# Variation: t-test Method ## What is the t-test? - The **t-test** is a statistical method used to **compare the means of two data sets** - It determines whether any difference between the two means is **statistically significant** or simply due to chance - The t-test is particularly useful when comparing biological measurements between two groups

The t-test compares two sample means to determine if they are significantly different from each other

## When to use a t-test - The t-test should be used when data meets specific criteria: - The data follows a **rough normal distribution** - The measurements are **continuous** (not categorical) - The **standard deviations** of the two data sets are approximately equal - Examples of suitable data include heights, masses, leaf lengths, or enzyme activity rates ## Setting up the test ### The null hypothesis - Before conducting a t-test, you must state a **null hypothesis** - The null hypothesis assumes there is **no significant difference** between the two means - Any observed differences are assumed to be due to **chance alone**

A typical null hypothesis: "There is no statistically significant difference between the means of sample 1 and sample 2"

### Calculating standard deviation first - The standard deviation ($s$) must be calculated for **each data set** before performing the t-test - Standard deviation measures the **spread of data** around the mean value **Formula for standard deviation:** $$s = \sqrt{\frac{\sum(x - \bar{x})^2}{n - 1}}$$ Where: - $x$ = individual observation - $\bar{x}$ = mean of the sample - $n$ = sample size (number of observations) ## Steps for conducting a t-test ### Step 1: Calculate the mean for each data set $$\bar{x}_1 = \frac{\text{sum of all values in sample 1}}{\text{number of values in sample 1}}$$ $$\bar{x}_2 = \frac{\text{sum of all values in sample 2}}{\text{number of values in sample 2}}$$ ### Step 2: Calculate the standard deviation for each data set - Use the standard deviation formula shown above - Calculate $s_1$ for sample 1 and $s_2$ for sample 2 - This requires finding the difference between each value and the mean, squaring these differences, summing them, dividing by $(n-1)$, and taking the square root ### Step 3: Square the standard deviation and divide by $n$ for each sample $$\frac{s_1^2}{n_1} \quad \text{and} \quad \frac{s_2^2}{n_2}$$ ### Step 4: Add the values from Step 3 and take the square root $$\sqrt{\left(\frac{s_1^2}{n_1} + \frac{s_2^2}{n_2}\right)}$$ ### Step 5: Calculate the t value - Divide the **difference between the two means** by the value calculated in Step 4 $$t = \frac{|\bar{x}_1 - \bar{x}_2|}{\sqrt{\left(\frac{s_1^2}{n_1} + \frac{s_2^2}{n_2}\right)}}$$

The t-test formula will be provided in the exam, but you need to know what each symbol represents

### Step 6: Calculate the degrees of freedom - The degrees of freedom ($v$) for the entire data set is calculated as: $$v = (n_1 - 1) + (n_2 - 1)$$ Where $n_1$ and $n_2$ are the sample sizes

The formula for degrees of freedom is NOT provided in the exam and must be learned

### Step 7: Compare the t value to critical values - Look up your calculated t value in a **critical values table** - Use the row corresponding to your degrees of freedom ($v$) - Compare your t value to the critical values at different probability levels **Example critical values table:** | **Degrees of freedom** | **Value of t** | | | | |---|---|---|---|---| | | **0.10** | **0.05** | **0.01** | **0.001** | | 5 | 2.02 | 2.57 | 4.03 | 6.87 | | 10 | 1.81 | 2.23 | 3.17 | 4.59 | | 15 | 1.75 | 2.13 | 2.95 | 4.07 | | 20 | 1.72 | 2.09 | 2.85 | 3.85 | | 28 | 1.70 | 2.05 | 2.76 | 3.67 | | **Probability that chance could have produced this value of t** | | | | | | **Confidence level** | **90%** | **95%** | **99%** | **99.9%** | ### Step 8: Draw a conclusion - The **greater** the calculated t value, the **lower** the probability that chance caused the difference - In biology, we typically use a **significance level of 0.05** (5% probability, 95% confidence) **If the t value is GREATER than the critical value at p = 0.05:** - There is a **statistically significant difference** between the means - The probability that chance caused the difference is **less than 5%** - The **null hypothesis is rejected** **If the t value is LESS than the critical value at p = 0.05:** - There is **no statistically significant difference** between the means - The probability that chance caused the difference is **greater than 5%** - The **null hypothesis is accepted** ## Writing a conclusion A complete conclusion should include: - Reference to the **calculated t value** - Reference to the **critical value** at the appropriate degrees of freedom and probability level (usually 0.05) - A statement about whether there **is or is not** a significant difference between the means - A statement about whether the **null hypothesis is accepted or rejected** **Example conclusion:** "The calculated t value of 2.38 is less than the critical value of 2.57 at 8 degrees of freedom and a probability level of 0.05. Therefore, there is no statistically significant difference between the means of the two samples, and we accept the null hypothesis."

Always state that the difference between data sets is significant, not just that "the data is significant"

## What you need to know for exams For exam questions on the t-test, you should be able to: 1. **Explain why** a t-test is being used to analyze data 2. **State the null hypothesis** for a given scenario 3. **Calculate the degrees of freedom** using the formula $v = (n_1 - 1) + (n_2 - 1)$ 4. **Interpret** a calculated t value using a critical values table 5. **Draw a conclusion** about whether differences are significant and whether to accept or reject the null hypothesis

You will be given the formula for calculating the t value in the exam, but NOT the formula for degrees of freedom

## Comparing t-tests to other statistical methods | **Feature** | **t-test** | |---|---| | **Purpose** | Compare means of two data sets | | **Data type** | Continuous data | | **Distribution** | Approximately normal | | **Standard deviations** | Should be roughly equal | | **Number of samples** | Two | | **Output** | t value, which is compared to critical values | The t-test is specifically designed for comparing whether two sample means differ significantly, making it ideal for experiments comparing a control group to an experimental group, or comparing measurements from two different populations.