Notes on Scientific Modeling, Evidence, and Explanation

Scientific Reasoning, Modeling, and Explanation

• Core workflow in science:

  • Every claim must be supported by evidence.
  • Evidence typically comes from experiments, observations, and the data produced.
  • A scientific principle can also serve as evidence.
  • With more involved experiments, you can collect more data and build models (e.g., graphs).

• The coffee and milk example as a learning model:

  • Our basic setup (the plate) is the starting point for evaluating claims.
  • Claim: Adding milk to hot coffee affects the temperature.
  • Evidence: Measure the temperature before and after adding milk; milky coffee has a lower temperature than the coffee without milk (as observed).
  • With more data points, e.g., adding 1 tablespoon of milk, recording the temperature after each addition, you can build a dataset.
  • A graph derived from this data becomes a model of the process.
  • A model can give predictive power: you can predict the final temperature if you add an untested amount of milk (e.g., 2.5 teaspoons).
  • Key takeaway: with sufficient data and experimentation, a model can generate testable, quantitative predictions.

• What a model is and the forms it can take:

  • A model is something that helps us make sense of a system by predicting outcomes.
  • Predictions must be testable and quantifiable.
  • Models explain how something happens; they do not just describe it.
  • Forms of models include:
  • Drawings
  • Graphs
  • Diagrams
  • Equations
  • Physical models
  • Mental (conceptual) models
  • The goal is to choose a model type that communicates the mechanism and yields useful predictions.

• Molecular-level modeling example in class:

  • Prompt: Draw what happens at the molecular level when you add cold drugs to water in a test scenario (referred to as "Adam's ale" in the transcript).
  • This kind of model focuses on molecules interacting, colliding, and exchanging energy to explain observed phenomena.
  • The model helps explain what is happening between molecules, illustrating the underlying process.

• Historical context: multiple models of the atom

  • The lecturer mentions that, on 10/01/1941, five different models of the atom were discussed.
  • All five models portray the same underlying reality but look very different.
  • The point: atomic theory changes over time as new data accumulate; models improve with evidence.
  • This illustrates how scientific knowledge evolves from one model to another while aiming to describe the same phenomena more accurately.

• Scientific explanation (CER framework): claim, evidence, reasoning

  • Scientific explanation is built from three parts:
    1) Claim: the target of the explanation (the statement you are trying to justify).
    2) Evidence: data from experiments, observations, or a scientific principle.
    3) Reasoning: the explanation of how the evidence supports the claim (the mechanism).
  • Reasoning is the hardest part because it links the evidence to the claim by detailing the mechanism or cause.
  • The goal is to craft explanations that are as detailed as needed while staying relevant and avoiding irrelevant information.

• Applying the CER framework to the coffee experiment:

  • Claim: Coffee cools down when milk is added.
  • Evidence: Measured temperature changes before and after adding milk.
  • Reasoning: Describe the mechanism by which the evidence supports the claim (e.g., heat transfer and mixing leading to a new, lower equilibrium temperature).
  • The lecturer asks students to work with peers for two minutes to draft a two-sentence reasoning: one sentence restating the mechanism and how it connects to the data, and a second sentence reinforcing the link between evidence and claim.

• Practical notes on building explanations:

  • Your reasoning should explain why and how the mechanism causes the observed outcome.
  • Consider what causes the temperature change when milk is added (the energy balance/heat transfer during mixing) and how that leads to a lower final temperature.
  • Don’t dump all knowledge; include only information relevant to the claim and evidence at hand.

• Important concepts to remember:

  • Claim: the hypothesis or conclusion being tested.
  • Evidence: data, observations, experiments, or scientific principles supporting the claim.
  • Reasoning: the logical connection and mechanism that explains how the evidence supports the claim.
  • Models: representations (graphs, drawings, diagrams, equations) that provide predictive power and explanations; they can be physical or mental.
  • Predictions: testable outcomes derived from the model; they should be measurable and quantitative when possible.
  • Testability and quantifiability: a hallmark of a good model.

• Connections to broader science practice:

  • Models can be used to make predictions about conditions not yet tested (e.g., different volumes of milk).
  • The same underlying principle can be represented by different models that are equally valid for the same phenomenon.
  • As data accumulate, models may be refined or replaced to improve explanatory and predictive power.

• Quick reference to the coffee experiment (specific values mentioned):

  • Additions mentioned: 1 tablespoon of milk; prediction example uses 2.5 teaspoons of milk.
  • Measured quantity: temperature before and after milk addition; milky coffee temperature is lower than the baseline hot coffee.
  • Predictive scenario: using a graph/model to predict temperature for a condition not explicitly tested (e.g., 2.5 teaspoons).

• Exercise prompt (in-class activity):

  • Take two minutes to work with peers and draft two-sentence reasoning for the coffee example:
  • Sentence 1: Explain the mechanism by which adding milk lowers the coffee temperature.
  • Sentence 2: Explicitly connect the mechanism to the observed data (the temperature change).

• Final note on learning goals:

  • Understand that a model is a flexible representation that can be drawn, graphed, diagrammed, or written as an equation.
  • Recognize that explanations require a clear claim, robust evidence, and a well-articulated reasoning that links the two.
  • Appreciate the historical context: models evolve with new evidence, improving our understanding of how the world works.