"A bar chart is similar to a line chart, except that each data point is replaced by a rectangle with a height proportional to the value. The rectangle is usually centered on the spatial attribute of the data, and its width is often uniform. When values are categorical or discrete and cannot be shown in a series, a bar chart may be a suitable alternative for the line chart. Similarly to the case of a line chart, it is possible to create multivariate bar charts by stack‑ing the bars on top of each other in a form of superimposition easy to interpret." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"A scatterplot is a data visualization graph that uses dots to represent the relationship between two quantitative variables. One variable, called the explanatory variable, is plotted on the x‑axis, and the other variable, called the response variable, is plotted on the y‑axis. It is also possible to include a third categorical variable, represented by different dot colors. Each dot represents an individual data point, and the colors, when used, represent the categories of the dots. Therefore, the data point is organized into two or three columns, one for each variable, and each data point is plotted on the graph using two coordinates, one for each variable, with various colors representing each category.,." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Closure is a feature related to our capability of completing (closing) an object or a shape that is incomplete, that is, one that has some parts missing. The preattentive processing of closure is also automatic, not requiring conscious effort. For example, when looking at any shape, e.g., a circle or a square, with a small part missing, our brain automatically and preattentively perceives whether the shape is incomplete and fills these gaps. Preattentive processing of closure can be used in visual communication to create recognizable symbols and logos." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Color is a powerful visual tool to encode data and convey different meanings, such as categories, magnitude, visual hierarchy, and even emotions. Using different hues, saturations, and brightness levels can help differentiate between categories or show patterns in the data." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Curvature is another preattentive feature that leads to a fast detection of changes in the degree of curvature, bending, or angularity of a shape or line, such as the presence of a more or less curved line in a group of otherwise similar lines. The degree of curvature in a line or shape can be used to represent different quantities or values, for instance, a smaller or larger number of peaks in a function." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Data visualization, by contrast, focuses on the visual representation of data in such a way that its values, structure, nature, type, and variability are accurately expressed by means of graphs. It aims to support the exploration and understanding of data, the identi‑fication of patterns, trends, distributions, correlations, and anomalies, the communicationof insights, and aid in decision‑making." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Differences in orientation can help us differentiate between items (e.g., data points, lines, objects, etc.) or extract information about the data. For example, using vertical bars in a bar chart can help differentiate between categories, while using horizontal bars can emphasize the magnitude of the data. Angles and direction can be used to convey information, such as trends, movement, sense of depth, or changes in values." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"In data visualization, texture is the visual quality of an object related to its roughness, pattern, or smoothness. It can be created using a variety of techniques, for example, using different line styles, brushes, patterns, and even special effects. Differences in texture can help distinguish between data points or objects, create visual hierarchies, or convey infor‑mation about the data. For example, using different textures for different categories can help viewers quickly identify and differentiate patterns. Like the other features described here, the texture is usually processed preattentively, without the need for focused attention." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Length is another preattentive visual property that can be used to create visual contrast, differences, importance, and proportions. The perception of differences in length normally occurs automatically and rapidly, without conscious effort or attention. It can be used in visual communication to quickly draw attention to important information or to create a visual hierarchy. For example, in a graph, longer bars may indicate larger values or quanti‑ties; in a map, longer lines may indicate longer distances; in a drawing, longer items may convey a sense of flow, etc." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Line charts are useful for identifying patterns and trends in a one‑dimensional sequence of univariate data, that is, continuous data over time with a single value per data item. They map the sequence data (e.g., time) to one dimension, typically the x‑axis, and the data value to another dimension, typically the y‑axis, forming a line; or to the color of a mark or region along the spatial axis, forming a bar. The data is adjusted in size to be within the limits of the display attribute." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Preattentive features, such as color, shape, orientation, and size, are those basic visual properties that are processed automatically, without conscious effort or attention. By understanding preattentive features, data analysts can create effective data visualization designs that make use of them to convey information more efficiently and accurately to the audience." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Size is a preattentive feature that exerts a similar effect in vision as that exerted by the line width, that is, to detect differences quickly and automatically in items (e.g., objects, data points, font sizes, etc.). Differences in size can draw attention to specific data points, indicate hierarchy, emphasize specific items, or convey information about the magnitude of the data. Variation in size can be used to represent different quantities or values, where larger sizes may indicate higher values or importance, while smaller sizes may indicate lower values or importance." