Class Notes (835,929)
Canada (509,507)
Geography (660)
GGR353H5 (32)
Lecture 4

Week 4 Readings.docx

3 Pages
Unlock Document

Grace Barakat

Week 4 Readings Reading #1 Quantitative -quantitative spatial data analysis typically involves one or more of three tasks -drawing a map for visual evidence (Visualization) -geographical and statistical methods to explore data -formal statistical model Visualization -geographers rarely map observed counts of cases by area, since these would simply reflect population distribution or the age structure of that population Exploratory spatial data analysis -modifiable areal unit problem: depends on how zones or areas are defined can be changes and redfined repeatedly -at finer scales we say that estimates of disease risk are “unstable”, since the addition or subtracton of only a single case can greatly alter the estimate. This is called the small number problem -number of possible solutions would be to extend the data collection period to cover further years -a second solution is to work at a coarser scale -another option is use probability mapping (where we map the probability of observing a count that is more extreme or unusual tha that actually observed) -after constructing a map; one question commonly asked is whether areas of high incidence cluster together. This is referred to as spatial autocorrelation (we speak of positive autocorrelation where similar values tend to occupy adjacent locations on the map, and negative autocorrelation where higher values tend to be located next to low ones. If the arrangement is completely random we refer to an absence of spatial autocorrelation Modeling health data in a spatial setting -ecological fallacy: making inferences about individuals based on the geographical area Method and Technique -Multi-level modeling (MLM): most siginificant contribution made by geographers to quantitative health research has been to demonstrate that data for both places and individuals can be brought together in order to shed light on health outcomes. -offers a potentially rich analytical perspective on the geography of health, helping us to establish whether health variations from place to place arise simply because different sorts of people live in different places (Compositional effects) or because places themselves differ in terms of environmental wality or other attributes (contextual effects). -less affluent neighborhoods have social and physical environments less conducive to maintaining a healthy bodyweight -Logistic regression model: standard statistical model. Where the variable to be explained takes on one of two values. Geographical information systems and health -GIS is a computer-based system for collecting, editing, integrating, visualizing, and analyzing spatially referenced data. Such data comprise two forms: the locational element and associated attributed -essentially environmental injustice is the disproportionate burden of exposure to environmental pollution or contamination faced by disadvantaged groups (typically, ethnic minority or low income population groups), and the possible effects on the health of such groups Interpreting the Geography of Health: Qualitative methods -the problem with quant. is that the dots are not inantimate objects; they are real people, and while the ways in which they are arranged spatially may shed some light on disease causation, the approached we have explored so far give no consideration at all to the feelings, experiences, beleifes, and attitudes of indivuals. Interviews -structured questionnaires seek to elicit information from a set of individuals, yielding data agreeable to quant. analysis, -in depth interview: engaging the respondent in a conversation, the purpose of which would be to allow them to talk about their own experiences of health and life -sample sizes used in qual. research are much smaller than those used in quant. research. Often a source of criticism for qual. -the aim is for an intensive understanding (as opposed to an extensive exploration -another potential source of criticism in qual. research is the nature of sample selection. That is, while in quant. research a random sample is selected typically, a qual. sample is often purposefully selected in order to represent not random variation, but MAXIMUM variation -also unusual for qual. research to define a sample size ahead of time. That is, qual. researchers will continue to collect data until saturation is reached (no new themes emerge from data) -
More Less

Related notes for GGR353H5

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.