Ethics in Data Analysis: 5 Guidelines for Conveying the True Meaning of Information
“Data can be manipulated to tell any story. It’s our responsibility to tell the truth.” — Unknown
Data analysis plays a fundamental role in making informed decisions in various fields, from science and technology to business and the entertainment industry. However, it is essential to recognize the responsibility that lies on data analysts when interpreting and presenting information. The intersection between the work of data analysts and journalistic ethics is a relevant topic to be discussed, especially when it comes to conveying the true meaning of data. This article arises from my journey as a data analyst and my critical social sense, at a time when the manipulation of information is a prominent and concerning issue.
Raw data, by itself, does not provide meaningful information. Proper interpretation of data is essential to reveal relevant insights and make informed decisions. One of the challenges faced by data analysts is to avoid hasty conclusions or incorrect interpretations. At the time of writing this article, I am working on a project that examines the top 100 games on Metacritic. In one of the visualizations, the correlation between higher scores on Metacritic and the number of sales was easily misinterpreted, suggesting a correlation between lower sales levels and higher-rated games. However, the correct conclusion is that games with higher scores are less numerous in the market.
Effective communication of data analysis results is crucial to avoid misunderstandings. Data analysts should be transparent about the limitations of the data and avoid unsupported inferences. It is essential to present clear and well-contextualized visualizations, providing detailed explanations that help the audience understand the true meaning of the data. In the example mentioned, it is important to clarify that the relationship between higher scores and lower sales numbers occurs due to the presence of fewer games in the higher positions.
Additionally, data processing work is necessary to obtain accurate insights. In the case of this example, I proceeded by dividing the total sales by the number of games to obtain the average per game, as the average allows for a fairer comparison between different ‘Metascore’ levels by considering the distribution of sales within each level, regardless of volume. This enables a more precise analysis of the impact of ‘Metascore’ on sales, identifying consistent patterns or trends. On the other hand, a sum could make direct comparison difficult, especially due to the relative disparity in the volume of games evaluated in each level. Therefore, the graph displays the average per game, allowing for a visualization that provides more balanced and fair insights into the average performance of games at each level.
The following 5 guidelines in this article are useful not only for those who wish to pursue a career as a Data Analyst but also serve as a reminder for professionals in the field about essential points that are sometimes overlooked.
5 Guidelines to Follow for Any Data Analyst
1 — Use Journalistic Ethics as a Guide to Communicate Accurate and Relevant Data
Journalistic ethics should, in my opinion, serve as a reflection in the work of a data analyst, as both share the responsibility of communicating accurate and relevant information to the public. The fundamental principles of journalism, such as truthfulness, impartiality, and transparency, should be applied when dealing with data and analysis. Data analysts should ensure that their conclusions and interpretations are based on solid evidence, avoiding biases or manipulation of results.
2 — Select Data Responsibly: Avoid Bias and Distortion in Analysis
Data analysts should exercise responsibility in the selection of data to be analyzed. It is crucial to avoid bias in the selection process by choosing only the data that confirms a desired narrative. Furthermore, the visual representation of data should be accurate and avoid distortions that may lead to misleading interpretations. Charts and visualizations should be clear, consistent, and provide relevant information in an accessible manner.
3 — Ensure Transparency in Data Presentation: Important Context and Limitations
When presenting data analysis, analysts should provide transparency regarding the data source, methodology used, and inherent limitations of the data and analysis. This allows the audience to understand the context in which the conclusions were drawn and make informed decisions. Without proper disclosure of limitations, there is a risk of misunderstandings and incorrect interpretations of the results.
4 — Promote Collaboration and Review to Improve Data Analysis Quality
To minimize errors and biases in data interpretation, collaboration between data analysts and reviewers is essential. Teamwork allows for the exchange of ideas, cross-validation, and validation of the results obtained. By subjecting the analysis to a peer-review process, data analysts benefit from the critical insights of colleagues who can identify errors, methodological flaws, or misconceptions. For freelance data analysts, LinkedIn or WhatsApp groups can be an excellent alternative for seeking peer feedback.
5 — Stay Updated and Ethical in an Ever-Evolving Era
The field of data analysis is constantly evolving, with new technologies and approaches emerging regularly. This places an additional responsibility on data analysts to stay updated and aware of the ethical implications of their work. As we move into the era of artificial intelligence and machine learning, complex ethical issues such as algorithmic biases and data privacy also need to be considered and properly addressed.
Conclusion:
The responsibility of a data analyst goes beyond mere technical analysis. It is essential for analysts to understand the importance of conveying the true meaning of data and avoiding misinterpretations that can lead to erroneous conclusions. Journalistic ethics provide a valuable set of principles that can guide the conduct of data analysts, ensuring transparency, impartiality, and accountability in the analysis and presentation of results. By doing so, they play a crucial role in promoting a culture of ethical data analysis and building trust in the information that is presented to the public.
