Lecture 1: Introduction to course (01/04/2017)

Lecture one outlines the importance of GIS and three major themes of the course. The three major themes are landscape ecology, health geography and crime analysis. GIS allows for the better understanding of how landscape structure affects different variables. GIS is also important in health geography because places, spaces, and communities shape health outcomes and health care delivery. GIS may also help assessing spatial patterns of the spread of diseases, determining social inequalities in access to healthcare, and ways the environment affect one’s health well-being. In terms of crime analysis, GIS allows for hotspot identification as well as providing information relative to crime pattens and trend correlations in assistance for both operation and administrative use. Finally to note, it is very important to ask questions, especially questions relating to space.

Lecture 2: Why is ‘geography’ important (01/11/2017)

In today’s lecture Dr. Klinkenberg discussed the importance of space and divided the lecture in two parts. The first part concerns the scale, grain and extent of a study area. Scale plays an important role in terms of identifying spatial autocorrelation which can be fine or coarse depending on the study. If one wish to analyze spatial patterns at a neighborhood level then a finer scale would be more suitable, but if one wish to analyze spatial patterns for an entire state, then a coarse scale would be more adequate. The second part involves modifiable areal unit problem (MAUP), the combination of both scale and aggregation problems. A change in scale can skew statistical results as a specific spatial pattern observed in one scale may not be apparent at a different scale. The aggregation effect refers to the various ways units are grouped at a given scale which can also skew statistical results. A good example of MAUP is gerrymandering, where election results are skewed due to the variations in district divisions. One reason that MAUP exists is that species/individuals/households tend to group with one another in terms of likeness rather than at random. The three main grouping models are grouping, group dependent and feedback.

Lecture 3: Understanding Landscape Metrics (01/18/2017)

Today’s lecture is about landscape ecology. Landscape ecology is the study of spatial patterning on the dynamics of interacting ecosystems. It is relatable to GIS because landscape ecology is concerned with the reciprocal interactions between spatial patterns and ecological processes. Spatial autocorrelation is a kind of spatial pattern that can be divided into two types: first-order process and second-order process. First-order process is when patterns develop as a result of a response to an environmental factor whereas second-order process is when patterns develop as a result of interactions between the objects and events themselves. On the other hand, stationarity is almost the opposite of spatial autocorrelation. Stationarity occurs when the processes that govern the placement of an object or event do not change over space. There are also first or and second order stationary, and both are influence by direction.

Besides methodological considerations, there are also different processes that occur on landscapes. Those can be abiotic (climate, topography, soil), biotic (competition, keystone species), anthropogenic (human impact), and disturbances (fire, volcanic eruptions, floods). Last but not least it is important to note that these processes can operate at a different temporal and spatial scale. So scale is important.

Lecture 4: A review on statistics (01/25/2017)

Statistic is very important in GIS. It is important because it allows GIS operators to summarize, explore, look for relations and to predict patterns. Data that exist in the world of GIS and statistics are nominal, ordinal, interval and ratio. One statistical tool that can be performed in ArcGIS is grouping analysis which performs a classification procedure that tries to find natural clusters. The user would have to define the numbers of groups and spatial constraints and grouping analysis will separate user’s data into groups where features within groups are as similar as possible, and all the groups themselves are as different as possible.

In terms of measuring association, one may use chi-square statistics or Pearson’s and Spearman’s R. In terms of regression analysis, one may use simple regression (linear) or ordinary least squares (OLS), or geographically weighted regression (GWR). The biggest difference I found between GWR and OLS is that GWR takes into account geographic distribution and explores local relationships while OLS models the data globally and is a bit more rigid. All in all, every statistical tool has its advantages and disadvantages, so it is important to know your data.

Lecture 5: What is health geography (02/10/2017)

Today’s lecture is about health geography. So what is health geography, and how it relates to GIS? Health geography is the application of geographical information, perspectives, and methods in the study of health, disease and health care. The three main themes in health geography are disease ecology, health care delivery and environments and health.

Disease ecology studies the interactions between viruses or bacterias and human or non-human hosts. It also studies the spatial distribution of infectious diseases in relations to meteorological, biological and cultural phenomenas, as well as political and economical barriers. The study of health care delivery studies the distribution of health services in relations to human settlement. It also addresses issues like social inequalities in the access to health services. Environment and health studies how the environment (produced both naturally and anthropogenically) affect people’s health well-being.

Lecture 6: GIS in health geography (02/17/2017)

This lecture discusses major applications of GIS in health geography and spatial epidemiology. Four major applications of GIS in health geography are spatial epidemiology, environmental hazards, modeling health services and identifying health inequalities. Spatial epidemiology is the analysis of geographical variations in disease with respect to demographic, environmental, behavioral, socioeconomic, genetic and infectious risk factors. Environmental hazards may include landfills and chemical plants that may affect one’s health well-being. Modeling health services involve the mapping of health providers (normal and specialized clinics, hospitals, etc.) with respect to socioeconomic, risk and demographic (and many more) factors. Last but not least, in identifying health inequalities, health care services are mapped primarily in relations to socioeconomic and demographic data.

To go into specifics, there are four major applications of GIS in spatial epidemiology. Those are disease mapping, cluster detection, spatial exposure assessment and assessment of risk of disease. Disease mapping allows GIS operators to describe patterns of disease, explore and analyze spatial patterns, and hypothesize about possible causal relationships. Clustering detection/analysis is also important because clusters are geographically and temporally bounded groups of occurrences which unlikely to have occurred by chance. Since the fact that similar people tend to aggregate and ecological attributes both can affect health patterns, it makes cluster detection an indispensable tool for spatial epidemiology. Spatial exposure assessment assesses the exposure of an element (dust, acid rain etc.) that may affect people’s health well-being to a population. Exposure is often measured by TTHM Exposure Index.

Lecture 7: Is crime related to geography (03/01/2017)

GIS can aid in crime analysis in many ways. Visualization through GIS, for example, allows law enforcement agencies to understand where crime is occurring as well as any spatial patterns of criminal activity. This in turn allows them to operate more effectively in combating crime. Crime is also a geographic phenomena. The field of environmental criminology describes criminal patterns in three distinct theories: routine activity theory, rational choice theory and criminal pattern theory. Routine activities theory is based on routine activities (work, gym etc.) of people (offenders and victims). It recognizes that sociodemographic and socioeconomic characteristics of people are not random, and therefore routine activities are also not random. Rationale choice theory suggest that offenders are influenced by daily activities and routines of their lives and therefore they tend to concentrate in areas that are known to them. It gives us some insight into what an offender is thinking when they decide to commit the crime. Last but not least criminal pattern theory helps us understand where and when the offense will occur. In addition to the three theories briefly described above, there are two main concerns in environmental criminology:1) the spatial distribution of offenses and 2) the spatial distribution of offenders.