II Scientific MethodII Scientific MethodA. IntroductionB. KnowledgeC. Types of ReasoningD. ExampleE. Formal ProcessA. IntroductionYou are the new biologistin charge of a deer herd.Harvest has declined for 3 consecutive years.What should be done?A. IntroductionWhat caused decline in herd?List possible causesGather relavant informationAnalyze itSelect a course of actionA methodical process to knowledgeB. KnowledgeDefinition: Knowledge is defined as the set of ideas that agree with or are consistent with the facts of nature.B. KnowledgeHow do we attain knowledge? Attainment of knowledgeLogical argument (model)Descriptive observationsExperiment Attainment of knowledgeExample: Number of mayfly larvae eaten by trout in an hourLogical argumentNo larvae eaten if none availableIf few available, few eatenIf many available, many eatenTotal eaten is limitedLogical argumentConclusion: Number of larvae eaten increases as density of larvae increases up to a maximum above which no more are eaten per hour.Logical argumentWe can express this more clearly in numeric form as C. S. Holling did in 1950's.Holling’s Modeln= number eatena= search rateX = densityt = time feedingts = time searchingth = time handling preyh = handling time per itemEncountersNumber eaten = search rate x prey density x time searchingn=a X tsTime feedingtime feeding = time searching + time handling preyt = ts + th ts= t - thHandling timetime handling prey = number eaten x handling time per itemth = n hHolling’s Model of Functional Responsets = t - thts = t - n hn = a X (t- n h) a X tn = ------------------ 1 + a X hLogical argument: Functional ResponseN 020406080024681012No. of larvae eaten per hr.Larval density2. Descriptive observationsFind streams with different densities of larvaeCollect fishCut open and count larvaeRelate number eaten to density of larvae in streamsDescriptive observationsSize 1Size 2Size 30 2 4 6 8 10 12020406080100Larvae per stomachLarval density3. ExperimentMaintain trout in experimental stream sections in laboratoryExpose individuals to a range of densities of larvaeDetermine consumption per hourExperiment0 2 4 6 8 10 12020406080100Larvae eaten per hourLarval density Attainment of knowledgeLogical argument (model)Descriptive observationsExperimentC. Types of ReasoningInductionDeductionC. Types of Reasoning (Romesburg)Inductive ReasoningRetroductionHypothetico-Deductive ReasoningArguments by AuthorityD. ExampleWinter distribution of partridges on the PalouseChukar (Alectoris chukar)Gray Partridge (Perdix perdix)Chukar and Gray Partridge on PalouseIntroduced speciesHuns on ridge topsChukar lower down, rocky areasNative distributionsConceptual ModelTemperature may be separatingProposition: Huns more cold hardy than ChukarsConceptual model from initial observations, literature, suggestions of experts, experience, insight, and logicConceptual ModelChukars occur in lower sheltered areas because they are less able to withstand cold temperatures than huns.HypothesisChukars will lose weight more quickly at -10 deg. C than HunsDesign a testCold hardiness of Chukar and HunsChukarHuns1 2 3 4 5 6 7 80510152025No. of birdsDays to 20% weight lossCold hardiness of Chukar and HunsWhat do you conclude?Is there really a difference?Statistical TestNull hypothesis Ho: There is no difference.Ha: There is a difference.May reject Ho.What if fail to reject Ho?Fisher (1947) said“It should be noted that the null hypothesis is never proved or established but is possibly disproved in the course of experimentation. Every experiment may be said to exist only in order to give the facts a chance of disproving the null hypothesis.”A FactA fact is something that has existence. It is an event, an occurrence, observation, or relation, the reality of which is manifest in experience or may be inferred with certainty.CertaintyHow certain of conclusion?Probability levelFacts never established with absolute certainty.InterpretationRe-evaluate the experimentWas it valid?How were subjects chosen?Re-evaluate in context of larger questionLogical assumptionsCold hardiness of Chukar and HunsChukarHuns1 2 3 4 5 6 7 80510152025No. of birdsDays to 20% weight lossReport ResultsPublicationWhy bother?Recycle to next hypothesisE. Formal Process1. Literature review and observations2. Conceptual model (theory)3. Formulate hypothesis4. Test hypothesis5. Data analysis6. Evaluation and interpretation7. Speculation and new hypotheses8. PublicationSchematic OutlineGarton, Ratti and Giudice (2005)Fuller, more comprehensive list of stepsAlternate HypothesesPlatt (1964) pointed out that we tend to be narrow-minded.Platt (1964) and Chamberlain (1897 reprinted in 1965) said we should formulate alternate hypotheses.Strong Inference (Platt 1964 after Chamberlin 1897)Consider all reasonable alternate hypotheses and design one experiment or set of observations which would rule out many hypotheses.Then design another experiment, etc.Multiple CausesStrong Inference has proven very powerful in molecular biology and other sciences where single causes predominate.Most population questions are multi-causal so use an approach directed at examining Multiple Competing Hypotheses (Caughley and Gunn 1996) similar to model selection of Burnham & Anderson (1998)Multiple Causes - ExampleH0Major FallaciesPopulations and samplesReplicationControlsScience and PlanningScience and planning are intertwinedDecision-makingScience and planning compared as processesModelsMultiple factorsForces us to be clear and systematicSmart People Believe Weird Things Scientific American (2002) article by Michael Shermer (Skeptic)“Rarely do any of us sit down before a table of facts, weigh them pro and con, and choose the most logical and rational explanation, regardless of what we previously believed.”“Rather, such variables as genetic predisposition, parental predilection, sibling influence, peer pressure, educational experience and life impressions all shape the personality preferences that, in conjunction with numerous social and cultural influences, lead us to our beliefs.” Smart People Believe Weird Things Scientific American (2002) article by Michael Shermer (Skeptic)“ We then sort through the body of data and select those that most confirm what we already believe, and ignore or rationalize away those that do not.” = confrimation