Visual Presentation Of Data Research Articles

Error Bars Considered Harmful: Exploring Alternate Encodings for Mean and Error.

When making an inference or comparison with uncertain, noisy, or incomplete data, measurement error and confidence intervals can be as important for judgment as the actual mean values of different groups. These often misunderstood statistical quantities are frequently represented by bar charts with error bars. This paper investigates drawbacks with this standard encoding, and considers a set of alternatives designed to more effectively communicate the implications of mean and error data to a general audience, drawing from lessons learned from the use of visual statistics in the information visualization community. We present a series of crowd-sourced experiments that confirm that the encoding of mean and error significantly changes how viewers make decisions about uncertain data. Careful consideration of design tradeoffs in the visual presentation of data results in human reasoning that is more consistently aligned with statistical inferences. We suggest the use of gradient plots (which use transparency to encode uncertainty) and violin plots (which use width) as better alternatives for inferential tasks than bar charts with error bars.

Spine Tango annual report 2012

Since the Spine Tango registry was founded over a decade ago it has become established internationally. An annual report has been produced using the same format as the SWEspine group to allow for first data comparisons between the two registries. Data was captured with the latest generation of surgery and follow-up forms. Also, the Core Outcome Measures Index (COMI) from interventions performed in the year 2012 with follow-up to June 2013 was analyzed. Groups of patients with the most common degenerative lumbar spine diseases and a single group of patients with degenerative cervical spine diseases were created. The demographics, risk factors, previous treatments, current treatment, short-term outcomes, patient satisfaction and complications were analyzed. Pre- and postoperative pain and function scores were derived from the COMI. About 6,500 procedures were captured with Spine Tango in 2012. The definitions and composition of all the degenerative groups could not completely be matched between the two registries with the consequence that the age and sex distributions were partially different. Preoperative pain levels were similar. The short-term outcomes available did not allow for evaluation of the final result of surgical intervention. This will be possible with the longer term data in the next annual report. There was a distinct disparity in reported complication rates between surgeons and patients. This is a valuable first step in creating comparable reports for SWEspine and Spine Tango. The German spine registry may be able to collaborate in the future because of similar items and data structure as Spine Tango. There needs to be more work on understanding the harmonization of the different degenerative subgroups. The Spine Tango report is weakened by the short and incomplete follow-up. The visual presentation of data may be a useful model for aiding decision making for surgeons and patients in the future.

Uma introdução à visualização de informações

Information Visualization is an emerging scientific field that studies ways of presenting abstract data visually, so that information about that data can be better understood or even discovered. This paper discusses several aspects regarding the visual presentation of data and introduces well known and effective techniques for information visualization. The techniques presented here have applications in many scientific fields. The paper also discusses some of the modern challenges in the Information Visualization area. Keywords: Abstract data, information presentation, information discovery

Visualizing Daily Hospital Routine

Introduction Hospitals, just like other businesses with a comparatively complex organizational structure, are often confronted with analyzing and evaluating complex sets of data that are hard to monitor. Standardized forms of presentation (e.g., pie or bar charts) are frequently used for the visual analysis and presentation of data in such cases. From a visual communication perspective, the potential that visualization offers has not been exhausted. Collaboration between statisticians and designers—with the pooling of themespecific practical and creative competences—is therefore a worthwhile objective. The research project, “Visual Atlas of the Daily Hospital Routine,”1 joins these competences to research the options of alternative forms of visualization of complex sets of data for process optimization. The specialist department for quality management of the Bern University Hospital Inselspital (Annekathi Bischoff), which provided the data for this research project from four selected sub-processes, is a contact point for quality management issues for the entire hospital. During process optimization, the goal is to analyze and optimize hospital-internal procedures. For data collection, qualitative procedures are often applied, including field research, participatory observation, and interviews. For purposes of analysis and evaluation, standardized quantitative representation models (e.g., bar graphs) are almost exclusively used. The visualization of qualitative data, in contrast, is hardly considered significant. As a sub-discipline of visual communication, “Knowledge Visualization”2 generates context-specific visual representations of quantitative and qualitative data. The goal of the visual representation is to provide access to complex sets of data, often with great advantages over linguistic or standardized visual representation models. Edward Tufte describes this potential as follows: “Often the most effective way to describe, explore, and summarize a set of numbers—even a very large set—is to look at pictures of those numbers.”3 The viewing of such images can help to present work processes more clearly and thereby to analyze procedures and manage them more efficiently. The goal of this research project was to develop alternative presentation forms for four organizational and communicative sub-processes in the environment of the Inselspital. As a secondary means of analysis, the images developed in the project were

Integrating Visualization and Interaction Research to Improve Scientific Workflows

Scientific-visualization research is, nearly by necessity, interdisciplinary. In addition to their collaborators in application domains (for example, cell biology), researchers regularly build on close ties with disciplines related to visualization, such as graphics, human-computer interaction, and cognitive science. One of these ties is the connection between visualization and interaction research. This isn't a new direction for scientific visualization (see the "Early Connections" sidebar). However, momentum recently seems to be increasing toward integrating visualization research (for example, effective visual presentation of data) with interaction research (for example, innovative interactive techniques that facilitate manipulating and exploring data). We see evidence of this trend in several places, including the visualization literature and conferences.

