Dot distribution map

Last updated
A bivariate dot density map showing the relative concentrations of the Black and Hispanic populations in the United States in 2010. Dot map black hispanic.png
A bivariate dot density map showing the relative concentrations of the Black and Hispanic populations in the United States in 2010.

A dot distribution map (or a dot density map or simply a dot map) is a type of thematic map that uses a point symbol to visualize the geographic distribution of a large number of related phenomena. Dot maps are a type of unit visualizations that rely on a visual scatter to show spatial patterns, especially variances in density. [1] [2] The dots may represent the actual locations of individual phenomena, or be randomly placed in aggregation districts to represent a number of individuals. Although these two procedures, and their underlying models, are very different, the general effect is the same.

Contents

History

Valentine Seaman's map of the 1796 outbreak of yellow fever in New York City, showing disease cases by numbered dots that were analyzed in the text. Valentine Seaman's map of yellow fever epidemic in New York City (Plate 2 of 2).jpg
Valentine Seaman's map of the 1796 outbreak of yellow fever in New York City, showing disease cases by numbered dots that were analyzed in the text.

The idea of using dots to show relative density came about during the Industrial era of England and France in the 1830s and 1840s, a time when most modern types of thematic map were developed into a relatively modern form. [3] They were enabled by the increasing availability of statistical data and growing recognition of their value for scientific understanding. As with other types, the earliest inventions of the technique often went unnoticed, with later publications garnering much more fame.

It has been claimed that the first dot distribution map was created by Valentine Seaman in a 1797 article analyzing a recent outbreak of yellow fever in New York City. Although the relatively small number of case locations is not in keeping with the typical use of this technique to visualize the overall distribution of a large number of individuals, it is still noteworthy as possibly the first instance of using a map as an analytical and communication tool for social science, of spatial analysis, and of epidemiology (even though his conclusions turned out to be incorrect). [4]

de Montizon's 1830 Carte Philosophique figurant la Population de la France, the earliest known dot density map. Carte Philosophique figurant la Population de la France.jpg
de Montizon's 1830 Carte Philosophique figurant la Population de la France, the earliest known dot density map.

The earliest known district-based dot density map was created in 1830 by Armand Joseph Frère de Montizon (1788–????), a Franciscan friar, schoolteacher, and printer. [5] It is a relatively simple map of population by département (administrative district) in France, with each dot representing 10,000 individuals. [6] The map appears to have been drawn using the same technique practiced for the next two centuries and still performed by computer today: a number of dots, calculated from the total population of each department, is spread randomly across each department. The result is an intuitive visual display of population density, as higher population levels within an administrative border exhibit a closer, denser pattern of dots. Since the dots are evenly spaced, it is evident that they do not represent the actual locations of where people live within a department. This is an example of an ecological fallacy, where a value for an area generalizes all within that area to exhibit that value. [7]

von Mentzer's 1859 dot density map of Sweden and Norway, probably the first fully-developed representative dot density map. Fysisk-geografiska kartor ofver Skandinaviska halfon for beskrifvande undervisning i faderneslandets geografi - Kungliga Biblioteket - 10372816-thumb.png
von Mentzer's 1859 dot density map of Sweden and Norway, probably the first fully-developed representative dot density map.

Although Montizon's map was the first published dot map of its type, his innovation had no effect on practice for nearly 30 years until the district-based dot density map was reinvented in 1859 in a map of the population distribution of Sweden and Norway by Thure Alexander von Mentzer, a Swedish Army officer. [8] The dots in his map (each representing 200 residents) appear to have been based on the 1855 Census, but clearly show adjustments based on additional knowledge of population distribution. [9]

Shapter's 1849 map of the 1832-1834 Cholera outbreak in Exeter, with different symbols for cases in each year. Thomas-Shapter-HistoryOfCholeraInExeter1832-map.jpg
Shapter's 1849 map of the 1832-1834 Cholera outbreak in Exeter, with different symbols for cases in each year.

