Curly chart. Line charts, charts and statistical maps

If, for example, you decide to use a figure chart to depict the structure of unemployed women, among which 57% are young women (20-24 years old) and. girls 16-19 years old with no work experience 28% - engineering and technical workers and employees with special education aged 25-49 years and 15% - skilled and unskilled workers aged 50 years and older, you must depict three female

As with bar charts, you can trace price support and resistance levels on figure charts. In addition, the figure chart especially clearly demonstrates the breakout moment (Fig. 2 at the top of the page).

Many analysts use shape charts to predict how far an increase or

According to those who use shape charts, the more often a stock's price moves at a certain level, the more likely it is to rise or fall from that level. As seen in fig. 1 (turning step 1 pip), the target buy price is set at 21 pip, as the price crossed the 24 pip level 6 times (shaded line). Just six lines below is the target price of 21 pips. Similarly, the target sell price is set at 23 1/2 points due to the fact that the price crossed the 21 1/2 level 4 times (shaded line). The beginning of the breakout is always taken as the basis for calculating the target price, regardless of whether the target price is to be determined - the price of buying or selling.

Fed) 195, 200, 225, 229-233 Shape charts 298-300 Financial responsibility 16 Financial statements 21, 65, 66, 68-

Curly comparison diagrams are intended mainly for popularization purposes. The indicators in them are drawn in the form of a certain number of standard figures, which are simplified images of objects characteristic of the corresponding phenomena. Their disadvantage should be considered some inaccuracy associated with the need to round off the displayed indicators.

The main forms of graphs that are used in AHD are diagrams. Charts in their form are bar, strip, circular, square, linear, figured.

The diagram of dynamics is intended for representation of change of the phenomena for the corresponding intervals of time. For this purpose, bar, circle, square, curly and other graphs can be used. But line charts are more commonly used. The dynamics on such a graph is presented in the form of a line that characterizes the continuity of the process. To build line graphs, a coordinate system is used on the abscissa axis to plot periods, and on the ordinate axis - the level of indicators for the corresponding periods of time, based on the accepted scale.

Curly (or picture) diagrams enhance the visibility of the image, as they include a picture of the displayed indicator. The size of the figure corresponds to the size of the indicator (Fig. 4.9).

Graphs can be very different. More often than others, diagrams and cartograms are used. In turn, charts can be line, bar, pie, ribbon, curly, distribution histograms

Digital data can also be shown using combined charts, which simultaneously use curly lines, columns, circles, symbols, images of objects, etc. However, it is difficult to read digital material using such charts.

With EA, tabular and graphical reflection of analytical data is used. Moreover, each table must have a common heading, a system of horizontal lines and vertical columns. The subject of the tables shows what is at stake (it contains a list of indicators characterizing the phenomenon), and the predicate indicates what features the subject is characterized by. Tables are simple, group and combined (subject material is divided into groups and subgroups). Graphical reflection of information is carried out using diagrams (bar, circular, figured, etc.), distribution curves, correlation field graphs, statistical cartograms, which allows you to get a general picture of the state of affairs in statics and dynamics.

The dynamics diagram is designed to depict the change in phenomena over appropriate time intervals. For this purpose, histograms, linear, curly and other graphs can be used. Line charts are the most commonly used. The dynamics on such a chart is presented in the form of a line, which is characteristic

Figure comparison charts are mainly for promotional purposes. The indicators in them are drawn in the form of a certain number of standard figures, which are simplified Oi

There is another method of technical analysis that is completely different from the construction of bar charts, which is to build figure charts. Although the x and o signs used do not accurately indicate the timing and volume of trade on the chart - the two most important components of bar charts - figure charts still have their merits. After learning the simple rules for constructing curly charts, an investor, as in the case of bar charts, should practice analyzing already built modelsbefore applying the theory in their own practice.

Before building a curly chart, it is necessary to study the previous range of fluctuations and the degree of instability in the stock price. In addition, you should select the appropriate scale, or rotation step, of the movement, which will be used when plotting the diagram. For high value stocks (eg $50 and up) a step of 2, 3 or maybe 5 points is usually appropriate. Stocks with an average price (say, $20 to $50) may be better suited to move 1, 11/2, or 2 points. To analyze low-cost stocks, as a rule,

Point-and-figure charts (tic-tac-toe) are characterized by m, that there is no time axis, and only one axis is used - for the target, the chart reflects only price changes (Fig. 3.6). The chart is filled from left to right, if there is a change in the direction of prices, it is customary to mark the price increase with a cross, and its decrease with a zero,

The simplest interpretation of the scatter chart is to buy (go long) if a cross appears above the top cross of the previous column of crosses, and sell (go short) if there is a but-j so below the zero of the previous column of zeros. . I

The most common way to graphically represent data is through charts. They come in different types - linear, radial, point, planar, volumetric, curly. The type of charts depends on the type of data presented (one variable or one indicator, several variables or indicators, quantitative or non-quantitative) and the task of plotting.

There are many types of charts. The most commonly used are line charts, pie charts, radial charts, figure charts, volume charts, and planar charts. Cartograms and cartograms are used to show the geographical distribution of data.

curly charts represent an image in the form of drawings, silhouettes, figures.

The use of graphs for the presentation of statistical indicators makes it possible to give the latter visibility and expressiveness, facilitate their perception, and in many cases helps to understand the essence of the phenomenon under study, its patterns and features, to see the trends in its development, the relationship characterizing its indicators.

Statistical graphs can be classified according to different criteria: purpose (content), method of construction and nature of the graphic image.

By content or purpose one can distinguish graphs of comparison in space, graphs of various relative values ​​(structures, dynamics, etc.), graphs of variational series, graphs of placement across the territory, graphs of interrelated indicators. Combinations of these graphs are also possible, for example, a graphical representation of variation in dynamics or dynamics of interrelated indicators, etc.

By construction method Graphs can be divided into diagrams, cartograms and cartograms.

By the nature of the graphic image There are point, line, planar (bar, hourly, square, circular, sector, curly) and volumetric graphs.

An example of a diagram is fig. 3.2.

Rice. 3.2. Oil reserves in selected countries in 1987

A variation of the bar chart is a strip (ribbon) chart, which is characterized by a horizontal orientation of the bars (bars) and a vertical baseline. The bar chart is especially convenient in cases where individual objects of comparison are characterized by indicators opposite in sign (Fig. 3.3).

