Assignment for seminar 3
1. Describe in your own words the main methodological challenges to conducting neighbourhood
effects research (see Santiago & Galster, 2014).
The first methodological challenge in neighbourhood effects research is clarifying the boundaries of a
neighbourhood. Often the census tract level is chosen. However, those administrative boundaries do
not always represent what people actually perceive as their neighbourhood. An example is given by
Sharkey & Faber (2014), who state that people who live close to the boundary of a certain
neighbourhood also have contacts with people who live in the adjacent neighbourhood. This is in line
with Santiago & Galster’s (2014, p. 31) statement that people often define their neighbourhood as
the areas that are bounded closely to their immediate surroundings. Another problem is that
different causal forces may operate over different spatial scales of the neighbourhood, making it
even more important to choose the appropriate spatial scale.
Besides, a broader range of social and physical conditions within neighbourhoods must be taken into
account in researches and there must be a reflection on how these mechanisms influence child
outcomes: what is the causal relationship? Also, other unobserved factors on for example the
household level might be of influence, causing a biased estimate of the causal relationship between
the explanatory variable and the outcome. This is known as the omitted variable bias.
(continues next page) …
1. Describe in your own words the main methodological challenges to conducting neighbourhood
effects research (see Santiago & Galster, 2014).
The first methodological challenge in neighbourhood effects research is clarifying the boundaries of a
neighbourhood. Often the census tract level is chosen. However, those administrative boundaries do
not always represent what people actually perceive as their neighbourhood. An example is given by
Sharkey & Faber (2014), who state that people who live close to the boundary of a certain
neighbourhood also have contacts with people who live in the adjacent neighbourhood. This is in line
with Santiago & Galster’s (2014, p. 31) statement that people often define their neighbourhood as
the areas that are bounded closely to their immediate surroundings. Another problem is that
different causal forces may operate over different spatial scales of the neighbourhood, making it
even more important to choose the appropriate spatial scale.
Besides, a broader range of social and physical conditions within neighbourhoods must be taken into
account in researches and there must be a reflection on how these mechanisms influence child
outcomes: what is the causal relationship? Also, other unobserved factors on for example the
household level might be of influence, causing a biased estimate of the causal relationship between
the explanatory variable and the outcome. This is known as the omitted variable bias.
(continues next page) …
