AQA A-Level Environmental Science (7447)

## Learning Outcomes

- Understand the stages of scientific methodology in environmental science
- Distinguish between hypothesis, null hypothesis, and different types of variables
- Explain the difference between validity, reliability, accuracy, and precision
- Identify suitable sampling methods for different environmental investigations
- Understand why preliminary studies, repeats, and standardisation improve data quality
- Apply statistical ideas such as significance and choice of test to environmental data
- Avoid common exam mistakes in planning and evaluating investigations

## AQA A-Level Environmental Science (7447) Syllabus

For the AQA A-Level Environmental Science (7447), you are expected to understand scientific methodology in environmental investigations, with a focus on the following syllabus points:

- Know the main stages of scientific method, from identifying a question to reporting findings
- Identify **independent**, **dependent**, and control variables in practical and field investigations
- Explain the difference between a hypothesis and a null hypothesis
- Understand how validity, reliability, accuracy, precision, bias, and representativeness affect data quality
- Compare random, systematic, and stratified sampling, including when each is appropriate
- Understand how sample location, sample size, number of samples, and timing affect representative data
- Explain why preliminary or pilot studies improve investigations
- Understand how data can be analysed to judge whether results are statistically significant rather than due to chance

## Test Your Knowledge

_Attempt these questions before reading this article. If you find some difficult or cannot remember the answers, look more closely at that area during your revision._

1. What is the difference between a hypothesis and a null hypothesis?
2. How does a control variable differ from an independent variable?
3. Why is a reliable method not always a valid one?
4. When would systematic sampling be better than random sampling?
5. What does it mean if a result is statistically significant?

## Introduction

Scientific methodology is the logic behind good environmental science. It is not just about collecting data. It is about asking a clear question, choosing a suitable method, collecting representative evidence, and deciding whether your conclusion is strong enough to trust.

This matters in Environmental Science because many real-world questions are difficult. You cannot measure every tree in a forest, every fish in a sea, or every change in climate at every point on Earth. You must therefore use samples, controls, repeats, and careful judgement. In the exam, you may be asked to plan an investigation, improve a method, identify variables, or explain whether conclusions are justified.

> **Key Term: scientific methodology**
> The structured approach used to investigate a scientific question, including forming hypotheses, collecting data, analysing results, and evaluating conclusions.

### Stages of scientific method

A standard environmental investigation usually follows a clear sequence.

1. Identify a topic or question.
2. Find out what is already known.
3. Form a testable hypothesis.
4. Design a method to test it.
5. Collect data.
6. Analyse the data.
7. Draw conclusions.
8. Report findings and refine the next question.

This sequence applies whether the study is small-scale fieldwork on a school site or large-scale Antarctic research by the **British Antarctic Survey (BAS)**. For example, Antarctic researchers investigating pollution risks around research stations still follow the same logic: define the problem, collect evidence, assess impact, and report conclusions. Careful management at research stations, such as secondary containment for oil tanks and waste removal, also shows how scientific work depends on controlled methods and monitoring.

> **Key Term: hypothesis**
> A testable prediction about the relationship between variables.
>
> **Key Term: null hypothesis**
> A statement that there is no significant difference, relationship, or effect between the variables being investigated.

A strong hypothesis is specific and measurable. In environmental science, it is often written as:

“If the **independent variable** changes, then the **dependent variable** will change.”

For example:

- If distance from a footpath increases, then plant cover will increase.
- If nitrate concentration rises, then algal growth will increase.

The null hypothesis must match the same variables:

- There is no significant relationship between distance from the footpath and plant cover.
- There is no significant relationship between nitrate concentration and algal growth.
  > **Test Tip:** If a question asks for a null hypothesis, do not just write “there is no effect”. Name both variables clearly and state that any pattern is not significant or is due to chance.

### Variables and fair testing

A good investigation depends on identifying the right variables.

> **Key Term: independent variable**
> The factor changed by the investigator or the factor whose effect is being studied.
>
> **Key Term: dependent variable**
> The factor that is measured or observed as the outcome.
>
> **Key Term: control variables**
> Factors kept constant, or monitored carefully, so they do not unfairly affect the results.

Example: investigating the effect of soil pH on seed germination

- Independent variable: soil pH
- Dependent variable: percentage germination
- Control variables: temperature, light, water volume, seed species, number of seeds, time allowed

In fieldwork, full control is harder. If you investigate the effect of tree cover on temperature, you may not be able to control wind speed or cloud cover. In that case, measure them if possible and comment on their influence.

This is why field investigations often have more uncertainty than lab investigations. That does not make them poor science. It means you must evaluate limitations honestly.

A fair test also depends on standardisation. If quadrat size changes between samples, or one observer estimates percentage cover differently from another, the method is less trustworthy.

### Quality of evidence: validity, reliability, accuracy and precision

These four terms are often confused, but they are not the same.

> **Key Term: validity**
> The extent to which a method measures what it is meant to measure or answers the question appropriately.
>
> **Key Term: reliability**
> The extent to which a method gives consistent results when repeated under the same conditions.
>
> **Key Term: accuracy**
> How close a measurement is to the true or accepted value.
>
> **Key Term: precision**
> How close repeated measurements are to each other.

A method can be reliable but not valid. For example, a badly calibrated pH meter might give very similar readings every time, so it is reliable, but the readings may all be wrong, so it is not accurate and may reduce validity.

