| Speaker | Time | Text |
|---|---|---|
| Democracy is always an unfinished creation. | ||
| Democracy is worth dying for. | ||
| Democracy belongs to us all. | ||
| We are here in the sanctuary of democracy. | ||
| Great responsibilities fall once again to the great democracies. | ||
| American democracy is bigger than any one person. | ||
| Freedom and democracy must be constantly guarded and protected. | ||
|
unidentified
|
We are still at our core a democracy. | |
| This is also a massive victory for democracy and for freedom. | ||
|
unidentified
|
Next, a discussion on absenteeism in schools, specifically how students' absence from class affects their futures and that of their teachers. | |
| This nearly hour and 15-minute discussion was hosted by the American Enterprise Institute in Washington, D.C. Good afternoon, everyone, and welcome to the American Enterprise Institute for today's event addressing the attendance crisis, new research on chronic absenteeism since the pandemic. | ||
| I'm Nat Malkus. | ||
| I'm the Deputy Director of Education Policy Studies here at AEI and I'm an affiliate of the James Q. Wilson Program and K-12 Education Studies, also here at AEI. | ||
| Today's event is going to include new research presentations on post-pandemic chronic absenteeism that flow out of AEI's chronic absenteeism research working group, affectionately called CARWIG, which we first met in September of last year. | ||
| And before we get into the research, which I'm excited to see, I want to lay out a little bit of the how and the why behind CARWIG and offer some thanks for those who made it possible. | ||
| First, why the focus on chronic absenteeism? | ||
| You're here, so you probably have some inkling of what I'm going to say, but I think it's useful to put out a little context. | ||
| Chronic absenteeism, of course, is the percentage of students missing at least 10% of the school year. | ||
| It was a challenge pre-pandemic. | ||
| It is an even larger challenge post-pandemic. | ||
| And the data I'm about to show you comes from AEI's Return to Learn tracker, which gathered data on a bunch of pandemic-related issues, school districts across the country, closures and mask mandates, enrollments, and so forth, and switched to chronic absenteeism around 2022 when we realized this was going to be a major aspect of pandemic fallout. | ||
| So here we see in 2018, that's about 14,000 school districts or so. | ||
| And in 2018 and 2019, again, chronic absenteeism was a problem. | ||
| You're not going to see much change when we move to 2019. | ||
| It's within a tenth of a percent. | ||
| About 15% of students were chronically absent. | ||
| That's one in seven K-12 students across the country. | ||
| We're going to skip 2021, 2020 and 2021. | ||
| Those were the pandemic years. | ||
| The data is a little questionable, but by 2022, you can see the damage. | ||
| It got much worse. | ||
| Red is bad on this map. | ||
| You want to be not red. | ||
| In 2022, the data that we have indicates 28.4, 0.5% rounds down to 28 of students. | ||
| So more than one in four were chronically absent, an increase of almost 90%. | ||
| In 2023, things got a little better directionally, but not a huge amount. | ||
| And in 2024, this data is going to be coming out in the paper in hopefully next week, maybe the next one, if we can get the I's dotted and T's crossed. | ||
| In 2024, the rate was 23.5%. | ||
| This is what the national picture looks like. | ||
| So, again, we had a fairly stable rate in 2018-19, a huge jump. | ||
| And then it's coming down in these post-pandemic years, sort of 2023 being the first really post-pandemic year, 2024, of course, following that. | ||
| I see these drops of 3% and 2% in a year as, you know, you could see them as a glass half full, glass half empty. | ||
| The three-point drop after 2022 is good progress. | ||
| However, in 2022, we had Omicron, the highest rate of COVID cases ever. | ||
| And in 2024, we had the slowing rate of decline. | ||
| So one aspect of this is how chronic absenteeism affects different populations. | ||
| So I'm just going to look at district types here. | ||
| And I want to illustrate two ideas at the same time that I think is important to understand. | ||
| We have, there's about 14.5 million kids in each of these buckets in 2019 in gray, 2022 in the orangish, and 2024 in red. | ||
| And what we see is pre-pandemic, students in low-achieving districts had relatively higher chronic absenteeism. | ||
| And we see that happened again, that same pattern in 2022 and 2024. | ||
| So disadvantaged students are affected. | ||
| They are chronically absent more often. | ||
| Full stop. | ||
| However, the nature of the change is more even than that. | ||
| If you look at the proportional difference, how much off the baseline the rise was and the decline was, you see that those in low-achieving districts, they increased about 75%. | ||
| In high-achieving districts, they increased about 95%, give or take, and have dropped a little bit. | ||
| You can see the same sort of proportional increases and decreases by achievement, by student poverty, or by minority percentage. | ||
| The point here is that chronic absenteeism affects disadvantaged students more often, but the rise in chronic absenteeism was an unfortunate tide where all boats rose. | ||
| So when I look at this curve, I have a question that keeps me up at night. | ||
| And that question is: what's the new normal going to be? | ||
| In talking with Todd Rogers, who's a professor at Harvard who's done some work on chronic absenteeism for some time, he brought up Kurt Lewin's theory of organizational change, where organizations can unfreeze their sort of stable habits, change, and then refreeze. | ||
| And I kind of see this trajectory as an illustration of that. | ||
| We had sort of a relatively stable track pre-pandemic, around 15%. | ||
| Again, that's not good, but it was stable. | ||
| And then we had this period of change. | ||
| I know when it happened. | ||
| It happened in mid-March 2020. | ||
| And we've had change with the increase. | ||
| And also, we're still in that zone of change, I think, where there is a decrease that we're seeing. | ||
| The question is, where will that trajectory flatten out? | ||
| And 23.5%, I certainly hope, is not the end. | ||
| But the question is how far we can get that down. | ||
| I think that it takes urgency to get at this because I think that the longer we wait act and the longer chronic absenteeism can become somewhat normalized, the more dangerous a long-term threat it is. | ||
| Last spring and summer, I was doing a lot of work trying to push states, trying to push state leaders, district leaders to set goals and really push on this problem, which I was arguing then and would still argue is one of, if not the most important, issues facing public schools. | ||
| But I got to admit, I felt a little sort of nagging problem, and that is I felt like, well, we're kind of fighting 2024 chronic absenteeism with a 2018 playbook, right? | ||
| The research takes time. | ||
| It just does. | ||
| It takes time to get the data. | ||
| It takes time to analyze it and to publish it. | ||
| But the urgency that I was asking state leaders to act with was something that I thought needed to be supplemented about what do we know about chronic absenteeism that we're seeing today. | ||
| It's elevated. | ||
| A lot more students are seeing it. | ||
| Is it different? | ||
| Is it a different phenomenon? | ||
| These are the questions that I wanted to chase. | ||
| And that is the genesis behind this working group that we put together. | ||
| The working group is basically in three steps. | ||
| And I want to talk briefly about those steps because they entail three groups of folks to thank. | ||
| The first thing was to get some data that we could use, and we wanted broad data. | ||
| So we went to states that had high-quality attendance data at the student level, so not at the school level, or maybe even the student day level, so we could sort of get further and further under the hood. | ||
| And the states were very responsive. | ||
| And today we're going to see data or data were made available from Indiana, Rhode Island, Virginia, North Carolina, Connecticut. | ||
| And subsequently, we were able to fold in some data from Maryland, Texas, and California. | ||
| So a broad set of states, more since, have offered to do additional work. | ||
| So I'm really thankful for those state partners. | ||
| I think state and federal data on education that is high quality is important. | ||
| I want more of it available. | ||
| And so I really appreciate the effort of these state leaders. | ||
| It reflects their urgency. | ||
| The second step was to get some folks who would analyze the data. | ||
| We put out a call to a bunch of professors and some researchers who had done this work before that knew their way around the block. | ||
| And they, too, overwhelmingly were willing to pitch in. | ||
| They showed up for a meeting in September on pretty short notice. | ||
| Put our heads together to see what are the low-hanging fruit that we can do with the data at hand. | ||
| There's lots of questions that take a long time. | ||
| We were looking for the ones where we could wring out the most information quickly enough to have an impact, hopefully, on the next school year. | ||
| So we met in September, got some good ideas together, had a great discussion, and I think a good time. | ||
| And we put together some projects. | ||
| We have nine papers today. | ||
| Eight of those were a direct result of the Carwig project. | ||
| And we have more that are ongoing that will come out. | ||
