Introduction to Data Handling and Statistical Tests
In Biology, we don't just observe nature; we measure it. Whether you are counting the number of daisies in a field or measuring the heart rate of a daphnia, you will end up with a pile of numbers (data). Statistical tests are the tools that help us decide if our results are meaningful or just down to "luck." This chapter will guide you through the essential mathematical tools you need for your Edexcel Biology B papers, especially Paper 3.
1. Descriptive Statistics: Summarising Your Data
Before we run complex tests, we need to describe the data we have. These are often called measures of central tendency and measures of dispersion.
Mean, Median, and Mode
- Mean: The arithmetic average. Add all values and divide by the total number of values (\(n\)). It is useful for further statistical analysis but can be "skewed" by extreme values (outliers).
- Median: The middle value when data is put in order. It is less affected by outliers.
- Mode: The most frequent value in a data set.
Range and Standard Deviation
How spread out is your data? That’s what dispersion tells us.
- Range: The difference between the largest and smallest values. It’s simple but easily affected by one single weird result.
- Standard Deviation (\(s\)): This tells us how much the data varies around the mean.
- A small standard deviation means the data is clustered closely around the mean (the results are consistent).
- A large standard deviation means the data is spread widely (the results are more variable).
Quick Review: If two sets of data have overlapping error bars (representing standard deviation), it usually means there is no significant difference between them.
2. The Logic of Statistical Testing
Every time we perform a statistical test in Biology, we start with a Null Hypothesis (\(H_0\)). Don't let the name scare you! It simply states: "There is no significant difference or correlation; any difference seen is due to chance."
The goal of our test is to see if we can "reject" this null hypothesis. To do this, we look at a probability value, known as the \(p\)-value.
- In A Level Biology, the "magic number" is \(0.05\) (or \(5\%\)).
- If the probability that the result happened by chance is less than \(5\%\) (\(p < 0.05\)), we say the result is statistically significant. We reject the null hypothesis.
- If the probability is greater than \(0.05\) (\(p > 0.05\)), the result is not significant. We accept the null hypothesis.
3. Choosing the Right Statistical Test
The syllabus requires you to know three specific tests. Choosing the right one depends on what you are trying to find out.
A. The Chi-squared (\(\chi^2\)) Test
Use this when: You have categorical data (data you can put into groups) and you want to see if your observed results match the expected results.
Example: In Topic 8, you might use this to see if the offspring of a genetic cross fit a \(3:1\) ratio.
The Formula: \(\chi^2 = \sum \frac{(O - E)^2}{E}\)
Where \(O\) is the Observed value and \(E\) is the Expected value.
B. Student’s t-test
Use this when: You want to compare the means of two separate groups to see if they are significantly different from each other.
Example: Comparing the mean height of plants grown in the sun versus plants grown in the shade (Topic 10).
Key Takeaway: If your calculated \(t\) value is greater than the "critical value" found in a stats table, the difference between the means is significant.
C. Spearman Rank Correlation Coefficient
Use this when: You want to see if there is a relationship (correlation) between two variables.
Example: Is there a relationship between the light intensity in a forest and the percentage cover of moss on a tree? (Topic 10).
The Result: The value will always be between \(+1\) (perfect positive correlation) and \(-1\) (perfect negative correlation). A value of \(0\) means no correlation at all.
4. Visualising Data: Graphs and Charts
The exam will ask you to interpret or plot data. Remember these specific Edexcel conventions:
- Bar Charts: Use for discontinuous/categorical data (e.g., blood types). Leave gaps between the bars.
- Histograms: Use for continuous data (e.g., height ranges). There are no gaps between bars.
- Scatter Plots: Use to look for correlations between two variables (used for Spearman Rank).
- Line Graphs: Use when one variable (usually time) affects another. Always use a line of best fit or join points with straight lines as instructed.
Analogy: Think of a graph like a map. If the axes aren't labelled with units, the map is useless! Always include units and a title.
5. Mathematical Skills and Conventions
You are expected to handle numbers with precision. Here are the "must-know" skills:
- Significant Figures: Always give your answer to the same number of significant figures as the least precise piece of data you were given in the question.
- Percentage Change: A very common calculation.
\(\text{Percentage Change} = \frac{\text{New Value} - \text{Original Value}}{\text{Original Value}} \times 100\) - Standard Form: Be comfortable converting small numbers, like the size of a virus, into standard form (e.g., \(2.0 \times 10^{-7}\) m).
- Equations of a Line: Remember \(y = mx + c\).
- \(m\) is the gradient (rate of change).
- \(c\) is the y-intercept.
Common Mistakes to Avoid
- Confusing Correlation with Causation: Just because Spearman Rank shows a correlation (e.g., as ice cream sales rise, shark attacks rise), it doesn't mean one causes the other!
- Misidentifying the Null Hypothesis: Always start by assuming there is no effect.
- Forgetting Units: If you calculate a rate, it must have units (e.g., \(cm^3 s^{-1}\)).
- Incorrect T-test use: Don't use a t-test if you are comparing more than two means or if the data isn't normally distributed.
Key Takeaways Summary
1. Use Standard Deviation to talk about the reliability and spread of data.
2. Use Chi-squared for categories and ratios.
3. Use t-test to compare two averages (means).
4. Use Spearman Rank to look for a link between two variables.
5. A \(p < 0.05\) means the result is "real" (significant) and not just a fluke.