Is the Age of the Earth One of Our “Sorest Troubles?” Students’ Perceptions About Deep Time Affect Their Acceptance of Evolutionary Theory

When Charles Darwin was developing his ideas for On the Origin of Species, the most widely accepted estimates of Earth's age were those of William Thomson (later Lord Kelvin). Kelvin used calculations involving thermodynamics to argue that Earth is only 20–100 million years old—an age far too brief to accommodate evolution by natural selection. Darwin referred to Thomson's claim as one of his “sorest troubles,” for Darwin understood that the history of life on Earth ultimately relies on geology. Darwin suspected that Earth was much older than Thomson claimed, but Thomson's enormous stature as a scientist obliged Darwin to reconcile his claims with Kelvin's data. To accommodate Kelvin's timeline, Darwin proposed pangenesis as an explanation of inheritance (i.e., every sperm and egg contained “gemmules thrown off from each different unit throughout the body”). Darwin's explanation sped evolution while avoiding Lamarck's quasi-spiritual sources of acquired traits. However, Darwin's explanation of inheritance was wrong (see discussion in Moore et al. 2009a). The age of Earth remains a divisive topic in the modern evolution–creationism controversy. Whereas mainstream science has long acknowledged that Earth is approximately 4.5 billion years old, a vocal group of citizens and religious activists continue to insist that Earth is less than 10,000 years old. Although most geocentrists and flat-Earth advocates have capitulated to scientific evidence, young-Earth creationists continue to reject scientific evidence in favor of religious dictum to claim that Earth is less than 10,000 years old. These antiscience claims have been surprisingly popular with the public. For example, a Gallup Poll in early 2009 reported that “On Darwin's [200th] Birthday, Only 4 in 10 Believe in Evolution” (Newport 2009), and Berkman et al. (2008) noted that “16%[of biology teachers] believed that human beings were created by God in their present form at one time within the last 10,000 years.” In another study, 12.5% of students were young-Earth creationists (Rutledge and Warden 2000), as are 10%–14% of biology majors (Moore and Cotner 2009). Answers in Genesis’ (AiG) Creation Museum, along with the $27 million in donations required to build it, attest to the appeal of young-Earth creationism. Indeed, AiG's income for 2005 exceeded $13 million, and that of the Institute for Creation Research (ICR, another religious organization based on young-Earth creationism) exceeded $7 million. For comparison, the 2005 income of National Center for Science Education—the nation's leading organization that defends the teaching of evolution in public schools—was $1.2 million (Moore et al. 2009a). Clearly, Earth's age remains one of the “sorest troubles” for many people today, just as it did for Darwin. In this study, we examined how college students’ self-described religious and political views influence their beliefs about Earth's age and how this may affect their knowledge and acceptance of evolution. To our knowledge, this is the first study to examine these factors in college students. In 2009, we electronically surveyed 400 students enrolled in several sections of a nonmajors introductory biology course at the University of Minnesota. Because the course is one of a few options required of all undergraduates at the University, we assumed the survey population is representative of all students (politically, religiously, and demographically), except for those enrolled in the College of Biological Sciences. The optional survey, which was completed before the start of classes, consisted of the 20-item Measure of Acceptance of the Theory of Evolution (MATE) developed and validated by Rutledge and Sadler (2007), our own 10-item Knowledge of Evolution Exam (KEE; Moore et al. 2009a), and several items intended to gauge students’ religious and political preferences. The MATE consists of statements such as “the age of the Earth is less than 20,000 years” and “humans are the product of evolutionary processes,” to which students noted their level of agreement on a five-point Likert scale (from “strongly agree” to “strongly disagree”). The KEE questions were modified from an internal exam database, were authentic to the nonmajors’ test experience, and were designed to evaluate students’ understanding of basic tenets of evolutionary theory—for example, how fitness is measured, natural selection as a mechanism for evolutionary change, etc. (Appendix I). None of the KEE items specifically addresses the age of the Earth or evolutionary time. We also asked students whether they were politically liberal, conservative, or middle-of-the-road, and whether they were not religious (atheist or agnostic), or, if religious, were they progressive/liberal, orthodox, or middle-of-the-road. Students could ignore questions, and their responses had no impact on students’ grades. Response rate varied by survey item, with as many as 195 students (almost 50% of the targeted group) completing the KEE, and as few as 124 responding to some of the MATE items. The survey and subsequent data collection were approved by the University of Minnesota's Institutional Review Board. We extracted six variables from the survey to explore the factors that contribute to holding young-Earth and