| The goal there was the third part of this, which is to get the data sharing agreements in place and get this done before the school year was out. | ||
| And we were able to do that. | ||
| So we have eight of those projects, more coming. | ||
| The papers are going to be forthcoming over the next couple of months. | ||
| And we'll certainly have those publicized, but we get a first look at the results at a time when I still say this school year is still going on, even if in some places in the country it's not. | ||
| So I have a lot of appreciation for all the efforts that went into that. | ||
| So a little bit of housekeeping of what to expect today. | ||
| We're going to have three panels. | ||
| So buckle up. | ||
| In each of the three panels, we're going to have three papers. | ||
| And they'll go through those papers quickly, kind of hitting the high points in about 10 minutes each. | ||
| Then we'll have an extended period of Q ⁇ A where our three chairs will ask some questions. | ||
| We'll also have an opportunity for audience Q ⁇ A. | ||
| The audience here, the audience on live stream, the audience on C-SPAN. | ||
| You're all welcome to participate. | ||
| But for those who are not actually here present, if you want to ask a question, you can email those questions. | ||
| And we're going to get this twice, so catch it now, at christopher.robinson at aei.org. | ||
| That's christopher.robinson at aei.org. | ||
| Or on Twitter X with a hashtag hashtag AEI chronic absenteeism. | ||
| We'll have a break between sessions and some light refreshments out in the hall. | ||
| A couple of, again, additional thanks. | ||
| I want to thank Greg Fournier, Christopher Robinson, and my data analyst and co-author Sam Holland for their work on this, making the internal AEI things work. | ||
| The Carwig researchers that participated, including those who are going to present up here today. | ||
| And the state leaders that made the data come in. | ||
| I'd like to thank the James Q. Wilson Center at AEI program at AEI for their support and the Arnold Foundation for support in making this possible. | ||
| And lastly, and just before I let them take the stage, are our three chairs. | ||
| So our first session today is going to be on the question, where does chronic absenteeism stand today? | ||
| And it will be chaired by Ajit Gopalakrishna. | ||
| He's the chief performance officer at the Connecticut State Department of Education. | ||
| He oversees data collection, student assessment, research, and I'm sure much more. | ||
| Ajit served previously for 15 years in the fields of adult education and literacy, managing initiatives on content standards, accountability, assessments, and he was also the GED administrator for the state. | ||
| I'm going to ask in a moment that Ajit and our other chairs, when they come up here, that they introduce the presenters so you can kind of keep track of names and faces. | ||
| Our second session centers on the question, why are students chronically absent? | ||
| And our chair is Liz Cohen. | ||
| Liz is the Vice President of Policy at 50 Can. | ||
| That's the 50 State Campaign for Achievement Now. | ||
| It's a national nonprofit that advocates for high quality education across the country. | ||
| Recently, Liz changed. | ||
| She was the policy director at Future Ed, and Liz has done a bunch of work on K-12 education, including on tutoring, school choice, and chronic absenteeism. | ||
| The last session of the day focuses on the question: what are the consequences of chronic absenteeism today? | ||
| And the chair there is Catherine Falconer. | ||
| Catherine's a visiting scholar at the Johns Hopkins University Bloomberg School of Public Health and a Harkness Fellow in Health Policy and Practice with the Commonwealth Fund. | ||
| Her research focuses on addressing chronic absenteeism in schools as a public health issue. | ||
| She's also the Deputy Director of Health Equity and Inclusion at the UK Health Security Agency, where she leads a team focusing on health education and childcare for high-risk groups. | ||
| So with that, I'm going to turn it over to Ajit and the first three panelists. | ||
| You can come on up and take the stage. | ||
| Thanks. | ||
| Well, thanks, Nat. | ||
| Appreciate that warm introduction. | ||
| As Nat mentioned, Ajit Gopalakrishna, I'm the Chief Performance Officer at the Connecticut State Department of Education. | ||
| Connecticut has a long history of focusing on attendance and chronic absence. | ||
| We've been tracking it since 2012, really, and reporting it and having it part of accountability since 2015. | ||
| So I really applaud Nat and his leadership for focusing on attendance because attendance really is the prerequisite for learning. | ||
| So if kids aren't there, learning becomes that much more challenging. | ||
| And we know that attendance has worsened during the pandemic and not returned to pre-pandemic levels. | ||
| There's a lot we can learn by going under the hood. | ||
| And that's what this first session is really designed to do by going deeper into the patterns of attendance and absenteeism. | ||
| And there are three presenters here, and I'll introduce them momentarily. | ||
| Each presenter will have 10 minutes. | ||
| I know you could go on for a long time, but 10 minutes to just give a high-level summary, and then we'll go into a panel discussion. | ||
| And there will be a chance for people to ask questions as well. | ||
| So please think about what questions you might want to ask. | ||
| So, first, we'll hear from Jacob Kersey, who's the Associate Professor of Education at Texas Tech. | ||
| He will present his findings using data from North Carolina, Texas, and Virginia on the underlying distribution of absences. | ||
| You know, typically, we tend to look at the binary measure or average measures, but obviously, there are distributions underneath them. | ||
| So, he'll share a distributional look and how that has changed over the years. | ||
| Then, we'll move to a team from Wayne State, represented by Jeremy Singer, here, Associate Director of the Detroit Partnership for Education Equity and Research, who, along with the team that includes Sarah Lenhoff, who's a Leonard Kaplan endowed professor, along with Nat and Sam from AEI, who will share their research using data from Rhode Island on day-level absences and what those patterns tell us about the nature of absenteeism. | ||
| And last, we'll hear from Morgan Polakoff, who's a professor of education at the University of Southern California, and his doctoral student who is not here but who was part of the work, Nicolas Pardot, who will share their work using data from North Carolina and Virginia on the predictors of chronic absenteeism and how those have changed pre to post-COVID. | ||
| let's welcome Dr. Jacob Kersey from Texas Tech. | ||
| All right. | ||
| Thank you for the opportunity to be here and speak to you a little bit about the distribution and absenteeism. | ||
| My talk is titled, How Has This Changed Over Time? | ||
| So, just brief history: my first project in graduate school was to look at student absenteeism with Dr. Michael Gottfried at the time, UC Santa Barbara. | ||
| So, it's really nice to be able to kind of zoom out and start looking at some data not only from my home state in Texas, but some other states, North Carolina and Virginia as well. | ||
| So, I don't even have to start with the problem. | ||
| You know why we're here. | ||
| We have seen a spike in chronic absenteeism, as Nat so lovely laid out for us at the national level. | ||
| This is spiked in some states as high as double and higher in some others as well. | ||
| But, what I'm going to spend my time talking to you about today is that common policy metrics like chronic absenteeism or average daily attendance can really skew some under-the-hood dynamics that are worth better understanding as to how we as policymakers or practitioners or interventionists can really think about combating this issue. | ||
| And a quick example is just to look at this graph on the right, where you see that post-pandemic, Texas saw a 40% increase as of 2023, or 60% increase in chronic absenteeism. | ||
| But if you shift this to look at average daily absences, you see that that statistic changes quite dramatically, down to 40%. | ||
| Still high, but very different. | ||
| And the reason that it's different is these two statistics tell us very different things, and I would argue don't give us the complete picture and thus are not sufficient for us to understand the dynamics of absenteeism that we're seeing today. | ||
| So, some motivating questions is just how many students are missing school now is not enough. | ||
| Have we seen just our good attenders having a few bit fewer absences now? | ||
| Are we seeing that some of our more concerning students who are exhibiting high absences before are even more extreme, or has just everyone become more absent? | ||
| And the reason that we ask these questions is because we want to better understand not just how many more students are chronically absent, we also want to understand how educational dynamics and absences dynamics have really changed as a result of the pandemic. | ||
| So, a quick example or some examples I'm going to point to today. | ||
| My colleagues are going to look at some other states, but I'm going to focus on data coming from Texas, Virginia, and North Carolina. | ||
| I'm simply going to look at aggregate levels of absences. | ||
| With Texas, I get a little bit more in-depth looking at some six-week periods, but otherwise, we're looking at just annual absence rates for students and some differences in demographic trends that we've seen. | ||
| So, what have we seen in terms of how absences have shifted looking at data from 2017 to 2023? | ||
