What are the five techniques discussed in this chapter for misleading with graphs?

What are the steps to create a scatterplot?

How can you convert a scatterplot into a range-frame?

What does it mean for two variables to be linearly related?

How can we tell whether two variables are linearly or nonlinearly related?

What is the difference between a line graph and a time plot?

What is the difference between a bar graph and a Pareto chart?

Bar graphs and histograms look very similar. In your own words, what is the difference between the two?

What are pictorial graphs and pie charts?

Why are bar graphs preferred over pictorial graphs and pie charts?

Why is it important to identify the independent variable and the dependent variable before creating a visual display?

Under what circumstances would the x-axis and y-axis not start at 0?

Chartjunk comes in many forms. What specifically are moiré vibrations, grids, and ducks?

Geographic information systems (GIS), such as those provided by computerized graphing technologies, are particularly powerful tools for answering what kinds of research questions?

How is a bubble graph similar to a traditional scatterplot?

How does a bubble graph differ from a traditional scatterplot?

Alumni giving rates, calculated as the total dollars donated per year from 2006 to 2016, represent which kind of variable—nominal, ordinal, or scale? What would be an appropriate graph to depict these data?

Alumni giving rates for a number of universities, calculated as the number of alumni who donated and the number who did not donate in a given year, represent which kind of variable—nominal, ordinal, or scale? What would be an appropriate graph to depict these data?

You are exploring the relation between gender and video game performance, as measured by final scores on a game.

Do the data in the graph below show a linear relation, a nonlinear relation, or no relation? Explain.

Do the data in the graph below show a linear relation, a nonlinear relation, or no relation? Explain.

What elements are missing from the graphs in Exercises 3.20 and 3.21?

The following figure presents the enrollment of graduate students at a university, across six fall terms, as a percentage of the total student population.

When creating a graph, we need to make a decision about the numbering of the axes. If you had the following range of data for one variable, how might you label the relevant axis?

If you had the following range of data for one variable, how might you label the relevant axis?

The scatterplot in How It Works 3.1 depicts the relation between fertility and life expectancy. Each dot represents a country (or, as in the case of Hong Kong, a special administrative region).

Based on the data in the bubble graph in Figure 3-18, what is the relation between physical health and positive emotions?

The colors and sizes of the bubbles in the bubble graph in Figure 3-18 represent the gross domestic product (GDP) for each country. Using this information, explain what the relation is between positive emotions and GDP.

Type of graph for the relation between height and attractiveness: A social psychologist studied the effect of height on perceived overall attractiveness. Students were recruited to come to a research laboratory in pairs. The pairs sat together in the waiting room for several minutes and then were brought to separate rooms, where their heights were measured. They also filled out a questionnaire that asked, among other things, that they rate the attractiveness of the person who had been sitting with them in the waiting room on a scale of 1 to 10.

Type of graph for the effects of cognitive-behavioral therapy on depression: A social worker tracked the depression levels of clients being treated with cognitive-behavioral therapy for depression. For each client, depression was assessed at weeks 1 to 20 of therapy. She calculated a mean for all of her clients at week 1, week 2, and so on, all the way through week 20.

Type of graph for comparative suicide rates: The World Health Organization tracks suicide rates by gender across countries. For example, in 2011, the rate of suicide per 100,000 men was 17.3 in Canada, 17.7 in the United States, 44.6 in Sri Lanka, 53.9 in the Russian Federation, 1.4 in South Africa, and 2.5 in the Philippines.

Scatterplot of daily cycling distances and type of climb: Every summer, the touring company America by Bicycle conducts the “Cross Country Challenge,” a 7-week bicycle journey across the United States from San Francisco, California, to Portsmouth, New Hampshire. At some point during the trip, the exhausted cyclists usually start to complain that the organizers are purposely planning for days with lots of hill and mountain climbing to coincide with longer distances. The tour staff counter that no relation exists between climbs and mileage and that the route is organized based on practical issues, such as the location of towns in which riders can stay. The organizers who planned the route (these are the company owners who are not on the tour) say that they actually tried to reduce the mileage on the days with the worst climbs. Here are the approximate daily mileages and climbs (in vertical feet), as estimated from one rider’s bicycle computer.