Rice. 3.3. Oil production in selected countries in 1986 compared to 1970

Square and pie charts are less visual than bar and bar charts, which is due to the difficulty of visually assessing the area ratio. Therefore, inside the squares and circles, the values ​​​​of the displayed indicators should be affixed (Fig. 3.4). Volumetric diagrams (for example, in the form of cubes) are even less clear, in which the limit sizes of the graphic image are proportional to the cube roots of the compared values.

Rice. 3.4. Population of China and Canada, million people

The main form of structural diagrams are pie charts (Fig. 3.5). The "working" geometric parameter in the sector diagram of specific gravity is the angle between the radii: 1% is taken to be 3.6° in the diagram, and the sum of all angles, which is 360°, is equated to 100%.

Rice. 3.5. Structure of assets of a commercial bank by degree of risk.

Dynamic diagrams are used to depict economic phenomena occurring over time. Unlike diagrams that display the comparative values ​​of individual objects or their structure, in dynamic diagrams, processes are the object of display.

Geometrically adequate form of their reflection are linear coordinate diagrams (Fig. 3.6.).

Rice. 3.6. The level of the average price of privatization checks at the RTSB auctions, rub.

Rice. 3.7. Distribution of apartments according to the number of people living in them.

Linear and planar diagrams constructed in a rectangular coordinate system are used to depict variational series. With a discrete variation of a feature, the distribution polygon serves as a graph of the variation series (Fig. 3.7.).

The distribution polygon is a closed polygon, the abscissas of the vertices of which are the values ​​of the variable attribute, and the ordinates are the corresponding frequencies.

Rice. 5.25. Cartogram of the population density of eight districts of the region

According to the data in Table. 5.9 all areas by population density can be divided into three groups:

1. areas with a population density of up to 4 thousand people;

2. from 4 to 12 thousand people;

3. from 12 to 17 thousand people.

Then the first group includes districts No. 1, 8; to the second - No. 2, 3, 7; to the third - No. 4, 5, 6. If we take for each group of districts the color of different saturation, then the background cartogram clearly shows how individual districts are located on the territory of the region in terms of population density (Fig. 5.25). Another example of a background cartogram is fig. 5.26.

Rice. 5.26. Population density in the regions of the Central region of Russia (persons per 1 m2)

The second large group of statistical maps are chart diagrams, which are a combination of diagrams with a geographical map. Chart figures (bars, squares, circles, figures, stripes) are used as figurative signs in cartograms, which are placed on the contour of a geographical map. Cartograms make it possible to reflect geographically more complex statistical and geographical constructions than cartograms.

Among cartodigrams, it is necessary to single out charts of simple comparison, graphs of spatial displacements, isolines.

On a cartographic diagram of a simple comparison, in contrast to a conventional diagram, the diagrammatic figures depicting the values ​​of the indicator under study are not arranged in a row, as in a conventional diagram, but are spread throughout the map in accordance with the region, region or country they represent.

Elements of the simplest cartographic diagram can be found on a political map, where cities are distinguished by various geometric shapes depending on the number of inhabitants.

As an example of a chart diagram, let's take the image of the gross grain harvest of the Central Region of Russia (Fig. 5.27).

Rice. 5.27. Gross harvest of grain in the Central region of Russia (conditional data)

Contours(from the Greek isos - equal, identical, similar) - these are lines of equal value of any quantity in its distribution on the surface, in particular on a geographical map or graph. The isoline reflects the continuous change of the studied quantity depending on two other variables and is used in mapping natural and socio-economic phenomena. Isolines are used to obtain quantitative characteristics of the studied quantities and to analyze the correlations between them.

The listed types of graphs are not exhaustive, but they are the most commonly used.

Line graph. For construction, a system of rectangular coordinates is used. Variants of the studied indicator (or time) are plotted on the abscissa (horizontal) axis, and the value of the studied indicator is plotted on the ordinate axis. When building a line graph, it is very important to choose the right scale. An important advantage of line charts is that several indicators can be displayed on the same field of the chart, which makes it possible to compare and identify the specifics of their development. An example of a line chart is shown in fig. 2.

A chart is a graph in which statistical information is displayed using geometric shapes. Diagrams are used for visual comparison of socio-economic phenomena in space and analysis of their dynamics. When plotting charts using software (including MS Excel), scaling is performed automatically. The user can additionally configure the formats of the axes and the coordinate grid (the frequency of category labels, in what value the axes should intersect, etc.). Most often in practice, bar charts are used. In MS Excel, bar charts are called histograms.

Bar charts are used to compare statistical indicators characterizing different objects or the same objects in different years. They can be used in flat (two-dimensional) and three-dimensional (three-dimensional) images.

When building bar charts, each value of a statistic is displayed as a vertical bar. The columns are built in a rectangular coordinate system. The bases of the columns are placed along the abscissa axis, the width and distance between which are chosen arbitrarily, but must be the same. The height of the bars varies depending on the value of the statistical indicator. It is possible to simultaneously display several indicators on one chart. An example of a flat bar chart is shown in fig. 3.

A more visual variety of bar charts is a three-dimensional chart, which makes it easy to compare statistical data with each other and at the same time see their development in dynamics. An example of a three-dimensional diagram is shown in fig. 4.

Strip (tape) charts. In strip charts, the bases of the columns are located vertically, and the scale scale is applied to the horizontal axis and determines the size of the strips along the length of the corresponding values ​​of the displayed statistical indicators. When constructing bar charts, the same requirements are met as when constructing bar charts. An example of a bar chart is shown in fig. 5.

Pie (pie) charts. Various types of pie charts are used to depict the structure of one statistical population. The area of ​​the circle is taken as the value of the entire population, and the areas of individual sectors reflect the specific gravity (share) of its constituent parts. It is best to display the structure as a percentage. Then the whole circle is 100%.

The pie chart reflects indicators that are parts of one whole. For example, using a pie chart, you can clearly show the structure of convictions for the main offenses for the required period (Fig. 6 and 7).

Comment. A common mistake is when a pie chart is used to display any values ​​of one or more indicators over a number of years. A bar chart should be used to graphically represent such data.

Radial diagrams. In radial charts, the origin is the center of the circle, and the scale scales are the radii of the circle. In the MS Excel application, this type of chart is called a radar chart, which is an analogue of a chart in a polar coordinate system. An example of a radial diagram is shown in fig. 8.

