Category: x2015

The Open Communities Alliance has developed a tool to help individuals with housing subsidies find better communities and homes. While it is an excellent initiative, the mobility app is not yet as effective as it could be in helping people find homes and communities to move into. Through a housing information event organized with the help of the Open Communities Alliance and the Middletown North End Action Team, our class was able to conduct interviews with local who would potentially find the application useful. The interviews and data from the housing event indicate that while there is a clear interest in information about housing, the Mobility App is not highlighting the information that many individuals are interested in.

One of the largest problems with the app as it is currently designed is that it does not supply information about housing opportunities that app users can actually inquire about renting. The app has good information about opportunity levels and assets surrounding certain addresses, but this only goes so far in helping people find places they may actually live in. The Open Communities Alliance website states, “The Alliance is dedicated to bringing resources to lower opportunity areas and linking to higher opportunity areas people who historically have not had access to them.” To this end, it would be helpful for the housing mobility app to have access to information about specific apartment or houses to rent in higher­ opportunity areas.

People who attended the community housing workshop we held at the North End Action Team on April 15 cited the app’s lack of real listings as one of its most significant flaws. One participant said, “What are these, like, spots you can move?” when ze saw the dots for the neighborhood assets (Participant 10A). The same participant commented that ze liked the security deposit on the Portland, CT listing we showed workshop participants. This suggests a demand for information about specific housing opportunities in addition to the information about neighborhood assets and opportunity levels that the app already provides.

In the next version of the app, it would be helpful to incorporate housing listings from a rental listings website, such as Trulia, and to have these listings represented as points on a map of the area. A comparison of listings of Middletown apartments on Trulia, Realtor.com, and Zillow suggested that Trulia might be the optimal choice for a source of data to use in conjunction with the app. Trulia had more listings overall for Middletown housing opportunities (38 postings, compared to 28 on Realtor.com and 26 on Zillow) and more listings for opportunities under $1000/month (15, compared to 12 on Realtor.com and 9 on Zillow). An advantage of Zillow, however, is that Zillow presents its listings on a map of the area, which is what we propose for the app to do as well. Therefore it might be easier to pull location data from Zillow listings than from Trulia, but this would need further investigation by someone more familiar with the technical demands this would impose.

To make the listings very visible and simultaneously address participants’ concerns about the opacity of the opportunity ­level orange shading, we propose that the opportunity levels for individual listings be represented in the color of the balloon indicating an apartment or house for rent. A really clear redesign would be to completely remove the census tracts, but to color code the available housing markers by census opportunity level. This would allow users of the app to see clearly where the apartment or house was located with respect to familiar streets or neighborhood assets while also benefitting from the information about opportunity levels. Here is what this might look like:

To distinguish the listings from neighborhood assets and to ensure that the listings were the most prominent feature of the app, we propose that they be represented as balloons akin to the ones Google uses. Users could click on the balloons to get more information (opportunity level as well as a link to the posting on Trulia or Zillow) about the house or apartment listing. The neighborhood assets could be dots as they are in the app currently. This change to the app would allow housing listings to come first, which could really help people looking for housing make their decisions based on actual listings and locations without sacrificing the important opportunity­ level data and neighborhood assets.

The feature that was least evident to most participants was the transportation information. In the current format of the app, transportation is limited to a “get directions” link that takes the user to a new Google Maps page. Before it was explained, none of the fourteen participants clicked on or described the presence of this feature. This is a terrible shortcoming of the app, considering how important access to transportation is to most individuals when considering a new home or school for their children. Participants 3, 4, and 14 all referenced a need to understand different aspects of transportation during the process of finding housing despite there already being a section of the tool dedicated to housing. Any redesign of this tool must address this issue, because the inclusion of neighborhood assets or housing offerings are not as enlightening when users can’t find out how accessible they are.

