Lesson 3.1.4.5.2
3.1.4.5.2 Strengths and weaknesses of correlations Quiz: AQA Psychology, Unit 4
20 questions
In partnership with Revision Ninja
Lesson 3.1.4.5.2, Strengths and weaknesses of correlations: 20 multiple choice questions for the AQA GCSE Psychology (8182), Unit 4: Research methods, written with Revision Ninja.
Host it live on the board and students join with a game code on their own devices, or revise alone with Free Play. The answers are revealed in the game.
The 20 questions
-
What is a key strength of using correlations to study real-life relationships?
- They always produce results that generalise to every population
- They prove which variable caused the other to change
- They can be carried out on existing data without manipulating any variable
- They eliminate all participant variables from the study
-
Which is a weakness of correlation as a research method?
- Possible third variables may explain the link
- Correlations always require each participant to complete several conditions
- Correlations are impossible to carry out with more than ten people
- Correlations cannot be displayed on any kind of graph or diagram
-
A correlation links reading ability and vocabulary size. Why might a researcher be cautious?
- Reading ability always causes vocabulary size to increase in children
- Vocabulary size always causes reading ability to decrease
- The data must be qualitative because they are about ability
- Both could depend on a third factor such as schooling
-
Which statement is a strength of correlational studies in ethical terms?
- They can study harmful variables such as stress without deliberately causing it
- They remove the need for informed consent in every case
- They allow the researcher to inflict stress on participants safely
- They guarantee participants will never be harmed by any measure
-
Why is a correlation useful as a starting point for research?
- It replaces the need for experiments in psychology altogether
- It allows the researcher to randomly allocate participants to groups
- It can suggest relationships that experiments can later test for cause and effect
- It proves a hypothesis is correct before any further testing
-
Which limitation applies if a correlation is based on a small, unrepresentative sample?
- The findings may not generalise to the wider population
- The findings prove that participants were randomly allocated
- The findings must be discarded because they cannot be graphed
- The findings will always show a perfect correlation
-
A psychologist studies parental income and children's test results using existing school records. Which is a strength of this approach?
- Existing data can be used cheaply and quickly without manipulating anything
- Existing records remove all extraneous variables from the study
- Existing records always involve random allocation of children to classes
- Existing records prove that income causes better test results
-
Why might a correlation between screen time and attention span be misleading?
- Children with short attention spans may choose more screen time
- Screen time always causes attention spans to become shorter in children
- Attention span always causes screen time to fall to zero
- Correlations cannot measure attention span because it is qualitative
-
A study finds a correlation between caffeine intake and anxiety. What is the most important next step?
- Publish the result as proof that caffeine always causes anxiety
- Run an experiment that manipulates caffeine intake to test whether it causes anxiety
- Allocate participants to the same caffeine level to confirm the link
- Stop collecting data because the correlation has already been found
-
Which situation best justifies using a correlation rather than an experiment?
- A researcher tests whether a new drug reduces anxiety
- A researcher compares two teaching methods on test scores
- A researcher manipulates sleep to measure its effect on memory
- A researcher studies the link between childhood trauma and adult anxiety
-
Which statement best evaluates correlational research?
- It is the only method that can establish causation in psychology
- It is always more valid than a controlled laboratory experiment
- It produces results that never need to be checked by replication
- It is useful for identifying relationships
-
Why do psychologists often use correlation alongside experiments?
- Correlations show whether variables are linked
- Experiments are only used when a correlation has been found to be zero
- Correlations replace experiments whenever participants are under sixteen
- Experiments can only be done after a correlation has proved causation
-
A survey finds that height and salary are correlated. Which conclusion is strongest?
- The survey shows that high pay makes people grow taller
- The survey is invalid because height and salary are both quantitative
- The survey shows that height directly causes higher pay in every job
- The survey shows an association
-
What does correlational research NOT allow the researcher to do?
- Plot the results on a scatter diagram for inspection
- Manipulate the independent variable to observe its direct effect
- Identify whether a relationship is positive or negative
- Measure two variables for the same group of participants
-
What is the biggest practical advantage of a correlation when studying a rare condition?
- Researchers can create the rare condition in a laboratory safely
- Researchers can study existing cases without waiting to create the condition
- Researchers can guarantee a random sample of all sufferers
- Researchers can eliminate all confounding variables in one study
-
Which term describes a variable that may explain a correlation between two other variables?
- A control condition
- A demand characteristic
- A third variable
- A dependent variable
-
What would a perfect correlation look like on a scatter diagram?
- Points form a U-shaped curve on the graph
- All points lie exactly on a straight line
- Points spread randomly across the whole graph
- Points form a single cluster in the centre
-
Which limitation affects the use of correlational findings for intervention?
- They reliably show what causes the link, so the findings can be applied immediately
- They may not show the direction or cause of the link, limiting practical action
- They are measured on a fixed scale, so the strength of the link cannot change between samples
- They are only useful when the researcher manipulates the variables under controlled conditions
-
A researcher argues that correlations are always more reliable than experiments. Which response is best?
- This is wrong, since correlations show patterns but cannot establish cause
- This is correct, because correlations use real data while experiments always use artificial data
- This is correct, because correlations use larger samples than experiments always do
- This is partly right, because correlations never need to measure anything
-
Why must researchers consider the direction of a correlation's causal link?
- A correlation shows that the first variable causes the second whenever the coefficient is strong enough
- A correlation alone cannot show whether A causes B, B causes A, or a third factor affects both
- A correlation is always negative when the direction of the causal link is unclear to researchers
- A correlation shows the direction of the link only when a control group and random allocation are used
Related quizzes
- Null and alternative hypotheses Quiz · 3.1.4.1.1 · 20 questions
- Independent, dependent and extraneous variables Quiz · 3.1.4.2.1 · 20 questions
- Target populations, samples and how to select them Quiz · 3.1.4.3.1 · 20 questions
- Random, opportunity, systematic and stratified sampling Quiz · 3.1.4.3.2 · 20 questions
- Strengths and weaknesses of each sampling method Quiz · 3.1.4.3.3 · 20 questions
- Experimental designs: independent groups, repeated measures, matched pairs Quiz · 3.1.4.4.1 · 20 questions
- Laboratory, field and natural experiments Quiz · 3.1.4.4.2 · 20 questions
- Interviews and questionnaires Quiz · 3.1.4.4.3 · 20 questions
- Case studies Quiz · 3.1.4.4.4 · 20 questions
- Observation studies: categories of behaviour and interobserver reliability Quiz · 3.1.4.4.5 · 20 questions