Primer on Natural Resource Science

Chapter 2. Hypotheses

CHAPTER 2 Hypotheses We shall . . . see that there are several kinds of hypotheses; that some are verifiable, and when once confirmed by experiment become truths of great fertility; that others may be useful to us in fixing our ideas; and finally, that others are hypotheses only in appearance. —Henri Poincaré (1952:xxii) The hypothesis is the principal intellectual instrument in research. —W. I. B. Beveridge (1957:71) There isn’t a scientific method. There are many—and none of them is foolproof. —Brian L. Silver (1998:23; emphasis in original) Scientists widely regard hypotheses as key elements of science; indeed , most would say hypotheses are central. A hypothesis is generally viewed as a speculative thought, especially one that might explain something and is worthy of further investigation. Words such as guess, conjecture, supposition, surmise, and speculation are approximate synonyms in the proper context. However, the word hypothesis has richer and deeper meanings in science than conveyed in the preceding paragraph. It often is accompanied by modifiers such as nonscientific or research. Each of these modified forms has subcategories. What I call a Poincaré hypothesis is an elegant mental construct, an unobservable, that scientists posit to explain or understand phenomena. This chapter defines and gives examples of the many types of hypotheses used in natural resource science. Nonscientific Hypotheses There are at least 3 kinds of nonscientific hypotheses. The first kind is a hypothesis that is always false. Although it seems weird that scientists would advance hypotheses that are always false, the practice is common. The statistical null hypothesis, widely applied in scientific research, is a priori false, in general. This will be explained in chapter 10. The second kind of nonscientific hypothesis is one that is vacuously true (tautological, or necessarily true). Consider these examples: * “Our objective was to test the hypothesis that blackbirds [Icteridae ] select roosting locations according to specific habitat features” (Lyon and Caccamise 1981:435). 15 Hypotheses * “We tested the hypothesis that habitat characteristics differed among summer communal, winter communal, summer solitary , and winter solitary roosts. We also tested the hypothesis that roost habitat differed from habitat available at random” (Buehler et al. 1991). * “We hypothesized that ocelots [Leopardus pardalis] would prefer large patches of closed canopy habitat [i.e., known preferred habitat] and avoid large patches of unsuitable habitat” (Jackson et al. 2005). * “We hypothesized that occupied habitat would differ from a random sampling of available . . . habitat” (Patten et al. 2005). * “Our hypothesis was that the mix of species in native seed mixes would establish as well as the mix of species in predominantly introduced species mixes, but that higher total seeding rates are necessary for them to do so” (Thompson et al. 2006:238). This hypothesis is vacuously true because native species have lower establishment rates than introduced species. If you plant native species at higher seeding rates, eventually establishment rates will equalize. These hypotheses are simply assertions of the obvious. Vacuous hypotheses often indicate simple descriptive studies camouflaged as hypothesis-driven studies. They probably also imply cultural in- fluence in that practitioners suppose in a tribal manner, and so do referees of articles, that it is good to state a hypothesis. Honesty and simplicity would be better served in natural resource science if researchers doing descriptive work simply listed their objectives. Here are some examples of objectives in descriptive studies: * “The objective of our study was to determine the settlement patterns of offspring in this [red squirrel (Tamiasciurus hudsonicus)] population, and their fates” (Larsen and Boutin 1994:215). * “The objectives of this study were to examine seasonal habitat use and home range size of gray partridge [Perdix perdix] in eastern South Dakota” (Smith et al. 1982:580). * “The present study . . . was designed to investigate the energy economy of the Canada goose [Branta canadensis] in relation to variations in temperature and photoperiod to which it is 16 Perspectives subjected during its annual cycle” (Williams and Kendeigh 1982:588). * “This paper describes mobility, home range, and habitat use by mallard [Anas platyrhynchos] broods” (Talent et al. 1982:629). The third kind of nonscientific hypothesis is one that is not testable (not falsifiable), as pointed out in chapter 1, and has no heuristic value. Generally, religious beliefs are not testable in a scientific sense. Here are some examples: * My totem protects me from evil spirits. * Souls are reincarnated in different...