Initial soil organic carbon stocks govern changes in soil carbon: Reality or artifact?

Changes in soil organic carbon (SOC) storage have the potential to affect global climate; hence identifying environments with a high capacity to gain or lose SOC is of broad interest. Many cross-site studies have found that SOC-poor soils tend to gain or retain carbon more readily than SOC-rich soils. While this pattern may partly reflect reality, here we argue that it can also be created by a pair of statistical artifacts. First, soils that appear SOC-poor purely due to random variation will tend to yield more moderate SOC estimates upon resampling and hence will appear to accrue or retain more SOC than SOC-rich soils. This phenomenon is an example of regression to the mean. Second, normalized metrics of SOC change—such as relative rates and response ratios—will by definition show larger changes in SOC at lower initial SOC levels, even when the absolute change in SOC does not depend on initial SOC. These two artifacts create an exaggerated impression that initial SOC stocks are a major control on SOC dynamics. To address this problem, we recommend applying statistical corrections to eliminate the effect of regression to the mean, and avoiding normalized metrics when testing relationships between SOC change and initial SOC. Careful consideration of these issues in future cross-site studies will support clearer scientific inference that can better inform environmental management.     

"Describing our whole experience": the statistical philosophies of W. F. R. Weldon and Karl Pearson

There are two motivations commonly ascribed to historical actors for taking up statistics: to reduce complicated data to a mean value (e.g., Quetelet), and to take account of diversity (e.g., Galton). Different motivations will, it is assumed, lead to different methodological decisions in the practice of the statistical sciences. Karl Pearson and W. F. R. Weldon are generally seen as following directly in Galton’s footsteps. I argue for two related theses in light of this standard interpretation, based on a reading of several sources in which Weldon, independently of Pearson, reflects on his own motivations. First, while Pearson does approach statistics from this “Galtonian” perspective, he is, consistent with his positivist philosophy of science, utilizing statistics to simplify the highly variable data of biology. Weldon, on the other hand, is brought to statistics by a rich empiricism and a desire to preserve the diversity of biological data. Secondly, we have here a counterexample to the claim that divergence in motivation will lead to a corresponding separation in methodology. Pearson and Weldon, despite embracing biometry for different reasons, settled on precisely the same set of statistical tools for the investigation of evolution.     

How important is the statistical approach for analyzing categorical data? A critique using artificial nests

Ecologists rely heavily upon statistics to make inferences concerning ecological phenomena and to make management recommendations. It is therefore important to use statistical tests that are most appropriate for a given data-set. However, inappropriate statistical tests are often used in the analysis of studies with categorical data (i.e. count data or binary data). Since many types of statistical tests have been used in artificial nests studies, a review and comparison of these tests provides an opportunity to demonstrate the importance of choosing the most appropriate statistical approach for conceptual reasons as well as type I and type II errors.
Artificial nests have routinely been used to study the influences of habitat fragmentation, and habitat edges on nest predation. I review the variety of statistical tests used to analyze artificial nest data within the framework of the generalized linear model and argue that logistic regression is the most appropriate and flexible statistical test for analyzing binary data-sets. Using artificial nest data from my own studies and an independent data set from the medical literature as examples, I tested equivalent data using a variety of statistical methods. I then compared the p-values and the statistical power of these tests. Results vary greatly among statistical methods. Methods inappropriate for analyzing binary data often fail to yield significant results even when differences between study groups appear large, while logistic regression finds these differences statistically significant. Statistical power is is 2–3 times higher for logistic regression than for other tests. I recommend that logistic regression be used to analyze artificial nest data and other data-sets with binary data.     

Brief communication: Effect of size biases in the coefficient of variation on assessing intraspecific variability in the prosimian skeleton

Lack of independence of data points or the pooling fallacy has been suggested as a potential problem in the study of handedness in nonhuman primates, particularly as it relates to whether hand use responses should be recorded as individual events or bouts of activity. Here, I argue that there is no evidence that the concept of statistical independence of data points or the pooling fallacy is a problem in the evaluation of population‐level handedness in previous studies in nonhuman primates. I further argue these statistical concepts have been misapplied to the characterization of individual hand preferences. Finally, I argue that recording hand use responses as bouts rather than events has no significant effect on reports of hand use in nonhuman primates and, in fact, may unintentionally bias hand use toward the null hypothesis. Several suggestions for improvement in the measurement and statistical determination of individual handedness are offered in the article. Am J Phys Anthropol 151:151–157, 2013. © 2013 Wiley Periodicals, Inc.     

Where's the evidence?

