Geographical map visualizations are a popular and powerful way to communicate spatial data, such as population, income, or climate. However, they can also be misleading if they are not designed carefully and ethically. In this article, you will learn how to avoid some common pitfalls and errors that can distort your map and misinform your audience.
A map projection is a method of transforming the curved surface of the earth into a flat plane. However, no projection can preserve all the properties of the original sphere, such as area, shape, distance, or direction. Therefore, different projections can create different impressions of the relative size and position of countries and regions. For example, the Mercator projection, which is widely used in web mapping and navigation, exaggerates the area of high-latitude regions, such as Greenland and Antarctica, while shrinking the area of low-latitude regions, such as Africa and South America. To avoid misleading your audience, you should choose a projection that suits your purpose and data. For example, if you want to show the true area of countries, you can use an equal-area projection, such as the Gall-Peters or the Mollweide. If you want to show the true shape of countries, you can use a conformal projection, such as the Lambert or the Stereographic.
While creating a geographical map visualization in Power BI, I'm mindful of selecting the right map projection to accurately represent the data. Power BI provides various map visualizations, and I'm selecting one that ensures the geography displayed aligns correctly with the data. In Excel, I can use Power Map or Power View to create geographical visualizations and ensure I choose the appropriate map projection within these tools.
Choosing the right map projection is crucial to avoid misleading geographical visualisations. Each projection distorts some aspect of the earth, so it's important to select the projection that aligns with the map's purpose. Equal-area projections are best for displaying spatial distributions or demographic data, while conformal projections work well for navigational or directional maps. Thoughtfully selecting a projection minimizes misinterpretation and ensures viewers receive accurate geographic data.
Besides projections, you have to think of place, audience, and discipline. To reiterate projections, different types can produce different interpretations. Like if a type preserves the shape, they might not preserve the area or distance. Place is obviously where the map is located. Audience and discipline goes hand in hand, as in how complex or simple you want the maps to be or what things you want to emphasize for ex. mountain ranges or individual streets. Now, with the internet and the whole online ecosystem, there are various niche or interesting "basemaps" based on your own personality or preference.
Choosing an appropriate map projection is paramount in ensuring accurate representation of geographical data. A map projection is a method of transforming the curved surface of the earth into a flat plane. Different projections serve different purposes, and the choice depends on the area of interest and the type of data being visualized. For instance, if your focus is on preserving the area, an equal-area projection like the Albers Equal Area Conic may be suitable. However, if you prioritize preserving shape, a conformal projection like the Mercator projection might be more appropriate. Understanding the strengths and limitations of various projections is key to avoiding misleading visualizations.
The color scheme of your map can have a significant impact on how your audience perceives and interprets your data. For example, if you use a diverging color scheme, which has two contrasting colors at the ends and a neutral color in the middle, you can highlight the differences between positive and negative values, such as temperature anomalies or political preferences. However, if you use a sequential color scheme, which has a single hue that varies in intensity or saturation, you can show the variation in a single variable, such as population density or elevation. To avoid misleading your audience, you should use a color scheme that matches your data type and distribution. For example, if you have a categorical data, such as regions or types, you can use a qualitative color scheme, which has distinct colors that are easy to differentiate. If you have a continuous data, such as numbers or percentages, you can use a sequential or a diverging color scheme, depending on whether your data has a natural midpoint or not.
Se você quiser utilizar um esquema de cor categórico, n?o é recomendado utilizar mais do que cinco/seis cores (pois gera o efeito cara-crachá, em que o usuário olha o mapa e a legenda diversas vezes, e pode n?o observar as nuances dos dados). Nesses casos, vale a pena destacar o top-5 e colorir as demais regi?es com cores neutras (categoria: outros). If you want to use a categorical color scheme, it is not recommended to use more than five/six colors (as this creates the map-legend effect, in which the user looks at the map and the legend several times, does not memorize the colors, and overlook data nuances). In these cases, highlighting the top 5 and coloring the other regions with neutral colors (category: others) is the best option.