A method can also be precise but inaccurate. If repeated temperature readings are all 2°C too high, they are close together but still wrong.

Environmental monitoring often requires you to think about all four:

- **Validity**: does a Secchi disc actually answer the question about water clarity well enough?
- **Reliability**: would repeated observers get similar results?
- **Accuracy**: was the instrument calibrated?
- **Precision**: does the instrument measure finely enough?

For example, a Secchi disc is quick and cheap for turbidity, but it is subjective because it depends on light intensity and the observer’s judgement. An electronic instrument may improve precision and objectivity.

> **Test Tip:** In evaluation questions, do not use “reliable” as a general word for “good”. Say exactly what improves validity, reliability, accuracy, or precision.

### Sampling, representative data and preliminary studies

Environmental scientists usually cannot measure everything, so they take samples. The key question is whether those samples are **representative** of the whole area, habitat, or time period.

> **Key Term: representative sample**
> A sample that reflects the real characteristics of the whole population or habitat being studied.

There are three common sampling methods.

> **Key Term: random sampling**
> Sampling in which every location or individual has an equal chance of selection, reducing bias.
>
> **Key Term: systematic sampling**
> Sampling at regular intervals or in a fixed pattern.
>
> **Key Term: stratified sampling**
> Sampling from sub-groups or sub-areas in proportion to their importance or size.

Use them carefully:

- **Random sampling** is best when the habitat is fairly even and you want to reduce investigator bias.
- **Systematic sampling** is useful along gradients, such as distance from a road, path, river, or pollution source.
- **Stratified sampling** is best when the area has clear sub-areas, such as woodland, grassland, and bare soil, and each must be represented fairly.

Transects are especially important in environmental science.

> **Key Term: transect**
> A line across a habitat or environmental gradient along which samples are taken to study change.

A line transect is suited to point measurements such as light intensity or soil pH. A belt transect uses quadrats along a strip and is suited to species data such as plant cover or frequency.

Quadrats are commonly used for plants and slow-moving organisms.

> **Key Term: quadrat**
> A frame of known area used to sample organisms in a standardised area.

Sample design also includes:

- sample location
- sample size
- number of samples
- timing of sampling

A single reading may be anomalous. More repeats improve reliability and help identify outliers. Timing matters too. Soil moisture, river flow, or air temperature can vary through the day and through seasons. If sampling times differ, comparisons may be unfair.

Preliminary studies are very useful. A pilot study can help you:

- choose a suitable sampling interval
- decide how many repeats are needed
- check whether equipment works
- estimate how variable the habitat is
- avoid collecting unnecessary data

This is especially useful in systematic sampling. If sample points are too far apart, you may miss a pattern. If they are too close, you waste time without improving the answer much.

### Data analysis, significance and judgement

Once data are collected, you need to decide what they show. A pattern in raw data does not automatically mean a real effect exists. Some differences happen by chance.

> **Key Term: statistical significance**
> A result that is unlikely to have occurred by chance alone.

In school science, significance is commonly judged at **p = 0.05**.

- If the result is significant, reject the null hypothesis.
- If it is not significant, fail to reject the null hypothesis.

Do not say “prove”. Science usually supports or does not support an explanation.

Common test choices include:

- **Spearman’s rank** for correlation between two ranked or paired variables
- **Chi-squared** for frequency data in categories
- **Student’s t-test** for difference between two means
- **Standard deviation** for spread of data

Examples:

- Distance from path and plant cover → Spearman’s rank
- Numbers of organisms in categories → Chi-squared
- Mean dissolved oxygen in two rivers → t-test

A statistically significant result is not always environmentally important. A tiny but significant temperature change may matter little biologically. On the other hand, a moderate change in dissolved oxygen may have major ecological effects. Always link the statistics back to the environmental context.

You should also think about bias and uncertainty. Ask:

- Who collected the data?
- How was it measured?
- Where and when was it measured?
- What is the likely margin of error?

This matters in large-scale statements too. For example, climate predictions, trends in Arctic ice, or claims about rainforest loss all need questions about methods, sample basis, and uncertainty.

## Key Point Checklist

_This article has covered the following key knowledge points:_

- Scientific methodology follows a sequence from question to conclusion and reporting
- A hypothesis predicts a relationship, while a null hypothesis states there is no significant relationship or effect
- The independent variable is changed or studied, while the dependent variable is measured
- Control variables must be kept constant or monitored to make the test fair
- Validity is about measuring the right thing, while reliability is about getting consistent results
- Accuracy means closeness to the true value, while precision means closeness of repeated values to each other
- Environmental investigations need representative data because whole populations or habitats cannot usually be measured
- Random, systematic, and stratified sampling each have different uses
- Transects are useful for environmental gradients, and quadrats are standard tools for plant sampling
- Preliminary studies help refine sample size, sampling interval, and method choice
- Repeats improve reliability and help detect anomalies
- Statistical significance shows whether a result is unlikely to be due to chance alone

## Key Terms and Concepts

- scientific methodology
- hypothesis
- null hypothesis
- independent variable
- dependent variable
- control variables
- validity
- reliability
- accuracy
- precision
- representative sample
- random sampling
- systematic sampling
- stratified sampling
- transect
- quadrat
- statistical significance


Research methods - Scientific methodology