Allyson Benton is a Reader in the Department of Government at the University of Essex. Allyson’s research lies within the field of political economy and has been published in a variety of academic journals. She is currently examining the impact of politics (most recently, politicians’ career ambition) on subnational fiscal policy as well as the impact of politics (most recently, political speech) on financial markets. Her previous research also included the origins and operation of subnational authoritarian regimes in Latin America. Allyson’s research has been enriched by her long time working as a Latin American political risk analyst in the private sector, as well as researching, living, and working in that region.
Course Description
This course introduces participants to the foundations of practical data storytelling. It focuses on examining data for meaningful insights, crafting clear data narratives, and translating these into actionable ideas. Modules can be taken independently or as a sequence, with each building on the skills developed in the previous one.
Module 1: From Data to Insight (5, 6, 7, 8 May, 14:00 – 17:00)
This module covers how to explore and interrogate data before drawing conclusions. Participants learn to make valid comparisons, choose appropriate measures, spot meaningful patterns, and distinguish genuine signals from noise and error.
Module 2: From Insight to Narrative (12, 13, 14, 15 May, 14:00 – 17:00)
This module covers how to present data visually so that key insights are immediately clear. Participants learn to match graph types to the comparison being made, make effective design choices, and communicate findings with clarity and integrity.
Module 3: From Narrative to Influence (19, 20, 21, 22 May, 14:00 – 17:00)
This module covers how to shape a data story for different audiences and purposes. Participants learn to adapt their story’s detail, framing, and format to context, draw appropriate inferences from their data, and maintain credibility through transparency about sources and limitations.
Course Outline
Module 1: From Data to Insight
This module covers how to explore and interrogate data before drawing conclusions. Participants learn to make valid comparisons, choose appropriate measures, spot meaningful patterns, and distinguish genuine signals from noise and error.
- The comparison question: Recognizing that every number needs context to become meaningful; understanding the logic of comparison (temporal, cross-sectional, counterfactual).
- Choosing appropriate comparisons and baselines: Selecting meaningful comparison groups (peer, historical, target) and understanding when different comparison frames help versus mislead.
- Denominators and measurement choices: Understanding when to use per capita, percentages, rates, or absolute figures depending on the insight being pursued.
- Identifying meaningful patterns: Distinguishing genuine signals from noise; separating seasonal patterns, long-term trends, and one-off events.
- Understanding outliers and anomalies: Techniques for spotting data errors versus genuine anomalies that deserve investigation and explanation.
- Interpretation traps and communication ethics: Recognizing false patterns, understanding regression to the mean, avoiding the “compared to nothing” fallacy, and communicating uncertainty responsibly.
Module 2: From Insight to Narrative
This module covers how to present data visually so that key insights are immediately clear. Participants learn to match graph types to the comparison being made, make effective design choices, and communicate findings with clarity and integrity.
- Matching graph types to comparison logic: Line graphs for trends, grouped bars for categorical comparisons, scatter plots for relationships, tables for precise values—and when common choices like pie charts mislead.
- Scale and axis decisions: When to start axes at zero, when to use truncated axes, when to use logarithmic scales, and how scale choices shape interpretation.
- Visual clarity principles: Using color to highlight key comparisons, maintaining accessibility, removing chart junk, and ensuring consistent scales.
- Effective titles and annotations: Writing titles that state the insight rather than describe the data, and using annotations to guide readers to key findings.
- Communicating uncertainty: Presenting confidence intervals, margins of error, and data limitations honestly while maintaining credibility; avoiding inappropriate comparisons.
- Critical interpretation and common mistakes: How audiences process visual information; identifying errors like inconsistent scales, truncated axes, and misleading comparison frames.
Module 3: From Narrative to Influence
This module covers how to shape a data story for different audiences and purposes. Participants learn to adapt their story’s detail, framing, and format to context, draw appropriate inferences from their data, and maintain credibility through transparency about sources and limitations.
- Audience adaptation: Tailoring the same data story for executives, technical staff, and general audiences based on their different needs and decision contexts.
- Narrative length and depth: Creating short, medium, and long versions of the same story; knowing when a single chart suffices versus when fuller context is needed.
- Framing trade-offs: Presenting options and implications that inform rather than dictate decisions; maintaining credibility while acknowledging uncertainty.
- Drawing appropriate inferences: Understanding correlation versus causation; knowing when data supports strong claims versus when it suggests further questions.
- Data ethics and transparency: Citing sources, acknowledging limitations, avoiding selective presentation, and maintaining integrity in communication.
- Expanding the toolkit: When to use mapping and spatial visualization; anticipating pushback on methodology or alternative interpretations.