As I customize the colors for my geographical map visualization in Power BI or Excel, I'm selecting a color scheme that is both visually appealing and effective in conveying the data. I'm making sure to avoid color schemes that might be difficult for viewers to interpret, especially considering color blindness. Power BI and Excel offer options to customize color schemes, ensuring I can choose one that works best for my audience.
Choosing appropriate colours for maps is crucial to prevent misleading interpretations. Colours should accurately reflect data differences without exaggerating or understating any aspect. Use sequential color schemes for quantitative data and diverging schemes for highlighting differences. For categorical data, use easily distinguishable colours. Consider colorblind-friendly palettes for accessibility. Consistency in colour usage across similar maps aids in comparison and reduces confusion. Proper colour selection ensures accurate and effective data communication to the audience.
For exercise, see if you can make it black/white or divergent/contrasting as possible (like what the introduction says). The key is simplicity, you don't want the audience or stakeholder to look at it for more time than needed! Also, avoid color-blind colors as much as possible but if you absolutely need them then now-a-days the internet has optional colorations for color-blindness.
The color scheme of your map can greatly influence how your audience perceives and interprets the data. It's essential to choose colors that are intuitive and meaningful, while also considering color blindness and cultural interpretations. A sequential color scheme, where colors gradually change in intensity to represent increasing values, is often used for showing ordered data. Conversely, a diverging color scheme can effectively highlight both high and low extremes around a central midpoint. Avoid using colors that may convey unintended connotations or bias, and ensure there is sufficient contrast between different shades for clarity and accessibility.
A choropleth map is a map that uses colors to represent the values of a variable in different geographical units, such as countries, states, or counties. However, a common mistake is to use choropleth maps for raw counts, such as the number of cases, deaths, or votes. This can be misleading because it does not account for the variation in the size and population of the units, which can create a false impression of the magnitude and distribution of the variable. For example, a choropleth map of the number of COVID-19 cases in the US states can make it seem like California and Texas have more cases than New York and Florida, when in fact the opposite is true when adjusted for population. To avoid misleading your audience, you should use choropleth maps for normalized values, such as rates, ratios, or percentages. For example, a choropleth map of the COVID-19 case fatality rate in the US states can show the true impact of the pandemic on different regions.
In Power BI and Excel, I'm mindful of the limitations of choropleth maps for displaying raw counts. Instead, I'm considering alternative visualizations like bubble maps or filled maps with proportional symbols to accurately represent raw count data. By leveraging features in Power BI and Excel, such as data modeling and calculated fields, I can transform raw count data into a format suitable for these alternative visualizations.
This is more of a statistics technicality! The reason you don't want to use raw counts is because you can't compare your counts throughout an area like a country or through multiple households. Raw counts are counts of course, but they didn't start from the same baseline or range to begin with. Maybe think of currency or measurement conversion factors.
Choropleth maps are commonly used to visualize the distribution of a variable across different geographical areas. However, using choropleth maps for raw counts, such as the number of cases or deaths, can be misleading. Raw counts are influenced by the size and population of the geographical units, leading to disparities in visual representation. Instead, consider using rates or proportions to account for population differences and provide a more accurate portrayal of the data. This ensures that the map accurately reflects the underlying patterns without distorting reality.
Any time I see a geographic map, the first thing I do is check to see whether it's really just a recolored population map. You'd be surprised how many times people pass these off as being meaningful results.
Sometimes, a geographical map may not be the best way to visualize your spatial data, especially if you want to emphasize other aspects of the data, such as the relationships, patterns, or trends. In such cases, you can consider using alternative map types, such as cartograms, flow maps, or dot maps. A cartogram is a map that distorts the size of the geographical units according to the values of a variable, such as the population or the GDP. A flow map is a map that uses arrows or lines to show the movement or direction of a variable, such as migration, trade, or traffic. A dot map is a map that uses dots or points to show the location or frequency of a variable, such as events, incidents, or occurrences. These map types can help you avoid misleading your audience by highlighting the relevant information and reducing the noise and bias of the geographical boundaries.
In both Power BI and Excel, I'm exploring various map types beyond choropleth maps. Depending on the nature of my data and the story I want to tell, I might choose to create cartograms, flow maps, or heat maps using custom visuals or advanced features available in Power BI and Excel. I'm experimenting with different map types to find the one that effectively communicates the insights hidden within my data.
