Images from this publication. See all images 10 Free text Figure 1 Hypothetical example of social networking application profile data provided by application users we used age, race, and distance in feet from our location in this analysis. J Med Internet Res. Figure 2 Map of Atlanta gray outline including major interstates black lines and selected major roads dark red lines ; points in the figure represent 79 data collection locations.
Figure 3 Example of the GeoLocation app used in this study, which relies on cell towers and an Internet connection and provides latitude and longitude in decimal degrees. Figure 4 Map of Atlanta showing 79 data collection points from profiles on a sex-seeking networking app; radii of yellow circles represent distance to user sample at the maximum distance from the sample point, and overlapping circles completely cover Atlanta, with smaller circular areas used for data collection where there were the largest numbers of application users.
Figure 5 Histogram showing distribution of observed density of social network application users per 1-mile circle for the 79 sampled locations in Atlanta inset includes statistics for the distribution, which is highly skewed with SD estimated to be larger than the mean; numbers above bars are the number of sample points with density along the X axis and the Y axis representing the percent of all points with this density.
Figure 7 Comparison of the density of black and white social networking application users in Atlanta. Panel A shows the absolute difference in users Density of black users — Density of white users color-coded so that areas with more black users appear red and those with more white users appear blue. Yellow regions are areas where the two densities are similar.
Panel A highlights a small section of the city the area shaded the darkest red where there are many more black than white application users. With this measure, Atlanta is divided nearly in half, with relatively more black users in the southwest and more white users to the north and east.
The yellow band in Panel B shows the region with the highest absolute excess of black users for comparison purposes. The yellow rectangle highlights the midtown area of Atlanta for reference. The yellow oval in Panel B highlights an area with high density of black users but not white users. The yellow circle in Panel C highlights an area with a high density of young black users, but not black users overall ie, an area highlighted in Panel C but not Panel B.
Figure 9 Difference between extreme values of estimated kernel densities of white and black users of a social networking application. Figure shows regions where the values of this equation are -1 green shading, indicating areas with extremes of density for white but not black users , 1 red shading, indicating areas with extremes of the density for black but not white users , and 0 white shading indicating areas that are either not extremes of either density or are extremes for both races.
Figure 10 Application of two density metrics to evaluate recruitment for HIV prevention studies in Atlanta, showing recruitment venues currently in use for two HIV prevention studies green triangles. Panel A shows few recruitment locations in the southwestern part of Atlanta, where there are relatively more young black application users than white users.
Regions where the values of this equation are -1 blue shading, indicating areas with extremes of density for young white but not young black users , 1 red shading, indicating areas with extremes of the density for young black but not young white users , and 0 white shading indicating areas that are either not extremes of either density or are extremes for both races can then be compared to the locations of current recruitment venues.