Scatterplot of gross domestic product and education levels: The Group of Eight (G8) consists of many of the major world economic powers. It meets annually to discuss pressing world problems. In 2013, for example, the G8 met in the United Kingdom with an agenda that included international security. Decisions made by G8 nations can have a global impact; in fact, the eight nations that make up the membership reportedly account for almost two-thirds of the world’s economic output. Here are data for seven of the eight G8 nations for gross domestic product (GDP) in 2004 and a measure of education. The measure of education is the percentage of the population between the ages of 25 and 64 that had at least one university degree (Sherman, Honegger, & McGivern, 2003). No data point for education in Russia was available, so Russia is not included.

Time series plot of organ donations: The Canadian Institute for Health Information (CIHI) is a nonprofit organization that compiles data from a range of institutions—from governmental organizations to hospitals to universities. Among the many topics that interest public health specialists is the problem of low levels of organ donation. Medical advances have led to ever-increasing rates of transplantation, but organ donation has not kept up with doctors’ ability to perform more sophisticated and more complicated surgeries. Data reported by CIHI (https://secure.cihi.ca/free_products/2011_CORR_Annua_Report_EN.pdf, 2015) provide Canadian transplantation and donation rates for 2001–2010. Here are the donor rates per million in the population; these numbers include only deceased donors.

Bar graph of alumni donation rates at different kinds of universities: U.S. universities are concerned with increasing the percentage of alumni who donate to the school because alumni donation rate is a factor in the U.S. News & World Report’s university rankings. What role might type of university play in alumni donation rates? U.S. News & World Report ranks private national universities, public national universities, and private liberal arts colleges (http://www.usnews.com/education/best-colleges/the-short-list-college/articles/2015/10/20/10-universities-where-the-most-alumni-donate). National universities tend to focus on graduate education and research, whereas liberal arts colleges tend to focus on undergraduate education. Here are recent alumni donation rates for the top 10 schools on this variable. With the exception of Princeton, a private national university, all of these institutions are private liberal arts colleges.

Bar graph versus Pareto chart of countries’ gross domestic product: In How It Works 3.2, we created a bar graph for the 2012 GDP, in U.S. dollars per capita, for each of the G8 nations. In How It Works 3.3, we created a Pareto chart for these same data.

Bar graph versus time series plot of graduate school mentoring: Johnson, Koch, Fallow, and Huwe (2000) conducted a study of mentoring in two types of psychology doctoral programs: experimental and clinical. Students who graduated from the two types of programs were asked whether they had a faculty mentor while in graduate school. In response, 48.00% of clinical psychology students who graduated between 1945 and 1950 and 62.31% who graduated between 1996 and 1998 reported having had a mentor; 78.26% of experimental psychology students who graduated between 1945 and 1950 and 78.79% who graduated between 1996 and 1998 reported having had a mentor.

Bar graph versus pie chart of student participation in community activities: The National Survey on Student Engagement (NSSE) has surveyed more than 400,000 students—freshmen and seniors—at 730 U.S. schools since 1999 (“America’s Best Colleges 2004,” 2003). Among the many questions on the NSSE, students were asked how often they “participated in a community-based project as part of a regular course.” For the students at the 19 institutions classified as national universities that made their data publicly available through the U.S. News & World Report Web site, here are the data: never, 56%; sometimes, 31%; often, 9%; very often, 5%. (The percentages add up to 101% because of rounding decisions.) Explain why a bar graph would be more suitable for these data than a pie chart.

Software defaults of graphing programs that portray satisfaction with graduate advisors: The 2000 National Doctoral Program Survey asked 32,000 current and recent Ph.D. students in the United States, across all disciplines, to respond to the statement “I am satisfied with my advisor.” The researchers calculated the percentage of students who responded “agree” or “strongly agree”: current students, 87%; recent graduates, 86%; former students who left without completing the Ph.D., 48%.