The point feature map was also reinvented in the mid-19th century, with epidemiology again being a leading driver, especially the search for the cause of cholera, which was recognized as occurring in clear geographic patterns. [10] Among the variety of maps created between 1820 and 1850 are some showing the locations of every case in a region. A notable example was an 1849 map by Thomas Shapter in his history of the 1832-1834 Cholera outbreak in Exeter. [11] The map is innovative in using different point symbols to represent the cases in each of the three years. Shapter did not go so far as identifying the cause of the disease clusters he observed, his map was influential; John Snow later cited it as an inspiration for his own work.

Original map by John Snow showing the clusters of cholera cases in the London epidemic of 1854. The pump is located at the intersection of Broad Street and Little Windmill Street. Snow-cholera-map-1.jpg
Original map by John Snow showing the clusters of cholera cases in the London epidemic of 1854. The pump is located at the intersection of Broad Street and Little Windmill Street.

When a large outbreak occurred in London in 1854, Dr. John Snow collected data about the individual cases, especially their location in Soho (London), using nascent methods of spatial analysis and contact tracing to conclude that contaminated water was the disease vector, and successfully had the source shut off. [12] The map that accompanied his 1855 report showed individual cases, stacked at each house location, clearly showing a concentration around the Broad Street Pump as well as gaps in locations that had other water sources. [13] The map is now hailed as revolutionary; although its role in the investigation itself and its effect on settling the debate over the cause of the disease is often overstated, [14] it does deserve recognition for Snow's insight that a map was the most effective tool for communicating the spatial patterns of the disease.

In later years, dot maps do not appear to have been as prolific as other types of thematic maps, possibly due to the time needed to create them. Many were seen as an accomplishment worthy of academic publication on their own. [15] A hybrid technique emerged in early 20th century maps of population density, using representative dots in rural areas with proportional circles to represent major cities. The dot density method became standardized during this period, and design guidelines developed, [16] such that the technique could be taught in the cartography textbooks of the mid-20th century. [17] [18]

Geographic information systems have made the generation of dot density maps relatively easy by automating the placement of the individual dots, although the results are often less satisfactory than those that were manually crafted. A significant technological advance has been the availability of very large datasets, such as millions of geocoded social media posts, and innovations in how to visualize them. The resultant maps are able to show detailed patterns of geographic distributions.

Animated dot density map of COVID-19 cases in Connecticut between March 21 and May 21 2020. COVID19 CASES Mar21 May21.gif
Animated dot density map of COVID-19 cases in Connecticut between March 21 and May 21 2020.

Recent advancements in dot maps include using dasymetric mapping techniques to more accurately place dots within zones, [19] scaling dot maps to show different rates of dots per person at different zoom levels, [20] and using temporal interpolation to animate dot maps over time [2]

Types of dot maps

Two very different types of dot maps have been developed, often leading to some confusion in terminology. In fact, many cartographers have suggested that they not be grouped into a single type of thematic map. Although they use very different techniques, based on very different data sources, with a different semantics in the result, the general purpose is the same: to visualize the geographic distribution of a group phenomenon (i.e. a large number of individuals).

One-to-one (point feature)

A one-to-one dot distribution map, identifying concentrations of homicides in Washington, D.C. DChomicides.jpg
A one-to-one dot distribution map, identifying concentrations of homicides in Washington, D.C.

A one-to-one dot map displays the a large number of point symbols representing the locations individual occurrences of a phenomenon. Many kinds of maps display geographic features as point symbols, such as cities; this category only applies to those that show a large number of instances, each portrayed anonymously (e.g. not labeled), so that attention is focused on the overall distribution rather than on individuals. For many years, this approach has been a core part of the field of crime mapping, in addition to its original use in epidemiology. It has become especially popular in the recent era of big data, such as mapping millions of geotagged social media posts or cell phone locations, although these maps have raised concerns over privacy. [21] [22] [23]

Various terms for this technique have been proposed to distinguish it from the other approach, such as nominal point map, point feature map, and pin map. [24] [25] :135 Another suggestion is to use the term dot distribution map exclusively for this type (with dot density reserved for the other type), although this does not clarify the confusion, since both maps aim to show distribution and density.