The values ​​of crime intensity indicators by federal districts are plotted on the radii.

Statistical maps are used to characterize the distribution of a phenomenon in a certain area. Statistical maps are divided into cartograms and cartograms. The difference between them lies in the way the statistics are displayed on the maps.

A cartogram is a geographical map or diagram, which, using some conventional symbols (shading, coloring or dots), shows the degree of distribution of a particular phenomenon in space (for example, the crime rate by district, population density, etc.). Software that allows the user to build cartograms usually includes geographic information systems tools (a set of electronic maps with administrative-territorial divisions) and a tool for setting up the display of a range of data gradations (color palette).

On fig. 9 shows an example of a cartogram for the absolute number of registered crimes in the constituent entities of the Russian Federation in 2008.

Comment. When constructing cartograms, situations are possible when the name of an administrative-territorial division cannot be placed on the cartogram (it goes significantly beyond its boundaries or a very small font must be used). In this case, the names of the labels are taken out as an explanation - the legend. Thus, some of the territories have names on the map, and some are indicated by numbers, the values ​​​​of which are presented in the table.

A chart diagram is a combination of a geographical map or its scheme with a diagram. At the same time, various figures are not placed in a row, as in a regular diagram, but are spaced on a certain scale throughout the map in accordance with the area they represent. The cartogram not only gives an idea of ​​the value of the studied indicator in different territories, but also depicts the spatial distribution of the studied indicator. With the help of cartograms, more complex statistical and geographical comparisons can be reflected in comparison with cartograms. An example of a chart diagram is shown in fig. 10.

The chart shows statistical data for 2002 for the Urals Federal District: in terms of industrial output - in terms of territory coloring, and in terms of wages - in the form of a bar chart in share terms. Comparison is carried out visually both between sectors of the economy within the region and between regions, while the values ​​themselves are not displayed.

art theory

UDC 766:003.63

V.V. Laptev

curly charts in infographics: scope, classification and construction rules

LAPTEV Vladimir Vladimirovich - Associate Professor, Department of Engineering Graphics and Design, Institute of Metallurgy, Mechanical Engineering and Transport, St. Petersburg State Polytechnic University; Ph.D. in History of Arts.

Russia, 195251, St. Petersburg, Politekhnicheskaya st., 29

e-mail: [email protected]

annotation

The article is devoted to figurative representation in infographics. The types of diagrams using figurative images are considered in a historical retrospective. Particular attention is paid to figured quantitative diagrams, their scope, classification and construction requirements.

Keywords

INFOGRAPHICS; DATA VISUALIZATION; ILLUSTRATIONS; pictorial statistics; FIGURED CHARTS; PICTOGRAMS.

Modern infographics, presented on numerous examples in popular science publications, business, management and journalism, demonstrates the variety of tools for graphical representation of numerical data. Currently, to increase the attractiveness, various visual means are increasingly used as decorative accompaniment and as bearing a functional load. The last type of use of images includes figure charts, in which the graphical representation of certain numerical data is made in the conditional form of artistic or symbolic images - figures. The observed growth in the popularity of such graphs indicates the relevance of generalization

experience of their use and classification. And the often encountered construction errors and discrepancies between the type of graphs and the ideas presented, the lack of understanding of the scope of curly diagrams reveal the need to formulate the rules for their construction.

Figure diagrams combine the attractiveness of the resulting image and the utility of transmitting statistical information, which makes it easier to understand models in an applied problem. Therefore, they satisfy two conditional classification criteria at once, according to which the graphical representation of data in information design can be divided into emotional and rational infographics in relation to the artistic decorativeness of the image.

Based on this, figure charts are widely used in non-fiction publications, corporate reports, business presentations, media and education because they are visually appealing. They can also be used for the rational presentation of statistical arrays of numerical data and the results of scientific research.

However, these are graphs of a very simplified type, having certain distortions and conventions that do not allow us to talk about the accuracy of visual representations. In shape charts, each group of data is displayed as a separate segment of the chart, consisting of an image, part of it, or a group of identical symbols. They are most effective for displaying simple comparisons in a variational series, proportional percentages in a data structure, short discrete time series.

Figure diagrams can be considered an artistic development of planar diagrams using geometric figures as a graphic image. Replacing abstract rectangles, triangles and circles with artistic or symbolic images, they inherited such negative features as inaccuracy and approximateness of the quantitative assessment of the indicator, but in addition received artistic and figurative expressiveness, which is based on the visibility of the object of comparison. At first glance, the reader can understand that we are talking about wheat or tractors, electricity or financial results. Images not only reinforce the meanings of the numbers in the diagrams, but also help to understand their meaning - political, economic, military, etc.

Two types of curly diagram classifications can be distinguished - predicative (conceptual representation of properties) and architectonic (comparison of construction plans). The most important and scientifically substantiated are the predicative typological classifications. These include, for example, multiple division by the type of diagrams (comparison diagrams, structural diagrams, dynamic time series diagrams, distribution diagrams - a statistical aspect) or by the type of image of ideogram figures (they can be used as

drawings, photographs, silhouettes, pictograms - the art criticism aspect).

The architectonic classification considers diagrams in terms of construction, describing them according to their formation: according to their constituent parts and connections of individual elements. This is reflected in the form of visual representation of numerical data, for example, linear or planar, vertical or horizontal arrangement of pictorial elements. This constructive aspect also makes it possible to divide shape diagrams into three architectonic types: planar (or scalable), normalized and quantitative, which will be discussed further.

Figured planar diagrams, in which the size of the image is proportional to the displayed values, have received the greatest distribution in the visual statistics of the past (Fig. 1). They have the main disadvantage of all types of planar diagrams - the implicit comparison of quantities. Different figures can be compared either according to some linear feature (height or base), or according to the occupied area. This dichotomy of comparison results in the fact that the reader does not know what to compare.

To associate data with a graphic image, an image corresponding to the topic is selected - a pictogram or an ideogram. For example, for visual representation of

Rice. 1. Figured planar diagrams (the image size proportionally depends on the displayed values)

in agriculture, it will be a tractor, for the electric power industry - a lightning bolt, for mechanical engineering - a gear, etc. Now, in order to visualize indicators that differ from each other by a factor of two, you should change the scales or proportions of the figures. This can be a simultaneous change in linear terms, when both the length and width of the ideogram are doubled. It is also possible that the image is enlarged in comparison with the original twice in area.

