This page is referenced by:

• 1 2021-06-21T15:33:59-04:00 How should students analyze data visualizations in social studies? 1 The challenges that data visualizations present, coupled with their prevalence in social studies texts, standardized assessments, in online social studies resources, and as sources of information in society, suggest that teaching with and about data visualizations in social studies is essential. This module provides guidance for how teachers can support students' data literacy for social studies. plain 2021-06-21T15:33:59-04:00 The challenges that data visualizations present, coupled with their prevalence in social studies texts, standardized assessments, online social studies resources, and as sources of information in society, suggest that teaching with and about data visualizations in social studies is essential. Making sense of multimodal information – that is, sources that combine verbal and various kinds of visual information – requires the ability to shift among strategies and skills that correspond to the words and images (Serafini, 2015; Serafini and Clausen, 2012). Without instruction aiming to build necessary strategies and skills, students will be poorly equipped to negotiate and critically analyze the complex information that they regularly face (Ainsworth, 1999; Bezemer and Kress, 2010; Gee, 2014; Peeck, 1993; Serafini, 2014; The New London Group, 1996). 
  
  Existing research suggests that the following principles should guide data literacy instruction:

  (1) Students should learn at an early age that data visualizations in texts are important and they need to pay attention to them. More often than not, timelines, graphs, charts, and maps provide information that is not provided in surrounding verbal text. As they move through school, students should continue to learn about how to integrate visual information with verbal information.
  (2) Students should receive carefully sequenced, scaffolded instruction on reading, analyzing, interpreting, and evaluating data visualizations throughout their schooling and across multiple disciplinary contexts. 
  (3) It is preferable that students learn about different data visualizations and their graphical conventions in a progressive sequence, moving from simple data visualizations to more complex or layered data visualizations over time. 
  (4) Students should learn to be critical consumers of data visualizations by looking for biases, omissions, distortions, and manipulations. 

  Each of these principles is explained in more detail in the following sections. 
  
  Teach Students to Pay Attention to Data Visualizations
  Good readers take inventory of all the contents of a text, paying attention to the written words, and all the visual features, including data visualizations (Peeck, 1993; Serafini, 2010). Literacy researchers (e.g., Duke et al., 2013; Serafini, 2010) have argued that beginning in early elementary, readers must understand that data visualizations are representations of information created with intentionality by illustrators or authors, and that they have relevance to surrounding information, often extending that which is found in the verbal text. Stated in another way, students should understand that somebody created a data visualization with a specific purpose in mind, either the same person or someone else put it in the text for a particular reason, and the data visualization likely provides information that is not in paragraphs or passages around it. If one student reads the data visualization, and another does not, the student who didn't read it is going to be at a disadvantage when it comes to extracting information from the text. We want all students to learn to pay attention to data visualizations – especially since research suggests that doing so can enhance their understanding.
  
  Good readers also know to how integrate information across verbal and visual modes. When they have a particular task in mind while reading, they look for and make connections among task-relevant data extracted from multiple modes (Mayer, 2009; Peeck, 1993; Walsh, 2006). Such integration supports deep learning and development of rich mental representations of the particular topic under study, particularly for learners who lack strong prior knowledge (Ainsworth, 1999; Mayer, 1997, 2009; Prangsma et al., 2008; Schnotz and Bannert, 2003). Teachers can support such reading practices by designing instruction to support extraction of task-relevant data and integration of the verbal and visual. If possible, they should present related verbal and visual information simultaneously to support connections. If simultaneous presentation is not possible, there is some research to suggest that asking students to read visual information before they read verbal information may be more effective. 
  
  Consider the example to the right. This 3-D multi-set area graph is from a popular middle school textbook, History Alive!. Even though it covers the years 1820 to 2000, it's from a section of the textbook that students would use to learn about how immigration patterns changed from the first half of the 19th century to the second half of the 19th century and affected American life. The mismatch in temporal focus – that is, the fact that the graph covers more than the 19th century, doesn't make the graph irrelevant, but it might be confusing to students if they are asked to make sense of it and connect it on their own. To support comprehension, teachers could focus students' attention on the parts of the graph that are relevant – helping them delineate between the first and second halves of the 19th century, and separate the data from the parts of the graph that are not relevant to the verbal text. 
  
  It is important to note that some texts are poorly designed and provide too many purely decorative or irrelevant visuals. Teachers should therefore take care in selecting texts, and point out when and why visual displays are poor choices. When choosing visuals themselves, teachers should be careful to make sure they have meaningful overlap with accompanying information. This also provides a good model for students when they are asked to integrate verbal and visual information for essays, presentations, and projects. 
  
  Teach the Basic Steps of Reading Data Visualizations
  Some researchers have articulated a research-based sequence to approaching data visualizations, emphasizing factors that can mediate comprehension such as knowledge of graphical conventions and background knowledge (e.g., Harsh and Maltese, 2019; Maltese et al., 2015; Verdi and Kulhavy, 2002). In this recommended sequence, students should first take account of all the visual elements in a data visualization. As stated in the previous module, visual elements represent the ways that the data visualization creators have chosen to encode the data or information they are trying to communicate. This includes shapes, colors, symbols, and text. Reading the data by taking note of all visual elements also requires attention to common visualization conventions so that students can effectively decode the meaning of the visual elements and determine the variables that are being displayed (Shah and Hoeffner, 2002). Consider all the visual elements in the graph above. Students may not only require instruction to see each and every element, they may also need help in understanding how the colored areas associated with each region are used to communicate overall quantities and how, even though there is no visible third axis, the upward sloping line means an increasing number of immigrants, and downward sloping line means a decreasing number of immigrants. 
  