One-to-many (representative point)

Representative dot density map of Acres of Harvested Wheat in Illinois in 2012, using county-level aggregate data. Acres of Harvested Wheat in Illinois in 2012.pdf
Representative dot density map of Acres of Harvested Wheat in Illinois in 2012, using county-level aggregate data.

In a one-to-many dot map, each dot on the map does not represent an individual instance, but rather typifies the presence of one or more individuals derived from aggregate data. The data are based on predefined geographical districts (e.g., counties, provinces, countries, census tracts), into which data about individuals have been aggregated as statistical summary variables, such as total population. That is, it is the same type of dataset used for choropleth maps and many proportional symbol maps. Unlike a choropleth map, the only valid variable used for a dot density map is the total count of the individuals. [24] Once a dot value (the number of individuals represented by each dot) is chosen, the number of dots needed in each district can be calculated, and the dots are randomly distributed across the district. This distribution of a total over area gives the visual impression of population density. [18]

Most cartography textbooks prefer to use the term dot density map or dot map only for one-to-many dot maps. [24] [26] [18] The term one-to-many has become problematic as interactive maps have been developed that use this method but with each dot representing a single person, [27] although this is often criticized for creating the illusion of knowing the location of each individual. Other terms that have been suggested to distinguish this technique include representative dot map, district-based dot map, choropleth dot map, and point spread map. [28]

Representative dot design

The design of either type of dot map involves balancing the design of the individual point symbol (especially its size) with the spacing between the points. In the one-to-one dot map, the latter is fixed by the distribution of the individuals and the map scale, but in the representative dot map, it is also influenced by the cartographer's choice of dot value, the number of individuals that each dot represents. It has long been recognized that these choices are interdependent, with several competing considerations: [18] [24]

The ideal balance of these factors occurs when the dots start to coalesce in the most dense areas, the individual dots are just large enough to be seen individually, and the dot value is small enough that even the districts with the lowest values have more than one dot. In 1949, J. Ross Mackay developed a set of guidelines for calculating this balance of dot size and dot value, including an innovative nomograph, which became the standard for the profession. [16] Since then, improving technology in generating dots and printing or displaying them has led to modifications of the balance calculation, which has been automated in most GIS software. [29]

However, this ideal range of apparent densities places some restrictions on the phenomena that can be mapped. If the range of densities is too low (say, a ratio between the most sparse and most dense of less than about 1:10), the map will appear too consistent to be informative. If the range of densities is too high (a ratio of more than 1:1000), too many districts will be solid unless the dot value is decreased so much as to become invisible. [24] Design technology improvements have helped alleviate this restriction somewhat, such as the use of translucent dots, which can show a distinction between densities where dots are just coalescing and higher densities where many layers of dots are on top of one another. [27] However, this has the side effect of making individual dots very faint.

This one-to-one dot map shows the 1,300 immigrants from Germany and Switzerland in Salt Lake City, Utah in 1900 in black, compared to all 55,000 residents shown in gray. Note the blocks in which residents of the same household have been spread into distinct points using the "Grid" renderer in QGIS. German inmmigrants in 1900 Salt Lake City.png
This one-to-one dot map shows the 1,300 immigrants from Germany and Switzerland in Salt Lake City, Utah in 1900 in black, compared to all 55,000 residents shown in gray. Note the blocks in which residents of the same household have been spread into distinct points using the "Grid" renderer in QGIS.

Another design challenge can happen with the one-to-one type of map when multiple points occur at the same location, giving a false impression of lower density (i.e., looking like one dot instead of many). While many users of GIS software do not account for this problem, several automated algorithms have been developed to mitigate it, usually based on the solution developed in the early maps of Shapter and Snow of spreading the points out slightly so that they are distinct but still appear densely packed. [30]

Criticisms

One concern with dot density that has been studied at length is how accurately map readers can interpret the apparent density. Since the 1930s, repeated studies have shown a tendency to underestimate the density of an area shown as dots. [31]