It’s clear that linking out of the application for transportation information isn’t a very obvious feature to many. It would be better to include transportation information directly in the app. This way, it would be easier to use alongside visuals for opportunity level, housing, and neighborhood assets. The first and most simple step is to use a street map for the application instead of the map of census tracts. Being able to visualize a neighborhood’s roads and streets is the first step to understanding how one’s day to day travels will pass. The census tracts give very little in the way of understanding scale, population density, and nearness to highways or public transportation, all of which have an impact on living.

The next truly helpful asset would be to add bus (and other public transportation) routes and stops directly to the app. A good redesign would include an overlay of the different routes and stops on the map. Users should be able to hide or show this feature, just like the neighborhood assets. Also like the neighborhood assets, when clicked, the line would show a text box with the name of the bus or stop, along with a relevant phone number for the area’s CT Transit. This feature would not replace the “get directions” link, but would rather enhance its function. While this tool would not be as effective as the “get directions” link in finding tangible directions, it would provide a quick and broad understanding of how transportation in different neighborhoods compare to each other. This is an important layer of understanding that is not covered in the neighborhood assets or the opportunity index – a neighborhood can have wonderful schools, low crime, and high average incomes, but if life there requires a car, these positives would be of little use to a family without one.

The neighborhood assets feature was by far the most utilized in our housing workshop. Ten of the eleven individuals who clicked on any part of the application without assistance first clicked on neighborhood assets, and all eleven clicked on neighborhood assets at some point. By comparison, five people clicked on the shapes representing census tract opportunity level at some point, and no one clicked on the “get directions” link. This implies two things: That the way in which the neighborhood assets are represented in the tool is highly effective, and that neighborhood assets as a category are interesting to individuals who the app is designed for.

Many of the participants used items that could be labelled as “neighborhood assets” to decide whether or not a potential new home was in a good area. For instance, Participant 5 was interested in living near a community center. Participant thought that adding more points for things like public parks. Other participants wanted to find libraries and hospitals. These all could be listed alongside the current neighborhood assets.

In many ways, individuals were looking to these points as indicators of a good area. This is understandable – it is often crucial to live within a safe walking distance to a school or grocery store, or at least a bus stop to get there. For many, this is a basic necessity when looking for housing. However, this information in itself cannot stand in for the opportunity data. One might assume that high opportunity neighborhoods are more likely to have more neighborhood assets, but a quick examination of the mobility app shows otherwise. For all of the urban centers in Connecticut, the city center tends to have a lower opportunity level and quite a lot more neighborhood assets than the surrounding suburbs. This makes perfect sense, but it means that neighborhood assets alone are not adequate indicators of high-quality neighborhoods. Opportunity levels are also important to show, but their representation must be reworked to improve engagement.

Additionally, a stumbling block to many users of the app was the app’s overall lack of intuitiveness. Many participants in our focus group indicated that they were confused about what to do with what they were seeing in the app.

One specific major issue common across several participants in our housing workshop was that they found it difficult to orient themselves in the map. People said things like “How did I lose High Street?” (Participant 1) and “like where would Main…does it show like streets? Like where would the Y be from there?” (Participant 13), which indicated confusion. By changing the opportunity levels to colors of the balloons, as suggested in our discussion of housing listings earlier, the map would be clearer, and participants would be able to see specific streets and familiar landmarks more clearly. Having features pop up when you hovered the mouse over them instead of when you clicked them would also help with increasing the ease of use and reducing the clutter of the app.

Part of this struggle we observed in some of our participants may be attributable to the digital divide. A majority of the participants in our housing-app test session were not familiar or proficient at using computers (only 43% of our participants could use the search tool unassisted), and this represents a major potential-user population that may need different features to help them use the app. Several participants who were not very familiar with computers indicated that something of this nature would be helpful. One participant said, As Participant 5 said, “I think if something that could explain it for people that don’t know the computers, that would be really helpful.” Other participants indicated that more guidance would help them, making statements like “So what are we looking for?” (Participant 9M) and “So what do I have to put there?” (Participant 11). If people can’t figure out how to use the tool on their own, they will not be able to take advantage of the valuable information the app contains.

