 19:05.915
particularly in the two year college sector,

19:06.035 --> 19:08.595
declines in the likelihood of en enrollment for black

19:08.595 --> 19:10.715
and Latin a students and pretty large declines,

19:10.715 --> 19:12.515
particularly in the community college sector.

19:12.985 --> 19:14.795
Many of the students who are in the community college sector

19:14.825 --> 19:17.395
tend to be students from working class families,

19:17.595 --> 19:18.675
students who are working full-time.

19:19.215 --> 19:20.915
Um, they might be parents themselves.

19:22.005 --> 19:24.745
And so I think one lesson that was learned,

19:25.005 --> 19:27.905
was the importance of providing, aid to students,

19:28.525 --> 19:33.345
the importance of providing aid to institutions,

19:33.365 --> 19:36.025
but also the importance of wraparound services.

19:36.825 --> 19:39.685
Um, some of the work that, you know,

19:39.685 --> 19:43.365
colleagues like Susanna Martinez, at UCSF are doing,

19:43.425 --> 19:46.885
and some work that I've done, in collaboration with her,

19:46.885 --> 19:49.805
is looking at the role of basic needs resources.

19:49.825 --> 19:52.365
And this is when, colleges, for example,

19:52.585 --> 19:56.325
PO provide access to food pantries where they provide small,

19:56.945 --> 20:00.365
grants to students to help them with housing

20:00.785 --> 20:02.845
or to, you know, repair a car.

20:03.985 --> 20:04.995
What is the role

20:05.575 --> 20:10.195
of basic providing wraparound basic needs services in,

20:10.545 --> 20:12.995
keeping students enrolled, particularly during times

20:12.995 --> 20:16.635
of crisis and decreasing the likelihood of students leaving

20:16.655 --> 20:19.235
for a semester or potentially leaving,

20:19.255 --> 20:20.835
the institutions permanently.

20:21.335 --> 20:22.915
Um, which has implications particularly

20:23.115 --> 20:25.355
'cause many students are taking on debt to attend

20:26.105 --> 20:28.395
with the expectation that they will graduate.

20:28.495 --> 20:32.515
And yet our graduation rates, for students,

20:32.515 --> 20:35.035
for four year institutions after about six years

20:35.135 --> 20:36.755
or around 60%.

20:36.975 --> 20:41.315
So, you know, it's over certainly over half of students,

20:41.415 --> 20:42.995
and this is for four year institutions.

20:42.995 --> 20:45.515
Rates are certainly lower for completion rates when I,

20:45.515 --> 20:46.795
completion rates are certainly lower

20:46.855 --> 20:49.125
for students attending two year institutions.

20:49.185 --> 20:52.165
So, the implications are how do we,

20:52.305 --> 20:53.765
how do we support our students?

20:54.265 --> 20:58.725
How do we support our families, with the recognition

20:58.725 --> 21:00.965
and how do we actually re-enroll students

21:01.025 --> 21:03.205
who might have left, particularly given

21:03.205 --> 21:05.005
that many students leave for fiscal reasons?

21:05.155 --> 21:07.045
What are ways to bring students back

21:07.065 --> 21:09.605
to complete their degree so that they actually,

21:10.045 --> 21:12.725
derive the benefits of higher education?

21:15.145 --> 21:17.565
Um, so that's sort of conclusions implications.

21:17.625 --> 21:20.765
And then finally, areas for future research.

21:20.905 --> 21:24.445
So, I think there's so many areas for future research.

21:25.145 --> 21:28.725
Um, I think some of the areas for future research include,

21:29.185 --> 21:32.965
as I said, the role of basic needs, the role

21:33.025 --> 21:37.365
of structural support, the roles of, you know,

21:37.365 --> 21:39.725
when courses are delivered, how they're delivered.

21:40.105 --> 21:43.805
Um, the, you know, some of the, a project that I did

21:43.805 --> 21:46.365
with a collaborator, Dominique Baker last year was looking

21:46.545 --> 21:50.445
at, political and other kinds of conditions.

21:50.445 --> 21:52.765
We looked at the role of hate crimes and whether

21:52.765 --> 21:57.005
or not we saw changes in where students enrolled in places

21:57.005 --> 21:58.725
that high, higher levels of hate crimes.

21:59.225 --> 22:01.045
Um, and so I think that there are all these really

22:01.445 --> 22:03.885
fascinating questions to be asked in higher ed,

22:04.425 --> 22:08.205
but for me it's how do we support students not just to get

22:08.225 --> 22:11.845
to higher ed, but to get through higher ed, to complete,

22:11.905 --> 22:16.605
to get their degree and to, finally, you know, much

22:16.605 --> 22:20.245
of my work has been moving past, students are making,

22:20.465 --> 22:21.605
you know, bad decisions

22:21.665 --> 22:23.885
or students are just choosing not to complete.

22:24.645 --> 22:26.645
Students exist within structures, right?

22:26.705 --> 22:28.925
So there's certainly agency, but there are structures.

22:29.145 --> 22:30.565
How do we explore the structures?

22:30.585 --> 22:32.525
How do we figure out what's working for whom

22:32.545 --> 22:34.165
and what condition under what conditions?

Faculti AI can make mistakes