02_Hyp_Testing_Exp_Design

02_Hyp_Testing_Exp_Design - Hypothesis Testing and...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Hypothesis Testing and Experimental Design Today’s questions: I. What is the nature of science? (reprise) II. How can hypotheses be tested experimentally? I. What is the nature of science? (reprise) Some pseudoscience hypotheses: 1. Touching crystals can protect you against bad energy. 2. Certain types of behavior bring good luck. 3. Disease is caused by people wishing ill on others. Predictions: 1. If you enter X’s house and don’t touch crystals in a prescribed order, bad things will happen. 2. If you don’t change underwear during March madness, your favorite team is more likely to win. 3. If someone has cast a spell on you, a specific ceremony can counter the spell and maintain your health. Caveats… Problems with negative results: What if people believe in the mechanism? Scientific theories usually have a pattern component and a process component. • The pattern component is a statement of fact— something that is • The process component is a mechanism that In terms of pattern and process, why is pseudoscience considered “pseudo?” Pattern: Process: II. How can hypotheses be tested experimentally? A. What is the nature of science? “The essence of science is evidence‐based decision making.” Carl Bergstrom* “You’d love to have truth with a capital T, but you’ve got truth with a little, provisional t.” Ben Kerr* * “We absolutely must leave room for doubt or there is no progress and there is no learning. There is no learning without having to pose a question. And a question requires doubt. People search for certainty. But there is no certainty. “ “Scientific knowledge is a body of statements of varying degrees of certainty—some most unsure, some nearly sure, none absolutely certain.” Richard Feynman 1999. The Pleasure of Finding Things Out. (Basic Books, New York) B. How do scientists find things out? What is the nature of scientific evidence? Start with some observations: • Nemoria caterpillars hatched in spring look like • Nemoria caterpillars hatched in summer look like Key observations: Spring caterpillars eat Summer caterpillars eat Hypothesis to be tested: Caterpillars develop different morphology in spring versus summer because they Null hypothesis: Experiment Experiment conducted in the lab: * Raise large numbers of caterpillars from both * generations on two different diets, under identical * physical conditions. * * * What are the identical physical conditions? Temperature at Photoperiod at Housed in identical cups, in the same environmental chamber Ad lib food Results of the feeding experiment: How many individuals developed into the “flower morph?” Flower diet Leaf diet What can you conclude from these data? Suppose that the results had been: Flower diet Leaf diet Would these results be as convincing as the actual results? An An alternative hypothesis is that caterpillars develop differently if they are exposed to spring‐like versus summer‐like physical conditions (light and temperature). Design an experiment to test this hypothesis. Null hypothesis: Experimental set‐up • treatments: • control for other variables: • what to measure: Prediction of temp/light hypothesis: Prediction of null hypothesis: I am working on a stream restoration project on the Olympic Peninsula and we are trying to reforest an abandoned pasture. Which type of mulch works best for planting tree saplings: Burlap, vispore, or wood chips? Design an experiment to answer this question. Coumadin is a widely prescribed as a blood thinner for people at risk of stroke. (It’s also a rat poison.) Suppose that Zymogenetics developed an alternative drug that is hypothesized to have the same beneficial effects with fewer side effects. Design an experiment to test this hypothesis. Mulching question—Experimental set‐up • treatments: • control for other variables: • what to measure: List predictions: Coumadin question—Experimental set‐up • treatments: • control for other variables: • what to measure: List predictions: ...
View Full Document

This note was uploaded on 05/25/2010 for the course BIO 180 taught by Professor Bradshaw during the Spring '09 term at University of Washington.

Ask a homework question - tutors are online