Applied analytic methods on a policy case: Urban Air Quality…

*Please read the attached instruction files as they are very crucial for accomplishing the assignment.

*the csv file is optional to open if you want to analyse data yourself or any of my data seems off to you.

*writing tone should not be too academic as this is a policy document, and it is for people who are not experts in this area to read in a short period of time. Making sentences simple doesn’t mean dumbing down here.

Aim

Apply, appraise and recommend a range of analytic methods in informing

decisions about a complex science, technology and public policy issue.

Objectives

Design and undertake exploratory data analysis using quantitative and/or

qualitative techniques.

Generate evidence describing the behaviour of policy systems influencing air

quality.

Evaluate how uncertainty affects your findings and propose ways to

communicate this in policy contexts.

Use data visualisation to enhance understanding and communication of data

insights.

Judge the suitability of analytic methods used for analysis.

Marking

This assignment is 50% of the module. The marking scheme is

available on school website and covers the following relevant marking criteria:

Conceptual Understanding (40%)

Reasoning & Critical Analysis (40%)

Communication, Structure & Clarity (20%)

Required Length

The word limit is 2,000 words. The penalties apply for going more than

10% above or below this limit are outlined in the MPA Handbook.

References and footnotes solely containing references are not included in

the word count.

1Project context

Imagine that London has recently joined a newly established ‘Attractive Cities’

network – a global network of the mayors of capital cities taking action to become

attractive places to both live and work. A small group of core member cities have been

asked to take the lead for developing shared research and policy action on public

policy themes for the network’s members. Yesterday the news came through that

London has been assigned as the thematic city lead on “Air Quality and Citizen

Health”.

This means that London’s Mayor now has responsibility for an entirely new policy

portfolio, as well as in essence overnight become an influential and expert global voice

on science, technology and innovation around air quality and health policy issues and

their interrelationship. The Mayor is very conscious that while London has its own

challenges and experiences with managing air quality, the real challenge at hand will

be the development of evidence-based insights into air quality improvement as a

common policy agenda, yet one with many different local realities and considerations.

As it turns out, air quality is a topic the current Mayor knows little about, nor has

previously had much policy exposure to. By background, she is a lawyer. She is,

however, a keen advocate for evidence-based approaches to policy analysis and has

often been frustrated when there has been insufficient rigour in analyses used to

design and inform policy decisions. She has informally been warned that her Analysis

team has recently had a few issues with analysis credibility. She has therefore asked

a special analysis adviser be appointed immediately to support her over the next

weeks in preparing for this new role. Your CV impressed and this is now your role.

Part 1. Exploratory data analysis [50%]

The Mayor would like to have insight into any potential interrelationships between air

quality and citizen health to best understand the agenda of the network she has joined.

She also wants to understand whether and how context matters and how any

interrelationships vary across cities. As a first step, she has asked you to undertake

some exploratory data analysis to understand, where possible:

a. What are key trends, patterns and/or anomalies in the air quality and citizen

health of cities?

b. How do the phenomena of air quality and citizen health interact or interrelate?

Are there potential causal relationships?

c. How do air quality and health differ between cities with different

characteristics (e.g., by population size, exercise levels and/or modal split)?

d. How does London compare to other cities around the world?

e. Are there any immediate issues with data uncertainties, outliers, and/or

results confidence?

2The Data

Your predecessor has left you a data file containing the latest data that was about to

be analysed to brief the Mayor before they switched roles. This file

‘aqhealthcities.csv’ is available for download under the Assignment 2 Moodle

page. Your predecessor also forwarded an annotated data dictionary/codebook as

well as a short annotated set of references. These two resources are included in

Appendix A as Tables 1 and 2 respectively.

