The “Not So Dirty” Dozen, Plus a Few: Ways Hartford Magnet Schools Influence Student Enrollment

Posted on

How do magnet schools influence student enrollment? Kevin Welner previously published “The Dirty Dozen,” which describes ways charter schools influence student enrollment (Welner, 2013). Do magnet schools, which are founded upon different goals, use the same strategies to sway their student body? This essay aims to determine if and how Welner’s “dirty dozen” also apply to magnet schools in Hartford, how they are used differently, and what additional approaches magnets use.

Hartford Public Schools is an “all-choice” program, in which families submit lottery applications to their preferred district school, including charter and magnet schools (Dougherty et al., 2014). Charter schools aim to improve academic achievement, provide for educational innovation, provide vehicles for the reduction of racial, ethnic, and economic isolation, and provide a choice of public education programs for students and parents (Cotto & Feder, 2014). Although charters can have their charter revoked if they do not attract, enroll, and retain low income, minority, emerging bilingual students, and students with disabilities, no standards exist to measure this (Cotto & Feder, 2014). Magnet schools aim to reduce, eliminate, or prevent the racial, ethnic, or economic isolation of public school students, and offer a high-quality curriculum that supports educational improvement (Cotto & Feder, 2014). Magnet schools provide special curricular themes to attract students from both city and suburban districts (Dougherty, 2015). Funds for magnets can be withheld if the student body does not contain between 25% and 75% students of color, as in accordance with the Sheff vs. O’Neil standards (Cotto & Feder, 2014).

Location and Transportation 

Welner argues that charter schools choose their location with the students they intend to serve in mind, taking into account the availability of transportation. The Capitol Region Education Center, CREC coordinates free transportation to magnet schools, even for students attending schools out-of-district (CREC Schools, 2015), whereas charters only guarantee transportation to in-district students (Connecticut State Department of Education, 2015). The wider transportation services offered to students attending magnet schools allows them to attract students from different backgrounds in order to uphold the Sheff standards, while charters are not held to such standards and so do not need to provide transportation to such a wide region

RSCO transportation zone map (Source: Regional School Choice Office)
RSCO transportation zone map (Source: Regional School Choice Office)

This theme is evident in field notes collected by my classmates from attending choice fairs and open houses. IF, at an open house, noted that the rep discussed “how they provide transportation regardless of location.” Additionally, when SS and JC spoke with a student attending a magnet school at the RSCO fair, the student “volunteered the information that buses are free.” AL and SG, also at the RSCO magnet fair, described an area solely for providing information about transportation for families interested in magnet schools in the Hartford area. Thus, at least in Hartford, it appears that magnets are not trying to exclude students based on their location as charters allegedly may do; instead, they are accommodating students from the city and surrounding suburbs.

Although these schools may be trying to accommodate far-away students, location will always play a role in enrollment. Students may opt out of a free but long commute to a magnet in favor of a quick commute to a neighborhood school. This location approach is not necessarily mal-intentioned, though. Charter schools in Hartford likely are not intentionally excluding certain students, but locating themselves in areas of high need.


Welner’s “Mad men” approach argues that charters shape their school’s demographics using advertising and images. Hartford magnets appear to use this approach more to achieve diversity and inclusiveness, while Welner implies that charters may target or deter specific student demographics. AM noted that the students who represented the magnet school he visited were of different races, white and Hispanic. AM described a video shown that featured “an anecdote about a white girl and a black girl who became good friends through school,” and another clip in which a teacher says “not one of my students look alike.”

Source: Hartford Magnet Trinity College Academy

Magnets appear to be emphasizing diversity in their advertising to encourage students of different races to feel comfortable applying, and not simply targeting or deterring students of a certain race. AG and RU observed an elementary magnet distribute brochures in both English and Spanish, and EK, at the RSCO fair, noted that school representatives of magnets sometimes spoke Spanish to Spanish-speaking families. Additionally, the RSCO catalog, which provides information about Hartford’s magnet schools, is available online in Spanish.

LS, at the RSCO fair, observed one booth for a while and “noticed that the presenter’s spiel was different based on who came over to the booth.” He observed that when a black, seemingly lower class family came over, transportation was extensively focused on, while when a white family came, the rep skipped the transportation portion of his spiel and instead focused more on curriculum. Thus, these Hartford magnet schools are not targeting their advertisements for or against a particular group, as Welner implied in “Mad men”, but are tailoring them to fit the needs of each of the many target populations, likely to adhere to the Sheff standards.

