Abstract
In recent years, a new form of journalism has developed a strong dynamic: journalism based on data. The success of the move is so great that technologies and applications that support it have begun to emerge. Although technologies are in a period of mature productivity, their evaluation is an area lagging in development. It is a fact that lack of evaluation based on a reliable methodology is evident in the literature. Finding a suitable methodology for this purpose is particularly important in order to (a) evaluate systems supporting data journalism resulting in a specific ranking of potential, and (b) make it easier for the user to choose the application that best suits his / her requirements and cognitive level. A comparative evaluation of 9 applications related to the visual imaging component was attempted with usability and functionality criteria. The aim of this study is to provide a quick evaluation method, open to proposed improvements and further refinement, in order to establish a framework for qualitative assessment of data journalism applications.
TopBackground
A system’s evaluation is the process of testing and validating whether the system achieves its predefined goals and in what degree. Although in literature there are several qualitative and quantitative techniques proposed for different kinds of systems, there is not yet enough research for data journalism systems evaluation, so a goal for the area is for an evaluation framework to be established. Relative background can be found in related areas (e.g. information visualization, HCI), while the most suitable criteria for the proposal of such a framework are: functional and usability features.
Yi et. al. (2007) have collected the techniques from various research papers by creating a classification for InfoVis (Information Visualization) related to the interaction techniques. The following three classifications focus on interaction techniques, while the fourth refers to the tasks of the system’s user, i.e. it focuses on his/her goals without paying attention to the interaction. The four proposed classified techniques with their related research work are:
Low-level Techniques. Buja et al. (1996) classified the interaction techniques into three classes: Focusing, Linking, and Arranging views. Chuah & Roth (1996) summarized a set of basic visualization interaction (BVI) operations: Graphical operations, Set operations, and Data operations. Dix & Ellis (1998) condensed the different ways in which interactivity adds value and resolves conflicts in representation: Highlighting and focus, accessing extra information – drill down and hyperlinks, Overview and context - zooming and fish-eye views, Same representation / changing parameters, Same data / changing representation, Linking representation – temporal fusion. Keim (2002) proposed that in addition to the visualization technique, it is necessary to use some interaction and distortion techniques for an effective data exploration: Dynamic projections, Interactive filtering, Interactive zooming, Interactive distortion, Interactive linking and Brushing. A categorization of interactions can also be found in Wilkinson (2005). Interaction techniques are classified into seven categories: Navigating, Manipulating, Brushing and linking, Filtering, Rotating, Transforming, Animating.
Key Terms in this Chapter
Rubric: A set of evaluation criteria that are valued based on a rating scale.
Storytelling: The social and cultural activity of sharing stories, with great application to journalism.
Functionality: A system evaluation criterion aimed at identifying certain functional characteristics.
Usability: A criterion for the evaluation of the system aimed at identifying usability features from a user's point of view.
Data Journalism: Journalism based on structured data in order to support a story that makes sense to the public.
Visualization: The visual representation of journalistic story that aims to convey the information to the reader in the best possible way.
Interactive Chart: A chart on which a user can achieve a certain degree of interaction in order to reshape it, for example, hover a point, zoom, play with axis, add a button, etc. It allows the reader to go deeper in its understanding of the data, since he can customize it and try to answer his own question.