Given the incredibly tight timeframes, the Mayor has asked you to focus your analysis

in the next two weeks on the contents of the inherited ‘aqhealthcities.csv’ data. She is

happy for you to seek out other theories, knowledges, data sets, etc. beyond this data

set if they help you with your analysis, but she has stressed she wants you to do this

proportionately – i.e., her priority is analysis of the .csv file data and she is not expecting

you to spend much time looking at other sources. Further analysis of other data

sources is welcome, but depending on your capacity may have to wait until the start

of the new year.

Q1. Produce for the mayor [~50%]:

1. Exploratory data analysis informing her interests (a.-e.) outlined

above. Explore the patterns, trends and possible observations and

inferences to make from the data. Choose a set of final summative

information points and messages you want to communicate to the mayor.

Include at least two visualisations, though feel free to use more than this

if appropriate.

2. A short opening or closing summary of key recommendations derived

from the exploratory data analysis. You may, for example, want to

highlight some of the similarities and differences between and within cities

she should be aware of whilst steering this network. Or you may have ideas

for policy action based on your analysis. Or, you may may want to raise key

issues about data, analysis and confidence you think the Mayor should be

aware of at this point. For any recommendation you make, make clear

your assessment of the quality and reliability of the data.

Communicate your work as a short analysis report in a way appropriate for quick

reference and understanding. This means you can use a mixture of prose, bulleted

text, tables, figures, headers, labels, etc. Use data visualisation effectively to

support your analytical narrative.

Part 2: Informing policy decisions [50%]

Following your initial exploratory analysis, the Mayor has asked for your advice on

some analytical and methodological questions that have emerged as she prepares to

lead the Attractive Cities network.

Q2. Interpreting Probabilistic Analysis [5%]

The Mayor is evaluating two policy interventions to improve air quality and health

outcomes in London. As she now leads the Attractive Cities network’s air quality

theme, she knows other network cities will be watching London’s choices with interest,

but her primary responsibility is to make the best decision for London. An external

consultancy has undertaken a Monte Carlo simulation comparing the two options:

Option A: Comprehensive Air Quality Monitoring Network

o Deploy high-density monitoring infrastructure across all boroughs

o Estimated 5-year cost: ?50-90M

o Provides real-time data for enforcement and public information

Option B: Clean Air Zones Expansion with Technology Fund

o Expand Ultra Low Emission Zone (ULEZ) to all Londo n boroughs

o Create innovation fund for clean transport solutions

o Estimated 5-year cost: ?30-120M

o Includes enforcement tech and electric vehicle charging infrastructure

The simulation results show probability distributions for total costs over 5 years:

Fig 1. Monte Carlo Simulation of total total cost profile for 2 policy options from the consultants report

6Based on these simulation results:

a. b. What do the distributions tell you about the relative uncertainty of each option?

Which option would you advise the Mayor to pursue for London, and why?

c. What caveats should the Mayor be aware of when interpreting these

probabilistic estimates?

Q3. Additional Analytical Advice [5%]

Before the Mayor makes her final decision between Option A and Option B for London,

what one additional piece of analysis or evidence would you recommend she

commission, and why?

Q4. Deliberative Approaches to Policy Evaluation [10%]

Beyond the immediate monitoring vs expansion decision, the Mayor faces ongoing

choices about how to allocate London’s air quality budget across multiple competing

priorities. She has ?5M available for the next financial year and must decide between

investments such as:

School street expansions (car-free zones around schools during drop-off)

Green infrastructure (green walls, urban greening for pollution absorption)

Low-emission zone enforcement technology upgrades

Public transport fare subsidies to encourage modal shift

Community air quality monitoring and engagement programmes

Support for small businesses to transition to low-emission vehicles

These priorities have different beneficiaries, different evidence bases, different time

horizons, and involve different values and trade-offs. They cannot all be funded fully.

Her team has suggested using Multi-Criteria Analysis (MCA) with an Expert

Advisory Board to evaluate these options. The Mayor has read briefings on MCA and

understands the methodology. She is also conscious that her approach to this decision

may inform how other Attractive Cities network members handle similar allocation

challenges.