| We've seen that overall, all students are more absent on average. | ||
| Not just our high attenders from before, not just our more extreme, problematic previous attendees, but everyone. | ||
| But what I want to highlight for you is it's actually a bit more of a complicated story. | ||
| Not have we overall seen a rightward shift in the amount of overall absences in these three states, we have seen a bit more of a skew, meaning that we are seeing higher amounts and proportions of students who are exhibiting more extreme rates of absences. | ||
| This is mirrored across all three states, but I'm going to point to the distribution for Virginia specifically. | ||
| What you can see here, the red line represents our kind of you know sketchier data years immediately following the pandemic, 2020-2021. | ||
| So, I want to draw your attention to the blue line, which represents absences back 2017 through 2019, and compared to the green line, which is looking at absences 2022-2023. | ||
| And what I first want you to notice is that over there on the left, in A, we're seeing that that blue line is a lot higher than the green line, which simply tells us that we're seeing fewer students with better attendance rates. | ||
| But if you look over there on the left, or on the right, we see over at B that we have that green line then exceeding the blue line overall, which suggests that there are greater proportions of students who are missing school. | ||
| But again, the green curve overall still shifts there to the right, which suggests that across the absenteeism distribution, we're seeing that on average all students are more likely to miss school. | ||
| Now, pulling the data from the three states together, I want to show you something that I found particularly interesting. | ||
| I started this Sankey chart, I love a good Sankey chart, because I wanted to see just for the same groups of students who have attended school both right before the pandemic hit and right after, how were their absenteeism patterns shifting? | ||
| Kind of getting at that question I was asking earlier around: have absenteeism patterns shifted just for our better attenders, our lower attenders? | ||
| Overall, this is just looking at snapshots between 2019 and 2023. | ||
| What I first want to draw your attention to is that 0 to 4 percent group that made up over 50 percent of students attending schools in these three states, meaning that there are about 50 percent of students with pretty good attendance rates in 2019. | ||
| About a moderate amount represented by the orange, and then about 10 percent of students are missing, or 17 percent of students are missing about 10 percent or more of the school year. | ||
| As you can see over there on 2023, that blue bar shifts quite significantly, suggesting, again, that we've seen some of those better attenders now have lower attendance rates. | ||
| But getting to my point earlier, these are snapshots. | ||
| And then I thought, okay, for a Sankey chart, I'm going to have to go adjust these bars, adjust some of my graph code. | ||
| But actually, the distribution stayed roughly the same. | ||
| And so now what I want to show you is the value of being able to follow students longitudinally from those, again, attending school right before the pandemic and right after. | ||
| Prior to the pandemic, we saw that about 73% of students who were chronically absent then were also chronically absent in 2023. | ||
| That makes sense. | ||
| But what you'll notice is those students only make up a proportion of the students who are now chronically absent in 2023. | ||
| They make up just a fraction. | ||
| We see that big chunks of students who were in that well-attending distribution and big chunks of students who were more moderately absent from school are now feeding into these more extreme absence levels. | ||
| And even students who were in that kind of good attendance range, we've now seen fall in the more moderate absence group. | ||
| And this is interesting because, again, that distribution changed even though we're losing some students in some grade levels. | ||
| And so, yeah, absenteeism has changed just about all across the distribution. | ||
| But what this ultimately means, whether you're looking at it for these groups of students longitudinally or overall, that about we've gone from 50% of students in these three states who had some pretty good attendance rates to 30%. | ||
| We have about the same percentage of students who are exhibiting moderate absenteeism rates, but they're made up of some of our who we would expect to have better attendance rates prior to the pandemic. | ||
| And we have increased the bucket of students who we would consider chronically absent or just exhibiting higher rates of absenteeism, going from about 17% prior to 37% now in these three states. | ||
| But of course, these changes were driven not just across the board demographically. | ||
| There were some demographic trends that are worth highlighting. | ||
| Showing you some data from North Carolina, we see that the change in attendance rates prior to the pandemic to now were driven through differences across both race and ethnic lines, but also through some of students receiving some special services from schools or more historically marginalized demographics. | ||
| One in particular I'll draw your attention to across all three states is that students who were experiencing homelessness saw the biggest shift in the amount of days that they were missing school prior to the pandemic. | ||
| And these same patterns are reflected also in Texas and Virginia. | ||
| And then perfectly to tee up Jeremy Singer in his next talk, I was able to look at in Texas data some shift in absence trajectories For the year. | ||
| So, meaning not just each year, but also within a year, both prior to the pandemic and now. | ||
| So, what this graph shows you is just the six weeks periods across the x-axis and then the daily absence rate. | ||
| And what you'll notice is in the pre-pandemic, the blue line, we see a pretty steady increase in what we know from absenteeism research, which is that absences tend to increase daily as you go through the school year. | ||
| But what was interesting is we actually see that absences spike in the fall in the second and third six-week period in Texas at least, suggesting that students could be struggling to reacclimate or became more engaged in school at the very beginning, which I think draws some important attention when we start to think about interventions and policies and how we think about overall absenteeism for students in schools. | ||
| So, what does this mean? | ||
| It means that absences are both more common for everybody, but they are also more extreme. | ||
| We're seeing that not only are we seeing more students crossing that 10% chronic absenteeism threshold, but more students overall are shifting and shifting to more extreme levels. | ||
| There's also signs that there may be greater absences within the beginning of the school year. | ||
| We see that chronic absenteeism really is no longer the concentrated piece. | ||
| We can look at differences in the patterns across all rates of absences. | ||
| We see overall fewer students are exhibiting strong attendance, in particular, recall the big shift in our previous good attender buckets down to the moderate and chronic absence levels. | ||
| And then, we knew this probably before the data, but some of our more historically marginalized groups are some of the ones who have unfortunately experienced the biggest shifts in missing the most school. | ||
| So, thank you very much. | ||
| I'm excited to engage some of the discussion and follow-up. | ||
| Thank you, Dr. Kercy. | ||
| Next up will be Dr. Singer from Wayne State, who will share day-level absence patterns and what that tells us using data from Rhode Island. | ||
| Dr. Singer. | ||
| There you go. | ||
| Hi, everybody. | ||
| Jeremy Singer from Wayne State University in Detroit, Michigan. | ||
| I'm here representing myself, my colleague Sarah Lenhoff. | ||
| Together, we wrote a book called Rethinking Chronic Absenteeism, where we discussed some of the insights from our prior research on the topic. | ||
| And of course, Nat and Sam from AEI, who the four of us worked together on the following presentation and paper on day-level patterns in attendance and absenteeism. | ||
| So, we set out to answer two questions. | ||
| Number one, what can we learn when we look at attendance day to day to day over the course of a school year and also over the course of many school years pre to post-pandemic? | ||
| And in particular, whether these day-level patterns of attendance have changed from pre- to post-pandemic. | ||
| To do that, we were able to work with day-level student attendance data from the state of Rhode Island. | ||
| And you can see reflected on the chart here, Rhode Island's attendance patterns look something like the rest of the nation, where we saw a spike coming out of the or through the COVID-19 pandemic and some recovery, though not to where we were at in pre-pandemic levels. | ||
| And to frame the conversation here, we wanted to note, I guess, two things. | ||
| The first is that we had an initial interest in patterns in attendance because, like Jacob mentioned, when we think about the specific patterns that you observe over the course of a year, that can really inform school leaders and district leaders and policymakers for orienting their interventions or the way that they allocate resources to try and support students. | ||
| They might also reveal some insights on what has changed for students over the course of the pandemic. | ||
| We've heard a lot of theories or stories about why students are missing more school. | ||
