Integrated student supports (ISS) are an approach by schools to address students’ barriers to being fully engaged and learning in school. The theory underlying ISS is grounded in child development research, including the recognition that children’s health, social, cognitive, and behavioral development are interconnected. To meet students’ and families’ needs, ISS models typically include partnerships with community organizations that provide wraparound services, including housing, food assistance, medical and behavioral health care, and extended learning opportunities, among other supports. ISS models are often implemented in schools that serve a high number of students from families experiencing low incomes, and can be considered a poverty-mitigation strategy.
The use of ISS approaches—whether informally as a loose set of practices or partnerships, or more formally as part of a defined model—has been increasing over time. The growth of formal approaches and models for delivering ISS—such as Full-Service Community Schools, Communities in Schools, and City Connects—has contributed substantially to the spread of integrated student supports. With this growth has come an increasing number of evaluations, including rigorous studies. However, few studies of ISS approaches have measured the full range of outcomes suggested by the model (such as behavioral health outcomes) and none have yet continued to track youth into adulthood.
To fill these gaps in our knowledge, we used the Social Genome Model—a microsimulation model that can be used as a forecasting tool to examine potential later-in-life outcomes that result from earlier interventions. We asked, “What if all children in families experiencing low incomes attended schools that use an ISS approach from early elementary through high school?” To answer this question, we changed child outcomes in the SGM which an ISS model should improve—such as math and reading scores and mental health—in one life stage (e.g., early elementary school) and then simulated how those changes might affect outcomes in later life stages (e.g., high school graduation). See Figure 1 for a visual depiction of the model and how the simulated changes we made rippled through the lifecourse.
Our simulations indicated that children in families experiencing low incomes (which we defined as income below or just at 200% of the federal poverty level) who experience an ISS educational model in elementary school have modestly larger lifetime incomes, but that children who attend ISS schools through middle school and high school enjoy substantially larger lifetime incomes.
For more details on the Social Genome Model, see below. For more details on how we determined which outcomes to change, and by how much, see the Methods appendix.
The Social Genome Model is a microsimulation model first developed by the Brookings Institution and now a partnership between Brookings, Child Trends, and the Urban Institute. The Social Genome Model allows researchers to simulate changes in the lives of children and examine how early-in-life changes can accumulate into long-term benefits, controlling for earlier life circumstances measured in the model.
The version of the Social Genome Model used to run this simulation was developed by matching individual-level data from two separate datasets (Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 and the National Longitudinal Survey of Youth, 1997 ). This creates a single matched panel data set, with some additional information about circumstances at birth and early childhood estimated using the Early Childhood Longitudinal Study, Birth Cohort (ECLS-B) initiated in 2001. A total of 400,040 people are included in the Social Genome Model following a series of matching and multivariate imputation analyses. The final Social Genome Model dataset is weighted by sex and race/ethnicity to represent the birth cohort in the year 2000.
The Social Genome Model includes important ecological factors (child, family, school, child care, and neighborhood) that affect different kinds of outcomes (academic, social, emotional, health, and relationships).
The simulations are run separately by child race and ethnicity and by sex, recognizing the historic and current social, political, and economic factors that have resulted in different life circumstances and returns to investments for different groups of people. It should be noted, though, that several important subgroups, including Native Americans and Asian Americans, are included in the “White and other” racial category due to small sample sizes.
Figure 1. Visual depiction of ISS simulation
Key Findings
Early exposure to integrated student supports could have modest long-term benefits.
We estimated that students in families experiencing low incomes (that is, whose incomes are below or just at 200% of the federal poverty level) who attend an elementary school that uses an ISS approach—but who don’t necessarily receive any additional supports at school after elementary school—could earn about $25,700 more across their lifetime than students in socioeconomically similar families who do not attend an elementary school that uses an ISS approach. However, we estimated small differences in other outcomes such as high school graduation (a 1 percentage point difference in the percent of students who graduate high school), BA degree attainment (a 2 percentage point difference), and earnings at age 30 (a $703 per year difference in earnings). Nevertheless, these small differences accumulate throughout adulthood to translate into modest increases in lifetime earnings—about 5 or 6 percent increases, depending on race, ethnicity, and gender.
Figure 2. Students who attend an elementary school with ISS are predicted to have higher lifetime earnings
Source: Social Genome Model
Notes: n = 184,323. Lifetime earnings are discounted present values in 2018 dollars; undiscounted lifetime earnings would be about twice as much. The White and other race group includes a small percentage of people who identify as Native American, Asian or Pacific Islander, or more than one race; because of sample size constraints, the SGM cannot estimate simulations separately for this group.