Provide advice to the Mayor addressing:

a) Should she use MCA for this London budget allocation decision?

Consider: Is MCA appropriate for this type of decision? Why or why not? What

are the key strengths of MCA that make it suitable (or not) for managing this

budget allocation for London?

b) If she proceeds with MCA, identify and discuss THREE critical design

choices from the following areas:

Expert Board composition (who participates?)

Deliberative process structure (how is it organixed?)

7 Criteria selection and weighting (how are priorities determined?)

Bias management (how to mitigate biases?)

Stakeholder inclusion (which London communities represented?)

Transparency and documentation (how is process communicated?)

For each of your three chosen areas, explain the specific choice she must make and

why it matters for quality.

Note: The Mayor does not need you to explain how MCA works procedurally. She

needs practical, context-specific advice on whether and how to use it effectively for

this London budget allocation challenge.

Q5. Critical Reflexivity on Your Analytical Choices [15%]

Throughout this STEP0020 module, we have explored how analytical methods are

not neutral tools but embody particular worldviews, values, and assumptions about

what counts as knowledge and how it should be produced.

Reflecting specifically on your exploratory data analysis in Part 1, reflect on

your key methodological choices: which methods did you use for your analysis

(e.g., visualisation types, statistical approaches, ways of grouping or comparing

data); which variables did you focus on and which did you set aside; how you

defined and measured relationships between air quality and health.

a. Share a reflection on what your choices privileged and what they

obscured. You can consider:

What did your analytical approach emphasise or make visible? What

might it have obscured or marginalised?

What assumptions were embedded in your choices? (e.g., about

causality, about what’s measurable, about relationships between

variables)

Whose perspectives were centred in your analysis? What voices or

experiences were excluded?

b. Consider whose knowledge matters.

Your analysis relied on aggregate city-level data. What might be missed

or obscured when relying primarily on quantitative, aggregate data?

Given that the Attractive Cities network represents diverse cities with

different contexts, capacities, and knowledge traditions, what additional

approaches (drawing on methods from this course) might complement

your quantitative analysis to incorporate broader perspectives?

c. Reflect on your own positionality

How might your own background, training, and assumptions have shaped

what you chose to analyse and how you interpreted it?

If someone with a different background or perspective (e.g., a community

health worker, an environmental justice advocate, a city official from a

8Global South city) were to analyse this data, what might they emphasise

differently?

Use specific examples from your Part 1 analysis to ground your reflections.

Move beyond generic critiques to engage substantively with the actual analytical

choices you made and their implications for a global network on air quality & health.

Q6. Managing Deep Uncertainty in Policy Analysis [15%]

In her role as thematic lead for the Attractive Cities network, the Mayor recognises

that significant deep uncertainties affect the network’s collective approach to air

quality and health policy:

Future air quality trends are uncertain (climate change impacts on pollution,

changing mobility patterns post-pandemic, technological disruptions in

transport and energy systems)

Health impacts are uncertain (emerging evidence on pollution exposure

pathways, changing population health trends, new epidemiological

understanding)

Political and economic contexts vary enormously across network cities and

are themselves changing (policy commitment, resource availability,

governance capacity)

Policy effectiveness is uncertain (what works in one city may not work in

others, unexpected implementation challenges, behaviour change

uncertainties)

These uncertainties cannot easily be quantified probabilistically. The Mayor cannot

assign reliable probabilities to different futures, nor can she rely on historical data to

predict unprecedented changes.

Advise the Mayor on using scenarios to manage these deep uncertainties:

a. b. What are scenarios and why are they useful for this type of uncertainty?

How could scenarios help her make more robust decisions for the

network?

c. What are key limitations or challenges of scenario-based approaches?

Ground your advice in the Mayor’s specific context – leading a diverse global

network of cities facing climate, health, and technological uncertainties. Explain how

scenarios would work as a practical analytical tool, not just a conceptual framework.