Introduction to Statistical Tests

In Biology, data is often "messy." If you count the number of bluebells in two different woods, the numbers will never be exactly the same. But how do you know if the difference is because the woods are actually different, or just because of random chance? That is where statistical tests come in. They give us a mathematical way to decide whether to "trust" our results.

In this chapter, we will look at the three key tests required for your OxfordAQA International A-level: the Chi-squared test, Standard Error, and Spearman Rank.


1. The Chi-Squared (\(\chi^2\)) Test

When do we use it?

We use the Chi-squared test when we have categorical data (data that falls into distinct groups) and we want to compare observed frequencies with expected frequencies. A classic example is in genetics (Unit 3.3.6), such as counting the number of tall vs. short plants resulting from a cross.

The Null Hypothesis (\(H_0\))

Before any test, we start with a Null Hypothesis. For Chi-squared, it usually states: "There is no significant difference between the observed and expected frequencies; any difference is due to chance."

The Formula

\( \chi^2 = \sum \frac{(O - E)^2}{E} \)

  • \(O\) = Observed frequency (what you actually counted)
  • \(E\) = Expected frequency (what you predicted)
  • \(\sum\) = The sum of (add them all up)

Interpreting the Result

Once you calculate your \(\chi^2\) value, you compare it to a critical value from a statistical table using degrees of freedom (\(df = \text{number of categories} - 1\)).

  • If your calculated \(\chi^2\) is greater than the critical value at \(P = 0.05\), there is a less than 5% probability that the difference is due to chance. You reject the null hypothesis. The difference is significant.
  • If your value is smaller than the critical value, you accept the null hypothesis. The difference is not significant and is due to chance.

Quick Tip: If the probability (\(P\)) is less than 0.05, it means the result is "important" (significant)!


2. Standard Error and 95% Confidence Limits

What is Standard Error (SE)?

While Standard Deviation (SD) tells us how much the data spreads out around the mean of a single sample, the Standard Error tells us how close our sample mean is likely to be to the true mean of the whole population.

95% Confidence Limits

In Biology, we use 95% Confidence Limits (CL) to show the range in which we are 95% certain the true population mean lies. We represent this on graphs using error bars.

The formula for the 95% Confidence Interval is: \( \text{Mean} \pm (2 \times SE) \)

How to interpret Error Bars on a graph:

  • If the error bars overlap: The difference between the means is not significant. Any difference is likely due to chance.
  • If the error bars do NOT overlap: The difference between the means is significant. There is a less than 5% probability that the difference is due to chance.

Analogy: Imagine trying to guess a friend's height. If you say "Between 170cm and 172cm," you have a narrow range (low SE). If you say "Between 150cm and 190cm," you have a wide range (high SE). The narrower the range, the more precise your data!


3. Spearman Rank Correlation Test

When do we use it?

We use Spearman Rank when we want to see if there is a correlation (a relationship) between two sets of continuous data. For example, "Does the light intensity affect the number of species in a habitat?"

The Correlation Coefficient (\(r_s\))

The result of this test is a number between \(+1\) and \(-1\).

  • \(+1\) = A perfect positive correlation (as one goes up, the other goes up).
  • \(-1\) = A perfect negative correlation (as one goes up, the other goes down).
  • \(0\) = No correlation at all.

The Null Hypothesis for Spearman Rank

\(H_0\): "There is no significant correlation between the two variables."

Testing for Significance

Just like Chi-squared, you compare your calculated \(r_s\) value to a critical value table. If your value is greater than the critical value at \(P = 0.05\), the correlation is significant and not due to chance.


Summary Table: Which test should I use?

Choosing the right test is half the battle in the exam!

Test Type of Data What it looks for
Chi-squared Categories / Frequencies A difference between Observed and Expected.
Standard Error Means of continuous data How precise the mean is and if means are significantly different.
Spearman Rank Two continuous variables A correlation or relationship between variables.

Quick Review: Common Mistakes to Avoid

  • Confusing "Significant" with "Large": A difference can be very small but still be "statistically significant" if the data is very consistent.
  • The 0.05 Rule: Always remember that \(P < 0.05\) means we reject the null hypothesis. It means there is less than a 5% probability that the results happened by accident.
  • Sample Size: Remember from Unit 3.1.6 that a larger sample size makes your statistical tests more reliable and reduces the effect of anomalies.

Did you know? We use \(P = 0.05\) (5%) as a standard in Biology because it strikes a balance. It is strict enough to be reliable, but not so strict that we ignore real biological trends!


For more information on collecting the data for these tests, see the chapters on "Random sampling" and "Required practicals".