Another criticism is that aggregate district data has inherent problems that can lead to the same misinterpretations as other types of thematic maps based on this kind of data, such as choropleth maps, including the ecological fallacy and the modifiable areal unit problem. In fact, the dot technique can exacerbate the problem, because the detailed look of the individual dots gives the illusion of more detailed data than the solid color of a choropleth. Furthermore, map readers can easily interpret the dots, especially in sparse areas, as the locations of actual settlements. [24]

As with choropleth maps, the modifiable areal unit problem can be mitigated someone by using districts that are as small as is feasible, although this can lead to an increase in the extreme density variation problem discussed above. Another solution in common with choropleth mapping is the dasymetric technique. In the dot density application, external knowledge about the distribution of the phenomenon is incorporated to adjust the dot placement. The simplest approach is the binary method, creating a layer of land known to have no individuals (in the case of human population, this might include features such as water bodies and government-owned land), and using it as a mask to exclude dots from being drawn there, forcing them to be placed more densely in the remaining area. [24] Traditionally, a more subtle approach was taken when manually placing the dots, concentrating them in parts of the district where the density was known to be higher, leading to the appearance of the density varying gradually rather than changing abruptly at district boundaries. [18] Automated algorithms have been developed that mimic this technique, using ancillary information such as city point locations to alter the distribution of dots across each district, although they are not widely implemented in GIS software. [32]

Related Research Articles

<span class="mw-page-title-main">Contour line</span> Curve along which a 3-D surface is at equal elevation

A contour line of a function of two variables is a curve along which the function has a constant value, so that the curve joins points of equal value. It is a plane section of the three-dimensional graph of the function parallel to the -plane. More generally, a contour line for a function of two variables is a curve connecting points where the function has the same particular value.

<span class="mw-page-title-main">Cartogram</span> Map distorting size to show another value

A cartogram is a thematic map of a set of features, in which their geographic size is altered to be directly proportional to a selected variable, such as travel time, population, or Gross National Product. Geographic space itself is thus warped, sometimes extremely, in order to visualize the distribution of the variable. It is one of the most abstract types of map; in fact, some forms may more properly be called diagrams. They are primarily used to display emphasis and for analysis as nomographs.

<span class="mw-page-title-main">Infographic</span> Graphic visual representation of information

Infographics are graphic visual representations of information, data, or knowledge intended to present information quickly and clearly. They can improve cognition by using graphics to enhance the human visual system's ability to see patterns and trends. Similar pursuits are information visualization, data visualization, statistical graphics, information design, or information architecture. Infographics have evolved in recent years to be for mass communication, and thus are designed with fewer assumptions about the readers' knowledge base than other types of visualizations. Isotypes are an early example of infographics conveying information quickly and easily to the masses.

<span class="mw-page-title-main">Choropleth map</span> Type of data visualization for geographic regions

A choropleth map is a type of statistical thematic map that uses pseudocolor, meaning color corresponding with an aggregate summary of a geographic characteristic within spatial enumeration units, such as population density or per-capita income.

<span class="mw-page-title-main">Health geography</span>

Health geography is the application of geographical information, perspectives, and methods to the study of health, disease, and health care. Medical geography, a sub-discipline of, or sister field of health geography, focuses on understanding spatial patterns of health and disease in relation to the natural and social environment. Conventionally, there are two primary areas of research within medical geography: the first deals with the spatial distribution and determinants of morbidity and mortality, while the second deals with health planning, help-seeking behavior, and the provision of health services.

<span class="mw-page-title-main">Dasymetric map</span> Hybrid type of thematic map

A dasymetric map is a type of thematic map that uses areal symbols to visualize a geographic field by refining a choropleth map with ancillary information about the distribution of the variable. The name refers to the fact that the most common variable mapped using this technique has generally been population density. The dasymetric map is a hybrid product combining the strengths and weaknesses of choropleth and isarithmic maps.

<span class="mw-page-title-main">Heat map</span> Data visualization technique

A heat map is a 2-dimensional data visualization technique that represents the magnitude of individual values within a dataset as a color. The variation in color may be by hue or intensity.