A thorough tutorial that explained in detail what information to put where could help self-described “computer-illiterate” people use the app. Ideally, the information in the tutorial would go down the the level of telling people what to click on and where to type in information to help people who were extremely unfamiliar with computers and might not have the same intuition-based abilities to figure this out as more computer-savvy users. This might look something like this: (note: We’d like to turn this into an embedded slideshow)

We believe these suggestions will improve the housing mobility app and increase its utility and ease of use for those most in need of better housing opportunities. With our additions and revisions to the app, people will be able to find specific places to live, get a realistic sense of their transportation options, see enriched neighborhood assets, and understand how to make use of all this information even with limited computer skills. The Housing Mobility App has the power to be a tremendously helpful resource to housing mobility counselors and people looking for better housing opportunities for their families. These suggestions would make the app even more closely aligned with CT Open Communities Alliance’s mission to help people who historically may have not have had access to high opportunities have a chance to succeed.

According to the Connecticut State Department of Education, both magnet and charter schools were founded on the basis of creating environments designed to “reduce racial, ethnic, and economic isolation.” However, of the two, only magnet schools are actually required to maintain a racial balance, whereas it is merely a suggestion for charter schools. In an attempt to reduce racially isolated schools, settlements that arose out of the Sheff vs. O’Neill case include a section outlining specific requirements for magnet schools to be considered desegregated. In our essay, we look to past school enrollment data to reveal the importance of having this defined standard of reduced isolation, and how the presence of this guideline is more effective in racially, ethnically, and economically integrating schools when strictly employed as in magnet schools than when simply encouraged as in charters. The data reveals that every year, the number of racially isolated magnet schools decreases while the number of racially isolated charter schools remains relatively the same, which accounts for the majority of charter schools. Without a specific standard they are required to adhere to, the majority of charter schools are, in terms of Sheff standards, considered racially isolated. This supports the effectiveness of the Sheff standard in encouraging the desegregation of schools. Without the standard, charter schools have a larger proportion of racially isolated student bodies, as compared to magnet schools. Why then, are charters not held to the same racial requirements as magnets are? This is a question we would like to further explore through data analysis and visualization.

Magnets Versus Charters

The magnet schools we have chosen to discuss in this essay are those that participate in the Sheff legislation and specifically abide by the clearly defined standards for environments of reduced isolation. That said, magnet schools are also public schools, but these operate under “a local or regional school district, a regional educational service center, or a cooperative arrangement involving two or more districts” ¹. Magnets have two main goals: to reduce racial, ethnic and economic isolation and to offer curricula that encourage and support educational improvement and achievement ². The Sheff Stipulation and Proposed Order from 2013 went into effect to remedy the Hartford schools’ violation of the Connecticut Constitutional order to reduce the racial, ethnic and socioeconomic isolation ³. Because schools were not effectively desegregating, Sheff proposed benchmarks to enhance incentives and promote reduced isolation. If magnet schools fail to meet these guidelines, they risk losing grant funding ⁴. In 2008, the order required that magnets enroll between 25% and 75% minority students, defined as non-white students ⁵ . In 2013, the range remained the same, but the definition of minority changed to only Black and Hispanic students ⁶. In 2013, Sheff included a section calling for the enrollment of at least 44% Hartford-resident minorities ⁷. Most recently, Sheff has increased that requirement to 47.5% ⁸. Together these two clearly defined and measurable standards function to increase integration, achieving the primary goal of magnet schools.

Charter schools, on the other hand, are public schools “that operate independently of local and regional boards of education” ⁹. One main goal for charter schools is “to reduce racial, ethnic and economic isolation” ¹⁰. With its goals at the forefront, charters establish their own methods and standards for achieving them. Because charters function autonomously, they are not held to any specific standard, as magnets are. This raises concerns for the incentives and accountability of charters and their goal of fostering diverse environments.

