In DC Schools, Mid-Year Student Pushout Holds Back Achievement

By Alex Peerman
Policy & Advocacy Associate, DC Lawyers for Youth

Changing schools in the middle of the academic year can be challenging for students, families, and their new teachers. Transfer students may have trouble adjusting, experience lower academic achievement, and ultimately be less likely to graduate. These are some of the reasons child advocates are fighting to reduce expulsion rates. In fact, DC public charter schools successfully reduced their expulsion rate by 25% from school year 2012-13 to 2013-14. However, looking only at expulsion rates ignores an important component of school pushout – voluntary withdrawal. Sometimes, when parents withdraw a child from his or her school, it is not truly voluntary, but the result of pressure from school administrators.

For example, school staff may tell a parent that the student is not a good fit for the school; they may suspend the student repeatedly; and, in the most extreme cases, they may threaten that a student will be expelled if the parent does not “voluntarily” withdraw. This can be an appealing option for parents because the student does not get a formal expulsion on his or her record, but it also means that we may underestimate the true scope of school pushout by looking at expulsions alone. This post uses data from the 2012-13 school equity reports to look at school suspensions, school withdrawals, and school entry. This data provides some initial evidence that schools with harsh disciplinary policies are disproportionately pushing students out, which ultimately harms academic progress at the schools who receive those transfer students.

First, the data show that schools enrolling more students from low-income households, as measured by the percent of students receiving free or reduced-price lunch, tend to have a larger percentage of their student body withdraw over the course of the school year. This probably doesn’t come as a big surprise, since low-income families are more likely to experience housing instability, need to move to find work, etc.

Figure 1: High-poverty schools see more of their students
withdraw throughout the school year.[i]

Second, even when controlling for the poverty level of the school, schools with high suspension rates still have higher withdrawal rates. In figure 2, the y axis represents the difference between the withdrawal rate predicted based on the percentage of free or reduced price lunch students and the actual withdrawal rate found in the data.[ii] You can see that predictions using free or reduced lunch alone overestimate the withdrawal rates of low-suspending schools, while they underestimate the withdrawal rates of high-suspending schools.

Figure 2: Controlling for poverty, high-suspending schools still have higher withdrawal rates than low-suspending schools.

 

This is certainly not definitive proof that high suspension rates produce more withdrawals. High-suspending schools may be different in other non-poverty ways that could contribute to high withdrawal rates, like having more special needs students, serving older grade levels, operating in high-crime neighborhoods, or offering weaker academic programs. At the same time, this correlation does suggest that we should take a closer look at the role that exclusionary discipline plays in families’ decisions to withdraw from schools.

Third, we need to remember that each student who leaves a school, whether due to pushout or true voluntary withdrawal, ends up at another school. The equity reports show schools that receive more new students in the middle of the school year produce lower average growth in their students reading and math scores. Again, this is not proof that mid-year student entry causes lower achievement, but it seems likely from a common-sense perspective: teachers have less time to instruct and build relationships with mid-year transfers. If those transfers arrive with above-average academic or behavioral challenges, the burden entering students place on schools would be even greater.

 Here it’s worth noting that, while there is no statistically significant relationship between a school’s sector (DCPS or charter) and its withdrawal rate, there is a large and significant relationship between its sector and its entry rate. That is, DCPS schools are far more likely to receive large numbers of new students throughout the year, a natural consequence of traditional neighborhood schools being “schools of right.” This difference is one that we should keep in mind when comparing outcomes from the two sectors.

Figure 3: Schools that receive high numbers of transfer students tend to have
lower growth in reading and math test scores.

Here are a few key take-aways from the data. Based on the above, it’s clear that mid-year transfers make it tough for schools to do their job. Some transfers are unavoidable, and many more have deep roots in poverty-related causes. Other transfers, however, are probably the result of exclusionary discipline policies and negative school climates.

Reducing those sort of transfers is part of what the Every Student, Every Day Coalition is working for. Research shows that keeping kids engaged in school decreases the likelihood that they end up in the juvenile justice system and increases the chance that they will graduate high school. It’s ultimately about ensuring that every child in the District gets an education that puts them on the path to a great career and a fulfilling life. If you’d like to see our ideas on how to get there, check out the ESED policy platform. And if you want to be a part of this effort, sign up to become a member of the Coalition.


[i] Note that to reduce over-plotting, small amounts of random noise were introduced into both variables in this chart using the “jitter” command in R.

[ii] Readers familiar with statistics may think of the y axis as a residual or error. That is, it represents the difference between the predicted value based on a model and the actual observed value. In this case, the model is a linear regression where mid-year withdrawal rate is predicted using a single variable, the percentage of enrolled students receiving free or reduced-price lunch.


Be the first to comment

Please check your e-mail for a link to activate your account.