The Scientific Method: Knowledge Acquisition and the Quality of Science

THE SCIENTIFIC METHOD: KNOWLEDGE ACQUISITION AND THE QUALITY OF SCIENCE

Learning Objectives

• Outline the four requirements for science to result in knowledge acquisition.

• Explain the components of a scientific argument: claim, evidence, reasoning, and the concept of inferential strength.

• Explain what the scientific method entails and its significance.

• Differentiate between a hypothesis, a prediction, a fact, and a theory.

• Contrast inductive vs. deductive reasoning.

• Explain why science proceeds via rejecting, not proving, hypotheses.

• Explain the difference between descriptive vs. hypothesis-testing claims and how they complement one another within the scientific method.

• Explain the concept of extrapolation and relate it to observational vs. manipulative studies.

• Explain the concepts of confounding variables, the role of controls in addressing them, and their connection to the type of study.

Science and Knowledge Acquisition

• Knowledge acquisition requires researchers to be:

  • Rational: Inferences must be guided by sound, logical reasoning based on accurate evidence.

  • Skeptical: Researchers must repeatedly scrutinize patterns (to determine if they are real), inferences (to check if they are logical), and hypotheses (to see if they are reasonable and data-consistent). They must be willing to reject or modify hypotheses and conclusions based on evidence.

  • Objective: Researchers must be unbiased by preconceived notions, beliefs, ideologies, and experiences.

  • Methodologically Materialistic: Explanations must rely on natural, rather than supernatural, processes.

• Researchers make inferences through scientific arguments and the scientific method.

Scientific Argument

• A scientific argument is a logical chain of reasoning that connects evidence to a claim. It consists of three components: Claim, Evidence, Reasoning.

• Researchers maintain skepticism toward scientific arguments, aiming to assess the quality of the science.

• The quality of science is determined by the strength of the scientific argument (sometimes referred to as inferential strength), which depends on:

  • Nature of the claim

  • Quality and quantity of evidence

  • Soundness of the reasoning connecting evidence to the claim.

Claim

• Definition: A specific and clear assertion supported by strong evidence and sound reasoning. A claim represents a statement about believed truth, subject to revision if evidence or reasoning changes.

• Claims are typically found in the Discussion/Conclusions section of research papers.

Evidence

• Definition: Information that is relevant to the validity of a claim, often represented by results of studies presented in research papers.

• The strength of the claim is directly influenced by the quality and quantity of data/evidence.

Reasoning

• Definition: The process of logically relating evidence to a claim. This is often detailed in the Introduction and Discussion sections of research papers, and its quality is a critical determinant of inferential strength.

Evaluating the Quality of a Scientific Argument

• Question prompts for evaluating a scientific argument include:

  • Is the claim clear and testable?

  • Are patterns in the data genuine?

  • Does the reasoning logically connect the claim to the evidence?

Types of Claims

• There are two types of claims:

  • Hypothesis-testing claim: An assertion linked to the validity of a hypothesis.

  • Descriptive claim: An assertion that describes or characterizes a pattern in the physical and/or natural world.

• A scientific hypothesis serves as a causal explanation for a pattern or observation.

The Scientific Method

• The scientific method is a systematic approach to knowledge acquisition aimed at ensuring evidence-based understanding that respects rational conclusions drawn from sound reasoning.

• Interplay between description and hypothesis-testing:

  • Description provides data that suggests possible explanations for patterns.

  • Hypothesis-testing interprets these patterns, enhances causal understanding, and directs future observational efforts.

Example: Eutrophication in Freshwater Lakes

1. Descriptive claim: Observations of increased algal growth in nutrient-rich waters.

2. Hypothesis generation: Phosphorus (P) is a limiting nutrient that, when available in higher concentrations, leads to increased algal growth.

3. Hypothesis-testing: Studies comparing ponds with artificially increased P levels to control ponds found no difference, rejecting the hypothesis.

4. Further characterization: Investigate correlations between primary production and additional factors.

Inductive vs. Deductive Reasoning

• Inductive Reasoning:

  • Moves from specific observations to general claims.

  • Example: Based on observations of swans, one may conclude that all swans are white.

  • Often the source of biological hypotheses but not used for testing them.

• Deductive Reasoning:

  • Moves from general claims to specific predictions.

  • If the premises are true, the conclusion must also be true.

  • Example:

  • Premise 1: All birds have feathers.

  • Premise 2: All robins are birds.

  • Conclusion: Therefore, all robins must have feathers.

Falsification of Hypotheses

• As argued by Karl Popper, science should advance by eliminating hypotheses instead of proving them, as hypotheses cannot be definitively proven:

  • Example of logical fallacy:

  • If humidity is high due to rain (hypothesis H), and a wet garden is observed (P), one cannot conclude that high humidity is solely due to rain, as there may be other causes.

  • A hypothesis can be disproven if the predicted outcome (P) does not occur.