Charters abide by the Connecticut State Statute ¹¹, which frequently mentions racial and socioeconomic integration. The first section discusses the approval process for a charter. The document states that the “State Board of Education shall consider the effect of the proposed charter school on the reduction of racial, ethnic and economic isolation in the region in which it is to be located” ¹². To say that the Board will “consider” granting a charter based on its proposal to reduce isolation becomes altogether meaningless when there is no specifically outlined goal for which charters should aim to achieve concerning the diversity in enrollment. Merely considering something does not hold charters accountable to any kind of consequence if they do not successfully reduce isolation. Whether or not a proposed charter contributes to reducing isolated environments, the Board of Education is simply asked to consider it, not to decline the charter on the grounds of lacking a racially and socioeconomically integrated student body.

The next section mentioning reduced isolation discusses it in the context of probation processes. Charters can be placed on probation if they fail to “achieve measurable progress in reducing racial, ethnic and economic isolation” ¹³. The question here is: what qualifies as measurable progress when there is no quantifiable value provided? “Measurable progress” is a subjective term, whereby charters decide on their own ratio of minority to non-minority students. They are then held only to their own personal standard for reduced isolation. However, with the autonomy that they are grounded in, charters can adjust their measure to prevent them from going on probation or having their charter revoked. With the freedom to judge for themselves what is considered “measurable progress” or not, charters would never have to be at risk of losing funding or risk termination because they are in control of their own progress and how that progress is viewed by the Board. The main issue with this language is that while the statute does require charters to actively work towards integration, it provides no specific boundary on which to hold them accountable.

Another section states that applicants, when applying to charters, will receive preference if they “reside in a district in which seventy-five per cent or more of the enrolled students are members of racial or ethnic minorities” ¹⁴. This means that charters will give preference to students coming from homogenous, minority communities. Again, giving preference to these students promotes integration and functions to help those most in need. However, that does not hold them accountable for any particular percentage and it does not consider the effects of racial and socioeconomic imbalance. Similar to the issue of language discussed above, with no quantifiable quota there is no way for these charters to effectively reach the type of student body conducive with high minority achievement.

The Sheff Standard: Old and New

In an attempt to reduce racially isolated schools, Sheff settlements include a section outlining requirements for schools to be considered desegregated. In 2008, Sheff categorized reduced isolation as schools whose minority percentage fell between 25%-75% of total enrollment ¹⁵. Minority was defined as non-white students, including Black, Hispanic, Native American, Asian, and Pacific Islander. Under this standard, about a third of magnets failed to comply. Only two of these non-compliant schools failed to meet reduced isolation standards because its enrollment population comprised of mostly non-minority students – the remaining schools were considered segregated due to an exceedingly high percentage of minority and low-income students. The non-compliant schools described in the latter comprised of anywhere from 75% to almost 90% minority students, a point at which the student body was hyper-segregated with near homogeneity. The frequency of this extremely segregated environment can be attributed to the “niche market approach,” where schools and their administrators hope to target low-income and minority groups to serve populations with the greatest need ¹⁶. By enrolling a larger number of minority students than of white, the schools achieve their goals of reaching more disadvantaged communities. However, as we will discuss later in the essay, this approach proves to be far more detrimental to those disadvantaged students; educating them in a racially and economically segregated environment seems to do more harm than good.

In 2013 Sheff amended its reduced isolation standards to refer solely to students identifying as any part Black or Hispanic ¹⁷. Under this new definition of minority, the number of non-compliant magnets in the 2013-2014 school year was cut almost in half.

When counting only Black and/or Hispanic students towards reduced isolation, 83% of the magnets (52 out of 63 schools) successfully achieved Sheff compliance, a number that significantly increased from its original 66% of compliant schools (41 out of 63 schools) in the 2011-2012 school year, using the previous definition of minority. Looking at the most recent school enrollment data for magnet schools reveals only 5 out of 45 listed schools out of compliance with Sheff standards, which further increases the percentage to 89% compliant (see tables 1 and 2 at the end of this section for the data used in this analysis). The annual enrollment data shows a consistent increase in compliance; by holding magnets to a measurable standard of reduced isolation, schools are held accountable and given real incentives to foster integrated student bodies.

