Type I and Type II errors | AQA A-Level Psychology Revision
- Revision Notes
- Aug 5
- 18 min read
Updated: 5 days ago
For 7182 specification, first teach in September 2025
AQA A-Level Psychology | Free Revision Notes
Estimated study time: 55 minutes
This Type I and Type II errors A-Level Psychology revision page explains the mistakes psychologists may make when drawing conclusions from inferential statistical tests. You will learn how a Type I error involves rejecting a true null hypothesis, while a Type II error involves retaining a false null hypothesis. You will also examine how changing the significance level affects the likelihood of each error. This lesson builds directly on probability and significance and your understanding of statistical decision-making.
Learning Objectives 🎯
By the end of this revision page, you should be able to:
Define a Type I error.
Define a Type II error.
Distinguish between Type I and Type II errors.
Apply both types of error to unfamiliar psychological research.
Explain how the chosen significance level affects the likelihood of each error.
Analyse whether a researcher may have rejected or retained the null hypothesis incorrectly.
Revision Notes 📚
Type I and Type II errors in A-Level Psychology
Inferential statistical testing allows psychologists to decide whether a result is statistically significant.
The researcher uses the result of a statistical test to make one of two decisions:
reject the null hypothesis;
retain the null hypothesis.
However, inferential testing is based on probability rather than certainty. This means that the researcher’s statistical decision may sometimes be incorrect.
The two possible statistical errors are:
Type I error;
Type II error.
These errors concern the decision made about the null hypothesis.
The null hypothesis
A null hypothesis states that there is no genuine difference, correlation or association, and that any apparent result occurred through chance.
Examples include:
There will be no difference in memory scores between participants tested in silence and participants tested with background noise.
There will be no correlation between hours of sleep and memory-test scores.
There will be no association between treatment condition and whether participants improve.
The null hypothesis is tested using an appropriate inferential statistical test.
The construction of null and alternative hypotheses is covered in aims and hypotheses.
Decisions about the null hypothesis
Following an inferential statistical test, the researcher makes one of two decisions.
Reject the null hypothesis
The null hypothesis is rejected when the result meets the selected significance criterion.
For example:
$$p\leq0.05$$
The researcher concludes that there is sufficient statistical evidence of a difference, correlation or association.
Retain the null hypothesis
The null hypothesis is retained when the result does not meet the selected significance criterion.
For example:
$$p>0.05$$
The researcher concludes that there is insufficient statistical evidence to reject the null hypothesis.
These decision processes are introduced in introduction to statistical testing.
Correct and incorrect statistical decisions
The researcher’s decision may be correct or incorrect.
Reality | Researcher rejects the null hypothesis | Researcher retains the null hypothesis |
Null hypothesis is true | Type I error | Correct decision |
Null hypothesis is false | Correct decision | Type II error |
A Type I error and a Type II error are therefore different kinds of incorrect conclusion.
Type I error
A Type I error occurs when a researcher rejects a null hypothesis that is actually true.
The researcher concludes that there is a statistically significant result when no genuine effect, difference, association or correlation exists.
A Type I error is sometimes described as a false positive.
The researcher detects a result that is not genuinely present.
Type I error as a false positive
The phrase false positive means that the statistical test appears to identify a positive finding, but this conclusion is incorrect.
The researcher concludes:
There is a genuine effect.
In reality:
There is no genuine effect.
The apparent pattern occurred through chance variation, but the researcher rejected the null hypothesis.
Type I error and the null hypothesis
A Type I error can be represented as:
$$\text{Reject a true null hypothesis}$$
The decision to reject is incorrect because the null hypothesis accurately describes reality.
Worked example: Type I error in an experiment
A psychologist investigates whether background music improves memory.
The null hypothesis is:
There will be no difference in memory scores between participants who revise with background music and participants who revise in silence.
The inferential test produces a statistically significant result:
$$p\leq0.05$$
The researcher rejects the null hypothesis and concludes that background music affects memory.
