Charter schools are created to provide more choices for students and their parents. As a group, charter schools distinguish themselves from traditional school by developing new curricula and trying out different pedagogical methods and practices. Students are offered many choices. Charter schools are not one size fits all, so we are told. By offering a plethora of school choices, parents and students are given more freedom to pursue the curriculum that fit them the most. The belief is that more freedom in school choices would lead to better learning experience, which would translate to better academic performance.

Another rationale for supporting the creation of charter schools is to promote diversity. Unlike traditional public schools, students’ eligibility is not determined by residence. As we know, zip code inequality is a prominent phenomenon in this country. Income inequality is reflected in where people live. It is well documented that schools in wealthy neighborhoods are more resourceful. And that is true even for schools within the same school district. Many charter schools by contrast use a lottery system of sorts to admit students. A random system by definition should any set patterns of stratification. There should be more class equality, and by extension, racial and ethnic diversity in charter schools.

Do charter schools deliver on the promises of promoting intra-school diversity, both racially/ethnically and socioeconomically? Do more diversified charter schools bring about better academic performance?

I gathered the demographic information of all public and charter high schools in San Diego county, where there is a substantial presence of ethnic minorities.

Degree of Segregation

To begin answering this question, I obtained the ethnic percentages through extensive research of the Education Data database for each of the following ethnicities for the 19 charter high schools and 63 public high schools: American Indian Alaska Native, Asian, Black or African American, Filipino, Hispanic or Latino, Native Hawaiian or Pacific Islander, None Reported, Two or More Races, and White. Next, I calculated the standard deviation for the percentages of each ethnicity, keeping the public school data and charter school data separate. Standard deviation, a measure of dispersion of a set of data from its mean (1), is a reliable indicator of how close the ethnic percentages for each school are to each other, providing the degree of segregation for the two types of schools. In this case, a high standard deviation of one ethnicity demonstrates that, overall, many schools exhibit either a highly concentrated or lowly concentrated presence of that ethnicity and a less evenly mixed composition of students of that ethnicity, and vice versa. In my study, I focused on the three nationalities that showed the most variety across schools: Black or African American, Hispanic or Latino, and White. My hypothesis was that there would be more segregation towards these three ethnicities in public schools than in charter schools. What I found was quite surprising:

For the Black or African American ethnicity, the standard deviation in charter high schools is 6.3233 %, while in public schools, it is only 4.9079 %. While the difference is not extreme, this tells us that there is some greater degree of segregation towards Black or African American presence in charter schools compared to non-charter schools. On the other hand, with Hispanic or Latino students, the standard deviation is 20.5948 % in charter high schools but 24.8165 % in public high schools--more segregation towards the Hispanic or Latino ethnicity in public schools than charter schools, the contrary for Blacks or African Americans. Similarly, for White students, there is a standard deviation of 19.3043 % at charter schools and 22.4229 % in public schools, demonstrating higher segregation in public schools than in charter schools for Whites.

In the final analysis, pieces of evidence in favor of my hypothesis include the Hispanic or Latino and White origins, which were more diverse in charter schools than non-charter schools. Contrarily, the Black or African American ethnicity is more greatly segregated by charter schools compared to non-charter schools. In my sample of San Diego, charter high schools’ mission of granting choice effectively fosters more inclusive learning environments for Hispanic and White students. Upon reasoning why this is the case, it is important to note the effect of the spatial factor: although many view segregation as culturally isolating, perhaps these outcomes are merely a result of how closely located the school is to one’s residency and not of intended ostracization conducted by the school’s part: where one resides depends on the socioeconomic status of the individual. To this end, it could be that what we need is more affordable housing across all communities to create more diversified student bodies within the entire education system.

Socioeconomic Status

The next variable I investigated was the socioeconomic status of the student population in the two types of schools. The percentage of students qualified for reduced-price lunch suggests that certain low-income students are eligible for cheaper-priced meals at school. My hypothesis was that students at charter schools come from lower-income families than those at public schools.

For each high school, I collected the proportions of students who were eligible for reduced-price lunch. I then averaged these values for all the charter schools and repeated for all the traditional schools, producing two averages. In charter high schools, 56.495 % of students were eligible for a reduction in lunch price, while in public high schools, 46.6 % of students were. The difference is almost 10 %. However, what appears to be a wide division between charter and public schools is actually not what it seems to be: after conducting a two-sided independent samples t-test comparing the two means, the p-value produced was 0.128, which is not significant at the 5% significance level or even the 10% significance level.

