Our project was developed on conducting research on urban and rural community school districts. We initially wanted to see if ACT scores were higher in urban communities compared to rural communities. We evaluated several variables that contribute to an individual’s test scores such as: student to teacher ratio, population size, free and reduced lunch enrollment and ethnicity. From this, we strategically chose our two school districts based on similar enrollment rates as well as evaluating the county population of each district so that we had one district represent an urban community and the other represent a rural community.
For the urban community, we chose the Fort Lupton school district which is located in Weld County, District RE-8. For the rural community, we chose the Alamosa school district located in Alamosa County, RE-11J. The ACT is a standardized test for high school achievement and college admissions in the United States. The ACT has historically consisted of four tests: English, Mathematics, Reading, and Science Reasoning. They also added an optional writing portion to the exam (ACT Overview). These portions of the test are individually scored on a scale of 1-36 and a composite overall score is provided which is the average f all four test scores.
The ACT assessment is used to measure high school students’ general educational development and their capability to complete college-level work. It is used to see the student’s readiness for college. The main source of our data was collected from the Colorado Department of Education (CDE). The Colorado Department of Education provides leadership, resources, support, and accountability to the state’s 178 school districts, 1,780 schools, and over 130,000 educators to help them build capacity to meet the needs of the states’ over 840,000 public school students (About the Colorado Department of Education (CDE)).
We also used School Review, which is an online resource through CDE that tracks all performance, human resource and student data. We also obtained direct information from both school districts in addition to our sources listed above. We were able to obtain the percentages of gifted/talented students as well as English Secondary Language (ESL) students from each district for the current 2011-2012 school year. We first started analyzing our data using simple descriptive statistics. We then separated each of our variables and placed them in tables to gain a better understanding of the data that was presented.
By implementing the multivariate approach, we were able to better compare different variables to the ACT scores in each school district. Process The effect of receiving an education in a rural community compared to receiving an education in an urban community was analyzed in this project. In addition to analyzing the communities, we analyzed which variable has the greatest effect on the ACT scores. Our hypothesis states that students living in urban communities will have higher ACT scores compared to students living in rural communities based on the fact that students in urban communities have more resources available to them.
We also feel that student to teacher ratio would have the greatest effect on overall ACT scores. Our data was compiled from the CDE website, School Review, Census Bureau, and school administrators. We strategically chose our two school districts based on county population, with one school district in a rural area and one district in an urban area. We wanted to make sure that both school districts were comparable with enrollment sizes. We compiled all of our information into a raw data sheet. We first used simple descriptive statistics that compared all the specific variables that affect ACT scores.
These two variables were student to teacher ratio and school enrollment. We also ran simple descriptive statistics on the ACT scores for both the Fort Lupton school district and the Alamosa school district. After the descriptive statistics, we chose to conduct a regression analysis on each school district by analyzing three pairs of variables. Our first test compared the relationship between student enrollment and student to teacher ratio for the 2008-2009, 2009-2010, 2010-2011 school years. Then, we compared the relationship of student to teacher ratio to ACT score for the same three school years.
Our last regression analysis showed the relationship between enrollment and ACT score. After the regression analysis, we ran the T-test analyzing the same three pairs of variables as listed above. For our final analysis, we decided to prepare graphs to evaluate the relationship between ethnicity and free/reduced lunch, as well as county population, school enrollment, student to teach ratio, and ACT scores. With the limitations of Excel, we had to compare each school district separately and then use graphs to analyze each district side by side. Analysis of Results
During the analysis of our data results, we were forced to revise our hypothesis so that it was specific enough to support our conclusions. After analyzing our data more in depth, we realized we needed to adjust our hypothesis to include what variable effects the ACT score the most. Our project would best be designed for a program like SPSS because we were trying to compare two variables at same time. Excel is not capable of doing this function. We decided to separate the two school’s variables to the ACT score, and then compare both schools side by side, to see the differences in rural and urban community’s effects.
Our final stated hypothesis is that students in urban communities will have higher ACT scores than students in rural communities because of the resources available to the students. We believe that student to teacher ratio will have the greatest impact on ACT scores. Another issue we ran into was having trouble coding ethnicity and gender in order to run regression tests on those variables. We decided to compare and analyze the variables using graphs. Through the course of evaluations, we used several tests including: regression analysis, descriptive analysis, graphs, and t-test.
Regression Analysis “The regression analysis is used to model the relationship between a response variable and one or more predictor variables” (Regression Analysis, 2006). We found this analysis to be most relatable to our project because it is a tool to use for investigating relationships between variables. Fort Lupton’s first regression analysis was between enrollment and student/teacher ratio. Our R2 = 0. 86032. This shows us that there is strong likelihood that enrollment and student/teacher ratio plays an important factor on how well the students perform on the ACT test.
Alamosa’s regression analysis between enrollment and student/teacher ratio gave us an R2 value of . 09492. This score confirms that there is a strong relevance on how students do on the ACT test. Descriptive Analysis / Graphs Descriptive analysis helped us better characterize our sample populations. Included in the descriptive analysis was the mean, standard error, median, standard deviation, sample variance and range. We mainly used these to compare the two school districts side by side to get a better understanding of all three years (2008-2009, 2009-2010-2010-2011)combined into one analysis.
For example, the mean ACT score for the past three years in Alamosa was 17. 46433, while Fort Lupton’s mean score was 15. 99467. This agrees with our hypothesis. Graphing our data was an easier way for us to use some of our variables because of some of the limitations we had in Excel. For example, for ethnicity, between both school districts we were able to use pie graphs to represent each ethnicity without having to code it. We were also able to use graphs with gender because this would have had to be coded as well.
Graphs were also used to analyze free/reduced lunch in each school district. T-Test “A T-Test assesses whether the means of two groups are statistically different from each other. This analysis is appropriate whenever you want to compare the means of two groups” (Trochim, 2006). With this test we were able to see that our t critical two-tail result was 2. 77644. According to a table of critical value for T, with a degree of freedom of 4 and significance . 05, for it to significant it would need a value of 2. 78. Our t-value’s were large enough to be significant. Conclusion
In conclusion, we found that student to teacher ratio was a significant indicator of how well students performed on the ACT test. The higher the student to teacher ratio, the lower the average ACT score was. Our regression analysis showed that all of our variables held significance in affecting the average ACT score, with student to teacher ratio holding the highest significance. Overall, we found our hypothesis to be incorrect because the average ACT score for the three years compiled was higher in the Alamosa school district than in the Fort Lupton school district.
These results can be explained through various assumptions such as the possibility of Alamosa having more qualified teachers than Fort Lupton and the percentage of gifted and talented students could be higher in Alamosa than Fort Lupton. Also, differences in curriculum, teaching methods, scheduling, and after school programs available between the two schools could all be valid arguments as to why ACT scores were higher in the Alamosa school district. To conclude, our project could validate the recommendation that each school district should improve the student to teacher ratio in an effort to raise the average student ACT score.