Scientific Facts and Theories

• While science cannot 'prove' hypotheses, certain claims are so strongly supported that they become accepted scientific knowledge:

  • Fact: A descriptive claim that withstands rigorous scrutiny, e.g., matter is made of atoms, DNA carries hereditary information.

  • Theory: A causal explanation, e.g., evolutionary theory, the theory of relativity, germ theory of disease.

Types of Scientific Studies

• There are two categories:

  • Observational studies: Researchers observe and measure without manipulation.

  • Manipulative studies (experiments): Researchers manipulate one or more variables and compare outcomes to controls or treatments.

• The type of claim is not dependent on the type of study (descriptive vs. hypothesis testing).

Example of Studies:

1. Observational Study: Assessing when hummingbirds arrive in spring.

2. Manipulative Study: Testing the impact of differing phosphorus levels on algal growth.

Extrapolation

• Extrapolation involves drawing inferences from model systems to actual systems, and higher levels of extrapolation reduce inferential strength.

• Types include:

  • Interspecies: Using rats as models for humans.

  • Spatial: Experimenting in simplified lab settings versus natural environments.

  • Temporal: Applying short-term results over longer periods.

Confounding Variables

• Defined as an external variable that may influence observed patterns. It may correlate with both independent and dependent variables and create apparent associations.

  • Example: Cities with more pubs may have more churches due to the confounding variable of population size.

Addressing Confounding Variables

• Can be controlled through:

  • Design controls: Techniques applied during the study design phase to reduce confounding influences.

  • Statistical controls: Measurements included in data analysis to account for confounding variables, common in observational studies.

Ideal Study Characteristics

• An ideal scientific study:

  • Manipulates only relevant factors (thus minimizing confounders).

  • Conducted as close to the actual system of interest as possible to limit extrapolation.

• Observational studies tend to offer less extrapolation and may allow for higher statistical power due to increased replication.

Summary of the Scientific Method

• Causal and falsifiable hypotheses arise from observations and prior studies.

• Deductive predictions are rigorously tested through observations or experiments while maintaining controls to reduce confounding variables.

• Statistical analysis determines if patterns are evident in results and valid.

• Outcomes dictate support or rejection of hypotheses, leading to further inquiries or characterization of new claims.

Case Study: Skin Colour Evolution in Humans

• Geographic variations in skin colour imply underlying biological causes.

• A hypothesis is proposed regarding the evolution of skin colour related to UV exposure, with potential competing hypotheses that must be tested.

Science and Knowledge Acquisition

Knowledge acquisition through the scientific method requires researchers to be:

• Rational: Inferences guided by sound, logical reasoning based on accurate evidence.

• Skeptical: Scrutinize patterns, inferences, and hypotheses, willing to reject/modify based on evidence.

• Objective: Unbiased by preconceived notions, beliefs, or experiences.

• Methodologically Materialistic: Explanations rely on natural, not supernatural, processes.
  Researchers make inferences via scientific arguments and the scientific method.

Scientific Argument

A scientific argument is a logical chain of reasoning connecting Claim, Evidence, and Reasoning. Its quality (inferential strength) depends on the claim's nature, evidence quantity/quality, and reasoning soundness.

• Claim: A specific assertion about believed truth, supported by evidence, subject to revision.

• Evidence: Information relevant to claim validity, often study results.

• Reasoning: The logical process relating evidence to a claim, crucial for inferential strength.
  Evaluating arguments involves assessing claim clarity, data genuineness, and logical connections.

Types of Claims and Reasoning

Claims are either hypothesis-testing (linked to the validity of a hypothesis) or descriptive (characterizing patterns). A scientific hypothesis serves as a causal explanation for a pattern or observation.

• Inductive Reasoning: Moves from specific observations to general claims; often forms hypotheses but not for testing them.

• Deductive Reasoning: Moves from general claims to specific predictions; if premises are true, conclusion must be true, used for testing hypotheses.
  Science advances by rejecting (falsifying), not proving, hypotheses, as they cannot be definitively proven. A hypothesis is disproven if a predicted outcome does not occur.

Scientific Facts and Theories

• Fact: A descriptive claim that withstands rigorous scrutiny (e.g., matter is made of atoms).

• Theory: A causal explanation (e.g., evolutionary theory).

Types of Scientific Studies and Challenges

Scientific studies are either observational (measure without manipulation) or manipulative/experiments (manipulate variables, compare to controls).

• Extrapolation: Drawing inferences from model systems to actual systems, reducing inferential strength (e.g., interspecies, spatial, temporal).

• Confounding Variables: External variables influencing observed patterns, addressed by design controls (during study) or statistical controls (in data analysis).
  Ideal studies manipulate only relevant factors and are conducted close to the actual system, minimizing confounders and extrapolation.

Summary of the Scientific Method

The scientific method involves:

• Causal and falsifiable hypotheses arising from observations.

• Rigorous deductive testing via observations or experiments with controls.

• Statistical analysis to determine valid patterns.

• Outcomes dictating support or rejection of hypotheses, leading to further inquiries.