Flow cytometry histograms: Transformations, resolution, and display

Flow cytometry data analysis routinely includes the use of one- or two-parameter histograms to visualize the data. These histograms have traditionally been plotted with either a linear or logarithmic scale. However, the recent trend of performing the logarithmic conversion in software has made apparent some limitations of the traditional visual presentation of logarithmic data. This review discusses the mathematics of presenting data on a histogram and emphasizes the difference between scaling and binning. The review introduces the concept of an effective resolution to describe how the bin width changes in a variable bin-width histogram. The change in effective resolution is used to explain the commonly observed valley and picket fencing artifacts. These result from the effective resolution of the display histogram being too high for the data being presented. Recently, several different binning transformations have been described that are becoming more popular because they allow one to view a large dynamic range of data on a single plot, while allowing the display of negative data values. While each of the transforms is based upon different equations, they all exhibit very similar properties. All of the transforms bin the data logarithmically at high channel values and linearly at low channel values. The linear scaling of the lower channels serves to limit the effective resolution of the histogram, thus minimizing the valley and picket fencing artifacts. The newer transformations are not without their own limitations and recommendations for the appropriate manner of presenting flow cytometry data using these newer transformations are discussed.

Guidelines for Graphing Data with Microsoft® PowerPoint™

Graphs are vital components for analyzing data in the experimental analysis of behavior using single subject research methods. This paper extends the previous literature on the construction of single subject graphs by providing instructions for using Microsoft® Power PointTM and Microsoft® PowerPoint for Mac®, and describes improved reporting practices for visual presentation of single subject experimental data. Detailed instructions are provided for constructing line graphs for displaying single subject data, specialized graphs (i. e., semi-logarithmic, dual ordinate, and line and bar combination graphs), and electronically submitting graphs to journals for publication.

Music notation: a new method for visualizing social interaction in animals and humans

BackgroundResearchers have developed a variety of techniques for the visual presentation of quantitative data. These techniques can help to reveal trends and regularities that would be difficult to see if the data were left in raw form. Such techniques can be of great help in exploratory data analysis, making apparent the organization of data sets, developing new hypotheses, and in selecting effects to be tested by statistical analysis. Researchers studying social interaction in groups of animals and humans, however, have few tools to present their raw data visually, and it can be especially difficult to perceive patterns in these data. In this paper I introduce a new graphical method for the visual display of interaction records in human and animal groups, and I illustrate this method using data taken on chickens forming dominance hierarchies.ResultsThis new method presents data in a way that can help researchers immediately to see patterns and connections in long, detailed records of interaction. I show a variety of ways in which this new technique can be used: (1) to explore trends in the formation of both group social structures and individual relationships; (2) to compare interaction records across groups of real animals and between real animals and computer-simulated animal interactions; (3) to search for and discover new types of small-scale interaction sequences; and (4) to examine how interaction patterns in larger groups might emerge from those in component subgroups. In addition, I discuss how this method can be modified and extended for visualizing a variety of different kinds of social interaction in both humans and animals.ConclusionThis method can help researchers develop new insights into the structure and organization of social interaction. Such insights can make it easier for researchers to explain behavioural processes, to select aspects of data for statistical analysis, to design further studies, and to formulate appropriate mathematical models and computer simulations.

Enhancing retail location decision support: The development and application of geovisualization

Geovisualization refers to the visual exploration, analysis, synthesis and presentation of geospatial data. This paper presents findings from research that has focused on developing and applying geovisualization techniques and technologies for use within retail location decision support. To date, retailers represent a major user group of geographic information system (GIS) -based decision support technologies, with applications ranging from trade area mapping to store portfolio planning. However, the ability to handle spatial–temporal data, visualize change, and explore the temporal dimension of spatial data is limited within conventional GIS. The paper details the development of a prototype geovisualization system that has been designed to enable visualization of spatial–temporal change of retail-related data. From this explicitly visual paradigm, a number of examples of potential analysis are examined at four different scales of analysis: national, regional, market and micro-level. The paper highlights both the challenges and potential to enhance retail decision support by integrating geovisualization techniques and technology within decision support activities.

