Centerline Miles (CLM)
This page provides interim analysis, analysis between the initial report and the final report, on the variable of centerline miles (CLM). Along with the TD population as measured by the 5-year American Community Survey (in table C18130), the CLM variable attempts to capture inherent demand for transportation from one Florida county relative to another. The idea behind including the CLM variable is similar in concept to the use of county square miles in the Trip & Equipment Grant Program’s current allocation methodology: that some counties may have fewer residents (or TD-eligible residents), but that residents in those counties typically have farther to travel for a standard trip, whether it be to a doctor’s appointment, a job, a grocery store, etc.
If this idea behind support for inclusion of the CLM variable is sound, then it would be expected that its inclusion works to the benefit of smaller counties more than it does larger counties (as measured by population). A good indication that inclusion of CLM works to the benefit of smaller counties is the relationship between CLM per capita and population, not the total CLM in a county.
Below is an interactive horizontal bar chart that switches between the total CLM miles and the CLM per capita in each Florida county. To switch between the two charts, click on the “CLM Total” and “CLM per Capita” located near the top of the chart.
Relationship of Population Size to CLM per Capita
Below is a scatterplot that charts the relationship between the total population of each county and the CLM per capita of each county. The overall relationship is one of a strong, reverse exponential relationship between the size of a county and the CLM per capita of the county. In other words, a larger population is strongly predictive of a county having a low CLM per capita, while a smaller population is strongly predictive of a county having a high CLM per capita.
Modeling the Omission of the CLM Variable in Model 1
To visualize the impact of leaving out the CLM variable in a statewide model, below is another scatterplot that charts the relationship between the size of a county’s population and the change in total allocation between Model 1 from the initial report and the same model, but with CLM taken out as a variable and only using TD population as a variable for capturing inherent demand.
|Variable||Model 1 Weight||Model 1 (no CLM) Weight|