If the comparison is made by linear dimensions, then the ratio of the values ​​presented will be violated (significantly exaggerated). If we compare the areas of the figures, then there is an underestimation of the differences in the indicators. It has been noticed that for the human eye the “fair” assessment of the figures is somewhere in the middle, which certainly requires experimental confirmation not only for regular geometric images, but also for images with a complex curvilinear contour.

Another option for visual representation of relationships is the use of figures in which only one linear dimension is changed. In most cases, this leads to significant optical deformations: the artistic sense of proportion is violated due to the fact that the figure acquires the character of a grotesque elongation or thickening. The best option for planar diagrams can serve as an image that can be changed in one linear dimension without optical deformation. If this is hardly possible for a tractor symbol, then a graphic designer can “stretch” a pencil, a train, a fence or a stack of coins without distorting the artistic image. This becomes possible due to the linear shape of the figure, which initially has a “floating” size in length or height. In fact, such an image replaces a rectangle in bar charts, which means that a one-dimensional comparison base is used.

Another type of pictorial graphs are curly normalized charts, in which parts of the image are fractions of the whole, i.e. the structure of the indicator. If a curvilinear image or a complex

geometric figure, the user may encounter difficulty visually comparing individual disproportionate parts. For example, for a triangle with a horizontal division, this would be a comparison of the lower trapezoids and the upper triangle. Such a comparison of shapes of different shapes would require mandatory numerical support to minimize visual distortions.

In some cases, the triangle is recognized as convenient for visual representation of an ascending series of level character (education level - primary, secondary, higher; income level - low, medium, high), when the structure has a pyramidal character. The bottom level is larger compared to each subsequent top level, so in such a diagram, the form corresponds to the content. But despite this, the use of a triangle as a figure representing the structure of the whole has not gained much popularity in infographics.

If the use of regular geometric shapes other than a square or rectangle as a whole structure required such efforts, then what can we say about other, more complex images. The division of a drawing, photograph or pictogram with a curvilinear contour into several parts is accompanied by even greater distortions and is characterized by the general approximation of this method. Images of a concentric shape (circle, ring or part of them) stand apart - in this case, it is possible to model a pie chart based on the presented figure.

The next view: curly quantitative charts, in which each of the symbols represents a certain constant value, and their total number corresponds to the data. The founder of the use of images as counting units is the American engineer Willard K. Brinton, who replaced the proportional scaling of figures with a change in their number. In his monograph Graphic Methods for Presenting Facts (1914), he called for the popularization of this method, the abandonment of multi-format images, and the adoption of repetitive pictograms to improve quantitative analysis, so

how, in his opinion, in this case, the image of a histogram with a large grid of divisions arises.

The Austrian sociologist Otto Neurath was the most consistent supporter of the use of shaped quantitative charts. Having headed the Socio-Economic Museum in the capital of Austria in 1925, he put the use of figurativeness in diagrams on stream. The Viennese method of pictorial statistics was proposed, focusing on pictograms. They were worked by such well-known artists as Gerd Arntz, Peter Alma, Augustin Tschinkel, Erwin Bernath, Rudolf Modley.

Despite the priority of W. Brinton in the use of conventional signs as counting statistical units, the name of O. Neurath is inextricably linked with pictograms. If an American engineer only outlined the method of figured quantitative diagrams, then in Vienna the formalization of the rules of application continued, the popularization of this method not only in statistics, but also in sign systems. The result was the figurative language ISOTYPE, which was based on pictograms, which became the new symbolic tool for information design.

In the 1930s, figure charts based on pictograms became

on foot conquered the media space of Europe, America, the Soviet Union. Science enthusiasts used them to popularize the latest research, propagandists - for the purpose of agitation (Fig. 2). In secondary education and health education, graphics of this kind acted as visual aids. Widespread use required a certain unification of their use.

The stylistic features of the Vienna method of isostatistics were based on a number of imperative requirements. Firstly, exclusively curly quantitative charts were used as a tool. Line charts, pie charts and histograms were considered unsuccessful for the perception of information. Secondly, as figures, emphasis was placed on the symbolic representation of images using the simplest pictograms. Sometimes the use of silhouette images was allowed. Thirdly, the pictograms were made in a flat form. The projection representation of images was denied, all hints of volume - shadows, as well as perspective. Fourth, the numerical information in the diagram was presented in an implicitly encrypted form. Each sign-pictogram corresponded to a certain numerical value, the image of a part of the sign was allowed - half or a quarter (the so-called "cut sign"). This concerned absolute and relative values.

Each sign stands for one million tons of pig iron

Rice. 2. Curly quantitative charts (pictograms are divided into fives for ease of counting)

Curly quantitative charts, as a rule, were built either like strip charts (symbols were placed horizontally, one- and two-way), or applied to maps. Compositionally, the construction of diagrams was carried out minimalistically, without decoration and any unnecessary information, only with the addition of the so-called guiding illustration. It was forbidden to use different shades in the pictogram - only monochrome shading.

Special requirements were placed on pictograms. When choosing a symbolic image, we followed the rules formulated by R. Modly, who had a great influence on the popularization of the Vienna method in the USA:

The image of the symbol must comply with the principles of good drawing established in the fine and applied arts;

The symbol must be suitable for both large and small images;

The symbol should give a generalized image, and not convey individualized features;

The symbol must be easily distinguishable from any other;

The symbol must be of interest;

The symbol is essentially a unit of account, and it must first of all be clear as a unit of account;

The symbol must be suitable for both outline and silhouette representation.

In the USSR, in parallel with foreign experience, the domestic method of Isostat, proposed by I.P. Ivanitsky. In it, the struggle with the main drawback of figured quantitative diagrams - the approximation and inaccuracy of graphical interpretation - resulted in the maximum refinement of the "cut sign". This was done by placing symbols

fishing in the form of a film - repeating frames (Fig. 3). This method combined curly and abstract geometric diagrams, both horizontal bar and vertical bar. Parallel to the signs, there was a division scale for each module, which made it possible to solve the problem of the accuracy of the graphical representation of statistical data.