old-Earth beliefs about the origins of the world, on the one hand, and the relation of those beliefs to students’ knowledge of evolution and their presidential vote. The first two variables are self-reported measures of religious and political beliefs. The variable measuring students’ religious views (MYRELVIEW) is based on the question, “In general, I would describe my religious views as conservative, middle-of-the-road, liberal/progressive, or none of the above/I’m not religious.” Responses were coded on a four-point scale from conservative (1) to not religious (4). Similarly, students’ political views (MYPOLVIEW) were determined by the question, “In general, I would describe my political views as conservative, middle-of-the-road, or liberal.” MYPOLVIEW consists of a three-point scale ranging from conservative (1) to liberal (3). We constructed the young-Earth and old-Earth variables from MATE items that represent perspectives aligning with those beliefs. For the young-Earth variable (YOUNGEARTH), we averaged responses to the following two items: “The age of the Earth is less than 20,000 years,” and “The theory of evolution cannot be correct because it disagrees with the Biblical account of creation.” The old-Earth variable (OLDEARTH) was constructed by averaging responses to how much students agreed or disagreed with the statements that “Organisms existing today are the result of evolutionary processes that have occurred over millions of years,” and “The age of the earth is at least 4 billion years.” Scales for YOUNGEARTH and OLDEARTH range from 1 (Strongly Disagree) to 4 (Strongly Agree). Factor analysis confirms that each pair of items load onto a single dimension representing young-Earth creationist and old-Earth evolutionist beliefs with rotated factor loadings of 0.6979 and 0.7787, respectively. Students’ recollections of their high school biology courses (HSBIO) were captured in responses to an item asking them to identify whether or not evolution or creationism was taught in their courses. For the purposes of our analysis, HSBIO is coded according the following scheme: included neither evolution nor creationism = 1; included creationism, but not evolution = 2; included both evolution and creationism = 3; and included evolution, but not creationism = 4. The variable measuring students’ level of evolutionary biology knowledge (EVOGRADE) is a summative index of the number of questions answered correctly about various facets of evolutionary theory. EVOGRADE ranges in value from zero to 10 with each increment representing a correctly answered question. The 2008 presidential candidate supported by each student was collected via the self-reported, retrospective response to the statement, “In the past presidential election, I voted/would have voted for John McCain or Barack Obama.” For analytical purposes, we constructed the dichotomous variable VOTE in which support for John McCain = 0 and support for Barack Obama = 1. Students who voted for other candidates constituted a small minority and were excluded from analysis. Although our initial goals included the construction of a structural equation model (SEM) that employed simultaneous equations, the nature of the data is prohibitive (none of our variables is either linear or continuous). Therefore, we employed several individual models, each of which is best suited to the particular dependent variable under consideration. We then developed a structural model that demonstrates the nature of the relationships between and among our variables (Fig. 1). With the exception of models with VOTE as the dependent variable, all models are ordered logistical regression. We employed logistic regression models when the dichotomous VOTE variable was dependent. A Monte Carlo simulation was used to transform the ordered logistical regression results into predicted probabilities. Structural equation model demonstrating the nature of the relationships among the variables identified in Table 1. Note that “+” and “−” refer to positive and negative relationships, respectively, and asterisks follow the code *P < 0.05, **P< 0.01, ***P < 0.001. Summary statistics for the variables under consideration are reported in Table 1. The first set of relationships includes the exogenous variables related to students’ religious and political views. These variables mutually reinforce one another at statistically significant levels (P < 0.001). That is, the more liberal one's political views, the more likely one is to be liberal, agnostic, or atheistic in their religious views and vice versa. Students’ religious views also served as significant predictors of their beliefs about the origins of the world, their knowledge of evolutionary theory, and for whom they cast their presidential votes. Specifically, the more conservative a student's religious views, the greater the likelihood of endorsing young-Earth beliefs (P < 0.05) and the less likely they are to endorse old-Earth evolutionist beliefs (P < 0.01). Additionally, more liberal, agnostic, or atheistic religious students were significantly more likely (P < 0.05) to correctly answer knowledge-based questions about theories and facts related to evolution. Finally, students holding less conservative religious views were considerably more likely to have voted for or supported Barack Obama in the 2008 presidential election (P < 0.001). The political views held by students