Scientific Skills & Statistics in Biology

Scientific Skills & Statistics

Experimental Design & Hypotheses

Understanding experimental design is essential for conducting reliable biological research. Hypotheses are testable statements that guide the direction of experiments and data analysis.

• Hypothesis: A testable prediction about the outcome of an experiment.

• Null Hypothesis (H0): States there is no effect or difference.

• Alternative Hypothesis (HA): States there is an effect or difference.

• Variables: Factors that can change in an experiment.

  • Independent Variable: The variable manipulated by the experimenter.

  • Dependent Variable: The variable measured in response to changes in the independent variable.

  • Controlled Variables: Factors kept constant to ensure a fair test.

• Replication: Repeating experiments to ensure reliability.

• Randomization: Assigning subjects randomly to reduce bias.

Example: Testing the effect of light intensity on plant growth. The independent variable is light intensity, the dependent variable is plant growth, and controlled variables include soil type and water amount.

Data Collection & Sampling

Accurate data collection and appropriate sampling methods are crucial for valid results in biological studies.

• Sample: A subset of a population used to represent the whole.

• Random Sampling: Every member of the population has an equal chance of being selected.

• Systematic Sampling: Samples are taken at regular intervals.

• Bias: Systematic error that can affect the validity of results.

Example: Measuring leaf size in a forest using random sampling to avoid location bias.

Describing Data

Data can be described using measures of central tendency and variability, which help summarize and interpret biological results.

• Mean: The average value. $\text{Mean} = \frac{\sum x}{n}$

• Median: The middle value when data are ordered.

• Mode: The most frequently occurring value.

• Range: The difference between the highest and lowest values.

• Standard Deviation (SD): Measures the spread of data around the mean. $SD = \sqrt{\frac{\sum (x_i - \bar{x})^2}{n-1}}$

Example: Calculating the mean and standard deviation of enzyme activity rates in a sample.

Comparing Two Unpaired Groups

Statistical tests are used to determine if differences between two groups are significant.

• Unpaired t-test: Compares means of two independent groups.

• Mann-Whitney U test: Non-parametric test for comparing medians of two independent groups.

• Assumptions: Normal distribution, equal variances (for t-test).

Example: Comparing blood glucose levels between treated and untreated mice.

Comparing Two Paired Groups

Paired tests are used when measurements are taken from the same subjects under different conditions.

• Paired t-test: Compares means of two related groups.

• Wilcoxon signed-rank test: Non-parametric test for paired data.

Example: Measuring heart rate before and after exercise in the same individuals.

Types of Data & Choosing Statistical Tests

Different types of data require different statistical approaches. Understanding data type is essential for selecting the correct test.

Data Type	Examples	Appropriate Test
Continuous	Height, weight, enzyme activity	t-test, ANOVA
Ordinal	Rankings, scales
Nominal	Species, gender		Mann-Whitney U, Wilcoxon
Chi-square test

Additional info: Choosing the correct statistical test depends on both the type of data and the experimental design (paired vs. unpaired, number of groups, etc.).

Interpreting Results

Statistical significance indicates whether observed differences are likely due to chance.

• p-value: Probability that the observed result occurred by chance. Typically, $p < 0.05$ is considered significant.

• Confidence Interval (CI): Range within which the true value is likely to fall.

• Error Bars: Graphical representation of variability in data.

Example: Reporting a significant increase in plant growth with a p-value of 0.03 and a 95% confidence interval.

Summary Table: Statistical Tests

Test	Data Type	Paired/Unpaired	Purpose
t-test	Continuous	Unpaired/Paired	Compare means
Mann-Whitney U	Ordinal/Continuous	Unpaired	Compare medians
Wilcoxon		Ordinal/Continuous	Paired	Compare medians
Chi-square	Nominal	Unpaired	Compare frequencies