It is important to acknowledge, however, that the way the word “minority” is defined has immense power in its ability to influence future demographics and enrollment standards in schools. Simply changing the definition for reduced isolation had a significant impact on whether or not magnet schools remained compliant with the racial standards set by Sheff. Had the definition remained unchanged, 18 magnet schools would have been noncompliant in 2013-2014 (compared to the 11 noncompliant schools under Sheff’s 2013 standard), showing little to no significant change from past years. Furthermore, of those 18 magnets that failed to meet Sheff’s 2008 standards, 8 of them became compliant under Sheff’s revised standards in 2013 without having to alter their student enrollment. Under the original Sheff standard, defining “minority” as non-white, these schools would not have been in compliance. However, the updated definition of minority automatically puts these schools in compliance without any active effort to adjust racial enrollment. Thus, while an increasing number of schools were considered desegregated, the racial distribution for some schools changed minimally or not at all. Regardless, a significant portion of these newly compliant schools did in fact make an active effort to reduce isolation, reinforcing the idea that this imposed standard of compliance encourages magnet schools to diversify and become compliant with Sheff standards.

Table 1: The spreadsheet above shows the percentages of reduced isolation for magnet schools in the 2011-2012 and 2013-2014 school years. For the 2011-2012 school year, reduced isolation was calculated in terms of nonwhite students, and in 2013-2014 it became only Black and Hispanic students. Reduced isolation is given as the inverse of these calculations. Schools are compliant if their reduced isolation percentage falls between 25% and 75%. (Source: Robert Cotto and Jack Dougherty)

Table 2: The spreadsheet above shows the percentages of reduced isolation for magnet schools in the current 2014-2015 school year. Reduced isolation is given as the inverse of the percent Black and Hispanic students. Schools are compliant if their reduced isolation percentage falls between 25% and 75%. (Source: Barbara Canzonetti)

What If?: Applying Sheff to Charters

While currently, the Sheff reduced isolation standard is only exercised by magnets, this is not out of exclusivity within the legislation, but rather because other types of schools have chosen not to participate. In the Sheff legislation, the state agreed to incorporate a variety of programs, some of which include “charter school initiatives, state technical high schools and vocational agriculture programs, and other new and progressive initiatives,” that could, if they chose, be considered a part of the reduced isolation goal ¹⁸. Participating in this goal would allow the state to count the school in their total number of integrated schools, as well as make the school eligible for funding. Given that charters were designed to promote diversity, such a standard of integration appears to directly align with the mission statement of charter schools. However, charters have yet to act in accordance with the Sheff legislation; instead, they follow their own, more lenient integration guidelines as previously discussed.

Without a concrete and measurable unit for integration, charters can often neglect entirely this goal of diversity within their student bodies. A lack in standard also creates more difficulty in determining which charters are succeeding or failing to create diverse and integrated environments. In order to test how integrated Connecticut Open Choice charter schools were, we applied the new Sheff reduced isolation standard to charter schools from the 2013-2014 school year. We found that only 28% (5 out of 18) would be compliant with reduced isolation, compared to the 89% of magnets currently in compliance (see graph for comparison and table 3 for data). Based on this disparity, we would argue that magnets are far more successful than are charters in creating environments of reduced isolation, a difference that is rooted in the strictly defined integration standard that magnets, but not charters, are held accountable to.