However, in reality, background music has no genuine effect on memory. The difference found in the sample occurred through chance variation.
The researcher has made a:
$$\boxed{\text{Type I error}}$$
This is because a true null hypothesis was rejected.
Worked example: Type I error in a correlation
A researcher investigates the relationship between daily social-media use and anxiety scores.
The null hypothesis is:
There will be no correlation between daily social-media use and anxiety scores.
The statistical test suggests that the correlation is significant, so the researcher rejects the null hypothesis.
In reality, there is no genuine correlation in the population.
This is a Type I error because the researcher has concluded that a relationship exists when it does not.
The interpretation of correlational results is covered in scattergrams and correlation coefficients.
Consequences of a Type I error
A Type I error may cause a researcher to:
report an effect that does not genuinely exist;
conclude that two conditions differ when they do not;
conclude that two co-variables are correlated when they are not;
conclude that two categorical variables are associated when they are not;
reject an accurate null hypothesis;
support an alternative hypothesis incorrectly.
The key idea is that the researcher has detected something that is not genuinely present.
Type II error
A Type II error occurs when a researcher retains a null hypothesis that is actually false.
The researcher concludes that there is no statistically significant result when a genuine effect, difference, association or correlation does exist.
A Type II error is sometimes described as a false negative.
The researcher fails to detect a result that is genuinely present.
Type II error as a false negative
The phrase false negative means that the statistical test appears to produce no significant finding, but this conclusion is incorrect.
The researcher concludes:
There is no genuine effect.
In reality:
A genuine effect exists.
The study failed to provide sufficient statistical evidence, so the researcher retained the null hypothesis incorrectly.
Type II error and the null hypothesis
A Type II error can be represented as:
$$\text{Retain a false null hypothesis}$$
The decision to retain is incorrect because the null hypothesis does not accurately describe reality.
Worked example: Type II error in an experiment
A psychologist investigates whether a relaxation activity reduces anxiety.
The null hypothesis is:
There will be no difference in anxiety scores before and after the relaxation activity.
The statistical result is not significant:
$$p>0.05$$
The researcher retains the null hypothesis and concludes that the relaxation activity does not affect anxiety.
However, in reality, the relaxation activity does reduce anxiety in the population. The study failed to detect this genuine effect.
The researcher has made a:
$$\boxed{\text{Type II error}}$$
This is because a false null hypothesis was retained.
Worked example: Type II error in a correlation
A psychologist investigates the relationship between sleep duration and concentration.
The statistical test produces a non-significant result, so the researcher retains the null hypothesis.
In reality, a genuine correlation exists between sleep duration and concentration.
The researcher has made a Type II error because the study failed to detect a relationship that was genuinely present.
Consequences of a Type II error
A Type II error may cause a researcher to:
overlook an effect that genuinely exists;
conclude that two conditions do not differ when they do;
conclude that no correlation exists when there is a genuine correlation;
conclude that no association exists when there is a genuine association;
retain an inaccurate null hypothesis;
fail to support an alternative hypothesis that is actually correct.
The key idea is that the researcher has failed to detect something that is genuinely present.
Comparing Type I and Type II errors
Feature | Type I error | Type II error |
Decision about the null hypothesis | Reject it | Retain it |
Reality of the null hypothesis | It is true | It is false |
Nature of the error | Detecting an effect that does not exist | Failing to detect an effect that does exist |
Alternative description | False positive | False negative |
Statistical outcome | Result appears significant | Result appears non-significant |
Basic summary | Reject a true null hypothesis | Retain a false null hypothesis |
A quick memory method
A Type I error can be remembered as:
I found an effect that is not really there.
A Type II error can be remembered as:
I missed an effect that really is there.
More formally:
$$\text{Type I}=\text{Reject a true null hypothesis}$$
$$\text{Type II}=\text{Retain a false null hypothesis}$$
Applying the errors to a research scenario
To identify the error, ask two questions.
