Neighborhood Selection
The 20 neighborhoods were selected using the randomization method. This method was used to prevent against selection bias and insure against accidental bias. Additionally, the use of randomization helped to reduce the likelihood of chance as a source for difference in outcomes. To reduce selection bias and accidental bias during the selection process, we chose to use a random number generator to select numbers for inclusion in the project. Before generating the number sequence the following parameters were applied to the initial 18 neighborhoods: (1) neighborhoods were numbered numerically; (2) the neighborhood needed to fall within a framework plan that was completed or is currently being worked on by the City; and (3) no more than one neighborhood could be in any one community. Thus, we used a pseudo random number generator. Pseudo random numbers are not random in the way one may expect from dice rolls or lottery tickets. Pseudo random number generators are algorithms that use mathematical formulas or pre-calculated sequences. Using a commercially available random number generator we generated a list of 18 numbers from a list of a possible 50 numbers (of those available based on completed framework plans). From this list we came up with the initial 18 neighborhoods. Following the selection of the initial 18 neighborhoods, two neighborhoods, Ensley and Wahouma, were added using parameters 1 and 2. The selection of these neighborhoods was constrained to the communities in the Eastern Area and Prat Ensley Framework Plans, since these plans will be the first to incorporate the quality of life health report cards. This was also the reason two additional neighborhoods were added.
As, this experiment is not a truly randomized control trial we did not evaluate the neighborhoods selected based on comparability at baseline, thus neighborhoods may differ on education, blight, and other domains describe herein. In truly randomized experiments, it is paramount to ensure that groups are equivalent at the onset, however since the objective of our project is to establish a baseline, we did not exclude neighborhoods based on how comparable they were to other neighborhoods.
As, this experiment is not a truly randomized control trial we did not evaluate the neighborhoods selected based on comparability at baseline, thus neighborhoods may differ on education, blight, and other domains describe herein. In truly randomized experiments, it is paramount to ensure that groups are equivalent at the onset, however since the objective of our project is to establish a baseline, we did not exclude neighborhoods based on how comparable they were to other neighborhoods.