| This presentation doesn't answer any of those questions, but the idea is that the patterns might lend or provide a little bit of insight to sort of orient or direct our thinking as we dig deeper into that question. | ||
| And so what I'm going to quickly go through is four different day-level attendance patterns that we wanted to hone in on based on the data. | ||
| The first, which Jacob teed up perfectly, is seasonal absenteeism patterns. | ||
| And what you're looking at here is a graph of attendance month to month in Rhode Island for every school year from 2016-17 through the most recent, sorry, from 20, no, I don't remember, sorry, but 23-24 is our most recent year, so 2015-16, I'm sorry, is the earliest school year. | ||
| I apologize. | ||
| So we've got a lot of years of data. | ||
| And you can see that cluster of blue and gray lines at the bottom represents these pre-pandemic school years. | ||
| So you can see a relatively consistent seasonal pattern. | ||
| Again, like Jacob mentioned, slow increase over time in absences. | ||
| And you can see a relatively similar pattern in the post-pandemic period as well, with the exception of during the years when COVID rates were particularly high, strong spikes in the periods where COVID spikes happened. | ||
| So looking at December and January in particular around Delta and Omicron waves. | ||
| Otherwise, what we're concluding is that on the whole, absence patterns tend to look relatively similar seasonally from pre- to post-pandemic, except at an elevated level because there are higher levels of absences overall. | ||
| Second pattern that we looked at was looking at absences by day of the week. | ||
| So this is probably not a surprise to anyone, but we found that absences were a little bit more likely on Mondays and Fridays. | ||
| So we're sort of talking about this as a weekend effect. | ||
| We also tried to explore some absences around holidays and other planned closures. | ||
| And I don't want to say too much more about that because we're still working on that analysis. | ||
| But we find something similar where there tends to be more absences around those days off from school. | ||
| And so you can see the patterns pre-pandemic in this slide. | ||
| And here's what it looked like in a snapshot of a post-pandemic school year. | ||
| And so you can see there does look like there's a bit of an increased level of absences on Mondays or Fridays relative to the other days of the week. | ||
| But on the whole, the distribution across day of the week is pretty similar to what it looked like pre-pandemic. | ||
| So we're not seeing a drastic spike in the amount of absenteeism concentrated on, say, Fridays or Mondays. | ||
| The third thing we looked at were streaks of absences. | ||
| And so in other words, as you can see in the chart here, the very first bar represents one-day absences. | ||
| So say a student is here today, misses the next day, is back in school the day after. | ||
| And just like before the pandemic, this is from the most recent school year 23-24. | ||
| Just like before the pandemic, at least plurality in this case of absences are from single-day absences. | ||
| But there's also times where students, as you can see, will miss two days in a row, or three days, or four days, or five, or even more than five days. | ||
| And so we wanted to investigate a little bit more closely how this streaks versus single-day absences looked over time. | ||
| What you can see is that, especially in the sort of height of the pandemic, single-day absences as a share of the total number of absences happening dropped substantially. | ||
| And that was replaced, or where we saw a greater increase in the distribution was for these very long absences, so five or more, or greater than five days of absences. | ||
| And we think that this pretty clearly reflects the public health conditions at the time, so both the actual rates of illness and infection from COVID-19 and also potentially public health measures that were in place, things like quarantining policies. | ||
| And so you can see that as we sort of exit that height of the pandemic, the streaks start to subside and we start to return to near pre-pandemic levels of single-day absences. | ||
| If you look closely, you'll notice that we do look like we have slightly elevated levels of two-day absences and slightly higher levels of one-day absences. | ||
| And Sam actually ran some simulations, and what we concluded from this is this has a lot to do with just the fact that there are more absences in general. | ||
| So I wouldn't read too much into where we're at right now. | ||
| We basically concluded that we're looking at more or less what you would expect at the level of absences that are occurring. | ||
| And so we're sort of a return to where we were more or less pre-pandemic. | ||
| The final pattern that we investigated here is what we talk about as acute versus persistent absenteeism. | ||
| So what we mean by this is for some students, their absences happen in a big clump or maybe a few spikes throughout the year, but otherwise they have pretty great attendance. | ||
| That's represented by the graph you can see on the left side. | ||
| So you can see these two example students, one blue and one red, had sharp spikes in absences. | ||
| They missed a lot of school, one of them in September, another one in December. | ||
| That could represent some sort of life change or illness or something like that. | ||
| But then you can see for the rest of the school year, in each month, they didn't miss that much school. | ||
| So that's an example of what one student's attendance patterns might look like. | ||
| On the other hand, you have what we would call persistent absences. | ||
| So on the right side, you can see two more examples. | ||
| And you have students who their attendance rates look relatively similar from month to month. | ||
| So maybe they're missing an extra day here or there from month to month, but on average, they're missing a similar amount of school. | ||
| All of the example students on this slide have missed the same amount of school. | ||
| They're all examples of students, real students in Rhode Island, who missed 15 days of school in this school year. | ||
| So you can see that the actual patterns for students over the course of the year can look different. | ||
| Some have more acute, some have more persistent patterns of absenteeism. | ||
| We looked at whether this acute versus persistent absenteeism changed over time. | ||
| And what we found is resonant with a lot of what we've already talked about in this presentation, which is during the height of the pandemic, we saw some change and then a sort of return to the baseline over time. | ||
| In particular, we saw that absences became more acute, and we think that aligns pretty clearly with the story that we were seeing earlier of spikes around what we would assume to be related to public health circumstances and measures. | ||
| So in other words, we started by thinking about can we get some insight into the various explanations for why attendance rates are lower and absenteeism is higher since the pandemic and why is that persisting. | ||
| And what we've concluded is that the patterns show, yes, some substantial differences, especially at the height of the pandemic, but now relatively similar patterns to what we were seeing before the pandemic. | ||
| So absence rates remain higher. | ||
| Or sorry, yes, absence rates remain higher, absenteeism remains higher, attendance rates remain lower. | ||
| But the patterns from day to day over the course of the school year look remarkably similar to what we were seeing in the pre-pandemic period. | ||
| So we're continuing to dig for answers to understand, but we're hoping that this at least gives some insight into what the shape and the dynamics of absenteeism look like today and help policymakers and school leaders respond. | ||
| Thank you, Dr. Singer. | ||
| Lastly, Dr. Morgan Pollockoff from USC on the predictors of chronic absence and how they've changed using data from North Carolina and Virginia. | ||
| All right, thank you. | ||
| It's very nice to be here. | ||
| I'm presenting on behalf of my student Nick Pardo, who has a very good excuse for not being here, which is he's getting married today. | ||
| So I will allow his absence. | ||
| So Jacob had a really nice slide where he showed demographic differences in absenteeism in North Carolina. | ||
| And I'm going to basically take that slide and make it 10 minutes. | ||
| So I'm going to talk to you about demographic differences in two of the states that he showed, but I'm going to really dive in deep. | ||
| So we look at absenteeism rates and how they vary across student demographic groups and years. | ||
| Which student and school characteristics are most strongly predictive of student absenteeism? | ||
| And I'm not actually going to present data on school characteristics because we don't find super strong school-level relationships. | ||
| It's mostly driven by students themselves. | ||
| Third, to what extent have the demographic predictors changed? | ||
| So, looking at some of those things again that Jacob was showing, and how sensitive are the results to the chosen measure of absenteeism. | ||
| So, most of what I'm going to present is using a measure of chronic absenteeism, but then at the very end, I'll talk about percent of days absent. | ||
| We use data from Virginia, North Carolina as well, and the years that are shown here, and the demographics that are shown there. | ||
| We have the only we have homeless and migrant in North Carolina, but not in Virginia, included in this analysis, and you'll see that those two groups do tend to show up as different from other groups. | ||
