Controlling extraneous variables (AQA GCSE Statistics): Revision Notes
Control groups and matched pair tests
When conducting scientific research, it's essential to control extraneous variables that might affect your results. Two powerful methods that researchers use are control groups and matched pair tests. These techniques help ensure that any differences observed are truly due to the factor being investigated, rather than other unwanted influences.
What are control groups?
A control group serves as a comparison standard when testing whether a treatment or intervention actually works. Here's how the process works:
The setup process:
- Researchers randomly divide participants into two groups
- The test group receives the actual treatment being investigated
- The control group receives either no treatment at all, or a placebo (fake treatment that looks identical but contains no active ingredients)
- Importantly, participants don't know which group they're in, so their expectations won't influence the results
The key principle behind control groups is creating a fair comparison. Without knowing which treatment they're receiving, participants can't unconsciously influence the results through their expectations or behaviour changes.
Why this matters: By comparing outcomes between these two groups, researchers can determine if any improvements are genuinely caused by the treatment itself, or if they would have happened anyway. Without a control group, you can't be sure whether your treatment is actually effective.
Critical Point: Random selection for control groups is essential. This ensures that both groups have similar characteristics on average, making any differences in outcomes more likely to be due to the treatment rather than pre-existing differences between the groups.
Understanding matched pair tests
Matched pair tests take a different approach to controlling unwanted variables. Instead of relying purely on random selection, researchers deliberately pair up participants who are very similar in most ways.
How matching works:
- Each person in the treatment group is carefully paired with someone in the control group
- The paired individuals share similar characteristics such as age, gender, health status, or other relevant factors
- The only significant difference between the paired individuals should be whether they receive the treatment or not
- This helps ensure that any differences in outcomes are due to the treatment rather than personal characteristics
Think of matched pair tests like studying twins - you want to compare people who are as similar as possible in every way except for the one factor you're investigating.
Worked example: Testing arthritis medication
Worked Example: Arthritis Drug Trial Using Matched Pairs
The study setup: Researchers want to test a new tablet for treating arthritis. They measure arthritis severity on a scale from 1 to 100, where higher numbers indicate more severe symptoms.
Creating matched pairs: Participants could be paired based on key characteristics that might affect arthritis, such as:
- Age (since arthritis often worsens with age)
- Gender (as arthritis affects men and women differently)
- Current severity level (to ensure fair comparison)
Example pairs:
- Pair A: Both males, both under 30 years old, both with severity levels below 15
- Pair B: Both females, both aged 30-40, both with severity levels between 15-25
One person from each pair receives the medication, whilst their matched partner receives an identical-looking tablet with no active medication.
Advantages and disadvantages
Advantages of matched pair tests:
- Excellent control over extraneous variables like age, gender, and initial condition severity
- More precise comparisons since paired individuals are so similar
- Smaller sample sizes needed compared to purely random studies
Disadvantages of matched pair tests:
- Very time-consuming to find enough suitable matched pairs
- Can be expensive due to the extensive screening required
- Sometimes impossible to find perfect matches for all participants
- May not account for unknown factors that could still influence results
Key principles to remember
Both control groups and matched pair tests aim to isolate the effect of the treatment by minimising the influence of extraneous variables. The choice between methods often depends on practical considerations like time, cost, and the nature of what's being studied.
Key Points to Remember:
- Control groups provide a baseline comparison by giving some participants no treatment or a placebo
- Matched pair tests control extraneous variables by pairing similar individuals who differ only in treatment received
- Both methods help ensure that observed differences are truly caused by the treatment being tested
- Random selection is crucial for control groups, whilst careful matching is essential for paired studies
- Neither method is perfect - each has advantages and limitations that researchers must consider