xiii
Statistics is hot. According to an article in the New York Times, statistics is perhaps the most promising, adventurous career option you can choose right now—
If you dare to embrace what your professor is teaching you, it will bring you to the brink of personal and social change. You will have to make many decisions about how you think—
We dare you to love this course.
In their classic and persuasive article, Marsha Lovett and Joel Greenhouse (2000) present principles to teach statistics more effectively (all based on empirical research from cognitive psychology). And other researchers continue to build on their helpful work (see Benassi, Overson, & Hakala, 2014). We look to this body of research as we create every edition of this statistics text, from designing the pedagogy to deciding what specific examples to include. Six principles emerge from this research on teaching statistics and drive our text:
Practice and participation. Recent research has shown that active learning, broadly defined, increases student performance and reduces the failure rate in science courses, including psychology courses (Freeman et al., 2013). This principle pertains to work outside the classroom as well (Lovett & Greenhouse, 2000). Based on these findings, we encourage students to actively participate in their learning throughout the text. Students can practice their knowledge through the many applied exercises, especially in the Applying the Concepts and Putting It All Together sections. In these sections, the source of the original data is often supplied, whether it is data from the Centers for Disease Control or a Marist poll, encouraging students to dig deeper. And students can take advantage of data sets from the General Social Survey and EESEE Case Studies to “play” with statistics beyond the exercises in the book.
Vivid examples. Researchers have found that students are most likely to remember concepts illustrated with a vivid instructional tool (VanderStoep, Fagerlin, & Feenstra, 2000). So, whenever possible, we use striking, vivid examples to make statistical concepts memorable, including the weights of cockroaches to explain standardization, destructive hurricanes in the discussion of confounding variables, entertainment by a clown during in vitro fertilization to teach chi square, a Damien Hirst dot painting to explain randomness, and a house purchase by Beyoncé to highlight celebrity outliers. Vivid examples are often accompanied by photos to enhance their memorability. When such examples are drawn from outside the academic literature, we follow with engaging research examples from the behavioral sciences to increase the memorability of important concepts.
xiv
Integrating new knowledge with previous knowledge. When connecting new material to existing student knowledge, students can more easily embed that new material into “a framework that will enable them to learn, retrieve, and use new knowledge when they need it” (p. 7, Ambrose & Lovett, 2014). Throughout the text, we illustrate new concepts with examples that connect to things most students already know. Chapter 1 includes an exercise that uses students’ knowledge of contemporary music, specifically the percentage of rhyming words in rap lyrics, to teach students how to operationalize variables. In Chapter 2, an example from Britain’s Got Talent uses students’ understanding of the ranking systems on reality shows to explain ordinal variables. In Chapter 5, we use students’ understanding of the potential fallibility of pregnancy tests to teach the difference between Type I and Type II errors. And in Chapter 14, we use the predictive abilities of Facebook profiles to teach regression. Learning in different contexts helps students to transfer knowledge to new situations, so we use multiple examples for each concept – typically an initial one that is easier to grasp followed by more traditional behavioral science research examples.
Confronting misconceptions. Conversely, some kinds of prior knowledge can slow students down (Lovett & Greenhouse, 2000). Students know many statistical words – from independent to variability to significant. But they know the “everyday” definitions of these words, and this prior knowledge can impede their learning of the statistical definitions. Throughout the book, we point out students’ likely prior understanding of these key terms, and contrast that with the newer statistical definitions. We also include exercises aimed at having students explain the various ways a given word can be understood. Plus, in Chapter 5, we introduce ways in which other types of misconceptions can emerge through illusory correlation, confirmation bias, and coincidence. Throughout the rest of the book, we highlight these types of flawed thinking with examples, and show how statistics can be the antidote to these kinds of misconceptions – whether it’s a belief that holiday weight gain is a serious problem, cheating is associated with better grades, or online personality quizzes are always accurate.
Real-
xv
Repetition. There is a growing literature on the role of “desirable difficulty” in learning – that is, students learn better when they struggle with new material with support (Clark & Bjork, 2014). The three techniques of spacing, interleaving, and testing – all based on the central idea of repetition – help to create the right level of difficulty to help students learn more efficiently.
Spacing involves repeated practice sessions with the same material with delays in between. Our book is set up to encourage spacing. For example, the Before You Go On sections at the beginning of each chapter offer students a chance to review previous material. Several sets of Check Your Learning questions are included across each chapter, and more exercises are included at the end of each chapter.
Interleaving refers to the practice of mixing the types of exercises the student is practicing. Rather than practicing each new task in one block of exercises, students mix exercises on a new topic with repeats of exercises on earlier topics. This repetition of practice with earlier concepts increases retention of material. We build in exercises that encourage interleaving in the Putting It All Together sections, which ask students to return to concepts learned in earlier chapters.
Testing is possibly the best way to learn new material. Simply studying does not introduce the desirable difficulty that enhances learning, but testing forces errors and drives efficient retention of new material. The tiered exercises throughout the chapter and at the end of the chapter provide numerous opportunities for testing – and then more testing. We encourage students to aim for repeated practice, completing more exercises than assigned, rather than by studying in more traditional, but less effective, ways.
Statistics and statistical reasoning are in the midst of profound changes. Here are two important trends:
Trend 1: Visual Displays of Data. On the one hand, Chapter 3 of this text reminds us that there is nothing very new about creating visual displays of data. On the other hand, the entire field has gone topsy-
xvi
Trend 2: Free Software. Although earning a college degree is pretty expensive, the Internet has created opportunities for particular forms of education to progressively become less expensive. Massive open online courses (MOOCs) are just one of the more obvious efforts. One of your coauthors, Tom, took one MOOC with 80,000 other classmates. Kahn Academy online tutorials are another excellent, low-