This is the second article in the methodology series; it focuses on different ways to analyze data graphically. As is the case for all articles in this series, the authors welcome comments from all readers who have suggestions about the way information is presented or questions about content. This treatment is not comprehensive and does not replace information found in textbooks or peer-reviewed articles. Beyond their experience, the authors have used textbooks, articles, and Web sites as sources, and they recommend that readers do likewise (see Bibliography) for more detailed explanations of the content of this article.Scenario: Raj, the junior member of an architectural firm we met in the first article in this series, has been asked by his boss to present the results of his analysis of the survey data previously collected. The chief executive officer (CEO) from the hospital where the data were collected will be there, too. Raj contacts his colleague, Sarah, for help about how to do this. Their conversation follows.Raj: Sarah, thanks for your previous help with understanding and computing descriptive statistics. My boss wants me to present the survey results to the hospital CEO and chief operating officer. I was planning to take some of the descriptive statistics I created and present them in tables. How does that sound?Sarah: That is certainly one way to present the information, Raj. Descriptive statistics are certainly valuable, and it makes sense to discuss these statistics in your presentation. To further strengthen your presentation and better explain the results, however, I would also include graphical displays of the data.Raj: Graphical displays? Would that be similar to the histograms we created previously?Sarah: Yes. In fact, histograms are one type of graphical display of data. For this discussion, I am going to use a data set I collected for a different project. First, though, I need to explain the advantages of graphical analysis of data.Graphical analyses of data are important because they communicate valuable information beyond that contained in descriptive statistics-and even in more complicated statistical analyses. Welldesigned plots of data stimulate comparisons that can lead to new understanding, and they can uncover patterns and structure in the data that might be missed in statistical analysis. Graphical displays allow the viewer to immediately see outliers (i.e., data points that are quite different from the majority of data points) as well as relationships between variables (curvilinear associations, for example) that would not be evident solely by examining statistics.Graphical displays also allow one to see trends or changes across a number of different factors. They can reveal distinct clusters (groupings) in the data related to a previously unknown variable. For example, the histograms below visually depict central tendency (where the data points tend to cluster) and variability (spread in the data) for a variable such as noise. In this example, noise is a built environment factor that is measured with three questions about whether noise is minimized in the hospital setting.Each of the questions contained a 1 = Strongly Disagree to 7 = Strongly Agree Likert-type rating scale. The histograms in Figure 1 are based on the average responses to the three questions for each of the participants who answered these questions. In this example, two factors are depicted. The first factor is the type of participant who answered the noise questions: family members of patients, patients, staff , and physicians. The second is the clinic in which the participants were located: A, B, C, and D.By examining the histograms, we can see that within participant type, responses tend to be similar. Notice that family member responses tend to be clustered near the high end of the response scale (i.e., responses tend to indicate agreement with the questions about noise). …