In the period 1931-1941, Soviet artists published dozens of books, brochures, illustrated albums, posters and postcards, giving the world excellent examples of the complex use of the entire spectrum of visual statistics tools. Post-war infographics in the USSR, unfortunately, were no longer at the forefront of information design. And not only in our country, but all over the world, during the period of the culmination of modernism, there was a transition from albeit to the limit schematized, but still artistic images in Viennese-style pictograms to dry visualization of data, which could be called infographics with a big stretch. There are new trends in the presentation of information - functional, geometrically accurate, with internal logic and mathematic aesthetics of construction. This can be seen in the example of Swiss posters, made solely on the basis of the laconicism of geometric figures, cleared of extraneous illustration and subjective feelings.

In academic circles, the opinion is being strengthened that charts and maps are special forms of communication - a visual language of communication. Thus, the French scientist Jacques Bertin analyzed diagrams, networks, signs and maps as a semiotic system, highlighting their features in the now classic monograph Semiology of Graphics (Semiologie Graphique, 1967). figurativeness in

presented concepts and examples was almost invisible.

According to the definition of the American philosopher Nelson Goodman, visualization must be syntactically written, that is, it must consist of discrete and incoherent visual symbols. This means that a line graph, a pie chart, a planar chart, or a bar chart has a certain number of points or segments connected by lines into the correct image. In this case, the size and location of the points, the length and shape of the lines may not be important for the observer.

Goodman's ideas about visualization as a special illustrative language provided their own symbols and conventions. Geometric elements acted as signs: points, segments, lines, which functioned as signs in writing. The reader does not focus on individual letters - he perceives the text as groups of words or sentences. So the diagrams are read in one piece. Strikingly, "we think of such diagrams more like schematized pictures." At the same time, the display of numerical data using visual techniques, including drawings, photographs, and even pictograms, was recognized by Goodman as a worse visualization translator than geometric abstraction.

Starting from the middle of the 20th century, statistical infographics demonstrated a movement from the design and artistic method that developed in the 1920s and 1930s to a formalized process of data visualization. In many publications that use statistical data, their graphical interpretation

based on geometric primitives. Linear, pie, bar (bar and strip), planar charts have become the de facto standard of visualization.

This process can be seen in the statistical yearbooks of the Netherlands (Statistischjaarboek), which provided the most important indicators of the life of Dutch society in the form of tables, diagrams and thematic maps. If until 1965 figured quantitative charts were an important part of the graphical representation of numerical data, then after the retirement of the artist G. Arnz, who was in charge of the design department at the Central Bureau of Statistics of the Netherlands, they were replaced by abstract graphic images. Numerous tables with numerical data began to play the leading role, which became the spokesman for information design, columns or cartograms rarely appeared. And only with the growing popularity of emotionally oriented infographics at the end of the 20th and beginning of the 21st century, curly normalized and quantitative charts began to timidly appear in these yearbooks. In them, silhouette images acted as ideograms denoting, for example, the unemployment rate, the average cost of housing or the export of goods. Often such signs were chosen randomly, without taking into account the angle of the image and did not fully reflect the object or phenomenon. So, to indicate the proportion of twins among newborns, they used the silhouette of the child (perhaps through the contour of the photograph) and, by shading with different intensities, they tried to show these values ​​at 1.24 and 1.59% (Fig. 4). As a result, the sign is read with great difficulty, especially

Rice. 4. Use of silhouette images as ideograms (the proportion of twins among newborns is shown using shading of different intensity)

when decreasing, semantically and quantitatively it does not correspond to the structure of births (error by 10 times!).

As a rule, pictorial elements are arranged in horizontal rows, so figured quantitative charts for the most part resemble bar charts. However, the latter symbolize a comparison of linear values, and curly quantitative ones - a numerical comparison. This is especially noticeable when grouping characters. On fig. 2 you can see that the pictograms for the convenience of counting follow the fives, between which there are intervals. But in the Dutch statistical yearbooks, it is precisely the linear, and not the counting, representation of graphic images that can be traced: the figures simply fill the space of invisible bar charts (Fig. 5). Hence - fragments of images, losing the value of the counting unit.

Similar shortcomings were noticed in many domestic and foreign publications, corporate reports, information

Werkzame personen, 2012

Werknemers Zelfstandigen

J 3 4

Aantal huwelijkssluitingen

W = 10,000 huwelijkssluitingen

Rice. 5. Linear representation of graphic images

messages. It becomes obvious that it is necessary to formulate and disseminate uniform rules for the use of curly quantitative charts, as well as to introduce construction algorithms into a software product for automating the design of quantitative infographics.

As a result of the generalization of the studies carried out in this area, it is possible to present the fundamental rules for constructing curly quantitative diagrams. The main challenges facing infographers when designing them are choosing or designing an appropriate shape symbol and finding the unit size to convert a bar chart to a figure chart. The clarity and visibility of the entire graphic image, the minimization of inaccuracies, the absence of a “cut sign” or the convenience of reading it depend on the right choice.

The first step is to analyze a set of numerical data to find out the very possibility of using this type of visualization. Among the restrictions are the following:

The number series or structure should not be long, i.e., no more than 5 components (short sets are most preferable for figure charts - one, two or three numbers can be effectively represented precisely by pictorial methods);

The numerical data of individual features should not change by more than 50 times due to the difficulty of visually comparing sharply different values ​​​​by counting (for such a comparison, another type of chart should be used).

At the second stage, the counting unit is searched, i.e. the diagram module:

A smaller indicator or part of it is considered as a counting unit (the selection of the coefficients for dividing the module by 1.5; 2; 2.5; 3, 4, etc.);

It turns out how accurately the remaining indicators are described by such a counting unit (the choice of the optimal approximation due to integers and "cut signs");

The largest indicator is examined for a sufficient number of modules - it cannot be depicted by 4 characters or less (then the diagram is disaggregated

by reducing the counting unit, for example, by 2 times).

The third stage involves the transformation of each member of the series or fraction of the structure into a discrete form:

An integer number of modules is determined, of which each number in the aggregate is composed;

Numeric values ​​are rounded to form end pieces for each member of a row or structure fraction: a whole module or a “cut mark” (i.e., divided into 2 or 4 parts).

