Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
ENVS257 and ENVS279
GIS for Human Geographers
GIS for Planners
Summative
Assessment 2
Use GIS and statistical analysis to identify key commuting patterns
in Liverpool and their push-and-pull factors.
Key assumptions:
- Various area characteristics (such as un-/self-/employment rates)
can be used to understand why flow movements occur.
- Principal flows are defined by your own threshold of importance.
- Movements are for work-related purposes (commuters); not re-
location residential purposes (migration).
Submission deadline: Check Module Handbook/CANVAS.
Project Tasks
You will write a short executive report (max. 500 words) that will answer the
following questions:
1. What are the main commuting patterns in Liverpool?
2. Where are flows largest? What might this suggest?
3. How does the geography of employment/unemployment relate to flows?
4. How important are distance, origin population, destination population
(employed people) and other variables
(e.g., unemployment) in explaining commuting flows between the areas of
Liverpool?
In addition to answering these questions in your report you should present, as a
minimum, the following information:
1. A map of percentages of unemployed people.
2. A map showing commuting flows between MSOAs in Liverpool (with your own
flow thresholds).
3. A table summarising Poisson regression outputs.
Overall, your report should not have more than 6 maps or charts. Maps can be
generated within R or QGIS.
Data
INFUSE
- Employment data.
- Lots of other variables: age, gender, car ownership etc.
CIDER
- Flow Data
CDRC
- Day and Night-time Populations
- Geodemographic Classification
- Index of Multiple Deprivation (IMD)
- Internet User Classification (IUC)
OS
- Infrastructure
Poisson Regression
- What factors have an influence on commuter flows? Employment-
related and other socio-economic and demographic factors.
- Define the variables (what they measure, units) in your regression
model.
- Explain the sign obtained (+/-) for each slope coefficient.
- Comment on the level of significance (highly, moderately, mildly or
none) of each coefficient which is the likelihood that the coefficient
obtained is a result of pure chance. A highly significant coefficient
shows that the relationship between the outcome and explanatory
variable is very unlikely to have occurred by chance.
- Comment on the model fit statistic (AIC) – Akaike information
criteria. How this changes when we add more variables to the model,
or change the flow threshold.
- Provide reasons for these relationships.
Poisson Regression
Variables Model 1 (Baseline
Model incl. all inter-
zonal flows)
Model 2
(e.g. Model with top
100 rows)
Model 3
(e.g. Model with top
50 rows)
Model 4
(e.g. Model with top
20 rows)
Model 5
(e.g. Model with top 5
rows)
Distance Est(z-stat.)** Est(z-stat)** Est(z-stat.)** Est(z-stat)** Est(z-stat.)**
OEmpoyed …
DEmployed …
OtherVar …
AnotherVar …
AIC: …
Significance Levels:
*: Slope coefficient estimate is mildly significant.
**: Slope coefficient estimate is moderately significant.
***: Slope coefficient estimate is highly significant.
Spatial Dependence
- Moran’s I statistic
- Clustering of unemployment values.
- Test if other explanatory variables also cluster.
- Discuss the magnitude – whether negatively,
positively autocorrelated.
- Discuss p-value: if smaller, this is highly significant.
The Report
Clear Structure (500 words, +/- 10%):
• Introduction
• Methodology
• Results
• Conclusions
• Caveats/Limitations
Provide a total of 6 figures:
• Maps (Min 4, Max 5) – well-labelled and clearly illustrated.
- One map showing unemployment rates.
- One map showing flows between MSOAs in Liverpool.
- Additional maps at your discretion – as deemed relevant to the analysis.
• Tables (Min 1, Max 2).
Use References (list is not included in word count).
Higher Marks
• Use GIS layers from earlier parts of the course.
• Test effects of additional variables downloaded from
INFUSE: prepare data by yourself instructions provided.
• Spatial dependence testing on unemployment and other
variables.
• Good explanations/reasoning for effects of area
characteristics on commuter flows.
• Evidence of additional reading – i.e. beyond what is
provided or recommended reading, and updated versions
of policy/planning documents.
Final Thoughts
• Help and Support is available outside labs on MS Teams - please
post questions there rather than by email – and read responses
to everyone else’s questions.