Fig. 1
From: Estimating exposure to neighborhood crime by race and ethnicity for public health research
To estimate area-level gestational crime exposure for participant X: (1) Count all reported crimes within 800 m that occurred between X’s last menstrual period and date of delivery. (2) Calculate the area of X’s 800 m buffer within Durham city limits (shaded portion of circle). For participants living ≥800 m from city limits, this is equal to π(800m)2. For X, this is equal to [π(800m)2 − Area(unshaded portion of circle)]. (3) Estimate the population within X’s 800 m buffer contained within Durham city limits (shaded portion of circle). For block group A, this is equal to: \( \left[\left(\frac{Area\left( shaded\ portion\ of\ A\right)}{Area(A)}\right)\ast Population(A)\right] \). Repeat this for block groups B-E, and sum these populations. (4) Dividing the crime count from [1] by the area in [2] results in a measure of crimes per km2. (5) Dividing the crimes per km2 derived in [4] by the population estimated in [3] results in a measure of crimes per 1000 people per km2
To estimate block group level gestational crime exposure for X: (1) Count all reported crimes within the block group of residence, C, that occurred between X’s last menstrual period and date of delivery. (2) Calculate the proportion of C’s area within Durham city limits (1, in this example). (3) Multiply the area of C by the proportion derived in [2]. (4) Multiply the population of C by the proportion derived in [2]. (5) Dividing the crime count from [1] by the area in [3] results in a measure of crimes per km2. (6) Dividing the crimes per km2 derived in [5] by the population estimated in [4] results in a measure of crimes per 1000 people per km2.