Science as a Way of Knowing

Introduction to Scientific Inquiry

Science is a systematic approach to understanding the natural world through observation, hypothesis formation, experimentation, and analysis. The process of forming and testing hypotheses is central to scientific investigation and allows researchers to draw meaningful conclusions about biological phenomena.

Forming Hypotheses

Developing a Hypothesis

A hypothesis is a testable statement that explains an observation or answers a scientific question. Hypotheses are often developed in response to observations in nature or experimental results.

• Observation: Noticing a phenomenon, such as increased ocean acidity affecting kelp growth.

• Forming a Hypothesis: Example: "Ocean acidification will have negative effects on kelp."

• Example: After reading a news article about ocean acidification, a scientist may hypothesize that lower pH levels in seawater will reduce kelp growth rates.

Making Predictions

Predictions are specific statements about what will happen if the hypothesis is correct. They often use "if...then..." logic.

• Example Prediction: "If the pH of seawater is reduced, then kelp will grow more slowly."

• Alternatively: "Reducing the pH of seawater will reduce the growth rate of kelp."

• Variables:

  • Independent variable: The variable that is manipulated (e.g., pH of seawater).

  • Dependent variable: The variable that is measured (e.g., kelp growth rate).

Alternative and Null Hypotheses

Defining Hypotheses

Scientific studies often involve two types of hypotheses: the alternative hypothesis and the null hypothesis.

• Alternative Hypothesis (HA): States that changing the independent variable will affect the dependent variable. Example: "Reducing the pH of seawater will reduce the growth rate of kelp."

• Null Hypothesis (H0): States that changing the independent variable will not affect the dependent variable. Example: "The pH of seawater will have no effect on the growth rate of kelp."

When using "if...then..." statements: Alternative: "If the pH of seawater is reduced, then the growth rate of kelp will be reduced." Null: "If the pH of seawater is reduced, then the growth rate of kelp will be unchanged."

Purpose of Null Hypothesis

• Null hypotheses are used to test whether observed effects are statistically significant.

• If statistical analysis allows us to reject H0, we can accept HA (assuming proper experimental design).

Testing Hypotheses

Main Methods

There are three primary ways to test scientific hypotheses:

• Experimental Study

• Observational Study

• Modeling

Experimental Studies

Experimental studies involve manipulating one variable to observe its effect on another.

• Independent variable: The variable manipulated by the researcher.

• Dependent variable: The variable measured to assess the effect.

• Experiments must have at least two groups:

  • Experimental group: Variable of interest is manipulated.

  • Control group: Variable of interest is not manipulated.

• All other variables should be kept constant between groups (controlled variables).

• Confounding variables: Variables that may affect the outcome and should be controlled.

Observational Studies

Observational studies involve measuring variables without manipulation to identify patterns or correlations.

• Measure variables in a population or location of interest.

• Look for correlations between variables (e.g., does kelp growth correlate with pH levels?).

• Compare populations or locations to see if variable differences are predictable.

• Important: Correlation does not imply causation. Spurious correlations can occur.

Observational Study Table

Features of Good Studies

• Use standardized methods to improve repeatability.

• Proper controls to ensure valid comparisons.

• Confounding variables do not differ between groups.

• Replication:

  • Within a study: Multiple individuals, plots, or sites.

  • Between studies: Test the hypothesis more than once.

Replication in Science

Importance of Replication

Replication increases confidence in scientific results and helps identify patterns in biological variation.

• Replicates in control and experimental groups are essential.

• Studying large numbers of individuals reveals real patterns.

• Random assignment to groups reduces bias.

• Replication of entire studies tests reproducibility and generalizability.

Using Models to Test Hypotheses

Role of Models in Science

Models are representations of how systems work and can be used to test hypotheses. They may be physical, visual, or mathematical.

• Inputs: Independent variables

• Model: Describes how the system works (the hypothesis)

• Outputs: Dependent variables

Modeling Table

Testing Models

• Compare model outputs to real-world data to assess validity.

• Use historical data to see if model predictions match reality.

• Use models to predict future outcomes based on forecasted inputs.

Model Testing Table

Summary

• Science relies on forming and testing hypotheses through observation, experimentation, and modeling.

• Alternative and null hypotheses are essential for statistical testing.

• Experimental, observational, and modeling approaches each have strengths and limitations.

• Replication and proper controls are critical for reliable scientific results.

• Models help scientists understand, validate, and predict biological phenomena.

Additional info: Replication and control of confounding variables are emphasized in modern biology to ensure the reliability and generalizability of experimental results. Models are increasingly used in biology to integrate complex data and predict outcomes under changing environmental conditions.