"Is There a Correlation between Low Test Scores and High Poverty Rates in Bowling Green/Warren County Public Schools?"
by Casey Button
It was my belief that standardized test scores could be correlated to poverty rates and racial composition of particular schools. However, through my research, which has been provided throughout this report, I have concluded just as the piece entitled “Test Scores and Discipline Vary by the Income of Students” in Updating the Dream, that test scores can be more closely correlated with poverty rates than can be correlated with racial compositions of particular schools.
The following is provided to give a brief overview of the data included in this study:
Tables one and two are general statistics about the Bowling Green City School District and the Warren County School District. Tables three and four break down the poverty and minority percentages in each school. Tables five through eight break down the racial composition in each school, as well as, the percentage of english as a second language (ESL) students. Tables nine through fifteen then provide the standardized test scores for all schools in both the city and county school districts, as well as, district and state averages. An additional table, table sixteen, has been provided to give the drop out rates of each high school and the funding given per student in both the city school district and the county school district. Though all of this information is at your disposal, I have decided to take a closer look at only three schools from each of the two districts.
As an analysis, I chose to observe three elementary schools from the city school district and three elementary schools from the county school district. There are several reasons for this and they are as follows:
1) Each school represents a different poverty level in its district - one having a low-level poverty rate, one having a moderate-level poverty rate, and one having a high-level poverty rate.
2) Generally speaking, how children perform at the elementary school level is a good indication of how they will perform as they advance in their education at the middle school and high school levels.
3) Poverty levels will more than likely remain relatively the same for these students throughout their schooling.
4) Fourth grade test scores were used because once again, how students will perform at a lower grade level is a good indication of how they will perform as they advance their education. Therefore, fifth and sixth grade test scores were not used from the elementary schools in the summary of my research.
Bowling Green Independent School System
The total enrollment of students in the Bowling Green Independent School System for the 2000-2001 school year was 3,469. Of those, 305 students attended Dishman McGinnis Elementary, 183 attended Parker Bennett Elementary and 233 attended T.C. Cherry Elementary. The previous three schools are listed in Table 1, in order of the highest poverty level to a lower poverty level. It should be known that these poverty levels are an estimate based upon the free and reduced lunch percentages at each of the three schools. However, they are an accurate indication of the poverty rates in each school. The test scores correspond accordingly, and can be seen in the last column of Table 1. Hence, the higher the poverty rate, the lower the test scores. One remarkable fact about the test scores of the city school system, is that there was a higher percentage of minorities at Parker Bennett than at Dishman McGinnis. However, as noted in Table 1, Parker Bennett scored higher than Dishman McGinnis on the standardized tests. Therefore, this provides a premiss in support of my conclusion that high poverty rates are more closely correlated to low test scores than are minority rates in schools.
Warren County School System
The total enrollment of students in the Warren County School System for the 2000-2001 school year was 11,044. Of those students, 635 went to Warren County Elementary, 541 went to Bristow Elementary, and 471 went to Rockfield Elementary. The previous three schools are listed in Table 2, in order of the highest poverty level to a lower poverty level. Once again, as with the city schools, these poverty levels are estimated on the basis of free and reduced lunch percentages at each of the three schools. However, as with the city schools, the higher the poverty rate in a particular school, the lower the test scores. As one can see from Table 2, Warren County Elementary School had the highest poverty rate and their test scores reflected that; they had the lowest test scores of all elementary schools in the Warren County School System.
In conclusion, it should be noted that there is a high correlation with poverty rates and low test scores, as the above information would indicate. This is not to say that the teachers employed by the schools with the lower test scores are bad teachers. It is to say, however, that students who attend poverty stricken schools with other poverty stricken students fail to advance themselves in order to perform well on standardized tests. I cannot say why in particular, that poorer students perform worse on standardized tests. A few things could be assumed, such as, the parents of these children do not help them and encourage them to do well in school. However, one cannot know for sure without extensive research on such matters. What has been concluded and previously stated is that the two (low test scores and high poverty rates) have a direct connection.