Application and Enrollment Requirements 

Welner suggests that charters sometimes place conditions on applications, with lengthy applications and mandatory visits. In Hartford, each charter requires its own application, while there is a single application for all magnets, managed by RSCO, in which up to five magnet schools can be ranked. This lengthy application  appears daunting at first, but only asks for basic information, and rankings of either open choice or magnet programs. Jumoke Academy, a charter school, requires five student essays on top of similar information, and Explorations Charter School requires incoming students to submit two essays. Odyssey Community School, another charter, requires interested students to attend an open house. Requiring essays and attendance at events weeds out unmotivated families with likely lower-achieving kids, and lower-income working parents who find it difficult to attend open houses and help their child write five essays. No such essays or visits are required to apply to any magnet schools. Auditions are not even required for the Hartford arts magnets, making them more accessible to lower income students who cannot afford private lessons. Thus, Hartford charters, with more extensive, more complicated applications that are unique to each school seem to make more use of this “Hooping it up” strategy than magnets with their simpler, single application.

No evidence was found for Welner’s “Illegal and Dicey Practices” approach, in which charters may ask for Social Security cards or birth certificates, even though these cannot be required by law. Nowhere on the RSCO application was there a request for social security numbers or birth certificates. This held true for the easily accessible Hartford charter school applications, indicating that Hartford choice schools may not employ this “dirty” approach.

There is also no apparent evidence of Welner’s “Send us your best” strategy in the RSCO application, in which conditions like needing a certain GPA or prerequisites are placed upon enrollment. Although few charter applications could be accessed because the application period recently closed, Explorations Charter School did employ this tactic, by requiring that students entering the high school not as freshmen already have certain number of credits. Once enrolled, students must pass 80% of their courses and be present 90% of the time. Again, although little data is available, Hartford charters may more readily place conditions upon enrollment.

Steering Away 

Charters also may use Welner’s “bum steer,” in which schools steer away less desirable students. This indirectly appeared in the field notes, but in a more positive way, with magnets emphasizing finding the right fit. IF explained that one magnet “offers a program in which a prospective and/or accepted student can shadow a current student because the school only wants students who want to be there.” EK, at the RSCO fair, observed, “several schools also encouraged parents to have their child experience a day at their school and decide for himself/herself if it would be a good fit.” Rather than the school representatives themselves steering away students who may not be the best match, they encourage students to make this “fit” judgment themselves. Thus, the magnets are, at least in theory, getting the students who are a good fit for the school without using a manipulative approach.

Relatedly, charters may use Welner’s “not in service” approach by simply not readily providing services for a group of higher-needs students. AM, at a magnet open house, explained, “these special arts programs received a lot of airtime when things like special education and ELL programs were skimmed over.” While state law requires schools enrolling more than twenty English language learners (ELL) to offer a school-wide bilingual education program, there is no requirement that choice schools offer such a program if ELL enrollment is under this threshold (Cotto & Feder, 2014). The school’s failure to mention any services for ELL students was possibly an attempt to discourage these students from applying, saving the school from bearing the financial burden of implementing these services. ELL students in Hartford are less likely to attend magnet and charter schools than local public schools (Cotto & Feder, 2014), perhaps due to the “not in service” approach.  However, as discussed above, magnet schools did make efforts to cater to ELL students by advertising in Spanish.

ELL students are the most underrepresented in charter schools, followed by magnet schools. (Source: Cotto & Feder, 2014)
ELL students are the most underrepresented in charter schools, followed by magnet schools. (Source: Cotto & Feder, 2014)
In Hartford, there is the smallest percentage of ELL students in charter schools, followed by magnet schools. (Source: Cotto & Feder, 2014)
In Hartford, there is the smallest percentage of ELL students in charter schools, followed by magnet schools. (Source: Cotto & Feder, 2014)


Welner’s “discipline and punish” tactic involves charters using harsh discipline regimes, leading to high expulsion rates. While data was unavailable for expulsion rates, a feature on CT Mirror compares disciplinary incidents at a given school to the statewide average. Data for the available Hartford magnet schools was perplexing; many schools reported zero disciplinary incidents in 2012-2013, however, even more schools reported numbers higher than the state average, most frequently school policy violations.  This could either indicate great variation in school policies and disciplinary strategies between Hartford magnets, or some magnets not reporting this information. Many charters with accessible data were also above the state average, again with the majority being for policy violations. While the data does not indicate if these incidents led to expulsions, the high numbers suggest that some of Hartford’s choice schools may employ harsh discipline regimes, perhaps in an attempt to remove disruptive students, as Welner suggested.