<span class="mw-page-title-main">Thematic map</span> Type of map that visualizes data

A thematic map is a type of map that portrays the geographic pattern of a particular subject matter (theme) in a geographic area. This usually involves the use of map symbols to visualize selected properties of geographic features that are not naturally visible, such as temperature, language, or population. In this, they contrast with general reference maps, which focus on the location of a diverse set of physical features, such as rivers, roads, and buildings. Alternative names have been suggested for this class, such as special-subject or special-purpose maps, statistical maps, or distribution maps, but these have generally fallen out of common usage. Thematic mapping is closely allied with the field of Geovisualization.

<span class="mw-page-title-main">Field (geography)</span> Property that varies over space

In the context of spatial analysis, geographic information systems, and geographic information science, a field is a property that fills space, and varies over space, such as temperature or density. This use of the term has been adopted from physics and mathematics, due to their similarity to physical fields (vector or scalar) such as the electromagnetic field or gravitational field. Synonymous terms include spatially dependent variable (geostatistics), statistical surface ( thematic mapping), and intensive property (physics and chemistry) and crossbreeding between these disciplines is common. The simplest formal model for a field is the function, which yields a single value given a point in space (i.e., t = f(x, y, z) )

<span class="mw-page-title-main">Multivariate map</span> Thematic map visualizing multiple variables

A bivariate map or multivariate map is a type of thematic map that displays two or more variables on a single map by combining different sets of symbols. Each of the variables is represented using a standard thematic map technique, such as choropleth, cartogram, or proportional symbols. They may be the same type or different types, and they may be on separate layers of the map, or they may be combined into a single multivariate symbol.

MacChoro was a computer program for choropleth mapping developed for early versions of the Apple Macintosh computer. A choropleth map shades areas, such as states or counties, to represent values and is mainly used for the mapping of statistical data. Released in 1986, MacChoro was the first computer mapping program to implement Macintosh's point-and-click user interface for the analysis and production of thematic maps. MacChoro II, released in 1988, was the first program to incorporate interaction in animated mapping.

<span class="mw-page-title-main">Flow map</span> Thematic map visualizing linear flow

A flow map is a type of thematic map that uses linear symbols to represent movement between locations. It may thus be considered a hybrid of a map and a flow diagram. The movement being mapped may be that of anything, including people, highway traffic, trade goods, water, ideas, telecommunications data, etc. The wide variety of moving material, and the variety of geographic networks through they move, has led to many different design strategies. Some cartographers have expanded this term to any thematic map of a linear network, while others restrict its use to maps that specifically show movement of some kind.

Borden D. Dent (1938–2000) was an American geographer and cartographer who served as professor emeritus and chairman of the Department of Geography and Anthropology at Georgia State University. His textbook, Cartography: Thematic Map Design, is one of the seminal texts in the field, and its sixth edition was reissued in 2009.

The Jenks optimization method, also called the Jenks natural breaks classification method, is a data clustering method designed to determine the best arrangement of values into different classes. This is done by seeking to minimize each class's average deviation from the class mean, while maximizing each class's deviation from the means of the other classes. In other words, the method seeks to reduce the variance within classes and maximize the variance between classes.

<span class="mw-page-title-main">Map symbol</span> Graphic depiction of a geographic phenomenon

A map symbol or cartographic symbol is a graphical device used to visually represent a real-world feature on a map, working in the same fashion as other forms of symbols. Map symbols may include point markers, lines, regions, continuous fields, or text; these can be designed visually in their shape, size, color, pattern, and other graphic variables to represent a variety of information about each phenomenon being represented.

<span class="mw-page-title-main">Makan Map</span> Online Interactive Atlas

The Makan Map is an online interactive atlas of the Xinjiang Uyghur Autonomous Region of China. This region covers an area of 1,664,897 km2, or 16% of China's territory. It allows to visualize geographic, administrative and socio-economic baseline data, to carry out statistical analysis and to create thematic maps for this region. The Makan Map is the first multi-language atlas of the Xinjiang. It has been created in four languages: Uyghur, Chinese, French and English. The current version is the second version of the Makan Map. Development is ongoing for future releases.