More years of schooling under an ISS approach could have more substantial long-term benefits.
While even some exposure to ISS at school could carry benefits, our simulation found that being enrolled for a longer time in an ISS school was associated with larger benefits later in life for students across racial and ethnic subgroups and genders (although benefits are not the same for all children, a finding we discuss further in the next section). For example, while attending an ISS elementary school could result in $25,700 more in lifetime income, attending through middle school could add an additional $40,800 and attending through high school could add an additional $10,400. Therefore, children who attend schools that use an ISS approach from early elementary school through high school could earn an estimated $76,800 more by age 65 than children who never attend such a school—about a 15 percent increase in lifetime earnings.
Figure 3. Students in families experiencing low incomes who attend more years of schooling with ISS have higher lifetime earnings
Source: Social Genome Model
Notes: n = 184,323. Lifetime earnings are discounted present values in 2018 dollars; undiscounted lifetime earnings would be about twice as much.
Importantly, the increasing returns to more years of schooling with an ISS approach are driven by improved outcomes throughout the transition to adulthood, including educational attainment, income, and mental and physical health. For example, while an estimated 14.4 percent of students from families experiencing low incomes who never attended a school that uses ISS earn a BA, we estimated that 16.4 percent would earn a BA if they attended an elementary school that uses ISS; furthermore, 19.6 percent would earn a BA if they attended an ISS elementary and middle school and 20.3 percent would earn a BA if they received ISS from elementary through high school. BA attainment is predicted to increase among all students as the years of schooling under an ISS approach increases (Figure 4).
Figure 4. Students in families experiencing low incomes who have longer exposure to ISS have higher rates of BA attainment by age 30
Percent of students estimated to earn a BA
Source: Social Genome Model
Notes: n = 184,323. The White and other race group includes a small percentage of people who identify as Native American, Asian or Pacific Islander, or more than one race; because of sample size constraints, the SGM cannot estimate simulations separately for this group.
Students who attend secondary schools that provide ISS could have better educational, health, and economic outcomes in adulthood.
Our simulation showed promising outcomes for Black, Hispanic, and White and other race students in families that experience low incomes—and who also attend ISS schools throughout their secondary school years. Not only are their lifetime earnings and educational attainment higher, but they experience better physical and mental health. Because the ISS model was provided to all students in low-income families, across race/ethnicity and gender subgroups, this simulation did not erase initial differences, such as the large difference in lifetime earnings between White and other race males and all other groups. However, we estimated that all students experience a double-digit percentage increase in lifetime earnings if they attend ISS schools. Figure 5 shows estimated lifetime earnings differences, which range from a 12 percent increase among Hispanic males to a 25 percent increase among Black males.
Figure 5. Students from families who experienced low incomes, and who attend ISS schools from elementary through high school, are predicted to have higher lifetime earnings
Source: Social Genome Model
Notes: n = 184,323. Lifetime earnings are discounted present values in 2018 dollars; undiscounted lifetime earnings would be about twice as much. The White and other race group includes a small percentage of people who identify as Native American, Asian or Pacific Islander, or more than one race; because of sample size constraints, the SGM cannot estimate simulations separately for this group.
Self-reported mental and physical health at age 30, by people from families with low incomes who attended ISS schools throughout secondary school, were also estimated to be substantially better than for those who did not attend ISS schools. The mental health “ratings” of people who attended ISS schools were estimated to be from 19 percent (among Hispanic males) to 135 percent (among Black females) higher than those of people who did not (Figure 6).
Figure 6. Self-rated mental health at age 30 of people from families experiencing low incomes who attended ISS schools is higher than those who did not attend
Percent difference in mental health rating
Source: Social Genome Model
Notes: n = 184,323. The White and other race group includes a small percentage of people who identify as Native American, Asian or Pacific Islander, or more than one race; because of sample size constraints, the SGM cannot estimate simulations separately for this group.
Discussion
These simulations indicate that ISS has the potential to carry long-term benefits well past graduation from high school. We found that students who attended elementary schools that provide ISS were predicted to have somewhat higher earnings across their lifetimes, even if they didn’t receive ISS after 5th grade. But, we also found that having multiple years of ISS after elementary school could have even larger educational, economic, and health benefits. It is unlikely that any single intervention at a single point in a child’s life will be able to change their trajectory significantly without more sustained and significant investment; the results of our simulations underscore the importance of consistent, sustained intervention to change children’s later life outcomes. There are, however, several important considerations for interpreting these findings.