Science is in large part the art of careful measurement, and a fixed measurement scale is the sine qua non of this art. It is obvious to us that measurement devices lacking fixed units and constancy of scale across applications are problematic, yet we seem oddly laissez faire in our approach to measurement of one critically important quantity: statistical evidence. Here I reconsider problems with reliance on p values or maximum LOD scores as measures of evidence, from a measure-theoretic perspective. I argue that the lack of an absolute scale for evidence measurement is every bit as problematic for modern biological research as was lack of an absolute thermal scale in pre-thermodynamic physics. Indeed, the difficulty of establishing properly calibrated evidence measures is strikingly similar to the problem 19th century physicists faced in deriving an absolute scale for the measurement of temperature. I propose that the formal relationship between the two problems might enable us to apply the mathematical foundations of thermodynamics to establish an absolute scale for the measurement of evidence, in statistical applications and possibly other areas of mathematical modeling as well. Here I begin to sketch out what such an endeavor might look like.     

Splitting the Difference? Principled Compromise and Assisted Dying

Compromise on moral matters attracts ambivalent reactions, since it seems at once laudable and deplorable. When a hotly‐contested phenomenon like assisted dying is debated, all‐or‐nothing positions tend to be advanced, with little thought given to the desirability of, or prospects for, compromise. In response to recent articles by Søren Holm and Alex Mullock, in this article I argue that principled compromise can be encouraged even in relation to this phenomenon, provided that certain conditions are present (which I suggest they are). In order to qualify as appropriately principled, the ensuing negotiations require disputants to observe three constraints: they should be suitably reflective, reliable and respectful in their dealings with one another. The product that will result from such a process will also need to split the difference between the warring parties. In assisted dying, I argue that a reduced offence of ‘compassionate killing’ can achieve this. I acknowledge, however, that splitting the difference can induce splitting headaches, as there remain certain questions to be answered. Hopefully, however, sufficient work is done here to defend attempts to occupy the middle ground, whether these relate to assisted dying specifically or to other disputed moral matters.     

The “survivor effect” and the evolution of parthenogenesis and self-fertilization

Parthenogenetic or self-fertilizing species are often reduced to small numbers of survivors or colonists. These types of reproduction may therefore evolve because they eliminate problems of finding a mate. It is argued here that outbreeding populations which are often reduced to a few survivors will experience heavy offspring losses when population expansion occurs. This ‘survivor effect’ results from the production of many individuals homozygous for deleterious mutations (i.e. inbreeding depression). Parthenogenetic and selfing species do not experience a “survivor effect”. This effect together with mate-finding problems will tend to cause the evolution of parthenogenesis and selfing in weedy or colonizing species.     

An alternative foundation for the planning and evaluation of linkage analysis. II. Implications for multiple test adjustments

The 'multiple testing problem' currently bedevils the field of genetic epidemiology. Briefly stated, this problem arises with the performance of more than one statistical test and results in an increased probability of committing at least one Type I error. The accepted/conventional way of dealing with this problem is based on the classical Neyman-Pearson statistical paradigm and involves adjusting one's error probabilities. This adjustment is, however, problematic because in the process of doing that, one is also adjusting one's measure of evidence. Investigators have actually become wary of looking at their data, for fear of having to adjust the strength of the evidence they observed at a given locus on the genome every time they conduct an additional test. In a companion paper in this issue (Strug & Hodge I), we presented an alternative statistical paradigm, the 'evidential paradigm', to be used when planning and evaluating linkage studies. The evidential paradigm uses the lod score as the measure of evidence (as opposed to a p value), and provides new, alternatively defined error probabilities (alternative to Type I and Type II error rates). We showed how this paradigm separates or decouples the two concepts of error probabilities and strength of the evidence. In the current paper we apply the evidential paradigm to the multiple testing problem - specifically, multiple testing in the context of linkage analysis. We advocate using the lod score as the sole measure of the strength of evidence; we then derive the corresponding probabilities of being misled by the data under different multiple testing scenarios. We distinguish two situations: performing multiple tests of a single hypothesis, vs. performing a single test of multiple hypotheses. For the first situation the probability of being misled remains small regardless of the number of times one tests the single hypothesis, as we show. For the second situation, we provide a rigorous argument outlining how replication samples themselves (analyzed in conjunction with the original sample) constitute appropriate adjustments for conducting multiple hypothesis tests on a data set.     

Interpreting evidence: why values can matter as much as science

Despite increasing awareness of the ways in which non-epistemic values play roles in science, many scientists remain reluctant to acknowledge values at stake in their own work. Even when research clearly relates to risk assessment and establishing public policy, contexts in which the presence of values is less likely to be contentious, scientists tend to present such research as merely involving empirical questions about what the evidence is. As a result, debates over policy-related science tend to be framed as purely epistemic debates over the state of the evidence. We argue that this neglects the important ways that ethical and social values play legitimate roles in judgments about what we take to be evidence for a particular policy. Using the case of recent disputes about the relative safety of home birth, we argue that although the debate has been framed as a purely scientific one about the empirical evidence for home birth, it actually involves disagreements about underlying value assumptions. If our claims are correct, then in order to move the debate forward, scientists will need to engage in a critical discussion about the values at stake.     

