In one line: Reducing bias is how a study earns the right to be trusted.
You have met the biases: from the design, from the researcher and from the participants.
Naming a bias is the first step. The second is knowing what would reduce it.
The study for this page: Does a meditation app lower stress?
The app's inventor tests it on fans of the app, who know what it is for. She measures their stress herself.
The aim is objectivity.
No study removes every bias. Good controls make the bias smaller, and an honest report says what is left.
Remember: Spot it, then fix it.
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
Key idea: A lot of bias can be removed before the first participant arrives.
Three things to settle in advance
A written plan
Everyone gets the same instructions, the same timing and the same measure (a standardised procedure).
Here: every person fills in the same stress questionnaire at the same time of day.
A trial run
The study is tried on a few people first (a pilot study). It shows up confusing instructions and questions that push people towards one answer.
Here: ten people try the questionnaire, and a leading question is rewritten.
A fair sample
Everyone in the target group has the same chance of being chosen (random sampling).
Here: names are drawn from the staff list of a whole company, not from fans of the app.
Plan · Pilot · Sample
Once the sample is chosen, chance also decides who gets the app and who does not. This is random allocation, and it makes the two groups alike at the start.
Exam tip: Sampling decides who is in the study.
Allocation decides which group each person joins.
See how examiners mark answers
Answer exam-style questions with model answers. Learn exactly what earns marks and what doesn't.
Key idea: People cannot act on what they do not know.
Single-blind
- The participants do not know which group they are in
- They cannot act to fit the aim
- Reduces participant bias
- Here: half get a fake app that plays plain sounds
Double-blind
- The person measuring does not know either
- She cannot nudge people or read results hopefully
- Reduces researcher bias as well
- Here: a helper measures stress without knowing who had which app
In the first version of the app study nothing was hidden. The fans knew the aim, and the inventor knew who had used the app. Both kinds of bias were free to work.
When you cannot blind: In a survey or an interview, people usually know the topic.
Use neutral questions, and let people answer in private and without their names, so there is less reason to try to look good.
Key idea: One person's reading of the results is checked against someone else's.
Four checks
A second researcher
Two or more researchers collect or analyse the data separately and compare (researcher triangulation).
If one sees only what she hoped for, the other will notice.
Reflexivity
The researcher asks how her own views could be shaping the study, and writes it down.
Here: the inventor states that she gains if the app works.
Replication
Another team repeats the study.
A result produced by one team's bias is unlikely to appear a second time.
Report everything
The plan is published before any data are collected (pre-registration), and the results are published whatever they show.
Failed studies can no longer disappear, so publication bias is reduced.
Second researcher · Reflexivity · Replication · Report everything
Triangulation can also mean using more than one method or more than one kind of data on the same question. When they agree, the finding is more believable.
Watch out: Trying hard to be fair is not a control.
Most bias is not deliberate, so it needs a check that does not depend on the researcher noticing it.
Study smarter, not longer
Most students waste 40% of study time on topics they already know. Our AI tracks your progress and optimizes every minute.
Key idea: Different biases need different fixes. A fix only works on the bias it was made for.
The bias
- Sampling bias
- Leading questions
- Participant bias
- Wanting to look good
- Researcher bias
- Confirmation bias
- Publication bias
A fix that fits
- Random sampling
- A pilot study, neutral wording
- Single-blind design
- Private answers without names
- Double-blind design
- A second researcher, reflexivity
- Pre-registration, publish all results
Cultural bias and gender bias have fixes of their own: studying a culture from the inside and back-translating, and balanced samples with results for each gender.
Exam tip: Do not list fixes.
Pair each one with the bias it targets, and say why it works.
How this is asked: You use ways of reducing bias in two places on Paper 2: in the 4-mark question on bias in your class practical, and in the 15-mark question on a study you are given.
Your class practicals
- Your class runs four small studies of its own, called class practicals.
- They are an experiment, an interview, an observation and a survey.
- Paper 2 asks about the ones your class ran.
- Keep a short note of each: aim, what you did, who took part, what you found, and the ethics.
The pattern for the 15-mark study
- Choose two concepts: bias, and one other.
- For each bias: name it and point to the detail in the study that lets it in.
- Say what it does to the finding, and what would reduce it.
- Show how the two concepts affect each other.
- Conclude: can the study's claim be trusted?
The trap: fixes with no bias: "She should use a double-blind design and a random sample" earns little.
Say which bias each one reduces in this study.
Discuss the following study with reference to two or more of the following concepts: bias, causality, measurement and/or responsibility.
A teacher wanted to find out whether a new revision method improves test scores. She taught it to her own tutor group of 24 students and told them she hoped it would help. After four weeks they took a test that she wrote and marked herself. She compared their scores with those of another class, which had not used the method. Her group scored higher, and she concluded that the method improves learning.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.