Question 1: What decision did the researcher make?
Did the researcher:
reject the null hypothesis;
retain the null hypothesis?
Question 2: What was true in reality?
Was the null hypothesis:
actually true;
actually false?
Use the following decision table:
Researcher’s decision | Reality | Error |
Reject null | Null is true | Type I |
Retain null | Null is false | Type II |
Reject null | Null is false | No error |
Retain null | Null is true | No error |
Worked decision example 1
A researcher rejects the null hypothesis and concludes that a therapy reduces symptoms.
Later evidence indicates that the therapy has no genuine effect.
The null hypothesis was true, but the researcher rejected it.
Therefore:
$$\boxed{\text{Type I error}}$$
Worked decision example 2
A researcher retains the null hypothesis and concludes that a therapy has no effect.
Later evidence indicates that the therapy genuinely reduces symptoms.
The null hypothesis was false, but the researcher retained it.
Therefore:
$$\boxed{\text{Type II error}}$$
Worked decision example 3
A researcher rejects the null hypothesis and concludes that two conditions differ.
A genuine difference does exist in the population.
The null hypothesis was false and was correctly rejected.
Therefore:
$$\boxed{\text{No Type I or Type II error}}$$
Worked decision example 4
A researcher retains the null hypothesis.
In reality, there is no genuine effect.
The null hypothesis was true and was correctly retained.
Therefore:
$$\boxed{\text{No Type I or Type II error}}$$
Significance levels and errors
The chosen significance level affects how easily the researcher rejects the null hypothesis.
Common significance levels include:
$$p\leq0.05$$
and:
$$p\leq0.01$$
The \(0.01\) level is stricter than the \(0.05\) level because it requires stronger statistical evidence before the null hypothesis is rejected.
Changing the significance level changes the balance between the likelihood of Type I and Type II errors.
The \(0.05\) significance level
The significance level:
$$p\leq0.05$$
allows a result to be considered statistically significant when the probability associated with chance is no more than:
$$5\%$$
This is less strict than:
$$p\leq0.01$$
A result can therefore reach significance more easily at the \(0.05\) level than at the \(0.01\) level.
The \(0.01\) significance level
The significance level:
$$p\leq0.01$$
allows a result to be considered statistically significant only when the probability associated with chance is no more than:
$$1\%$$
This makes it harder to reject the null hypothesis.
A result that is significant at:
$$p\leq0.05$$
may not be significant at:
$$p\leq0.01$$
The meaning of significance levels is covered fully in probability and significance.
Lowering the significance level
Changing the significance level from:
$$p\leq0.05$$
to:
$$p\leq0.01$$
makes the criterion more stringent.
All other factors being equal, this has two main effects.
Effect on Type I errors
The likelihood of a Type I error is reduced.
The researcher requires stronger evidence before rejecting the null hypothesis, so they are less likely to reject a null hypothesis that is actually true.
Effect on Type II errors
The likelihood of a Type II error is increased.
Because it is more difficult to reject the null hypothesis, the researcher is more likely to retain it even when a genuine effect exists.
Therefore:
$$\text{Stricter significance level}\rightarrow\text{Lower Type I risk}$$
$$\text{Stricter significance level}\rightarrow\text{Higher Type II risk}$$
Raising the significance level
Changing the significance level from:
$$p\leq0.01$$
to:
$$p\leq0.05$$
makes the criterion less stringent.
All other factors being equal, this has two main effects.
Effect on Type I errors
The likelihood of a Type I error is increased.
The null hypothesis is easier to reject, so the researcher is more likely to reject it when it is actually true.
Effect on Type II errors
The likelihood of a Type II error is reduced.
A genuine effect is more likely to meet the less stringent significance criterion.