contribute significantly to their disposition toward evolutionary theory and to their choice of presidential candidate. Here, the more liberal a student's political views, the more likely they are to hold old-Earth beliefs about the origins of the world (P < 0.05) and to have cast their vote for Obama (P < 0.001). However, although more liberal political views are negatively related to acceptance of young-Earth views and positively related to knowledge of evolutionary theory, political views fail to predict significantly either of those variables. Turning to the intermediate variables measuring the impact of beliefs about the origins of life on Earth, we find mutually reinforced inverse relationships between young-Earth and old-Earth views that are highly significant (P < 0.001). Specifically, those who hold young-Earth views are significantly less likely to accept an old-Earth rooted in evolutionary theory and vice versa. We also find that holding old-Earth beliefs contributes significantly to the ability of students to comprehend complex theoretical and factual tenets of evolutionary theory (P < 0.05). Furthermore, acceptance of old-Earth beliefs also significantly increased the probability of voting for or supporting Obama. Conversely, holding young-Earth beliefs appears to impact negatively, but not significantly, one's ability to grasp cognitively evolutionary theory. And while holding creationist beliefs did predict a greater likelihood of voting for John McCain, they did not do so in a statistically significant manner. Finally, in keeping with previous findings (Moore and Cotner 2009), the content of students’ high school biology courses was a significant predictor of their acceptance of evolutionary theory. The content of students’ high school biology course affects knowledge of evolution as well: students whose high school biology course included only evolutionary theory had approximately a 70% chance of answering half of the questions or more correctly whereas those with courses teaching only creationism had an approximately 50% chance of doing so (for discussion, see Moore et al. 2009b); those with neither evolution nor creationism had a 42% chance of scoring 50% or above, suggesting that creationism-only education is comparable to no education at all with respect to evolutionary biology. In combination, students who recall being taught evolution only in high school, and who are religiously and politically liberal, were more likely to earn any score above 50% on the exam than were their counterparts with more conservative educational, religious, and political backgrounds. College students have a variety of religious and political views that have been shaped collectively by their families, their communities, and the institutions with which they have come into contact during the course of their lives. These deeply rooted views influence students’ receptiveness to theories about life's origins. And because these beliefs about creationism and evolution are firmly grounded in, and are expressions of, their worldviews, as instructors we are naïve to assume that students in college biology courses are necessarily open to scientific inquiry. Although student views of evolution are subject to multiple influences, our data indicate that their views about Earth's age are a strong predictor of several different factors. But is the age of Earth one of our “sorest troubles?” Our research and historical and contemporary literature suggest the following: (1) Deep time is conceptually difficult to grasp, for the general public, science educators, and students throughout the educational spectrum. Charles Darwin himself had a hard time grasping “deep time”—the geologic concept of time, requiring billions of years—but knew that it was essential for his proposed evolutionary mechanism. Darwin didn't publish any estimates of Earth's age, but in the first edition of On the Origin of Species he did estimate the time needed to erode the Weald (a region in south England stretching from Kent to Surrey), “say three hundred million years.” He then explained: “I have made these few remarks because it is highly important for us to gain some notion, however, imperfect, of the lapse of years. During each of these years, over the whole world, the land and the water has been peopled by hosts of living forms. What infinite number of generations, which the mind cannot grasp, must have succeeded each other in the long roll of years” (Darwin 1859). More recently, attention has focused on the inability of scientists (Brush 2001), science educators (Petcovic and Ruhf 2008), college students (Catley and Novick 2009; Libarkin 2006; Truscott et al. 2006), and the general public (Hofstadter 1996) to comprehend time in billions of years. Our deficiencies involve concepts about both absolute time (time in and Novick and time in an Libarkin Students’ inability to accept an Earth is a to evolution only is an of time to our understanding of life's and (Catley and Novick 2009; it is a concept for understanding and (see in that age for the Earth is the of and in years, focused their to an Earth et al. et al. in an Earth is only if in Earth's age is to the and is, in the of “the that the whole Furthermore, evolution is of to about time et al. a that students to about evolution value as to modern For example, Libarkin et al. describe students who Earth's age correctly but fail to