The process of creating a figured quantitative diagram is completed by turning the modules (and their parts) into figurative ideograms - drawings, silhouettes, pictograms. For a better perception of signs as counting units, they should be grouped by 5 or 10 characters with the designation of the gap between the groups. In the case of sharp differences in numerical data, large values ​​\u200b\u200bcan be displayed as a snake, that is, several rows one under the other. In this case, not a linear representation is formed, but a planar one - in the form of rectangles. As a “cut sign”, you can use an image divided in half (in extreme cases into 4 shares) with a complemented contour

or transparency like those used in Soviet infographics of the 1930s.

Curly quantitative charts are not recommended to be used in conjunction with other graphical images: columns, lines, etc. To compare the data shown in the figure chart with another number series, you must use a separate chart field with its own system of scale and spatial references.

The result of this study was the identification of three architectonic types of shape diagrams and the definition of their scope in the visualization of numerical data. This is a graphical representation of comparison, structure, and distribution for shaped quantity charts. For other types, there are more serious restrictions on use. Figured planar diagrams can be used only in the case of a simple comparison, and curly normalized diagrams can be used to identify a single structure. Curly quantitative diagrams are one of the most expressive means of conveying information with the participation of artistic images. The above rules for their construction will help to eliminate errors, take a step towards automating the processes of designing infographics.

bibliography

1. Gelman A., Unwin A. Infovis and Statistical Graphics: Different Goals, Different Looks // Columbia University. official website. 11 June 2011. URL: http://www.stat.columbia.edu/~gelman/research/published/vis14.pdf (reference date: 01/10/2014).

2. Laptev V.V. Infographics: basic concepts and definitions // Nauch.-tekhn. ved. St. Petersburg. state polytechnic university Humanite. and societies. Sciences. 2013. No. 4 (184). pp. 180-187.

3. Gillan D.J., sapp M. Length and area estimation with visual and tactile stimuli // Proceedings of the Human Factors and Ergonomics Society 48th Annual Meeting. 2004. P. 1875-1879.

4. Modley R. How to Use Pictorial Statistics. N. Y.: Harper and Brothers, 1937. 170 p.

5. Laptev V.V. Soviet information graphics of the 1930s // Vestn. St. Petersburg. university Ser. 15: Art history. 2013. No. 1. S. 224-232.

6. Vashchuk O.A. Concert posters by J. Müller-Brockmann (from the history of the Swiss school of graphic design) // Design. Materials. Technology. 2012. No. 4 (24). pp. 84-89.

7. Goodman N. Languages ​​of Art: An Approach to a Theory of Symbols. 2nd ed. Hackett Publ. Company, 1976. 291 p.

8. statistisch jaarboek. Den Haag-Heerlen: Central Bureau voor de Statistiek, 2013. 241 p.

9. Laptev V.V. Infographic design. St. Petersburg: Publishing House of SPGUTD, 2013. 127 p.

pictorial charts in infographics: scope of application, classification and rules of construction

LAPTEV Vladimir V. - St. Petersburg State Polytechnic University. Politekhnicheskaya st., 29, St. Petersburg, 195251, Russia e-mail: [email protected]

This article focuses on figurativeness in infographics. The author considered the chart types that use figurativeness from the historical perspective. Particular attention is given to quantitative pictorial charts, their application, classification and requirements for construction.

INFOGRAPHICS; DATA VISUALIZATION; ILLUSTRATIONS; PICTORIAL STATISTICS; PICTORIAL CHARTS; PICTOGRAMS.

1. Gelman A., Unwin A. Infovis and Statistical Graphics: Different Goals, Different Looks. www.stat. columbia.edu. 11 June 2011. Available at: http://www.stat. columbia.edu/~gelman/research/published/vis14.pdf (accessed 01/10/2014).

2. Laptev V.V. Infografika: osnovnyye ponyatiya i opredeleniya. St. Petersburg State Polytechnical University Journal. Humanities and Social Sciences, 2013, no 4 (184), pp. 180-187. (In Russ.)

3. Gillan D.J., Sapp M. Length and area estimation with visual and tactile stimuli. Proceedings of the Human Factors and Ergonomics Society 48th Annual Meeting, 2004, pp. 1875-1879.

4. Modley R. How to Use Pictorial Statistics. N. Y., Harper and Brothers, 1937. 170 p.

5. Laptev YV Sovetskaya informatsionnaya grafika 1930-kh godov. Vestnik Sankt-Peterburgskogo universiteta. Seriya 15: Iskusstvovedeniye, 2013, no 1, pp. 224-232. (In Russ.)

6. Vashchuk O.A. Kontsertnyye plakaty J. MullerBrokmann (iz istorii shveytsarskoy shkoly grafichesko-go dizayna). design. material. Tekhnologiya, 2012, no 4 (24), pp. 84-89. (In Russ.)

7. Goodman N. Languages ​​of Art: An Approach to a Theory of Symbols. Hackett Publishing Company, 1976. 291 p.

8. Statistischjaarboek. Den Haag-Heerlen, Centraal Bureau voor de Statistiek, 2013. 241 p.

9. Laptev YV. Proyektirovaniye infographics. St. Petersburg, SPGUTD Publ., 2013. 127 p. (In Russ.)

© St. Petersburg State Polytechnic University, 2014

Consider the construction of the main types of diagrams on

specific numerical examples.

On bar charts statistical data

shown as vertically elongated

rectangles.

When constructing bar charts, you must perform

the following requirements:

1) the scale by which the height of the column is set,

must start from zero;

2) the scale should be, as a rule, continuous;

3) the bases of the columns must be equal to each other;

columns can be placed at the same distance

from each other, close to one another or in an influx, with

in which one column partially overlaps another;

4) along with the marking of the scale with the corresponding digital

The columns themselves should also be labeled.

Example. Let's display graphical data about a number

non-state general education schools in Russia for

the following academic years (at the beginning of the year), units: 1997/98 -

570; 1998/99 - 568; 1999/2000 - 607; 2000/01 - 635.

We study non-state educational

institutions using a comparison bar chart.

On the horizontal axis we place the bases of six columns

at a distance of 0.5 cm from each other. The width of the columns is 1 cm.

Scale on the vertical axis - 10 units. by 1 cm (Fig. 5.5).