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Preattentive processing of 3D (three‑dimensional) properties allows us to detect the depth and spatial relationships between objects, such as the presence of an object that appears to be closer or farther away than the others, without the need for focused attention. Perspective, lighting, size, or shading can be used to create the illusion of depth and convey information, such as relationships between variables." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The histogram is a useful visualization technique to explore the pattern of a single variable distribution, where the x‑axis represents the range of values, and the y‑axis represents the absoluteor relative frequency of data points within each bin. Histograms allow the exploration of cen‑tral tendency measures, such as the mean and median; dispersion measures, such as the stan‑dard deviation; and range, and shape, such as skewness and kurtosis. It also helps to identify outliers or unusual values and to reveal potential biases or errors in the data collection process." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The preattentive processing of density occurs automatically and rapidly, without conscious effort or attention, and can be used in visual communication to create contrast and emphasize importance or relevance. This feature can be swiftly detected by the presence of varying numbers of objects (e.g., data points or shapes) in a given region of the space, rep‑resenting different quantities or values. For instance, in a chart or graph, a higher density of data points can be used to represent a larger quantity, a more significant trend, or a more exciting or energetic area. By making use of the preattentive processing of density, design‑ers can create effective visual designs that convey information quickly and efficiently to the viewer." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The preattentive processing of markings (e.g., stripes, dots, crosses, stars, hatchings, etc.) includes various visual properties, such as texture, shading, and patterns. These properties allow us to swiftly detect differences and similarities between objects or regions, such as the presence of a repeating pattern in a group of otherwise random shapes. The presence or absence of certain markings, such as dots or squares, can be used to represent different categories or values." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The principle of closure states that incomplete objects are perceived as complete because our brain tends to fill the gaps to create the complete image. Note that closure is also a pre‑attentive feature and thus plays a key role not only in the quick filling of gaps or completion of shapes, but also in the organization of the information to be conveyed."(Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The principle of common fate proposes that objects that move together or change similarly tend to be perceived as a group or a pattern. In this case, graphs that allow visualizing data obeying this principle will have to embody a type or a sense of motion. To illustrate this principle, let us consider a motion chart, a streamgraph, and a force‑directed graph. The motion chart is a visualization method that shows how data changes over time; the streamgraph is a stacked area graph that shows the changes in a set of data over time; and the force‑directed graph is a network visualization that shows the relationships of nodes in a graph. In all cases, there is a sense of common fate in the data." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The principle of continuity states that objects that are arranged in a smooth, continuous way are more likely to be perceived as a single object, even if their pattern is interrupted. The line chart, the Sankey diagram, and the scatterplot are good examples of the principle of continuity in the use of Gestalt theory in data visualization." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The principle of figure‑ground, also called figure‑field, states that objects are perceived as either being in the foreground or the background. One way of forcing this principle is by using contrasting colors in the background and foreground of an image, for instance, black and white, blue and orange, green and purple, red and green, yellow and purple, pink and green, and others. However, many of these pairs are not suitable for technical and scientific works, and thus, the recommendation is to use colors with parsimony." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The principle of proximity proposes that objects that are close to one another tend to be perceived as a group or a pattern. In data visualization, the heatmap, the scatterplot, and the bar chart are good examples of methods that account for the principle of proximity. The heatmap is a graph in which the values of a matrix are represented by colors, which are a preattentive feature, and neighboring cells in the matrix convey a sense of organization and relationship. The scatterplot places similar data values close to one another, grouping them in the plot. In a bar chart, related data values are placed close together in the bars, allowing a visual association among them." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The principle of similarity proposes that objects that share similar characteristics, such as color or form, tend to be perceived as a group or a pattern. Examples of data visualization techniques that account for the similarity principle in Gestalt theory include a line chart in which lines representing different categories have the same style, a bar chart in which the bar patterns or colors indicate the same group or category, and a scatterplot with different markers representing different categories of categorical variables." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"The principle of symmetry states that objects that are symmetrical, or have a balanced appearance, tend to be perceived as a group or a pattern. Some data visualization graphs that can be used to explore this principle are the boxplot with boxes symmetrically placed around the median (Q2), the radar chart displaying multivariate data as a bidimensional chart with quantitative variables, and the mirrored bar chart with two sets of bars with mirrored values displayed." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
"Preattentive processing of position allows us to quickly detect changes in location, such as the presence of a dot or other object that is slightly displaced from the others. The spa‑tial location of visual elements can also be used to guide the viewer’s attention or encode information, such as ranking, hierarchy, or relationship (grouping)." (Leandro N de Castro, "Exploratory Data Analysis: Descriptive Analysis, Visualization, and Dashboard Design", 2025)
No comments:
Post a Comment