Table 3: The schools highlighted in purple are the only 5 charters that would be compliant with Sheff’s reduced isolation standard. (Source: Jack Dougherty)

So why, then, do charters not adopt the Sheff standards? When our seminar spoke with the authors of A Smarter Charter, Kahlenberg and Potter explained that such a set standard was not in the nature of charters. Placing specific criteria or regulations on charters goes against Albert Shanker’s initial vision of charters as laboratory schools. For Shanker, charter schools were to function as experiments with varying approaches and curricula to discover what works best for all students ¹⁹. Charters do not want to hinder or bind themselves by requiring a standard for racial balance and integration because perhaps it is not the best way to educate its students. An education system built on freedom of exploration strives to hold on to that independence and self-governance. But while they exercise their freedom, charters fail to produce the rich and diverse environments Shanker had hoped would come naturally.

(Note from the instructor: Students were ask to write essays under 2500 words, which would be about here. Evaluators may read as much as they wish, but should not rate the essay beyond this point.)

So What?

What is the big deal about having racial, ethnic, and economic integration in school populations anyways? Integration on all levels plays a strong role in developing better, moral characters, leading to a “dramatic decrease in discriminatory attitudes and prejudices” ²⁰. Research has found that children are at higher “risk of developing stereotypes about racial groups if they live in and are educated in racially isolated settings,” further perpetuating the idea that integration is crucial ²¹.

Not only is integration important for character, but also for the creation of an environment producing higher academic achievement. Research has found that “children from socioeconomically deprived families do better academically when they are integrated with children of higher socioeconomic status and better-educated families” ²². The motivation and ambition of middle- and upper- class peers are contagious; deprived students are able to accelerate their learning through informal interactions with peers ²³. Integrated academic environments also allow low-income students and their families to profit from the presence of middle- and high-income parents, who tend to “volunteer in the classroom, have high standards, hold school officials accountable, apply political pressures to ensure adequate funding, and provide financial support” ²⁴. Increasing diversity is not a zero-sum game; the achievement levels of middle class, white students will not decrease due to the presence of low income, minority students ²⁵. In fact, their achievement will most likely remain the same, and their ability to interact with individuals of diverse backgrounds will improve.

The importance of integrated environments is highlighted by indisputable data regarding the higher academic achievement and increased social awareness of minority students educated in diverse environments. Looking at recent school enrollment data, we have found that a larger number of magnet schools have integrated student bodies than do charters. Should charters be held to a similar, strictly defined standard for creating environments of reduced isolation as magnets are held to Sheff, we strongly believe that they too can have more integrated student bodies.

Introduction

In a seminar at Wesleyan University called “Choice: A Case Study in Education and Entrepreneurship,” students visited various “choice” school fairs and open houses. Choice schools in Connecticut are part of a movement to allow parents and students to choose their school, rather than the only option being the traditional neighborhood public school, and include district-wide open choice programs, interdistrict city-suburban transfer, charter schools, and interdistrict magnet schools. ¹ In Connecticut, the choice movement exploded after the landmark Sheff v. O’Neill ruling in 1996, in which the State Supreme Court ruled that the racial and socioeconomic segregation of Hartford’s school children violated the Connecticut Constitution. However, from the field notes collected at the school fairs and open houses we visited, racial and socioeconomic integration were rarely mentioned by school representatives, parents, or students.

This is not because race and socioeconomic status don’t matter anymore; on the contrary, there is still a great deal of variation in racial balance even in Connecticut choice schools. In this paper, I will examine how racial balance plays out in Hartford-area interdistrict magnet schools based on school theme and school location. I am focusing on magnet schools because these schools are designed to have a specific racial balance in order to achieve the goal of desegregating Connecticut’s public schools, yet demographics play out differently depending on the school, and some schools are struggling to maintain their required desegregation standards.

Desegregating Hartford’s schools: The controversy and the lack of transparency

Why does integration matter? In Connecticut today, there are 40,000 children attending chronically failing schools where most students are far below grade level. At these schools, nearly 90% of students are African American or Hispanic/Latino and come from low-income households, on average. ² Students of color bear the burden of Connecticut’s failing schools, in spite of the American ideal in which all children should have equal opportunity to learn and grow together.