| We do some very simple descriptive analyses for the first research question, and then we run some regressions. | ||
| Basically, we dump all these demographic variables in a model and we say, okay, well, which ones are significant predictors of chronic absenteeism, controlling for the other variables in the model. | ||
| And we do this separately pre- and post-COVID and then compare the coefficients to say, well, has the magnitude of the difference, say, between low-income students and non-low-income students changed from pre- to post-COVID. | ||
| So, those are our analyses. | ||
| And at the end, as I said, I'll also talk about the percent of days attended dependent variable. | ||
| So, I'm going to jump into results. | ||
| I'm going to present a lot of different results here and tell you sort of the story of what we observed in the data. | ||
| So, first, we see that chronic absenteeism rates are up sharply in both states and have not fully recovered to pre-COVID levels. | ||
| There's nothing new here because we've already heard that three different times. | ||
| There is substantial variation across student groups and current chronic absenteeism rates. | ||
| So, this is Virginia. | ||
| You might not be able to see the labels of the individual groups, but on the left side of the graph there, you can see the groups with the lowest chronic absenteeism rates are gifted students, Asian students, students in later elementary grades, like two, three, four, and five, and white students. | ||
| So, those groups have the lowest chronic absence rates. | ||
| And on the right side, you can see the groups that have the highest chronic absence rates, and those are mobile students, foster youth, low-income students, 12th graders who have substantially more absences than other grades, students with disabilities. | ||
| So, this is the graph for Virginia. | ||
| This is the graph for North Carolina. | ||
| And without going into detail, I can tell you the order is almost exactly the same. | ||
| So, the demographic groups are very, very similar across the states in terms of their rates of chronic absenteeism. | ||
| We also see homeless, as I mentioned earlier, as the top group there in terms of chronic absenteeism in North Carolina. | ||
| So, we see very consistent patterns between the states. | ||
| When you calculate the chronic absence rates pre-COVID and post-COVID, and then just take the difference, so that just tells you how much chronic absenteeism has increased in percentage point terms, that's what this graph shows. | ||
| And this is for Virginia. | ||
| And so, what you can see is that, for the most part, more disadvantaged groups mostly had larger increases in percentage point terms in chronic absenteeism. | ||
| So, on the right side, you can see the groups that had the largest increases in chronic absenteeism in Virginia are English learners and Hispanic students, mobile students, low-income students, foster youth, and black students. | ||
| On the left side, you can see the groups that had smaller percentage point increases are gifted students and white students, and Asian students as well. | ||
| But interestingly, and this kind of goes back to what Nat said at the beginning, to some extent the opposite is true if instead of taking the difference between the chronic absence rates, you take the ratio. | ||
| So, you sort of consider where the group started, and then you consider where the groups ended. | ||
| And so, if you now are looking at percent increases in chronic absenteeism, so relative to pre-COVID rates, you see that some of the groups that had the lowest chronic absence rates pre-COVID actually had the largest increase. | ||
| So, for instance, Asian students had the largest increase in percent terms. | ||
| Now, in percentage point terms, they still had quite low increase, but that's an interesting contrast. | ||
| And we went back and forth with AEI about whether to focus on one or the other. | ||
| I like percentage point. | ||
| I think Nat and Sam preferred percent. | ||
| And anyway, so it's my paper, so I'm going to mostly talk about percentage points. | ||
| So, when we put all these things in a regression model and we say, okay, well, which are the groups that are significantly different from other groups in terms of chronic absenteeism post-COVID, what we see is mobile students are the largest coefficient there, so they're 18 percentage points more likely to be chronically absent than non-mobile students. | ||
| Low-income students, 12 percentage points more likely than non-low-income students, 12th graders, 4 to 14 percentage points more likely, depending on the comparison grade. | ||
| So, it's like 12th and 11th graders and kindergartners have high absence rates, and all the grades in the middle have lower. | ||
| Gifted students are less likely by seven percentage points than non-gifted students, and students with disabilities are five points more likely to be chronically absent. | ||
| In Virginia, English learner racial gaps are generally quite small once you control for these other variables, once you control for income and other things that I've mentioned. | ||
| In North Carolina, the patterns are very similar, but the magnitudes are a little bit larger. | ||
| So, homeless students and low-income students, and mobile students, more than 10 percentage points more likely. | ||
| Gifted students and foster students, somewhat less likely. | ||
| The foster student one is the, I think, the only coefficient in any of the regressions that's interesting, meaning that it's the opposite of what you might expect. | ||
| And I don't know, I don't study foster students that much, but I'm sure you could probably come up with an explanation about what kinds of families those students are placed in. | ||
| And the racial gaps are slightly larger in North Carolina than in Virginia. | ||
| So, then, as I mentioned, what we do is we look at the magnitude of these coefficients from pre to post, and we say, okay, well, have the gaps gotten wider? | ||
| And by and large, the answer to that question is yes. | ||
| The gaps have gotten either the gaps have gotten wider or things have tended in kind of a worse direction. | ||
| So, looking, for instance, at the right graph there, you can see that the gap between low-income and non-low-income students increased from pre-COVID, the blue bar, to post-COVID, the red bar. | ||
| And the little star tells you that that's a significant increase. | ||
| And similarly, the gifted coefficient was negative before and it became more negative. | ||
| Gifted students are even less likely to be chronically absent post-COVID. | ||
| And the patterns are very similar in North Carolina as well. | ||
| I'll just leave this slide up for one sec. | ||
| And then I'll summarize. | ||
| Looking across those two states, in both states, the low-income and non-low-income gap widens significantly. | ||
| In both states, the gifted, non-gifted gap widens significantly. | ||
| In most racial groups, coefficients became more positive, meaning that they are more likely to be chronically absent relative to white students and relative to Asian students. | ||
| And those changes were almost always significant. | ||
| And just looking at levels, post-COVID, low-income students, mobile students, and homeless students remain by far the most absent subgroups. | ||
| There's a lot of words on this slide. | ||
| If instead we do the logistic regression thing that Sam and Nat wanted me to do and consider the relative increases, the story actually is interestingly different, and I encourage you to take a look at the paper once that becomes public. | ||
| But So, for instance, like black students typically had lower odds of chronic absenteeism than white students, controlling for income and other things pre-COVID, and they have still lower but closer to even odds post-COVID. | ||
| That's just one example of these differences. | ||
| If instead we use the percent of days absent outcome instead of chronically absent indicators, the story is basically the same. | ||
| So, the answer to the question is yes. | ||
| Some interesting facts: percent of daily absence increased from about 5 to 7 percent in Virginia and about 5.5 to 8 percent in North Carolina. | ||
| In percentage point terms, the increases were highest for foster Hispanic EL and mobile students in Virginia and mobile and homeless students in North Carolina. | ||
| And the same group showed up in the regressions as being the most different: mobile students, low-income students, homeless students, and 12th graders. | ||
| So, overall, absenteeism is up. | ||
| It's up for all groups across the board. | ||
| I think that's a very consistent finding among our papers. | ||
| When looking in absolute terms, the most disadvantaged groups are typically more likely to have seen larger increases in chronic absenteeism. | ||
| So, low-income and mobile or homeless students, especially, had large increases. | ||
| Racial gaps are not overly large, controlling for income and other things in the model. | ||
| But in both states, regression-adjusted racial gaps have tended in the wrong direction, meaning that underrepresented minority students are seeing attendance getting worse relative to white and Asian students. | ||
| And the patterns are broadly very similar between the states, but the level of the problem seems, at least in these data as we analyzed it, although I noticed it was different from the percentages on Jacob's graphs, a tiny bit worse in North Carolina. | ||
| Thanks. | ||
| Great, thank you. | ||
| Thank you, everyone. | ||
| You've given us a lot to chew on in 30 minutes here. | ||