Be Careful before doing them! You need to know your data and maybe your audience really well before creating these alternatives. Although these alternatives are very creative and more simplistic, they could mislead or have people question it more since they are not necessarily familiar with them on a daily basis or know what to make of it.
In addition to choropleth maps, several alternative map types can be used to effectively visualize geographical data and avoid misleading interpretations. Cartograms, for example, distort the size of geographical units according to the values of a variable, such as population or GDP, providing a more proportional representation. Flow maps illustrate the movement of goods, people, or information between different locations, offering insights into spatial relationships and connectivity. Dot maps use dots to represent individual data points, allowing for precise localization and density analysis. By exploring alternative map types, you can choose the most suitable visualization method for your data and avoid potential distortions
The final and most important step to avoid misleading geographical map visualizations is to test your map with your audience. You should ask for feedback and suggestions from people who represent your target audience, such as your colleagues, clients, or users. You should also evaluate your map according to some criteria, such as accuracy, clarity, relevance, and aesthetics. You should check if your map conveys the correct and complete information, if your map is easy to understand and interpret, if your map answers the questions and needs of your audience, and if your map is visually appealing and engaging. By testing your map with your audience, you can identify and correct any errors or flaws that may affect your message and credibility.
Just because you know it very well as the creator, doesn't mean other people will. Sometimes we make assumptions in our heads what the audiences know based on our expertise or knowledge. A good rule of thumb is to test it on children or grade school students. If they need further explanation, then it means you need to make the map better. A good idea is to create a survey along with the map so people can easily fill it out and provide valuable feedback. This may take a couple times and maybe exhausting for you, but if you do it right then you will help alot of people in the long run.
Before finalizing my geographical map visualization in Power BI or Excel, I'm sharing it with stakeholders to gather feedback. I'm using Power BI's collaboration features or sharing Excel files with colleagues to ensure they can provide input on the clarity, accuracy, and interpretation of the map. By involving my audience early in the process, I can make necessary adjustments to improve the effectiveness of the visualization.
The final and most crucial step in avoiding misleading geographical map visualizations is to test your map with your audience. Solicit feedback from individuals who represent your target audience, such as colleagues, clients, or users. Conduct usability tests to assess how well your map communicates the intended message and whether it is easily understood by the audience. Pay attention to any misconceptions or misinterpretations that arise during testing and make necessary adjustments to improve clarity and accuracy. By involving your audience in the validation process, you can ensure that your map effectively conveys the intended information without leading to confusion or misunderstanding.
Consider scale, interactivity, and context when crafting geographical visualisations. The correct scale ensures that data isn't too aggregated or too detailed. Interactive features empower users to explore data, while context enriches maps by linking data with background information. Balancing these elements creates insightful narratives and informative tools for analysis and storytelling.
Throughout the visualization process in Power BI and Excel, I'm paying attention to details like legends, scales, and annotations. I'm ensuring these elements are clear and informative to aid interpretation. Additionally, I'm transparent about any data limitations or uncertainties, providing context for viewers to understand the insights presented in the map visualization.
Maps are excellent in depicting geographic relevance. However, when analyzing data, the choice of representation depends on the user's main objectives and how they analyze data. For instance, bar charts trump maps for region-to-region comparisons devoid of spatial context. So, it is essential to keep this in mind while choosing the right representation. Mapas s?o excelentes para representar a relevancia geográfica. Porém, ao analisar os dados, a escolha da representa??o depende dos objetivos principais do usuário e de como ele analisa o dado. Por exemplo, gráficos de barras superam mapas para compara??es entre regi?es desprovidas de contexto espacial. é essencial ter isso sempre em mente quando for escolher um gráfico.
Maintain a consistent scale and aspect ratio across the map to avoid distorting the relative sizes and shapes of geographical features to ensure accurate comparisons between regions and choose appropriate classification methods and symbols for representing data on the map. Avoid using visuals like bubble maps that vary in size but don't accurately represent the data values.