In education reform, integration is a debated concept. Now, charter schools, which do not focus on integration and tend to be very segregated in order to give spots to the children who are most in need. Kahlenberg and Potter critique the devolution of the charter school movement away from founder Albert Shanker’s initial vision to create racially and economically integrated alternatives to traditional public schools; they quote Shanker saying, “children from socioeconomically deprived families do better academically when they are integrated with children of higher socioeconomic status and better-educated families,” and “when children converse, they learn from each other. Placing a child with a large vocabulary next to one with a smaller vocabulary can provide a gain to one without a loss to the other.” ³

On the other hand, frustrated parents argue that the focus on integration forces schools to put their resources into attracting students from whiter, wealthier towns. ⁴ Darien Franco, 2011 graduate of Capital Preparatory Magnet School told me, “I think that the desegregation goal is a bit superficial because what I assume is the whole point is part of an effort to make sure everyone’s getting a similar, quality education. I think whoever wants/needs a spot the most should get it. Of course Hartford schools are going to have a high percentage of black/Latino kids, because that’s who lives in Hartford. I don’t see exactly how sending in non-black/Latino children to a school alleviates any particular issue, other than they’ll be more used to seeing them in everyday life.” ⁵

As a result of Sheff v. O’Neill, in the past ten years, the state has spent $1.4 billion to renovate and build new magnet schools, which are designed as reduced isolation schools that draw students from the city and suburbs. ⁶ Magnet schools in the Hartford area have special themes designed to draw in students from both the city and suburbs, and they are required to have a student body that is 25%-75% racial minority students (newly defined as African American or Hispanic/Latino). While magnet schools are public schools open to all residents of Connecticut and appear to select students randomly based on a lottery, in truth, there are many subtle factors determining which students end up at different schools. Magnet schools have incentives to be academically successful and are required to maintain a racial balance, so they are never truly random their selection of students.

 One important factor in attracting a certain student body is the way schools communicate with parents and prospective students. Our class observed many schools present themselves to parents and children in order to better understand how they choose to communicate. Perhaps surprisingly, school representatives rarely brought up integration or provided information on their school’s demographic statistics or goals, and parents did not usually ask about it either.

Unwrapping Hartford magnet school demographics

The fact that there is little transparency or emphasis on racial balance in schools does not mean demographic factors are unrelated to parents’ choice of schools for their children. In their field notes from the Regional School Choice Office (RSCO) fair on February 7, 2015, Alix Liss and Sara Guernsey observe that “despite the fair being particularly minority heavy in attendance, the individuals looking at the specialized schools, whether performing arts or science based, were predominantly white.” ⁷ Does this hold true for actually enrollment in magnet schools? In order to find out, I assigned a theme to each Hartford-area magnet school and analyzed the demographics of each category. In order to create my categories, I drew on a list compiled by Mira Debs and Jack Dougherty, and looked at the website and mission statement of each school. Though some schools have more than one theme, for the sake of this analysis, I chose the theme I thought was most prominent. I have divided the schools into the following categories: STEM, college prep, career prep, alternative pedagogy, arts, global/international studies, liberal arts, character education, and early childhood only. Using data from the 2014-2015 CSDE Sheff Compliance report from October 1, 2014, I calculated the average percent of black/Latino students in each category. ⁸ I found that large differences in demographics do exist among these schools. The following table shows all of the Hartford-area magnet schools, categories, and demographic data.

Table 1: Hartford-Area Magnet Schools, themes, and 2014-2015 demographic data. Demographic data compiled from CSDE Sheff Compliance Report.

The categories with the highest percentage of minority students were character education* (79.2%), college prep (77.5%), and career prep (77.0%), while the categories with the lowest percentage of minority students were early childhood* (53.1%), STEM (55.8%), arts (56.8%), and liberal arts* (59.0%) (I have marked categories with only one or two schools with an asterisk here and later in this essay; all other categories have at least four schools). The following graph shows these percentages.