Correlation Between Minority Presence and Socioeconomic Status

To measure socioeconomic status, the percentage of students qualified for reduced-price meals is an insightful proxy because it is most likely the low-income students who would be eligible. The correlation between socioeconomic status and minority presence tells us if schools with higher minority presence tend to have higher reduced-price meal percentages as well or vice versa. The outcome would be whether charter schools or traditional schools demonstrate the stronger correlation between the two variables.

The correlation coefficient r suggests the strength of the linear dependence between two variables: ethnic percentage and percentage of students qualified for reduced-price lunch. For the charter schools, I produced 19 coordinate values, the x-value being the ethnic percentage and the y-value being percentage eligible for reduced-price lunch. For the public schools, I produced 63 coordinate values with the same variables. My hypothesis was that there would be a strong correlation coefficient when relating the presence of ethnic minorities, including Blacks or African Americans and Hispanics or Latinos, to socioeconomic status indicated by lowered meal prices.

In charter schools, there was a strong positive association of 0.702 between percentage of Hispanic students and percentage of reduced price lunch: the greater the presence of Hispanic/Latino students, the greater the percentage of students who qualify for reduced price lunch. Likewise, there was a moderate positive association of 0.524 between percentage of Black students and percentage of reduced price lunch: the greater the presence of Black students, the greater the percentage of students who qualify for reduced price lunch. On the other side, there was a strong negative association of -0.797 between percentage of White students and percentage of reduced price lunch: the greater the presence of White students, the smaller the percentage of students who qualify for reduced price lunch. In public schools, there is a moderate positive association of 0.687 between percentage of Hispanic students and percentage of reduced price lunch: the greater the presence of Hispanic students, the greater the percentage of students who qualify for reduced price lunch. There is a moderate positive association of 0.469 between percentage of White students and percentage of reduced price lunch: the greater the presence of White students, the greater the percentage of students who qualify for reduced price lunch. While the correlation coefficients for charter schools seem more extreme than those for public schools, the significance tests indicate otherwise.

We can effectively conclude that charter schools do not exacerbate the problem of socioeconomic class separation. However, it is just as fair to say that charter schools do not fix the problem either. The significance tests comparing the correlation coefficients relating ethnic proportion of minority and White students to the proportion of students qualified for reduced-price lunch were not significant, indicating that any difference we see in the correlation coefficients between public and charter schools could have been a result of randomness. Therefore, we cannot eliminate the idea that there, surprisingly, may not be a difference between charter and public schools in this aspect. Promoting school choice certainly does not resolve discrepancies due to socioeconomic status, but it does not worsen them either. What we thought to be a by-product of choice of education does not seem to be the case in reality when utilizing percentage of students qualified for reduced price lunch as a representation of socioeconomic status.

Academic Achievement

The last variable to investigate is the output, the academic performance. To measure this variable, I analyzed the CAASPP proficiency test scores, required for all 11th graders to take. While CAASPP only tests one grade level, it does target the undoubtedly most academically rigorous year of high school and tests students who have already accumulated almost three-quarters’ worth of high school education. The test measures both literacy and mathematics, and the standards are met if the student received a score of three or higher on a scale of four. I collected the percentages of students who have met the standards for each school, keeping the charter school data and public high school data separate. Then I calculated the mean for each the charter school percentages and non-charter school percentages.

In terms of literacy, the charter schools averaged 54.67 % of students passing the standards; the public schools averaged 65.09 % of students passing the standards. In terms of mathematics, the charter schools averaged 23.65 % of students passing the standards; the public schools averaged 37.97 % of students passing the standards. I conducted the independent samples t-test to test if the differences between charter and public schools are as a result of randomness. For literacy, the two-sided t-test revealed a p-value of 0.0504. For mathematics, the p-value produced is 0.0034, indicating significant results at the 5 % significant level.

Discussion and Conclusion

Do charter schools in San Diego county help integrate low-income students? The short answer is no. In terms of socioeconomic status and racial/ethnic degree of segregation, traditional schools and charter schools lack any notable differences. We can conclude that this is not a class story nor a race story. However, upon analyzing academic success based on standardized test scores, traditional schools perform better than charter schools. So should parents just send their students to public schools then? Again, not necessarily. Ultimately, it is up to the students and their families to decide which type of school best supports their circumstance.

Citations

Bennet, Coleman. “Definition of Standard Deviation | What Is Standard Deviation ? Standard Deviation Meaning.” The Economic Times, Economic Times, 2018.