The trouble with isolation by distance

The genetic population structure of many species is characterised by a pattern of isolation by distance (IBD): due to limited dispersal, individuals that are geographically close tend to be genetically more similar than individuals that are far apart. Despite the ubiquity of IBD in nature, many commonly used statistical tests are based on a null model that is completely non-spatial, the Island model. Here, I argue that patterns of spatial autocorrelation deriving from IBD present a problem for such tests as it can severely bias their outcome. I use simulated data to illustrate this problem for two widely used types of tests: tests of hierarchical population structure and the detection of loci under selection. My results show that for both types of tests the presence of IBD can indeed lead to a large number of false positives. I therefore argue that all analyses in a study should take the spatial dependence in the data into account, unless it can be shown that there is no spatial autocorrelation in the allele frequency distribution that is under investigation. Thus, it is urgent to develop additional statistical approaches that are based on a spatially explicit null model instead of the non-spatial Island model.     

Consent and confidentiality in the light of recent demands for data sharing

Many attempts have been made to formalize ethical requirements for research. Among the most prominent mechanisms are informed consent requirements and data protection regimes. These mechanisms, however, sometimes appear as obstacles to research. In this opinion paper, we critically discuss conventional approaches to research ethics that emphasize consent and data protection. Several recent debates have highlighted other important ethical issues and underlined the need for greater openness in order to uphold the integrity of health‐related research. Some of these measures, such as the sharing of individual‐level data, pose problems for standard understandings of consent and privacy. Here, we argue that these interpretations tend to be overdemanding: They do not really protect research subjects and they hinder the research process. Accordingly, we suggest another way of framing these requirements. Individual consent must be situated alongside the wider distribution of knowledge created when the actions, commitments, and procedures of researchers and their institutions are opened to scrutiny. And instead of simply emphasizing privacy or data protection, we should understand confidentiality as a principle that facilitates the sharing of information while upholding important safeguards. Consent and confidentiality belong to a broader set of safeguards and procedures to uphold the integrity of the research process.     

Traveling questions: uncertainty and nonknowledge as vehicles of translation in genetic research participation

In this paper, I argue that uncertainty and nonknowledge, and not just research results, can be important vehicles of translation through which genetic research participation comes to affect the lives of research participants. Based on interviews with participants in a genetic research project, I outline epistemic, emotional, relational and moral implications of research participation. Many of them resemble what the literature has described as the social implications of genetic counseling, but here they stem from interaction with knowledge-in-the-making or what I simply call nonknowledge. While policies aimed at stimulating translation from bench to bedside tend to build on the assumption that research only works when knowledge translates into technological ability and creates utility, I suggest acknowledging that research has implications long before any clinical applications are at hand. Research questions, and not just results, may serve as a generative form of knowledge that can travel as fast as any answer.     

An alternative foundation for the planning and evaluation of linkage analysis. I. Decoupling "error probabilities" from "measures of evidence"

The lod score, which is based on the likelihood ratio (LR), is central to linkage analysis. Users interpret lods by translating them into standard statistical concepts such as p values, alpha levels and power. An alternative statistical paradigm, the Evidential Framework, in contrast, works directly with the LR. A key feature of this paradigm is that it decouples error probabilities from measures of evidence. We describe the philosophy behind and the operating characteristics of this paradigm--based on new, alternatively-defined error probabilities. We then apply this approach to linkage studies of a genetic trait for: I. fully informative gametes, II. double backcross sibling pairs, and III. nuclear families. We consider complete and incomplete penetrance for the disease model, as well as using an incorrect penetrance. We calculate the error probabilities (exactly for situations I and II, via simulation for III), over a range of recombination fractions, sample sizes, and linkage criteria. We show how to choose linkage criteria and plan linkage studies, such that the probabilities of being misled by the data (i.e. concluding either that there is strong evidence favouring linkage when there is no linkage, or that there is strong evidence against linkage when there is linkage) are low, and the probability of observing strong evidence in favour of the truth is high. We lay the groundwork for applying this paradigm in genetic studies and for understanding its implications for multiple tests.     

Surviving as an underrepresented minority scientist in a majority environment

I believe the evidence will show that the science we conduct and discoveries we make are influenced by our cultural experience, whether they be positive, negative, or neutral. I grew up as a person of color in the United States of America, faced with challenges that many had as members of an underrepresented minority group. I write here about some of the lessons I have learned that have allowed me to survive as an underrepresented minority ­scientist in a majority environment.     