Optimal decisions in combining the SOM with nonlinear projection methods

Visual data mining is an efficient way to involve human in search for a optimal decision. This paper focuses on the optimization of the visual presentation of multidimensional data. A variety of methods for projection of multidimensional data on the plane have been developed. At present, a tendency of their joint use is observed. In this paper, two consequent combinations of the self-organizing map (SOM) with two other well-known nonlinear projection methods are examined theoretically and experimentally. These two methods are: Sammon’s mapping and multidimensional scaling (MDS). The investigations showed that the combinations (SOM_Sammon and SOM_MDS) have a similar efficiency. This grounds the possibility of application of the MDS with the SOM, because up to now in most researches SOM is applied together with Sammon’s mapping. The problems on the quality and accuracy of such combined visualization are discussed. Three criteria of different nature are selected for evaluation the efficiency of the combined mapping. The joint use of these criteria allows us to choose the best visualization result from some possible ones. Several different initialization ways for nonlinear mapping are examined, and a new one is suggested. A new approach to the SOM visualization is suggested. The obtained results allow us to make better decisions in optimizing the data visualization.

Visual decisions: Geovisualisation techniques within retail decision support

Geovisualisation refers to an established area of research that provides the theory, methods and tools for visual exploration, analysis, synthesis and presentation of geospatial data. The aim of geovisualisation is to facilitate the knowledge-creation process by providing techniques and technologies that allow decision makers to interact visually with data by displaying the data in ways that promote visual-cognitive learning. The applications of geovisualisation within the retail industry are many and varied, for example, rendering complex retail demand surfaces (eg probabilistic trade areas) over time around a network of stores, sequencing changes in retail metrics (eg sales per square foot, profitability by store) and monitoring change in commercial structure (eg retail vacancy/turnover, new format retail development). To date, geographical information systems (GIS) technology in use within major retail organisations has relatively limited ‘visual’ capabilities, and finds dealing with temporal dimensions (ie the dimension upon which retail change occurs) particularly difficult. The increasing availability of time-series data within retail organisations provides many challenges and opportunities to the retail location analyst. It allows decision makers to move away from simply generating static ‘snapshots’ to undertaking ‘dynamic’ visualisation of change, for example, 3D animation of data sets over a number of time periods. This paper details the development of a prototype geovisualisation system that has been designed to enable geospatial visualisation of retail change. A number of potential application areas are examined. The paper concludes by highlighting the challenges that are faced in developing spatial–temporal visualisation systems and the potential to integrate such technology within retail decision support systems.

Application of Fuzzy Functions for Visual Presentation of Medical Data

While interpretation of medical data is very often an ambiguous process, computers usually display results of recommendations, provided by both human experts and computer algorithms, as concrete data. This study proposes a visual presentation of relative degrees of uncertainty along with "standard" concrete medical data. A medical parameter (mean arterial pressure) is dynamically evaluated during surgery for being too low or too high. Fuzzy membership functions are utilized to display degrees of deviation in the form of a clear and concise pie chart. Thus, in the Operating Room the anesthesiologist can be provided with an easy statistical assessment of uncertainty of existing recommendations.

Visualization of correlation‐based environmental data

AbstractA method for the visualization of environmental data via the analysis of correlations has been proposed. As an example, the data in correlation matrices of environmental parameters that describe air pollution in Vilnius city (10 parameters) as well as the development of coastal dunes and their vegetation in Finland (16 parameters) are visualized. These applied problems are very urgent because of their ecological nature: a visual presentation of data stored in the correlation matrices makes it possible for ecologists to discover additional knowledge hidden in the matrices and to make proper decisions. The method consists of two stages: building of a system of vectors based on the correlation matrix and its visualization. Sammon's mapping and the self‐organizing map (two its realizations having different features for visualization) were applied in the visualization. The advantage of the method lies in the possibility to restore the system of multidimensional vectors describing variables from the correlation matrix—one vector for one variable. Copyright © 2004 John Wiley & Sons, Ltd.

The Search for Simplicity: A.K. Romney and Building Models From Systematic Data

For more than 50 years, A.K. Romney and his many collaborators have published a steady stream of methods, ideas, critiques, and findings. This influential body of work has been characterized by a strong interest in quantification, extensive use of graphics and visual presentations of data, a focus on empirical issues, a search for cultural universals, intensive collaboration with other researchers, and a striving for conciseness and clarity. Of these, the search for simplicity is perhaps the most notable and pervasive characteristic of his work. This article reviews some of Romney’s early work on kinship, tracing the way in which Romney’s treatment of kinship data was influenced by his basic notion of humans as accurate and principled categorizers.

Inventing discovery tools: combining information visualization with data mining

The growing use of information visualization tools and data mining algorithms stems from two separate lines of research. Information visualization researchers believe in the importance of giving users an overview and insight into the data distributions, while data mining researchers believe that statistical algorithms and machine learning can be relied on to find the interesting patterns. This paper discusses two issues that influence design of discovery tools: statistical algorithms vs visual data presentation, and hypothesis testing vs exploratory data analysis. The paper claims that a combined approach could lead to novel discovery tools that preserve user control, enable more effective exploration, and promote responsibility.