CT Mirror data for the Sport and Medical Sciences Academy, with the highest number of disciplinary incidents of all Hartford magnet schools for which data was available.
CT Mirror data for the Sport and Medical Sciences Academy, with the highest number of disciplinary incidents of all Hartford magnet schools for which data was available.(Source: CT Mirror)

Approaches with Limited Data 

Welner discusses additional approaches, however limited data for these is available. Such strategies include “flunk or leave”, in which school officials threaten to hold the student back a grade if they remain in the school, and “the fitness test,” in which parents of less successful students are counseled to consider alternative options. It cannot be determined if these apply to Hartford magnets. There is also little data available for Welner’s “going mobile (or not)” strategy, in which charters decide whether or not to backfill the students they lose, either during the year or for higher grades. AG and RU, at a magnet open house, explained that openings in grades other than the school’s starting grade “depended on how many people leave,” suggesting they do replace students the following year. However this is an elementary magnet, so it is not known if magnets backfill in higher grades or allow new students to enter mid-year.

Magnet Themes 

Welner explains that charters decide early on what types of students to cater to when determining what niche the school is designed to fill, such as focusing on rigorous academics or special needs students. Expanding on this approach, magnets more indirectly influence enrollment with their special curricular themes. Hartford magnet themes include arts and humanities, STEM, college prep, vocational, and character education. These themes, which are intended to attract students, may actually attract certain students while repelling others.

STEM schools, for instance, may attract more higher-income, well-educated families who understand the importance of such curriculum. LS, observing a STEM magnet booth at the RSCO fair, described how an Asian mother from an affluent suburb had already researched the school, and asked the representative about specific opportunities and “what the school would offer her son.” In contrast, a black mother from Hartford asked more general and “practical questions, including after school care options and transportation.” Although both expressed interest in the school, the class difference likely explains this difference in questions, and the Hartford mother’s focus on after-school care and transportation suggests that practicality, not special programs, is most important to her family. While the curricular theme may be the selling point for an affluent family, lower-income families may have to settle for a more convenient school. AL and SG, at the RSCO fair, also commented, “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.” Magnet schools serve the lowest percentage of free and reduced lunch eligible students in Hartford (Cotto & Feder, 2014).  Whether or not this holds true for STEM schools, the specialized themes may partly explain why the least low-income students enroll in magnet schools.

Hartford magnet schools enroll the lowest percentage of students eligible for free or reduced price meals. (Source: Cotto & Feder, 2014)
Hartford magnet schools enroll the lowest percentage of students eligible for free or reduced price meals. (Source: Cotto & Feder, 2014)

Magnets also influence enrollment by not offering particular themes. For instance, there are no Hartford magnets with a dual language theme, which also may explain the lower ELL enrollment in Hartford magnets than local public schools (Cotto & Feder, 2014). Perhaps if magnets expanded the breadth of their themes, a wider population will be attracted.

Availability of Information 

Another bias in choice schools not accounted for by Welner is the availability of information about the schools. Information can often be found online on school websites. However, not every family in Hartford owns a computer, or knows how to use one, creating a bias against low-income families and families with illiterate parents, also biasing against likely lower achieving children. Valuable information is also gathered from attending open houses and school fairs that some parents cannot easily attend. HH, at the RSCO fair, noted “some parents expressed concern that they may not be able to make it to the specific info sessions offered by some schools due to the fact that those sessions are normally offered during weekdays.”  Working, lower-income parents likely have less flexible hours, or are unable to afford childcare, and cannot attend open houses to get the important information. Although this likely impacts student enrollment, this is probably unintentional, and thus more “accidental” than “dirty” in biasing access to choice schools.