Accurate delineation of biogeographical regions depends on the use of an appropriate distance measure

The use of analytical techniques to delineate biogeographical regions is becoming increasingly popular. One recent example, Heikinheimo et al . ( Journal of Biogeography , 2007, 34 , 1053–1064 ), applied the k -means clustering algorithm to define the biogeography of the European land mammal fauna. However, they used the Euclidean distance measure to cluster grid cells described by species-occurrence data, which is inappropriate. The Euclidian distance yields misleading results when applied to species-occurrence data because of the double-zero problem and the species-abundance paradox. We repeat their analysis using the Hellinger distance, a measure appropriate for species-occurrence data and which has been shown to outperform other such measures. Our results differ substantially from those presented by Heikinheimo et al. We argue that the rigorous application of appropriate statistical techniques is of crucial concern within conservation biogeography.     

Statistical tests as inappropriate tools for data analysis performed on non-random samples of plant communities

Many authors apply statistical tests to sets of relevés obtained using non-random methods to investigate phytosociological and ecological relationships. Frequently applied tests include thet-test, ANOVA, Mann-Whitney test, Kruskal-Wallis test, chi-square test (of independence, goodness-of-fit, and homogeneity), Kolmogorov-Smirnov test, concentration analysis, tests of linear correlation and Spearman rank correlation coefficient, computer intensive methods (such as randomization and re-sampling) and others. I examined the extent of reliability of the results of such tests applied to non-random data by examining the tests requirements according to statistical theory. I conclude that when used for such data, the statistical tests do not provide reliable support for the inferences made because non-randomness of samples violated the demand for observations to be independent, and different parts of the investigated communities did not have equal chance to be represented in the sample. Additional requirements, e.g. of normality and homoscedasticity, were also neglected in several cases. The importance of data satisfying the basic requirements set by statistical tests is stressed.     

Industrialization and ethnic change in the modern world

Despite the large recent attention given to ethnicity within the social sciences, the sources of modern ethnic change have remained opaque. Drawing upon social theory from Marx and Gellner, I argue here that industrialization incentivizes ethnic homogenization by lowering the relative value of land. Using carbon emissions per capita as a proxy for industrialization, I show that cross-country changes in ethno-linguistic fractionalization between 1961 and 1985 are negatively correlated with industrialization, and that this result is robust to the use of a variety of control variables, sub-samples and alternative measures of industrialization such as cement production, urbanization and agriculture as a percentage of GDP. In particular, I find no evidence for the direct role of the state in promoting ethnic homogenization, which adds to other recent evidence on how economic incentives may trump political ones as regards identity change, at least in the short- to medium term.     

Abortion, value and the sanctity of life

In Life's Dominion Dworkin argues that the debate about abortion is habitually misconstrued. Substantial areas of agreement are overlooked, while areas of disagreement are, mistakenly, seen as central. If we uncover a truer picture, then hope of a certain accord may no longer seem vain. I dispute many of these claims.
Dworkin argues that both sides in the debate are united in believing that life is sacred, or intrinsically valuable. I disagree. I maintain that only in a very attenuated sense of intrinsic value will this be agreed upon. I consider how an account of such value might be further fleshed out, but suggest, if this is done on any plausible lines, agreement will fall away. Dworkin argues, also, that the issue of personhood, does not, contrary to widespread belief, keep the parties apart. Again I disagree. We need to distinguish the question of whether there is in fact dispute over this issue from that of whether there is, in truth, good reason for dispute. And I argue that, rightly or wrongly, the issue of personhood remains central. Dworkin suggests that the purported proximity between the two sides offers some hope of an eventual reconciliation. At least, they will agree to differ, accepting that in this area freedom of choice is paramount. I am sceptical. Even this measure of reconciliation depends upon conservatives giving up positions which, I argue, they will continue to maintain.
There is a further point. Dworkin appears to be, in many ways, cautiously optimistic. I appear, in contrast, to be pessimistic. I argue, however, that only so long as we do disagree over matters of substance is there much hope that our differences might be resolved.     

Elastic priors to dynamically borrow information from historical data in clinical trials

Use of historical data and real-world evidence holds great potential to improve the efficiency of clinical trials. One major challenge is to effectively borrow information from historical data while maintaining a reasonable type I error and minimal bias. We propose the elastic prior approach to address this challenge. Unlike existing approaches, this approach proactively controls the behavior of information borrowing and type I errors by incorporating a well-known concept of clinically significant difference through an elastic function, defined as a monotonic function of a congruence measure between historical data and trial data. The elastic function is constructed to satisfy a set of prespecified criteria such that the resulting prior will strongly borrow information when historical and trial data are congruent, but refrain from information borrowing when historical and trial data are incongruent. The elastic prior approach has a desirable property of being information borrowing consistent, that is, asymptotically controls type I error at the nominal value, no matter that historical data are congruent or not to the trial data. Our simulation study that evaluates the finite sample characteristic confirms that, compared to existing methods, the elastic prior has better type I error control and yields competitive or higher power. The proposed approach is applicable to binary, continuous, and survival endpoints.