Inventing Discovery Tools: Combining Information Visualization with Data Mining

The growing use of information visualization tools and data mining algorithms stems from two separate lines of research. Information visualization researchers believe in the importance of giving users an overview and insight into the data distributions, while data mining researchers believe that statistical algorithms and machine learning can be relied on to find the interesting patterns. This paper discusses two issues that influence design of discovery tools: statistical algorithms vs visual data presentation, and hypothesis testing vs exploratory data analysis. The paper claims that a combined approach could lead to novel discovery tools that preserve user control, enable more effective exploration, and promote responsibility.

Structure and dynamics of a DNA duplex containing single alpha-anomeric deoxyadenosine residue.

Structure and dynamics of an undecamer DNA duplex containing a single alpha-anomeric deoxyadenosine residue placed in opposition to a thymidine unit have been studied using simulation of molecular dynamics in aqueous solution. Despite several noticeable deviations from the B-DNA duplex structure caused by the anomerisation, such as: West type puckering of the alpha-anomeric sugar, disrupted base stacking pattern and unstable duplex bending, the formation of a non-classical alpha-dA-T pair was observed. A novel way of visual presentation of trajectory data allowing high throughput screening of the conformational parameters is presented.

Supporting communication in rehabilitation engineering teams.

The objectives of this project were to examine how members of a colocated rehabilitation engineering team communicate during their work and hence deduce the implications of these communications for the design of video-based technologies to support communication among members of a virtual rehabilitation engineering team. Twenty-four assessment clinic sessions conducted by rehabilitation engineering team were recorded on videotape over a period of 3 years. These tapes were analyzed in considerable detail using a schema to identify and classify the talk and actions of the team members. Combining talk and actions with artifacts is a mechanism used by designers to develop ideas and communicate them to others. Speakers rely on actions to support and make their talk lucid. Cooperation based on sharing artifacts is a strength of face-to-face interaction. Participants can experience artifacts and observe others using the artifacts. Tools such as videoconferencing to support virtual rehabilitation teams will have to provide the participants with the ability to see often quite subtle gestures and actions if they are to grasp the meaning of the talk. Increased understanding how a team communicates visually complex data may (1) aid development of next generation videoconferencing equipment to better support distributed designers and rehabilitation engineers and (2) guide development of techniques to enhance the quality of visual data presentation in current videoconferencing systems.

Yield Map Interpretation

Yield monitoring and mapping are becoming commonplace with many crops throughout North America. Some fields now have a 3- to 5-yr yield map history. A yield map, however, only documents the spatial distribution of crop yield and does not explain what factor(s) caused the variation. The goal of yield map interpretation is enhanced profitability through better understanding and control of natural and management-induced sources of yield variation. Numerous causes of crop yield variation have been documented, including climate, soil-water relationships, soil physical and chemical properties, site attributes, crop pest infestations, crop inputs and condition, field history, and cultural practices. Proper visual presentation of yield monitor data in yield map form and the accurate identification of characteristic patterns of yield variation are essential for meaningful interpretation of the yield map. Unfortunately, all yield monitor data sets and maps contain inherent error, some of which cannot be easily corrected. Error-induced patterns must be separated from real yield variation in order to make correct interpretations. In general, irregular areas, blotches or, speckles, and elliptical patterns are the result of naturally occuring yield-limiting factors. Conversely, rectangles, abrupt boundaries, circles, arcs, and streaks or lines reflect management-induced patterns of yield variation. In addition, divergence of parallel swaths, missing data points, and a repeating sawtooth pattern along field margins usually result from errors associated with Global Positioning System (GPS) signal reception and yield monitor data collection. Yield map interpretation is greatly enhanced by ongoing grower involvement and the insightful use of auxiliary agronomic, spatial, and historical site information. Geographic Information System (GIS) tools are virtually essential to evaluate multiple layers of spatial data. GIS software can be used to systematically and quantitatively evaluate the relationships between yield data and other spatial features. In the future, tools such as data-mining software, and other sophisticated mathematical and spatial models may provide additional power to interpret single-year as well as multi-year yield map information.

Summarizing clinical psychiatric data.

The authors propose a method of summarizing clinical data to serve patient care. The proposed one-page summary uses diverse visual presentations of data, including small graphs for ratings and drug dosages, time lines for clinic visits and hospital stays, and genograms for inherited illness, as well as a textual presentation of recent clinical notes. The summary depicts a patient's recent and lifetime clinical experience. It allows the viewer to assess relationships between interventions and outcomes for psychiatric and medical problems. Computerized patient information systems, which are increasingly being used, can present data in virtually any form. The authors hope to encourage mental health professionals to reshape psychiatric records.