| But I'd say that overall it seems that absenteeism is getting worse across the board. | ||
| Any patterns that or any disruptions in traditional or sort of historical patterns that occurred during the pandemic seem to have reverted back to pre-pandemic patterns, if you will. | ||
| And when looking at predictors, it seems like while there's overall increase, the past predictors of historical underperformance, under-engagement just seem to have become more pronounced through the pandemic or after the pandemic. | ||
| So, I guess just to get the ball rolling here, what surprised you in your findings here? | ||
| Like, was there anything in your findings that actually sort of said, huh, that's different, that's not what I expected. | ||
| And so, what surprised you here? | ||
| And anybody can take that one. | ||
| I mean, I'll start. | ||
| I didn't expect to see the shift as much as I saw the shift. | ||
| And again, I think some of the difference is I just was looking kind of like raw descriptive statistics, but I mean, looking at those high attenders from before, there's a big chunk who are now in that moderate rate, which like you could argue, you know, | ||
| I know we have papers later on that are going to talk more about like what does absenteeism mean in terms of what it predicts, but it is a pretty significant shift to think about across these three states that there was such a big chunk of students who were pretty good attenders before, both in snapshots and longitudinally, who are more absent now. | ||
| And so, the implications there, I think, then inspire a lot more papers that you're going to hear about earlier. | ||
| I'll just say I don't know that any of the patterns that we looked at alone were surprising, but when you put them all together, just the fact that the patterns of attendance across the school year do look so similar pre-deposed that there was like no substantial differences, that taken as a whole was pretty surprising to me. | ||
| Still trying to think through what exactly that implies for how we make sense of the problem. | ||
| But at minimum, I think it at least helps provide some suggestive evidence that some of the pet theories that folks have about what might explain these changes, at least on average, don't necessarily hold up because they're not reflected in some of the patterns from day to day or from week to week over the school year. | ||
| And what might be the pet theories you're thinking of, just if you don't mind? | ||
| Well, so I'll try and think, like one that was on the slide, for example, is seeing more common students just taking vacations during the year or skipping school more often to extend their weekends. | ||
| Again, we don't necessarily see those as like the major patterns that are reflected. | ||
| It doesn't invalidate broader shifts that we I mean I think I think it lends creatives to the fact that there's broader disruptions or challenges that students and families are facing or that they're experiencing in school that might explain why the overall trajectory of the patterns look the same even if the sort of level of absences have slightly increased over time. | ||
| I just think that the income gap really was the main driver that showed up over and over again and that swamped the racial differences. | ||
| I think also that income indicator of course is super crude and I think if we had a more detailed measure of students' income I'm sure we would see all along the distribution that more disadvantaged kids would be the ones missing school more. | ||
| We just don't have the data for that. | ||
| Although we do have a hint of that because for instance homeless students are by far the most likely to be chronically absent. | ||
| So that's another thing that I wouldn't say it was surprising but the fact that student level income is a main driver here seems to be really important. | ||
| Thank you. | ||
| And I know the data that you're working with is attendance, whether it be daily, monthly data. | ||
| It doesn't necessarily tell you what's going on underneath why are students absent. | ||
| So the next question, I'll sort of keep it at the data level, which is given what you've done, what you've seen with the data, what advice would you give a district or a school leader or even an SEA at a state level in terms of how they should think about tracking and monitoring data? | ||
| And what are a few things you would suggest to an SEA or a school or district in terms of how they should look at their own data so as to actually make an impact during the year and not just do post-mortems? | ||
| If I can start, if you don't mind. | ||
| I mean the first is pretty consistent evidence over time that students who belong to certain subgroups are more often chronically absent. | ||
| One thing that we didn't talk about too is just students' prior chronic absence status. | ||
| But so if you're a school leader and you're here in May and June, I'm sorry, and you're thinking about what am I going to be doing for next year, start with a list of who your chronically absent students were already this year, who your students who switched schools this year were, who your homeless students or low-income students, and start to think of a plan to support those students now. | ||
| You don't need to wait to find out that they're missing school next year. | ||
| You can anticipate that a lot of them might have challenges. | ||
| The other thing is, we did want to mention that, especially I think that acute versus persistent absence pattern is a really helpful one for practitioners as they're supporting students to think about where and how to allocate resources. | ||
| So if students have like a streak of absences, they should receive a lot of attention right away. | ||
| But if a student missed a ton of school last month and missed basically no school this month, it might be an indicator that they had a one-time issue and they should try and find out what that was to make sure. | ||
| But whereas students who have more persistent absences over the school year might need more attention because there's some sustained issue that isn't being addressed that is continuing to cause them to miss that amount of school. | ||
| So I think those are some of the insights from the there's more for sure in this patterns over the school year when it comes to seasonality and day of the week. | ||
| Schools might be thinking strategically about that. | ||
| I think, yeah, I mean, collect and use better data, sort of big D data and small D data, right? | ||
| So try and understand the rest try and systematically understand the reasons why kids are out. | ||
| You know, I don't know whether that's if the kid is out, you just ask the parent which of these five things is it and try and track that and use that to inform. | ||
| Because really there's a big difference between the kid has an illness and is chronically sick versus the kid is super disengaged and doesn't want to show up. | ||
| And I think also small D data, which is, you know, have the have teachers reach out or have school counselors reach out to kids and try and understand for this particular child what is the factor because your response is going to be quite different if you know you have one reason for absence versus another. | ||
| And I'll just say, you know, another project that we're doing is that I'm doing is we're interviewing families and also teenagers about why they're absent and learning really tremendous things about I think the high levels of disengagement and also illness that seem to be major factors that are that I would guess those are the two main factors that are that are really driving these increases. | ||
| Yeah, so I'm going to start with more work that I've done with our research practice partners in West Texas and then I'll connect that to the data that I think we've seen today. | ||
| So what I have seen in my experience is that more people within the school system are talking to each other about absences. | ||
| So we have absenteeism being talked about in central offices more. | ||
| We have central office talking to principals more about absenteeism data, principals talking to teachers more. | ||
| And so I think connecting that to the data we've seen on the demographic differences in the extremes that we've seen in how much more certain marginalized groups are to be chronically absent or absent more generally post-pandemic, I think it then raises the urgency for these conversations to then talk about and not discount experiences from these groups. | ||
| So one example is we work with one school district that now is actually talking about transportation for the first time. | ||
| Like one of the anecdotes they bring up is a kid is, a family has said, hey, we live across really close to the school. | ||
| So we don't qualify for the bus route that you offer, but we live across a major street. | ||
| And so I'm sorry, I don't feel like when I can't take my kid to school that that kid, that I'm going to let that kid go by themselves or is understandable that the kids don't want to cross this major street with huge speeds and all that good stuff. | ||
| So I think it just connects some of those experiences that we're hearing about with more conversations around absenteeism to then, okay, let's talk about it more and let's triangulate more of what's happening because that's the only way we're going to get to solutions. | ||
| Yeah, I think this notion of the seasonal attendance that you brought up, Dr. Singer, in your paper, I was intrigued by that. | ||
| And you said just a moment ago that you were thinking that maybe it was not related to holidays or vacations as much. | ||
| Did I miss that? | ||
| Well, just that we didn't see, for example, we didn't see that acute absences were more likely since the pandemic. | ||