Therefore:
$$\text{Less stringent significance level}\rightarrow\text{Higher Type I risk}$$
$$\text{Less stringent significance level}\rightarrow\text{Lower Type II risk}$$
Comparing the effect of significance levels
Change in significance criterion | Effect on Type I error | Effect on Type II error |
From \(p\leq0.05\) to \(p\leq0.01\) | Less likely | More likely |
From \(p\leq0.01\) to \(p\leq0.05\) | More likely | Less likely |
There is a trade-off between the two errors.
Reducing the likelihood of one type of error can increase the likelihood of the other, assuming other aspects of the study remain unchanged.
Why a stricter level reduces Type I errors
A Type I error involves rejecting a true null hypothesis.
At:
$$p\leq0.01$$
the researcher requires stronger evidence before rejecting the null hypothesis than at:
$$p\leq0.05$$
This makes an incorrect rejection less likely.
The researcher is being more cautious before concluding that a statistically significant effect exists.
Why a stricter level increases Type II errors
A Type II error involves retaining a false null hypothesis.
At:
$$p\leq0.01$$
a genuine effect must produce stronger statistical evidence before the null hypothesis is rejected.
A real but less clearly detected effect may fail to reach this strict criterion.
The researcher may therefore retain the null hypothesis even though it is false.
Worked example: changing the significance level
A statistical test produces:
$$p=0.03$$
At the significance level:
$$p\leq0.05$$
the result is significant because:
$$0.03\leq0.05$$
The null hypothesis is rejected.
At the significance level:
$$p\leq0.01$$
the result is not significant because:
$$0.03>0.01$$
The null hypothesis is retained.
The same result therefore leads to different statistical decisions depending on the predetermined significance level.
Possible error at the \(0.05\) level
Using:
$$p\leq0.05$$
increases the likelihood of rejecting the null hypothesis compared with using:
$$p\leq0.01$$
If the null hypothesis is actually true, the decision to reject it would be a:
$$\boxed{\text{Type I error}}$$
Possible error at the \(0.01\) level
Using:
$$p\leq0.01$$
makes it harder to reject the null hypothesis.
If a genuine effect exists but the result does not meet the stricter criterion, the researcher may retain a false null hypothesis.
This would be a:
$$\boxed{\text{Type II error}}$$
Type I error and the significance level
The significance level represents the researcher’s accepted probability criterion for making a Type I error.
For example, selecting:
$$p\leq0.05$$
sets a more permissive criterion than:
$$p\leq0.01$$
A lower value of \(p\) makes a Type I error less likely because the evidence required to reject the null hypothesis is stronger.
However, no conventional significance level removes the possibility of error completely.
Inferential testing remains probabilistic.
Type II errors and sensitivity
A Type II error occurs when a genuine result is not detected.
This may be more likely when:
the significance level is very strict;
the pattern in the sample is not sufficiently clear;
there is substantial variation in the scores;
the study provides insufficient statistical evidence.
For this lesson, the main required relationship is between Type II errors and the selected significance level.
All else being equal, a stricter significance level increases the risk of failing to detect a genuine result.
Balancing the risks
Researchers must balance:
the risk of reporting an effect that does not exist;
the risk of overlooking an effect that does exist.
The most suitable significance criterion may depend on the consequences of each possible error.
However, in an AQA examination question, focus closely on the details provided rather than assuming which type of error is more serious.
The important statistical relationship is:
$$\text{Reducing Type I risk tends to increase Type II risk}$$
and:
$$\text{Reducing Type II risk tends to increase Type I risk}$$
when the significance threshold is changed and other factors remain constant.
Type I and Type II errors with observed and critical values
Statistical significance may be determined by comparing an observed value with a critical value.
A stricter significance level changes the critical value that must be reached.
This makes the threshold for rejecting the null hypothesis more demanding.
The exact comparison rule depends on the statistical test.
For some tests:
$$\text{Observed value}\geq\text{Critical value}$$
is required for significance.
For other tests:
$$\text{Observed value}\leq\text{Critical value}$$
is required.
The use of observed and critical values is covered in introduction to statistical testing.