Software defaults of graphing programs for depicting the “world’s deepest” trash bin: The car company Volkswagen has sponsored a “fun theory” campaign in recent years in which ordinary behaviors are given game-like incentives to promote pro-social behaviors such as recycling or obeying the speed limit (http://www.thefuntheory.com/). In one example, Volkswagen created a seemingly super-deep trash bin; when a person would throw something out, a highpitched whistling sound played for 7 seconds, culminating in a far-off-sounding boom, as if the item had fallen for hundreds of feet! The fun-theory people collected data; in a single day, 72 kilograms of trash were thrown out in their “deep” bin, versus just 31 kilograms in a nearby trash bin.

Multivariable graphs and college rankings by academics and sexiness: Buzzfeed.com published a multivariable graph that purported to rank colleges by academics and “hotness.” The data from this graph are represented here.

Types of graph appropriate for behavioral science research: Give an example of a study—real or hypothetical—in the behavioral sciences for which the researchers could use each type of graph. State the independent variable(s) and dependent variable, including levels for any nominal variables.

Creating the perfect graph: What advice would you give to the creators of each of the following graphs? Consider the basic guidelines for a clear graph, for avoiding chartjunk, and regarding the ways to mislead through statistics. Give three pieces of advice for each graph. Be specific—don’t just say that there is chartjunk; say exactly what you would change.

Graphs in the popular media: Find an article in the popular media (newspaper, magazine, Web site) that includes a graph in addition to the text.

Interpreting a graph about two kinds of career regrets: The Yerkes–Dodson graph demonstrates that graphs can be used to describe theoretical relations that can be tested. In a study that could be applied to the career decisions made during college, Gilovich and Medvec (1995) identified two types of regrets—regrets of action and regrets of inaction—and proposed that their intensity changes over time. You can think of these as Type I regrets—things you have done that you wish you had not done (regrets of action)—and Type II regrets—things you have not done that you wish you had done (regrets of inaction). The researchers suggested a theoretical relation between the variables that might look something like the graph below.

Thinking critically about a graph of the frequency of psychology degrees: The American Psychological Association (APA) compiles many statistics about training and careers in the field of psychology. The accompanying graph tracks the number of bachelor’s, master’s, and doctoral degrees conferred between the years 1970 and 2000.

Thinking critically about a graph about international students: Researchers surveyed Canadian students on their perceptions of the globalization of their campuses (Lambert & Usher, 2013). The 13,000 participants were domestic undergraduate and graduate students—that is, they were not recently from countries outside of Canada. The pie chart here shows the responses to one survey item: “The increasing number of international students attending my institution has led to improvements in the university’s reputation and image.”

Interpreting a graph about traffic flow: Go to http://maps.google.com/. On a map of your country, click on the traffic button in the upper left corner.

Interpreting a graph about political support: As we learned in this chapter, pie charts are often not the best choice to present data.

Multivariable graphs and shipwrecks: The cliché “women and children first” originated in part because of the Titanic captain’s famous directive as his ship was sinking. Yet, the cliché is not grounded in reality. See the multivariable graph on page 76.

Type of graph describing the effect of romantic songs on ratings of attractiveness: Guéguen, Jacob, and Lamy (2010) wondered if listening to romantic songs would affect the dating behavior of the French heterosexual women who participated in their study. The women were randomly assigned to listen to either a romantic song (“Je l’aime à Mourir,” or “I Love Her to Death”) or a nonromantic song (“L’heure du Thé,” or “Tea Time”) while waiting for the study to begin. Later in the study, an attractive male researcher asked each participant for her phone number. Of the women who listened to the romantic song, 52.2% gave their phone number to the researcher, whereas only 27.9% of the women who listened to the nonromantic song gave their phone number to the researcher.

Developing research questions from graphs: Graphs not only answer research questions, but can spur new ones. Figure 3-9 on page 55 depicts the pattern of changing attitudes, as expressed through Twitter.