Visual Summary of "Main-Street Bias"

The Lancet article describes the final stage of their sampling procedure as follows:

"The third stage consisted of random selection of a main street within the administrative unit from a list of all main streets. A residential street was then randomly selected from a list of residential streets crossing the main street. On the residential street, houses were numbered and a start household was randomly selected. From this start household, the team proceeded to the adjacent residence until 40 households were surveyed."

The following picture captures the essence of this scheme. There is a main street as well as cross streets to the main street. Starting from a household that is on a cross street to the main street it is possible for the survey team to spill over onto streets that are not cross streets to main streets as the interview team moves from adjacent household to adjacent household. Accordingly the picture allows such spillover although its extent depends on how close the initial household is to a street corner. However, even allowing for spillover there is still much territory tucked out of range of the survey.

The next picture shows a portion of central Baghdad. While many Iraqi street layouts are grid-like, this satellite shot hints at the complexity that often arises. Ask yourself which streets you think are the main streets? Try to decide what the cross roads to the main streets are. Bearing in mind the previous schematic diagram, it is clear that many parts of central Baghdad lie outside the survey area. Exactly how much territory cannot be sampled depends on precisely which streets qualify as main streets.

courtesy of Google Earth^TM

Please double-click on the images above to provide enlarged, high-quality versions

Please not that these maps are not meant to include a single administrative unit. They are merely to show the complicated, non-grid-like structure of the roads and neighborhoods of Baghdad.