First, as with all simulation modeling, the results of these simulations should not be viewed as causal. While we can project the potential benefits of ISS into adulthood, we cannot say for sure that those benefits are the direct result of attending ISS.
Second, the SGM is configured in a way such that effects tend to be additive (that is, the more early outcomes that we improve, the more later life outcomes also improve; and, the more we improve an earlier outcome, the larger the predicted later-life changes in outcomes). Therefore, our simulated finding that receiving the benefits of ISS across multiple life stages results in better long-term outcomes is a reflection of both the model and of the understanding that investments early in the life course generally enhance later outcomes. Hence, these findings are valuable in giving us a sense of the breadth and magnitude that could result from additional years of attending a supportive school. However, we believe the magnitude of changes in long-term outcomes that resulted from our simulations are not unrealistic. Additionally, the benefits that the SGM simulates and that we have reported in this brief are for a single cohort of people, and do not capture the potential benefits to society more broadly that could come from having more people with better incomes, education, and health.
Third, in the absence of better information on how sub-groups of students have benefitted from ISS interventions in the real world (see the Methods section for more detail on how we incorporated real-world findings into this simulation), we assumed that all people would benefit from ISS similarly. For example, we increased all students’ math scores in elementary school by the same amount (a set fraction of a standard deviation), rather than increasing the math scores of different students by different amounts. For example, we could have increased math scores of students with low scores by a large amount, and increased the math scores of students with average scores by just a little. Given the model behind ISS, in which students’ unique needs are met with different levels and configurations of supports, it is entirely possible—or even likely—that some students would benefit more from ISS than others. However, we did not have sufficient information to allow us to model these differing effects for some students. This has multiple implications, including that any disparities or gaps in outcomes that existed between people of different races or ethnicities and genders are unlikely to close in our simulations.
Even with these limitations, the simulation results presented in this brief indicate that ISS approaches continue to be important to meeting the needs of children, youth, and families. As models are strengthened, their potential to set young people up for success in adulthood will also increase. And, as more ISS initiatives evaluate their approaches, collect additional data on nonacademic outcomes, and provide disaggregated data on how different groups of young people benefit from ISS interventions, we may see additional evidence about positive long-term outcomes in the near future.
Suggested citation: Sacks, V., Moore, K.A., McClay, A., & Piña, G. (2022). Integrated student supports in schools may boost lifetime incomes for students in families with low incomes. Child Trends. https://doi.org/10.56417/2754c1596w
To run a simulation in the SGM, we need to know what we intend to change and by how much we should change it. For our simulation of long-term outcomes from ISS approaches, the “what” was based on outcomes that have been found in real-world evaluations for participants in ISS programs or students at schools that use an ISS approach, and on outcomes that reasonably align with an outcome in the SGM. To answer how much we would change these outcomes, we drew on evaluations of programs that have some elements of an ISS approach (even though they may not have all the characteristics), including ISS approaches, social-emotional learning (SEL) interventions, and meta-analyses of SEL or behavioral programs that targeted similar outcomes as ISS interventions. For early elementary school in particular, there are a limited number of studies for ISS, and available studies typically don’t include nonacademic outcomes, so we supplemented with high-quality or highly rated programs for elementary-aged students that aim to affect social-emotional outcomes that align with an ISS theory of change. This is necessary because, while the theory behind the ISS approach—that student needs will be identified and met through a coordinated system of service provision—implies that students’ physical and mental health would improve, few ISS evaluations include these outcomes.
One additional challenge was that the magnitude of effects found by studies of ISS programs varies considerably, which made it harder to select a reasonable effect size to use in simulations. Therefore, we avoided using effect sizes that appeared to be complete outliers, or we took the more conservative (smaller) of several effect sizes, if applicable. For example, one implementation of Communities in Schools in Wichita, Kansas was found to improve 9th grade standardized math test scores by an impressive 0.8 standard deviations. However, we did not identify other studies of ISS programs—even other implementations of Communities in Schools—that were able to achieve that magnitude of change, so we relied on a small effect size that was more typical of those found in several evaluations.
The target population for this simulation was all children in our model data set who were in families experiencing incomes at or below 200 percent of the federal poverty level when the person was in middle childhood.