<span class="mw-page-title-main">Chorochromatic map</span> Thematic map visualizing a discrete field

A Chorochromatic map, also known as an area-class, qualitative area, or mosaic map, is a type of thematic map that portray regions of categorical or nominal data using variations in color symbols. Chorochromatic maps are typically used to represent discrete fields, also known as categorical coverages. Chorochromatic maps differ from choropleth maps in that chorochromatic maps are mapped according to data-driven boundaries instead of trying to make the data fit within existing, sometimes arbitrary units such as political boundaries.

A visual variable, in cartographic design, graphic design, and data visualization, is an aspect of a graphical object that can visually differentiate it from other objects, and can be controlled during the design process. The concept was first systematized by Jacques Bertin, a French cartographer and graphic designer, and published in his 1967 book, Sémiologie Graphique. Bertin identified a basic set of these variables and provided guidance for their usage; the concept and the set of variables has since been expanded, especially in cartography, where it has become a core principle of education and practice.

<span class="mw-page-title-main">Cartographic design</span> Process of designing maps

Cartographic design or map design is the process of crafting the appearance of a map, applying the principles of design and knowledge of how maps are used to create a map that has both aesthetic appeal and practical function. It shares this dual goal with almost all forms of design; it also shares with other design, especially graphic design, the three skill sets of artistic talent, scientific reasoning, and technology. As a discipline, it integrates design, geography, and geographic information science.

<span class="mw-page-title-main">Proportional symbol map</span> Thematic map based on symbol size

A proportional symbol map or proportional point symbol map is a type of thematic map that uses map symbols that vary in size to represent a quantitative variable. For example, circles may be used to show the location of cities within the map, with the size of each circle sized proportionally to the population of the city. Typically, the size of each symbol is calculated so that its area is mathematically proportional to the variable, but more indirect methods are also used.