| So, there aren't like sudden spikes in random instances of taking three or four or five days off or something like that. | ||
| At least it doesn't seem that way. | ||
| Okay, great. | ||
| Appreciate that clarification. | ||
| Let me poke in the grade levels a little bit, because I don't know if all of your papers necessarily got into grade levels, but I know, Morgan, in your work, grade 12 came up, and we've all heard of senioritis and what that entails. | ||
| But I was just curious, and anyone can start off. | ||
| In your work, did you notice any sort of patterns? | ||
| And I was curious on the daily attendance front, too: are the patterns for daily attendance different in any way across grades? | ||
| And because why students might miss school and who is the driving force behind the student getting to school are different across those grades. | ||
| So, I was just curious if you had any further insights to add around grades and the impact of and how absenteeism behaves across the grades. | ||
| Yeah, so and I should have put that graph in the slides, but I didn't. | ||
| But we also, yeah, we looked at those differences by grade level pre and post-COVID, and we do find that the grades, I would say, kindergarten and 12th grade, those are basically the grades where there was more absences before things got worse, right? | ||
| And that I think is, you know, could be a story of disengagement for the 12th graders, probably not a story of disengagement for the kindergartners, but it might be a story of lots of different factors, including transportation and illnesses and things for kindergarteners. | ||
| There's also other things that happen that sometimes I think schools contribute by, for instance, having like a half-day policy where on the Fridays they do the half-day for the whatever, and the parents are like, Well, I'm not going to sell my kid for a half day, that's a waste of their time. | ||
| And so, there are lots of different factors that I think are contributing, but we did see the pattern of, again, the gaps widening consistently. | ||
| You're all renowned researchers. | ||
| You've done a lot of work in this space. | ||
| So, now I'm going to ask you to maybe, and Morgan, you started to do that, is connect this work that you did to other work that you have done. | ||
| How does what you learned here align with other research that you might have done or is ongoing? | ||
| And again, I'm going to ask you to hypothesize a little bit and what, in terms of like what direction might you give in terms of advice for schools and districts in terms of where to focus energies for tackling this work? | ||
| Because I know there are next sessions we'll get into that a little bit, but just given your wealth of knowledge, I just wanted to give you a moment to sort of extend what you've learned here and connect it with some of your other work. | ||
| So, we have one report that came out, I think, last year that was connecting. | ||
| So, I co-direct a nationally representative survey of American households called the Understanding America Study, and we've been tracking folks since COVID and their educational experiences. | ||
| And in that study, that report looked at connections between parent-reported absences and parent reports of kids' mental health or well-being using a validated instrument. | ||
| And we found strong connections between kids who were scoring higher on that mental health scale, meaning they have more concerns about their mental health, and their propensity to be chronically absent. | ||
| And that finding isn't surprising, but nonetheless, I think did validate some of the theories that people have to some extent. | ||
| And we're taking that, and as I just said, we're now moving on to continue to study. | ||
| We just got a new grant to study chronic absenteeism by surveying parents and kids. | ||
| Now, we're going to be surveying teenagers in those households for the first time. | ||
| And we're also interviewing both parents and kids from our sample about the reasons for it. | ||
| And let me tell you, reading these transcripts of the kids and parents who've been selected because they're highly chronically absent is really difficult. | ||
| A lot of these kids have really severe traumas, lots of very legitimate reasons for missing school, really chronic disengagement. | ||
| The school is not serving them well. | ||
| And it's very hard, very hard to read through that. | ||
| So we're trying to understand this over time and longitudinally and get some real insights into the reasons. | ||
| Jeremy, your yeah, so I mean, I think maybe to a bit more positive outlook, I mean, I want to think about there are bright spots that are happening, right? | ||
| So in some of our work, we find that there are some really effective teachers who are driving down absenteeism among students who statistically we would think are more likely to be absent. | ||
| So that's just one example of an in-school factor that is within your control. | ||
| Again, we just looked at it statistically, like we don't have qualitative data from these teachers, although we are collecting it as to what they are doing that seems to be moving the needle a little bit on these students. | ||
| But I also think too, like from our experience, school districts sometimes feel isolated, right? | ||
| And so I think we as researchers, one, what that means is we need to be able to support districts with data if we have access to statewide data to look for, okay, are there similar districts that seem to be bright spots that we can look towards around what they could be doing on the ground? | ||
| Are they hiring teachers with better qualifications, which seems to be the case in Texas? | ||
| Are we having conversations with principals around why it's important that you don't just unenroll your student to hit the chronic absenteeism metric or keep them in school that you hit the graduation metric? | ||
| Like, are we having the conversations as an ecosystem? | ||
| Which I know, Jeremy, that is a lot of your work. | ||
| And I just want to point to, I think there are some bright spots, and we just need to actually tap in and learn a lot from them. | ||
| I'll just add three things if that's okay. | ||
| So first, and this is something that actually is really echoed throughout all of our comments, is focus on the reasons that students are visiting school, not just the rates at which they're missing school. | ||
| You can even see in our presentations, we focused on attendance rates and absence rates because that's the easily accessible data. | ||
| That's also true for schools. | ||
| It's harder to get to the root causes, but that's going to be the best leverage to help resolve issues that students and their families are having is actually understanding the drivers, the root causes. | ||
| So that's one. | ||
| The second is, I would say, focus on fundamentals if you're a school or a district. | ||
| We are doing ongoing work understanding what sort of discrete strategies that schools across the state of Michigan are using, but some of our work in Detroit and elsewhere suggests that you don't necessarily need to focus on new initiatives. | ||
| You might want to focus more on fundamentally the culture and climate of your school, the strong relationships that you can foster, and those excellent academics and teaching. | ||
| And just last thing, if you don't mind, is again to echo what Jacob said. | ||
| Our book is called Rethinking Chronic Absenteeism: Why Schools Can't Solve It Alone. | ||
| So I think the subtitle says it all. | ||
| We think that it's important to bring in other sectors to this conversation because, especially with higher rates across the nation, the resources and strategies that we need to serve families and their students are, of course, what's in school, but also what's outside of them. | ||
| Great, thank you. | ||
| Now it's time for audience questions. | ||
| Anybody have a question in the audience over there? | ||
| Thank you all for your comments. | ||
| They've been really interesting. | ||
| Jacob, I wanted to ask you a little bit more about your Senke chart. | ||
| So I noticed that obviously things are looking worse, but some students who were struggling before the pandemic have actually improved. | ||
| I don't think that that's like cause a causal relationship with the pandemic, but I'm curious, it suggests that there's just like a natural, like sometimes kids become more chronically absent, sometimes less. | ||
| What's kind of the latent chronic absentee changes that we see? | ||
| Like, does it move around a lot? | ||
| Like, can one student be really chronically absent one year and then less the next year? | ||
| What does that pattern look like outside of COVID? | ||
| Yeah, so I mean, I'm not sure entirely if I have an answer to your question, but I guess I would point to, you know, it is variable. | ||
| So if you were to put a Sankey charges year to year to year to year to year, you would see lines doing all sorts of things. | ||
| And so I think it just draws more questions into it, honestly, than it has answers. | ||
| But, you know, when you throw that together, for example, you see differences for different students in one year to the next. | ||
| And I think what that suggests is there's going to be some of these structural factors that are just fundamentally going to be different for students at different points in their educational journey. | ||
| I also think it's important that if you, and I didn't do this, but, or show this, but if you shrink it to just certain grade levels, then you see even more interesting patterns and differences. | ||
