Tuesday, November 25, 2014

Factors Influencing Student Outcomes and How They Interact - 14 - Population Density

In response to the feedback received on "Educational Funding and Student Outcomes: The Relationship as Evidenced by State-Level Data," the KASB Research Department is working on a "Part II" which will dig further into other factors that impact student outcomes, and how funding impacts when these other factors are taken into consideration.

In this series of blog posts, I will describe the preliminary correlation analysis comparing these factors (independent variables) with each other in an attempt to show how closely tied to each other they are.  

Today's topic is Population Density

The US Census bureau provides information on the population per square mile by state and year.  This is used as a measure of overall population density for each state.

In terms of other independent variables, I will list the interactions in terms of the strength of the highest correlation between variables observed; Strong (+/- 1.0 to 0.5), Moderate (+/- 0.5 to 0.3), Weak (+/- .03 to 0.1), and None or Very Weak (+/- 0.1 to 0.0).

Strong

  • There is a strong positive correlation between RPP and population per square mile; indicating states with more people per mile have higher cost of living.
  • There is a moderate negative correlation between percent taking the ACT and population per square mile, and a strong positive correlation between percent taking the SAT and population per square mile; indicating that students from less densely populated states are more likely to take the ACT and less likely to take the SAT. 
  • There is a strong positive correlation between teacher salary and population per square mile; indicating states with higher teacher salaries are more densely populated.
  • There is a moderate to strong correlation between school spending and population per square mile; indicating states that spend more per pupil are more densely populated.
  • There is a weak to strong positive correlation between population per square mile and family income; indicating that incomes are lower in less densely populated states. 

Moderate

  • There is a moderate positive correlation between percent of students served under IDEA and population per square mile; indicating that more densely populated states have higher percentages of special education students.  
  • There is a weak to moderate negative correlation between poverty and population per square mile; with higher percents of poverty tied to lower population density.  This implies that there is more poverty in less densely populated states.

Weak

  • There is a weak negative correlation between percent of students eligible for free or reduced-price lunch and population per square mile; indicating that states with lower percents of at-risk students have higher numbers of persons per square mile. 
  • There is a weak negative correlation between population per square mile and percent of white and American Indian or Alaska native students, a weak positive correlation between population per square mile and percent of black students, and no correlation between population per square mile and Hispanic students.  This indicates states with lower percents of white and American Indian or Alaska native students and higher percents of black students have more people per square mile. 
  • There is a weak negative correlation between student teacher ratio and persons per square mile; indicating that state with more student per teacher also tend to be less densely populated.
  • There is a weak positive to no correlation between population density and the percent of the population with at least a high school diploma, and a strong positive correlation between percent of 25 year olds and up with college degrees and population per square mile; indicating more densely populated states tend to have higher percents of the population with bachelors and graduate degrees. 

None or Very Weak

  • There is no correlation between school year (period) and population density.
  • There is no correlation between percent of students participating in programs for English Language Learners and population density.
  • There is no correlation between spending on instruction as a percent and population density.

No comments:

Post a Comment