Philosophy of Science: Understanding Scientific Reasoning LIBS 7006

Course offerings

Learning Outcomes

Upon successful completion, the student will be able to:

• Distinguish theoretical models from real-world objects of study.
• Abstract brief descriptions of theoretical models from popular or semi-technical science reports.
• Recognize the empirical predictions of such models.
• Note whether these predictions are actually fulfilled in observation or experiment (negative or positive evidence).
• Judge whether there are other plausible models that can (also) explain the facts.
• Judge whether to accept, reject or suspend judgement on whether the proposed model accurately represents the world.
• Judge whether the proposed model is credible, but needs further development or testing; or judge that the proposed model is implausible, and (currently, at least) not worth pursuing.
• See how historically eminent scientific models triumphed over then-plausible contemporary views.
• The students will have read through James Watson's classic, The Double Helix, as well as the brief historical episodes discussed in the texts.
• The students will also have been provided with further details regarding the historical episodes discussed in the text. Such nuances can help reveal how and when theory evaluations are tied to theoretical traditions.
• See that 'marginal science' models are generally not worth accepting or pursuing because (i) they make vague or multiple predictions; (ii) their predictive 'successes' are also explained by more plausible models; or (iii) they are deeply inconsistent with well-established theories.
• Recognize and intelligently engage with some alternative philosophies of science.
• Distinguish sample and statistical models from the larger population they are designed to represent.
• Understand statistical proportions, distributions, correlations and variables.
• Understand basic mathematical probability models (addition, multiplication rules, conditional probabilities, the structure of random sampling, and standard deviation).
• Quickly compute margins of error from sample sizes.
• Recognize the connection between margin of error and confidence level.
• Construct simple statistical models of reported proportions, distributions and correlations.
• Recognize and evaluate samples according to how well they approximate random sampling.
• Distinguish statistical significance from 'significance.'
• Judge whether a proposed statistical model should be accepted, rejected or treated as unsupported.
• Distinguish causation from correlation.
• Distinguish deterministic from probabilistic models of causation.
• Distinguish causal models for individuals from those for populations.
• Understand causal 'effectiveness.'
• Understand the role of control and experimental groups in establishing causal hypotheses.
• Understand the theoretical superiority of Randomized Experimental Designs for supporting causal hypotheses.
• See when Prospective and Retrospective causal models are required, and when they can support a causal hypothesis.
• Understand controlling for other variables, matching control and experimental groups, and constructing control groups (in Retrospective studies).
• (If time permits.) Understand the elements of decision-making models (options, states of the world, outcomes and values).
• Distinguish ranked from measured values.
• Recognize when a situation calls for decision making with certainty, uncertainty, or risk.
• Grasp better, worse and satisfactory options and correlated strategies.
• Combine probabilities and values, in terms of expected values in situations involving known risks.

Effective as of Fall 2003

Subscribe