The scientific method and models

Reading

Scientific method: An introduction

Pease, C.M. and J.J. Bull. 2000. Science for Business, Law and Journalism.Read Chapter 1. The need for making good choices Chapter 2. A template for scientific inquiry Chapter 3. How non-scientists use the scientific method Note: This is a simple introduction to the scientific method. This semester, we will apply these concepts to the Grandjean et al. (1998) paper.

Models: An introduction

Pease, C.M. and J.J. Bull. 2000. Science for Business, Law and Journalism. Read Chpt. 4 Models are the building blocks of science and Chpt. 5 All models are false. But some are still useful. Note: You read these chapters a couple weeks ago. Take another look at them. They provide a broad introduction to the concept of a model. In this class, we will find numerous examples of models in the Grandjean et al. (1998) paper, and we will use these models to help us understand and simplify this paper.

A key scientific paper on methylmercury

Grandjean, P. et al. 1998. Cognitive performance of children prenatally exposed to "safe" levels of methylmercury. Environmental Research, Section A 77: 165-177. Note: Don't even think about reading this paper from front to back. Scientists generally do not read scientific paper from front to back, and you should not either. Read it with a view of answering the following questions. Note well that these question ask only that you understand this paper at the very broadest level.

Questions for class discussion

Asking questions about science

1. Bring to class any questions you have about the science in the Grandjean et al. paper. (I really do call on students, and ask them to ask me a question about the science.)

Scientific method (on asking the right questions)

2. Science for Business, Law and Journalism presents a 5 part template for the scientific paper: goal, model, data, evaluation, revision. Give an example of each part of the scientific method, drawn from the Grandjean et al. (1998) paper.

3. What is the goal of the Grandjean et al. (1998) paper? What study design do they employ to answer their question? What are their major conclusions?

Models in the Grandjean et al. (1998) paper

4. Here are a very few of the many models employed by Grandjean et al. (1998),:

(i)The hypothesis, "methylmercury causes neurological defects" is a model.
(ii) The hypothesis, "PCB causes low scores on the Boston Naming Test" is also a model.
(iii) The authors use mercury concentration in the hair as a model.
(iv) The Faroe Island residents who participated in this study is a model.
(v) The Abstract of the paper is a model.
(vi) Indeed the paper itself is a model.
(vii) A child's score on the Boston Naming test is a model.

For each model listed above, complete the following model template:
Model:
What this model substitutes for:
Type of model: (Physical, abstract, sampling)
Major assumption of this model:
Status of model: (Accepted, rejected, etc.)

Physical, abstract and sampling models

5. Provide examples that illustrate the distinction between physical, abstract and sampling models.

Why bother with models at all?

6. Why did employ the models listed above? Did they have any choice? Can a scientist simply refuse to employ models, on the grounds that any model must necessarily make untested and obviously wrong assumptions?