  The ability to see the data will impact the degree to which students are able to next make connections between visual elements, or read between the data and extract information (Curcio, 1987; Harsh and Maltese, 2019; Maltese et al., 2015). This is about helping students to see how all the visual elements are connected to communicate facts. In the graph above, for example, we want students to recognize that the number of immigrants from southern and eastern Europe rose in the latter half of the 19th century. At the same time, we would want them to recognize that overall numbers of immigrants increased in the late 19th century. The ability to read between the data is mediated by students' understanding of graphical conventions, so teachers would need to be aware of whether or not students understand how to read and make sense of area graphs, and why a designer might choose a 3-dimensional area graph instead of a stacked line graph. Supporting students as they make connections among visual elements and try to extract information from an area graph also requires that teachers pay attention to vocabulary that students might find difficult or that might be interpreted in multiple ways. For example, what countries likely make up southern, eastern, and northern Europe in the graph? What country makes up the bulk of Asian immigrants?
  
  The next step in the process of reading data visualizations is helping students read beyond the data. This step entails connecting the data visualization to the question under investigation or integrating it with other sources of information (Harsh and Maltese, 2019; Maltese et al., 2015). Students may need to be reminded of the question they are trying to address, or they may require the teacher to help them make explicit links between different modes of text. This step is also about making inferences – combining the facts from the data visualization with reasoning about the topic to draw conclusions.  For example, if a student was trying to use the graph above to understand what impact changes in immigration had on American society, they would have to think beyond raw numbers and consider how waves of immigration might affect places and people involved. They could do this by looking at primary source documents are reading historical accounts. Ideally, they would also ask questions about where immigrants are settling, and how concentrations of immigrants in regions, states, or cities might have a differential impact on different parts of the country and the different people living there. On the topic of 19th century immigration, they could find such information from additional sources, such as the University of Oregon's Mapping History Project shown below. In other words, reasoning within the data visualization itself – and only with the data visualization and its surrounding text – will be insufficient for drawing conclusions. Students need to see behind and beyond the numbers in the graph provided. 
  
  
  Consider the Complexity of the Data Visualizations You're Teaching; Teach in Progressive Sequence if Possible
  Some researchers have recommended that because reading data visualizations is so complex, students should learn about them in a progressive sequence (Friel et al., 2001). With graphs, for example, younger elementary students might learn about pictographs and bar graphs, with stacked bar graphs, multi-set bar graphs, and pie charts introduced in later elementary grades, and line graphs in middle school (Friel et al., 2001). As discussed in the sections under How do students learn with data visualizations?, progressions in reading data visualizations are usually poorly represented in state standards. For this reason, it's necessary for teachers to consider what kinds of data visualizations students are working with at different points in their schooling. 
  
  Researchers have made similar recommendations about map reading (Bausmith and Leinhardt,1998; Mohan and Mohan) They have suggested, for example, that maps with multiple layers are challenging so students should begin with simple layers containing less information before progressing to maps containing information-rich layers. In early elementary, students can work with simple maps of familiar places (e.g., classroom, house, neighborhood), and then, in later elementary, with maps that have a grid system and limited amounts of information. By the end of elementary school they should be comfortable working with thematic maps at multiple scales that have an overlay of data (e.g., a simple connection or distribution map). This will prepare them for working with maps that have multiple layers in middle school and beyond. 
  
  One thing to keep in mind as you're choosing data visualizations is that animated or interactive data visualizations – though potentially engaging and interesting to students – do not necessarily support learning more than static data visualizations. Animated data visualizations with no opportunities for students to pause and process, may be particularly challenging. As teachers, we should take care to provide learning environments that will adequately support students' comprehension of dynamic data visualizations, and not assume that interactivity by itself will lead to greater learning.  
  
  Teach Critical Data Literacy
  Several experts (e.g., Eaton, 2019; Maier and Imazeki, 2013; Shah and Hoeffner, 2002) have emphasized that students must learn to view data visualizations with a critical eye to determine what meaning can be drawn from them. After all, data can lie and mislead (Johnson and Gluck, 2016; Maier and Imazeki, 2013; Monmonier, 2018). Experts in the field of history have been particularly concerned with this lately because so-called big data has become more widely available, and data visualizations are being used as evidence in historical narratives and arguments despite a lack of training in statistics among many practicing historians (Eaton, 2019; Gibbs, 2016). Given the potential of data to mislead, it is therefore essential that students pay attention to sources of data visualizations, and ask questions about how data has been gathered, aggregated, and visually manipulated to communicate what might be a biased or deceptive message. 
  
  The data visualization analysis guide below is a tool that could be printed off for classroom use that tries to incorporate the principles discussed in this section, including aspects of critical data literacy. In its entirety, it represents the questions that should be asked of data visualizations when students reach a sophisticated level of analysis. But it can also be thought of as representing guideposts for progressive instruction, with elementary students working on mastery of the questions that are part of seeing the data, middle school students working on mastery of next section, and so on. 
  
  Yet, analysis of data visualizations may not be enough for students to achieve sophisticated levels of data literacy. Creating their own data visualizations is also important because it allows them to recognize the choices that go into data visualization, the ways that data can be manipulated, and the importance of clear terminology and labeling. The next section provides a variety of tools that students can use to create data visualizations, and manuals that will help them do it.