Values ​​displayed on the bar chart

proportional to the length of the columns. It can be seen from the diagram that

number of non-

Rice. 5.5. Number of general educational non-state

Russian schools for 1997-2001

Example. Let's build a square chart for comparison

the number of teachers and students in non-state

schools for 2001 (at the beginning of the year). For building

diagrams need to extract the square roots of the following

values: the number of teachers - 16 thousand people; number

students - 61 thousand people. This will be respectively 4; 7.81.

To construct squares from these data, it is necessary

select scale. Let's take 1 cm for 0.8 thousand people.

The sides of the squares on the graph will be segments,

proportional to the numbers obtained (Fig. 5.6). So

way square-

Rice. 5.6. Number of students and teachers in

non-state schools in Russia at the beginning of 2001 (thousand students)

Example. Let us depict the dynamics of watch production in one of

regions of Russia for 1999 - 2002 using a chart

figure-signs. Let's conditionally accept one drawing for 1000 pieces

hours. Then the number of hours: in 1999 in the amount of 4717 pieces.

should be depicted in the amount of 4.7 drawings; in 2000

in the amount of 3672 pcs. - 3.7 drawings; in 2001 in the amount of 3987 pieces

3.99 drawings; in 2002 in the amount of 2189 pcs. - 2.2 drawings

Rice. 5.8. Watch production in one of the regions of Russia in

Pie charts it is convenient to build as follows:

the entire magnitude of the phenomenon is taken as 100%, calculated

share of its individual parts as a percentage. The circle breaks into

sectors in proportion to the parts of the displayed whole.

Thus, 1% accounts for 3.6°. For getting

the central corners of the sectors depicting the fractions of parts of the whole, it is necessary to

multiply the percentage by 3.6°.

Example. Let's depict the number using a pie chart

students of non-state universities in Russia at the beginning of 2000/01

academic year according to the forms of education. On a daily basis

39% of students study; in the evening - 9%; in absentia -

51%; on external studies - 1% of students. Let's build a circle

arbitrary radius. According to the number of students for

constructing sectors, we determine the central angles: for

daytime form, the central angle was 140.4 "(41.0 ¦

3.6); for the evening - 32.4 ° (9 3.6); for part-time -183.6° (51

3.6); for the external - 3.6 ° (1 ¦ 3.6). With help

protractor divide the circle into the appropriate sectors

Rice. 5.9. The structure of the forms of education of state students

and non-state universities in Russia at the beginning of 2000/01

school year

If data on the structure of a phenomenon are expressed in

absolute values, then to find the sectors

it is necessary to divide 360 ​​° by the value of the whole, and then

quotient of division successively multiplied by absolute

parts values.

For simultaneous comparison of three quantities related

among themselves in such a way that one value is

by the product of the other two, diagrams are used,

called the "sign of Barzar".

Sign of Varzar is a rectangle with

one factor is taken as the base, the other as the height, and

the whole area is equal to the product.

Example. There are data on the collection of spring wheat in one

from the regions of Russia in 2003, in which, during the sowing

on an area of ​​14.5 million hectares, the yield was 1.16 t/ha.

In our case, the base of the rectangle is

spring wheat yield, height - sown area, and

the area of ​​the rectangle is the gross harvest of spring

wheat. The correctness of the chart readings can be

check with simple mathematical calculations:

sown area \u003d gross harvest / yield \u003d 16800000 /

1.16 = 14482758 ha (Fig. 5.10).

Rice.

Rice. 5.10. Dependence of the gross harvest of spring wheat

from the yield and sown area in one of the regions

Russia 2003

Line charts widely used for

characteristics of changes in phenomena over time, performance

planned tasks, as well as to study the series

distribution, revealing the relationship between phenomena. Linear

diagrams are built on a coordinate grid. Geometric

signs in line charts are points and

straight line segments connecting them in series, which

fold into broken curves.

Example. Line charts can be used to represent

data on the competition for entrance examinations to higher

educational institutions in Russia for 1996 - 2000; for one

enrolled accounts for those who have passed exams:

Year 1996 1997 1998 1999 2000

Competition, people 1.8 1.7 1.8 1.9 1.9

In a rectangular coordinate system, put on the y-axis

data on the competition of applicants (Fig. 5.11). Scale - 1 cm

0.05 people It can be seen from the graph that the position of the curve

determined not only by competition data, but also

time intervals between dates.

It is not uncommon for a single line chart to show several

curves that give a comparative characteristic of the dynamic

Rice. 5.11. Competition for entrance examinations to higher

educational institutions of Russia for 1996-2000. (for one

enrolled, accounts for those who passed exams, people)

mics of different indicators or the same indicator

for different territories. The technique for constructing such curves is not

differs from the plotting in Fig. 5.11. From data

rice. 5.11 shows how the competition for universities changes in 1996 - 2000

gg. In 1997, the competition decreased markedly compared to

competition in 1996. However, since 1997, competition for higher

educational institutions increased and in 1999 exceeded the competition

1996 From 1999 to 2000, competition for Russian universities remained

unchanged.

Distribution series are most often depicted as

polygon or histogram . The landfill is built primarily for

images of discrete series. When it is built on the axis

the abscissa represents the values ​​of the variable attribute, and

on the y-axis - absolute or relative numbers

population units (frequency or frequency). The polygon in fig.

5.12 is based on (conditional) data on

distribution of families according to the number of children.

Rice. 5.12. The polygon for the distribution of families by the number of children in

one of the regions in 2003.

The distribution histogram is most often used for

images of interval series. To build it along the axis

the abscissa shows the intervals of the feature, and along the ordinate

The number of units of the population. On the segments

depicting intervals, build rectangles, areas

which are proportional to the numbers of units (Fig. 5.13).

Rice. 5.13. A histogram of the distribution of firms in one of the

industries by the cost of fixed production assets

In some cases, to display variational series

a cumulative curve (cumulate) is used. For her

constructing the value of a variable attribute is postponed

on the abscissa axis, and on the ordinate axis are placed the accumulated

totals of frequencies or frequencies (Fig. 5.14).

From Eliseeva

4.2. Main types of charts

Statistical tables are supplemented with graphs in the event that

when the goal is to emphasize some feature of the data,

to compare them. Charts are the most efficient

form of data representation from the point of view of perception.

Graphs are often used outside of the table. WITH

with the help of graphs, the visibility of the characteristics is achieved

structures, dynamics, interconnections of phenomena, their comparison.

Statistical graphs are conditional

representations of numerical quantities and their ratios through

lines, geometric shapes, patterns or geographic

maps.