Figure 1: Average percent black/Latino students in Hartford-area magnets schools by theme

The observation at the RSCO fair that white families migrated toward the more niche-themed schools makes sense considering the actual enrollment in these schools (although, it is interesting to observe that they cited those families as “predominantly white” even though the schools with the highest percentage of non-minority students are still all over 50% black and Latino). Of course, this data cannot explain whether the reason for these demographic differences is that certain types of schools appeal to certain demographics or that these schools are actively working to attract different demographics due to their philosophies or institutional goals.

However, one way schools might alter their applicant pool is through the location of their school. Because transportation is an issue for many parents and busing for interdistrict schools can be quite complicated and time-consuming, a school far outside the city in the suburbs may be less accessible to many families. In fact, using the data on the Hartford resident students from the 2014-2015 CSDE Sheff Compliance Report, I calculated that magnet schools located in the city of Hartford have an average of 43.7% Hartford resident students enrolled, while only 34.2% of students are Hartford residents in schools located outside of Hartford (in towns such as East Hartford, Bloomfield, Avon, Enfield, Glastonbury, New Britain, Manchester, Rocky Hill, South Windsor, and West Hartford). Balancing Hartford and suburban students is a main goal of magnet schools because Hartford is the second poorest city in the country, and the vast majority of Hartford is black or Latino, with only 15.8% of the city’s population whites of non-Latino background in the 2010 census. ⁹ This caused the extreme segregation of Hartford public schools that fueled the Sheff movement. For race, though, the difference between magnet schools in Hartford and the suburbs is much less: schools in Hartford are on average 65.8% black/Latino, while schools located outside of the city are 62.7% black/Latino. Though a difference still exists, the fact that the numbers are pretty close probably indicates that schools located in the suburbs likely enroll more suburban black/Latino students than schools in Hartford. The following interactive map shows the location of Hartford-area magnet schools and their percentage of minority students:

Direct link to the map above

Based on my analysis, I can conclude that theme of the school is more important than school location in determining the racial demographics of schools. However, when looking at the percentage of Hartford students in each school, location is much more important. Theme may be important too; looking at the table above, we can see that the themes with the highest percentage of minority students, character education*, college prep, and career prep, enroll 48.7%, 41.5%, and 51.2% Hartford students, respectively, while the themes with the lowest percentage of minority students, early childhood*, STEM, and arts, enroll 28.6%, 34.6%, and 37.6% Hartford students respectively. But the themes do not follow the exact same order in terms of their Hartford student population as they do minority student population.

Since location does seem to affect demographics at least to some degree, do schools with themes that enroll fewer minority students and fewer Hartford students tend to locate themselves further outside the city? In order to answer this question, I compiled the addresses of each magnet schools and used Google Maps to calculate their distance from Hartford (I used the location point that Google Maps automatically associates with “Hartford, CT,” which is in the center of the city, so even schools located in Hartford have a “distance from Hartford” that is based on this common centerpoint).

 Based on this analysis, there is not an obvious correlation between school location, theme, and demographics. For example, STEM schools, which on average enroll only 55.8% minority and 34.6% Hartford resident students, are only 4.3 miles from Hartford center, on average. On the other hand, career prep schools, which on average enroll 77.0% minority and 51.2% Hartford resident students, are 6.7 miles from Hartford center, on average. Therefore, there is no clear indication that different themed schools are choosing their location based on the type of students they attract or wish to attract. Of course, this analysis is limited in that it is based on a small number of schools and does not take into account the demographics of different suburbs. Moreover, because many magnet schools were founded relatively recently, many schools have changed location in the past few years or are currently in temporary locations while permanent sites are built.

In conclusion, though school representatives and parents rarely talk about demographics, there are clearly many factors that affect racial composition of schools. In this analysis, I found that theme is a key indicator, while location also has a smaller influence. Further analysis would be necessary to understand why theme affects the racial balance of schools. Clearly, though, as schools develop their themes in order to attract students from Hartford and its suburbs, it is important to keep in mind how different themes relate to school demographics.