Introduction to Biological Statistics

In Biology, we rarely get the exact same result twice. Whether you are measuring the height of pea plants or the heart rate of Daphnia, there will always be variation. Statistics are the tools we use to decide if the differences we see are "real" (significant) or just down to chance. Don't worry if you find the numbers intimidating at first; for OCR A Level Biology, the focus is on choosing the right test and interpreting what the result means for your experiment.

Measures of Central Tendency

Before we look at how data spreads out, we need to find the "center" of our data. These are the three averages you likely know from GCSE, but they have specific uses in Biology:

  • Mean: The arithmetic average. You calculate this by adding all values and dividing by the number of values (\(n\)). This is used for most further statistical tests.
    Formula to recall: \(\bar{x} = \frac{\sum x}{n}\)
  • Median: The middle value when data is placed in order. It is useful if your data has "outliers" (extreme values) that would skew the mean.
  • Mode: The most frequent value. This is used for categorical data (e.g., the most common flower color in a field).

Measures of Dispersion

Dispersion tells us how spread out our data is. Two sets of data could have the exact same mean but look very different.

The Range

The range is the difference between the largest and smallest values. While simple to calculate, it is easily affected by a single anomalous result.

Standard Deviation (\(s\))

This is a much more powerful tool. Standard Deviation measures the spread of data around the mean. Why use it? A small standard deviation tells us the data is clustered closely around the mean, suggesting our results are reliable. A large standard deviation suggests the data is very spread out.

The formula is provided in your exam, so you don't need to memorize it, but you should recognize it:
\(s = \sqrt{\frac{\sum(x - \bar{x})^2}{n - 1}}\)

Quick Review: If the standard deviation bars on a graph overlap, the difference between the means is likely not significant!

Choosing the Right Statistical Test

One of the most common exam questions asks you to "Select and justify a suitable statistical test." You generally have four main options to choose from:

1. Student's t-test (Comparing Means)

Use this when you have continuous data (things you measure, like length or mass) and you want to see if the means of two groups are significantly different.

  • Unpaired t-test: Used when comparing two completely separate groups (e.g., the height of oak trees in a wood vs. the height of oak trees in an open field).
  • Paired t-test: Used when you measure the same individuals twice (e.g., heart rate before and after drinking caffeine).

2. Spearman’s Rank Correlation Coefficient (\(r_s\))

Use this when you want to see if there is a relationship or correlation between two different variables. For example: "Does the abundance of lichen increase as we move further away from a factory?"

3. Chi-squared (\(\chi^2\)) Test

Use this when you have categorical data (counts of individuals in categories) and you want to see if your observed results match the expected results. This is very common in genetics (e.g., counting the number of purple vs. white flowers).

Interpreting "Significance"

Once you run a test, you get a value. You then compare this value to a critical value in a table to find the p-value. In Biology, we use a significance level of 0.05 (or 5%).

The Rule of Thumb:

  • If \(p < 0.05\): There is a less than 5% probability that the results are due to chance. We reject the null hypothesis. The result is significant.
  • If \(p \geq 0.05\): There is a greater than 5% probability that the results are due to chance. We accept the null hypothesis. The result is not significant.

Analogy: Imagine a coin flip. If you flip heads 5 times in a row, you might think it's chance (\(p > 0.05\)). If you flip heads 100 times in a row, the chance of that happening by accident is so tiny (\(p < 0.05\)) that you would conclude the coin is rigged (a significant result!).

Common Mistakes to Avoid

  • Confusing 'Correlation' with 'Causation': Just because Spearman's Rank shows a relationship between two things doesn't mean one causes the other.
  • Using the wrong t-test: Always check if you are looking at the same individuals (paired) or different groups (unpaired).
  • Incorrect Rounding: Always keep your significant figures consistent with the raw data provided in the question.
Key Takeaway Summary

Statistical tests help us move past "the means look a bit different" to "there is a 95% certainty that this difference is real." Always identify your data type (categorical or continuous) and your goal (comparing means, looking for a relationship, or checking frequencies) before picking your test.