The graphical method facilitates the consideration of statistical

data. The range of change is immediately visible on the graph

indicator, the comparative rate of change of different

indicators, their volatility. However, the schedule is

certain restrictions: first of all can not include

as much data as can be included in the table; besides, on

it always shows rounded data - not exact, but

approximate. Thus, the graph is used only

to depict the general situation, not the details. Last

minus - the complexity of construction. But this shortcoming can

be overcome by application packages

(PPP) for computer graphics, such as PPP "Harvard

Graphs are divided into diagrams according to the way they are constructed.

cartograms and charts.

The most common are charts. They

are of different types: linear, radial, point,

flat, voluminous, curly. The appearance of the chart depends on

type of data to be presented (one variable or one

indicator, several variables or indicators,

quantitative or non-quantitative) and construction tasks

Rice. 4.1. Dynamics of emissions of harmful substances into the atmosphere

in St. Petersburg

defined.

correlations).

is the distribution polygon, the second is line. 4.1. Dynamics of emissions of harmful substances into the atmosphere

and index of physical volume of industrial production

in St. Petersburg

In any case, the schedule must be accompanied

heading - above or below the graph field. In the title

indicate which indicator is shown, in what units

measurements, over what territory and for what time it

defined.

Line charts are used to represent

quantitative variables: characteristics of their variation

values, dynamics, relationships between variables.

Data variation is analyzed using a polygon

distributions, cumulates (curve "not less than") and ogives

(curve "greater than"). Line charts are used in

solving problems of data classification. Line charts

used in the analysis of the dynamics of relationships. In analysis

scatter plots are used (the so-called field

correlations).

It is advisable to divide line charts into used

to represent data by one variable - one-dimensional

or in two variables - two-dimensional. An example of the first

is the distribution polygon, the second is the regression line.

It is possible that the graph contains several variables (indicators), but it still

is not multidimensional (Figure 4.1).

In order for the dynamics of two or more indicators to be

comparable, they should be provided with a “single start”, as in

rice. 4.1, where the figures for 1990 are taken as 100%.

;

О--------assessment of the changes in the economic

situation in Russia;

O- - assessment of expected changes in the economic situation in

L-- - assessment of the changes in personal

financial situation;

-*-¦ -assessment of expected changes in personal material

provisions;

- - ¦ - - - assessment of favorable conditions for large purchases

sq. - May, III quarter. - August, IV quarter. - november)

The dynamics of two indicators on the same chart can

be presented without bringing them to 100%, if these

indicators are related to some functional

ratio (for example, the dynamics of the total

indicator and indicator, which is one of its

components). An example of such a graph is Fig. 4.2.

With a graphical representation of the dynamics along the x-axis

time is shown (years, quarters, months); along the y-axis

Values ​​​​of indicators or indicators (Fig. 4.3, a). Wherein

The y-axis must have an origin at the point "O". Sometimes instead

zero point as the starting level on the y-axis

shows the level of any year. This is done in

if the displayed indicator changes

significant - by 8-10 times or more during the considered

It is more correct to specify the zero point, and then (if necessary)

"break" the y-axis as shown in Fig. 4.3b.

Sometimes, with large changes in the indicator, they resort to

logarithmic scale. Suppose the values ​​of the indicator

change from 1 to 100 (100 times); it may cause

Difficulties in plotting. If you go to

logarithms, then their values ​​for the minimum (maximum)

indicator values ​​will not differ so much: log 1 =

Among 2D charts by frequency of use

column charts are highlighted, in which the indicator

represented as a column, the height of which corresponds to

indicator value. Bar Chart Example

shown in fig. 4.4. Often on a bar chart

relative values ​​are shown: when comparing

indicators for groups, for different populations, one of

which can be taken as 100%.

The proportionality of the area of ​​a particular geometric

figures the value of the indicator is the basis of other types

planar diagrams: triangular, square,

rectangular. In a triangular diagram, you need to choose like this

sides and height of a triangle so that its area corresponds to

indicator value. To build a square chart

you need to set the size of one side, rectangular - two __

sides. You can also use a comparison of the areas of a circle; V

In this case, the radius of the circle is given.

A strip chart represents measures in the form

horizontally elongated rectangles. Like columnar

and strip charts can be used not only for

comparing the quantities themselves, but also for comparing their parts (Fig.

A special type of strip chart is used for

presentation of data with different nature of changes:

positive and negative (Fig. 4.7).

The diagram shown in fig. 4.7 can be used,

for example, to represent regions with different sizes and

the nature of the migration balance (positive and

negative) enterprises, which increased and

wages went down, etc.

Of the planar charts, the pie chart is often used.

diagram. It is used to illustrate the structure

studied population. The whole set is taken as

th indicator. The area of ​​the figure corresponds to the value

indicator (Fig. 4.10).

If, for example, you choose to use a shape chart

to depict the structure of unemployed women, among

of which 47% are young women (20-24 years old) and girls

16-19 years old with no work experience; 28% - engineering

technical workers and employees with special

education at the age of 25-49 and 15% - female workers

skilled and unskilled labor aged

50 years and older, then must depict three female figures,

and the first of them must be twice as large as the second, and

the second is almost twice as large as the third.

When plotting a graph, everything is equally important - the right

selection of the type of graphic representation of proportions, compliance with

formatting rules. All these issues are covered in more detail in

Various types of graphs allow you to get the RFP for

PC "Harvardgraphics", "Supercalc", "Statictica", "Statgraphics

» and others. Based on the graphical representation

some procedures for classifying (grouping) data,

analysis of dynamics: identification of trends, comparison of dynamics

different indicators, etc.

Finally, the process of summarizing statistical data can be

present graphically (Fig. 4.11). The entire array is displayed

collected data, i.e. table "object-attribute" obtained

for a number of periods. For example, data are collected on industrial

businesses in the area for many

features for each month. This can be represented in

in the form of a parallelepiped, as shown in Fig. 4.11.

The third dimension may not be time, but a certain

territory, i.e. each table "object-attribute" refers to

to a certain territory (district, region, etc.). On

the following figures show that each of the subarrays,

taken from fig. 4.12, a, can be distinguished and developed

independently (b); in fig. 4.12, vag shows that the data

can be subdivided by regions, quarters and, finally, by

division of data on three grounds: by time,