comprehend the of time Earth was or the of that likely These may to a to historical and or, more an on the nature of science et al. into a more for teaching about evolution and the age of for life's are to and than are scientific that on an understanding of time. For evidence of one to the for throughout In a to claim evidence of a in John to a and in William a and of every of which in the in the of nature the scientific required to from to but the basic of (Moore et al. 2009a). more contemporary evidence of the of In that scientific creationists no of data to their is not on their In the their must only to their that time, of the public believed that living were created in their present form within the last 10,000 years more than two of science education that remains we are by or et al. 2008), to in a designed for a a natural to in with a more model of Earth's age than of see et al. a of time. In asking our students to the the and the for the old, the and the we may be our own evolutionary about time teaching for about time is a for most instructors must the that Earth's age to student understanding of evolution. In and in with of teaching for et al. 2008), may to students to a that their of Earth's age, both absolute and are and in of Although several have for time and on the is to accept an Earth has We do not that to creationist for natural has However, in this study, we specifically on voting in and a significant between presidential candidate choice and is likely by contemporary for example, in several support some form of teaching creationism as a theory to evolution by natural selection support Creation Science with other theories of the of evolution is it be as only a The political for creationism is Gallup the different levels of acceptance of evolution among and (Newport 2009). In one of claim that were created in their present form by God within the last 10,000 years, a by of and of those that and God had no are are and are These data are with findings that are more likely to and who are highly likely to in creationist for human origins (Newport For example, of those who 70% that God created in their present form (a of young-Earth or of those who or are (Newport These data suggest that with one's to vote and one's likelihood to evolutionary of as an Earth that is billion years old. We do not to claim that of science is to nor that are no creationist creationism more than do their counterparts and have in years by creationism a Our research that students who are liberal, agnostic, or atheistic in their religious views, are politically liberal, were taught evolution in high school, and accept the science evolutionary theory are more likely to the theoretical concepts and findings related to evolution than those who are more conservative politically and religiously, either no evolution or a form of evolution in high school, and who do not accept an However, our findings also a exception to this holding young-Earth views may not significantly a student's ability to facts about evolutionary theory. although it is not the of biology instructors to in political or religious remains the of students about evolution via A student's vote is likely a for religious views, and the between their voting and their acceptance of evolution appears and that to young-Earth beliefs a not only of modern but also and such may as a for scientific in a to in Earth's age, the scientific and evolutionary theory could have that less than who we for political of the following support the theory of selection as an of natural selection and of among be that no the evolutionary history of be all of the above evidence to support the theory of evolution to a variety of has in many of are on the are the of these and the are by natural selection no are just or is not evolution by natural selection have so these are to are than of the following is the most in an evolutionary a who is at but has no a who has many of which to a who a and to have three a who for his two of whom to a who has a of many and one a a of has a occurred as a result of in each of the population the of this a with than needed to to the and over time, they to this some of the population had than was they would be more likely to and increased in to small the of this a with than that by stretching their the would and this was on to their time, the none of the above of the following statements about natural selection is natural selection to within a population natural selection to likelihood of for based on The from the population all within a population have an chance of and is based on choice natural selection results in those within a population who are and more natural selection to the is evidence that evolution is evolution has occurred evolution all are from a life millions of years of the following statements evolution by natural selection is natural selection on natural selection is a small time natural selection result in the of from a are important as the of which natural selection A in the of a population of time is evolution evolution natural selection of the following is the of in natural natural selection of the following best the between evolution and natural natural selection is one mechanism that result in the of evolution natural selection in whereas evolution natural selection is a whereas evolution toward a natural selection is of or in the evolution of individual they are the