References

  1. Pearson Education, Inc. "Key Terms." Making Maps With GIS. Pearson Education, Inc. 8 December 2009. http://wps.prenhall.com/esm_clarke_gsgis_4/7/1848/473320.cw/index.html Archived 2010-01-01 at the Wayback Machine . 1 May 2010.
  2. 1 2 Allen, Jeff (2021-05-11). "Temporal transitions of demographic dot maps". International Journal of Cartography. 8 (2): 208–222. doi:10.1080/23729333.2021.1910184. ISSN   2372-9333. S2CID   236567004.
  3. Robinson, Arthur H. (1982). Early Thematic Mapping in the History of Cartography. University of Chicago Press.
  4. Altonen, Brian (30 January 2012). "Valentine Seaman, 1797 (1804) - The Black Plague or Yellow Fever of New York City". Public Health, Medicine and History. Retrieved 17 November 2020.
  5. Frère de Montizon, Armand Joseph (1830). Carte philosophique figurant la population de la France.
  6. Gilles Palsky (1984). "La naissance de la démocartographie. Analyse historique et sémiologique". Espace, populations, sociétés. 2 (2). Université des Sciences et Technologies de Lille: 25–34. doi:10.3406/espos.1984.956. ISSN   0755-7809.
  7. Konvitz, Josef W., Cartography in France, 1660–1848: Science, Engineering, and Statecraft. University of Chicago Press, 1987. p. 147.
  8. von Mentzer, Thure Alexander (1859). Fysisk-geografiska kartor öfver Skandinaviska halfön för beskrifvande undervisning i fädernes landets geografi.
  9. Kant, Edgar (Jan 1970). "Über die Ersten Absoluten Punktkarten der Bevölkerungsverteilung". Annales Societatis Litterarum Estonicae in Svecia. 5.
  10. Jarcho, Saul (April 1970). "Yellow Fever, Cholera, and the Beginnings of Medical Cartography". Journal of the History of Medicine and Allied Sciences. 25 (2): 131–142. doi:10.1093/jhmas/XXV.2.131. JSTOR   24622309. PMID   4914376.
  11. Shapter, Thomas (1849). The History of the Cholera in Exeter in 1832. Exeter: Churchill.
  12. Johnson, Steven (2007). The Ghost Map: The story of London's most terrifying epidemic-- and how it changed science, cities, and the modern world. Riverhead Books.
  13. Snow, John (1855). On the Mode of Communication of Cholera. John Churchill.
  14. Rosenberg, Matt. "Map Stops Cholera: John Snow's Map of London." About.com:Geography. N.p., 1 May 2010. http://geography.about.com/cs/medicalgeography/a/cholera.htm Archived 2010-12-04 at the Wayback Machine . 1 May 2010.
  15. Coulter, Wesley (Apr 1926). "A Dot Map of the Distribution of Population in Japan". Geographical Review. 16 (2): 283–284. Bibcode:1926GeoRv..16..283C. doi:10.2307/208684. JSTOR   208684.
  16. 1 2 Mackay, J. Ross (1949). "Dotting the Dot Map". Surveying and Maping. 9 (1): 3–10.
  17. Raisz, Erwin, General Cartography, 2nd Edition, McGraw-Hill, 1948, p.250
  18. 1 2 3 4 5 Robinson, Arthur, Elements of Cartography, Wiley, 1960, pp.156-161
  19. Dmowska, Anna; Stepinski, Tomasz F. (May 2019). "Racial Dot Maps Based on Dasymetrically Modeled Gridded Population Data". Social Sciences. 8 (5): 157. doi: 10.3390/socsci8050157 .
  20. Walker, Kyle E. (2018-09-01). "Scaling the Interactive Dot Map". Cartographica: The International Journal for Geographic Information and Geovisualization. 53 (3): 171–184. doi:10.3138/cart.53.3.2017-0021. ISSN   0317-7173. S2CID   135059941.
  21. "Tweetmap". Omnisci. Retrieved 17 November 2020.
  22. "Showing the Location of Tweets and Flickr Photos". New York Times. 15 July 2011.
  23. Leetaru, Kalev (6 Mar 2019). "The Era of Precision Mapping of Social Media is Coming to and End". Forbes.
  24. 1 2 3 4 5 6 7 Dent, Borden D.; Torguson, Jeffrey S.; Hodler, Thomas W. (2009). Cartography: Thematic Map Design (6th ed.). McGraw-Hill. pp. 119–130.
  25. Kraak, Menno-Jan; Ormeling, Ferjan (2003). Cartography: Visualization of Spatial Data (2nd ed.). Prentice Hall. pp. 116–121. ISBN   978-0-13-088890-7.
  26. T. Slocum, R. McMaster, F. Kessler, H. Howard (2009). Thematic Cartography and Geovisualization, Third Edn, page 252. Pearson Prentice Hall: Upper Saddle River, NJ., pp.318-324
  27. 1 2 Cable, Dustin. "The Racial Dot Map: One Dot Per Person for the Entire United States". Cooper Center for Public Service, Demographics Research Group. University of Virginia. Retrieved 17 November 2020.
  28. Imhof, Eduard (1972). Thematische Kartographie. De Gruyter. pp. 154–163.
  29. Kimerling, A. Jon (2009). "Dotting the Dot Map, Revisited". Cartography and Geographic Information Science. 36 (2): 165–182. Bibcode:2009CGISc..36..165K. doi:10.1559/152304009788188754. S2CID   121869966.
  30. Chua, Alvin; Moere, Andrew Vande (2017). "BinSq: visualizing geographic dot density patterns with gridded maps". Cartography and Geographic Information Science. 44 (5): 390–409. Bibcode:2017CGISc..44..390C. doi:10.1080/15230406.2016.1174623. S2CID   124131704.
  31. Provin, Robert W. (1977). "The Perception of Numerousness on Dot Maps". The American Cartographer. 4 (2): 111–125. doi:10.1559/152304077784080374. hdl: 10211.2/5170 .
  32. Hey, Annette; Bill, Ralf (2014). "Placing dots in dot maps". International Journal of Geographical Information Science. 28 (12): 2417–2434. Bibcode:2014IJGIS..28.2417H. doi:10.1080/13658816.2014.928822. S2CID   205793873.