Many of these approaches mentioned in “The Dirty Dozen” that charter schools may use seem to be less readily employed by Hartford magnet schools. With the magnets’ inclusive approaches in advertising and transportation and a convenient application, for instance, magnets seem to be well-intentioned and trying to create a diverse student body catering to different types of families and students rather than purposely trying to exclude certain types of students. However, the attempts do not always accomplish this inclusiveness, and some factors, like themes and availability of information, may unintentionally leave some students behind. While magnets may advertise in Spanish, for instance, they still enroll a smaller ELL population than the neighborhood schools. This indicates that magnets do not intend to exclude this population, but other factors play a role. If choice schools wish to prevent these “accidental” factors that make certain students less likely to enroll from occurring, these populations must be identified so that choice schools can better recruit these students.

(Note from the instructor: Students were ask to write essays no more than 2500 words, which would be about here. Evaluators may read as much as they wish, but should not rate the essay beyond this point.)


Attend and Apply (2015). Explorations Charter School. Retrieved from Charter School Questions and Answers. (2015).Connecticut State Department of Education. Retrieved from

Cotto, R. & Feder, K. (2014). Choice Watch: Diversity and Access in Connecticut’s School Choice Programs. CT Voices for Children. Retrieved from

Dougherty, J., Zannoni, D., Block, M., & Spirou, S. (2014). Who Chooses in Hartford? Report 1: Statistical analysis of Regional School Choice Office applicants and non-applicants among Hartford resident HPS students in grades 3-7, Spring 2012. Trinity College Digital Repository. Retrieved from

Dougherty, J. (2015). A Vocabulary for Understanding School Choice in Connecticut. Presentation for CSPL 341: Innovation in Education- Choice Seminar.

How to Apply (2015). Odyssey Community School. Retrieved from

Explorations Charter School Application (2015). Retrieved from

IF, SS, JC, AL, SG, AM, EK, LS, HH, AG, RU (2015). Class Field Notes.

Jumoke Academy Application (2015). Retrieved from

RSCO Catalog (English). (2015). The Regional School Office. Retrieved from

RSCO Catalog (Spanish). (2015). The Regional School Office. Retrieved from

RSCO Lottery Application (2015). The Regional School Choice Office. Retrieved from

The RSCO Transportation Zone. (2015). The Regional School Choice Office. Retrieved from

Transportation. (2015).CREC Schools. Retrieved from

Video. (2015).  Hartford Magnet Trinity College Academy. Retrieved from

Welner, K. G. (2013). The Dirty Dozen: How Charter Schools Influence Student Enrollment. Teachers College Record. [online],

Your School (2015). The CT Mirror. Retrieved from

2 thoughts on “The “Not So Dirty” Dozen, Plus a Few: Ways Hartford Magnet Schools Influence Student Enrollment”

  1. Excellent work, Alyssa. You’re taking on an important question – whether magnet schools control their student demographics using similar tactics to charter schools (according to Werner) – and exploring it in a very thorough way. You’re making good use the web-essay medium (I love that you included the video). Your essay is also very well organized, and your conclusions are well supported with evidence. I think a little further reorganization might be beneficial. To begin, I would clearly state Werner’s “dirty dozen” early on in your essay, so the reader knows what to expect. Then, I would split the remainder of your essay into subsections, separating out practices you find evidence that magnet schools do engage in from practices you find that magnet schools do not engage in (possibly with a third section for practices for which you have limited or insufficient evidence and would like to see further research done). That way, a reader can have no confusion about which of Werner’s practices magnet schools do and do not employ. Finally, at the end, summarize which practices magnet schools do and do not engage in. I also think a naïve reader might be confused about what a magnet school actually is, why these schools exist, and how the application process works. All this information is prerequisite to understanding your argument. Also, a brief methods section might be valuable, explaining what RSCO is, and how you collected field notes. Finally, while perhaps beyond the scope of this essay, I would encourage you to give some thought to why schools may try to structure their student body at all. After all, public schools are supposed to serve all students, right? So why is it that specific schools may want to serve only some “types” of students? What is the incentive structure, and what are the social forces, that drive schools to try to reshape their student demographics instead of taking whomever comes? Good luck!

  2. To follow up on our more substantive conversation with your guest evaluator, remember that one quick option would be to summarize Welner’s 12 points near the front of your revised essay, possibly as a list of bullet points or a numbered outline. It’s easy to create either of these with the tools in your WordPress visual editor toolbar.

Comments are closed.