| And so, yeah, I think your question raises more questions that, yeah, I'd love to continue to dig into this data and have more conversations about. | ||
| Other questions over there? | ||
| Oh, sorry. | ||
| It's hard to see from the lights in your face. | ||
| Yeah, so I think a couple of you talked about how there are things outside of schools that are influencing chronic absenteeism. | ||
| And I think, so I work in school mental health, so I think mental health in schools is great. | ||
| But Morgan, you talking about how kids who need the help are also not going to school. | ||
| And so a lot of times we kind of are putting a lot of our eggs in the school-based mental health basket, for example. | ||
| But I think some of the other kind of outside of school solutions, I'm just curious what you all have seen in terms of where we're not putting everything upon the schools to fix the problem, but there's partnerships and recognition that it has to be kind of a group effort. | ||
| I'd just like to say you raise a really good point, which is there's sort of a spectrum of you can integrate resources into the school and use the school as an important hub. | ||
| In some cases, you might actually make schools formally responsible for delivering those things, like a school counselor or a social worker or a school-based mental health service. | ||
| On the other hand, you might have completely separate initiatives. | ||
| I think the right answer is probably having some kind of mix, because to your point, you're asking schools to do increasingly more and more, and things that are outside of the core focus of instruction. | ||
| And so some kind of mix of like integrating services and resources while also improving the sort of direct delivery of resources or services to families makes sense. | ||
| I had something else I wanted to say about that, but if you want, did you want to chime in? | ||
| Because you had the well, I was going to give a very unsatisfying answer to that question, which is to say another thing that I wanted to say about what schools do that is not maybe not good, but I think is definitely a contributing factor, which is it is just way easier to be absent now and make up for it. | ||
| So I mentioned in our interview study, we interviewed 40 families, and do you know out of the 40 families how many said it's so easy to make up for being absent? | ||
| 39 of them. | ||
| And then we asked, well, would you like to make it not so easy? | ||
| Like, could we not have Google Classroom? | ||
| And so then you'd have to, like in my day, you had to get a packet and do the work at home. | ||
| And do you know how many of the 40 said that's a good idea? | ||
| Let's make it harder. | ||
| Zero. | ||
| So people like that everything's available online and convenient. | ||
| And then also, I think there is absolutely no question in my mind that doing that, which is well-intentioned, and certainly the pandemic made more places do that, makes it much easier for people to be absent and feel like they can make it up. | ||
| So, just to put a button on what I was saying earlier, I'm sorry to bring you back, but I think you want to be strategic about what changes and improvements to services and things like that can you do outside of schools. | ||
| I mean, obviously, one of the big answers is just greater investment, more resources into some of these things, like transportation and health and housing. | ||
| But in lieu of that, maybe we can reduce some of the administrative barriers, invest more in things like navigators, help people accessing those things. | ||
| I would just add, and not to sound like a cynical academic, but I think if you were to show districts a bunch of data like this, they would be overwhelmed and they would just be paralyzed as to what to do. | ||
| So, I think we like looking at this data a lot, and perhaps at the state level, I think it's interesting to look at this data. | ||
| But there is no bigger way that you can make an impact at a school district than to go at their board meeting and present data in a very accessible way. | ||
| Because school board members, if they are invested in the community, are outside of school. | ||
| They are the ones who can make more structural changes more quickly about things like changing the bus policy of the school, which affects an out-of-school factor around transportation. | ||
| So, that's what I've learned a lot on the ground: to, you know, sorry, put it on researchers' backs to go outside of our computers and talk to a school board, talk to a superintendent, talk to a principal. | ||
| Thank you. | ||
| One more question. | ||
| I think the lady over there had her hand up, actually. | ||
| Thank you. | ||
| That's okay. | ||
| It's more of a comment. | ||
| So, we are the boots on the ground. | ||
| This is the left boot, and I'm the right boot. | ||
| So, appreciate everything that we saw. | ||
| We're like bobbleheads. | ||
| Yes, yes, yes. | ||
| It's great validation that we can bring back. | ||
| We hope that you share the slides out so that we can also be the ones talking. | ||
| We're exhausted. | ||
| We didn't know if it was because we were getting older, but it's because of all that we, the data that you presented. | ||
| And again, I appreciate it. | ||
| You presented it the objective, and we could tell you subjectively it's dead on. | ||
| Thank you. | ||
| I guess our last question. | ||
| Yep. | ||
| I was wondering if you could say anything about the predictive ability. | ||
| Like, could you know something by the end of October about how students are likely to do over the course of the year? | ||
| Or did any of you have access to partial day attendance? | ||
| I know partial day attendance is also highly predictive of other outcomes that we care about. | ||
| So, if you could comment on sort of the predictive ability that you could look at in your data Yeah, so I'd say both looking at district data we use as well as what you saw in Texas, we have found that yes, by the time that you get to around that six-week period, that you can leverage those absences that students are missing to predict whether they are more likely to be absent from school later. | ||
| We also show that in a paper that I did in graduate school back in 2017, looking at fall semester, spring semester, 30 days before next 30 days. | ||
| So, I think they're highly predictive of each other. | ||
| And not only can you use absences from the first period to predict, but you can use common sense and research-based evidence to know which students on your roster are likely to be chronically absent before the first day of school as well. | ||
| Just going back to some of what we were talking about earlier. | ||
| Great. | ||
| I think that concludes our presentation. | ||
| I want to thank the panelists. | ||
| Let's give them a round of applause here. | ||
| Fantastic. | ||
| You find the reasons for absence through relationships. | ||
| I think it absolutely starts with that. | ||
| And yes, early chronic absence does predict later chronic absence. | ||
| I'll just put on my hat as a Connecticut researcher, data person from Connecticut. | ||
| We did look at that and we did actually see that for different modes of learning for in-person, hybrid, and remote, it absolutely does. | ||
| So, again, thank you very much to the panelists, and thank you to the audience for the questions as well. | ||
| C-SPAN's Washington Journal, our live forum inviting you to discuss the latest issues in government, politics, and public policy from Washington and across the country. | ||
| Coming up Tuesday morning, the CEO of the nonprofit Open the Books, John Hart, talks about the efforts of the Department of Government Efficiency and federal spending cuts. | ||
| Then, McCain Institute Executive Director Evelyn Farkas on the latest in efforts to forge a ceasefire between Russia and Ukraine and the future of U.S. involvement. | ||
| C-SPAN's Washington Journal. | ||
| Join in the conversation live at 7 Eastern Tuesday morning on C-SPAN, C-SPAN Now, our free mobile video app, or online at c-SPAN.org. | ||
| Here's a look at what's coming up live Tuesday on the C-SPAN networks. | ||
| On C-SPAN at 10 a.m. Eastern, Education Secretary Linda McMahon testifies before a Senate subcommittee on the Trump administration's budget request for her agency. | ||
| Then at noon, the House of Representatives is back in session to consider several small business-related measures. | ||
| Over on C-SPAN 2, the Senate gavels back in at 10 a.m. Eastern to consider nominees for the defense and state departments. | ||
| And on C-SPAN 3 at 2:30 p.m., the chair of the Securities and Exchange Commission, Paul Atkins, testifies before a Senate Appropriations Subcommittee on the President's 2026 budget request for the SEC. | ||
| The proposal calls for a flat budget of $2.1 billion and aims to reduce nearly 10% of full-time positions. | ||
| You can also watch live on the C-SPAN Now video app or online at c-SPAN.org. | ||
| Mr. President, no doubt about it. | ||
| This is today's historic in many ways. | ||
| The proceedings of the United States Senate are being broadcast to the nation on television for the first time. | ||
| Today, we mark the 39th anniversary of the U.S. Senate's first live television broadcast on C-SPAN 2. | ||
| Join us as senators take to the floor to reflect on this landmark moment in American democracy. | ||
| Thanks to C-SPAN 2, this public service allows our constituents to see the swearing in of newly elected members, watching all-night sessions during Votoramas, and tune in to history being made. | ||
| That's why on its 39th birthday, Senator Grassy and I wanted to highlight how important it is for all television providers, including major streaming services like YouTube, |