Worked example using a statistical decision
A researcher uses a statistical test for which the observed value must be equal to or greater than the critical value.
The observed value is:
$$7.4$$
The critical value at:
$$p\leq0.05$$
is:
$$6.8$$
The comparison is:
$$7.4\geq6.8$$
The result is statistically significant, so the null hypothesis is rejected.
If the null hypothesis is actually true, the researcher has made a Type I error.
If the null hypothesis is false, the researcher has made the correct decision.
The statistical outcome alone cannot tell the researcher whether an error has occurred because the true population situation is not known with certainty.
Why errors cannot usually be identified from one study alone
A researcher can describe the possibility of a Type I or Type II error, but cannot normally know with certainty that one has occurred.
For example, after rejecting the null hypothesis, the researcher knows that a Type I error is possible.
However, they do not automatically know that the null hypothesis was actually true.
Similarly, after retaining the null hypothesis, a Type II error is possible, but the researcher does not automatically know that a genuine effect exists.
Exam questions may provide additional information, such as later evidence about the true effect, allowing you to identify the error.
Writing about a possible Type I error
Suppose a result is statistically significant and the null hypothesis is rejected.
A suitable answer is:
A Type I error may have occurred because the researcher rejected the null hypothesis. This would be an error if the null hypothesis were actually true and the apparent difference had occurred through chance.
Writing about a possible Type II error
Suppose a result is not statistically significant and the null hypothesis is retained.
A suitable answer is:
A Type II error may have occurred because the researcher retained the null hypothesis. This would be an error if a genuine difference actually existed but the study failed to detect it.
Applying errors to a directional study
A researcher predicts that participants will recall more words after a revision strategy.
The statistical result is significant, so the null hypothesis is rejected.
If the revision strategy has no genuine effect, this is a Type I error.
If the statistical result is not significant and the null hypothesis is retained, but the strategy genuinely improves recall, this is a Type II error.
The direction of the hypothesis affects the statistical test but does not change the definitions of Type I and Type II errors.
Applying errors to the sign test
A researcher uses the sign test to compare related scores before and after an intervention.
If:
$$S_{\text{observed}}\leq S_{\text{critical}}$$
the result is significant and the null hypothesis is rejected.
If the null hypothesis is actually true, this is a Type I error.
If:
$$S_{\text{observed}}>S_{\text{critical}}$$
the result is not significant and the null hypothesis is retained.
If a genuine difference actually exists, this is a Type II error.
Writing a full explanation of significance and errors
A strong explanation should connect:
The significance level.
The decision about the null hypothesis.
The type of possible error.
What is true in reality.
For example:
Using the stricter significance level of \(p\leq0.01\) reduces the likelihood of a Type I error because stronger evidence is required before a true null hypothesis is rejected. However, it increases the likelihood of a Type II error because a genuine effect may fail to meet the stricter criterion, causing a false null hypothesis to be retained.
Summary of the relationship
$$\text{Type I error}=\text{Reject a true null hypothesis}$$
$$\text{Type II error}=\text{Retain a false null hypothesis}$$
$$p\leq0.01\rightarrow\text{Lower Type I risk and higher Type II risk}$$
$$p\leq0.05\rightarrow\text{Higher Type I risk and lower Type II risk}$$
These comparisons assume that other factors remain constant.
Key Words 🔑
Key word | Student-friendly definition | How it may be used in an exam |
Type I error | Rejecting the null hypothesis when it is actually true. | You may identify a false claim that a significant effect exists. |
Type II error | Retaining the null hypothesis when it is actually false. | You may identify a failure to detect a genuine effect. |
False positive | Concluding that an effect exists when it does not. | This is another way of describing a Type I error. |
False negative | Concluding that no effect exists when a genuine effect is present. | This is another way of describing a Type II error. |
Null hypothesis | A statement that there is no genuine difference, correlation or association. | The decision to reject or retain it determines the possible statistical error. |
Alternative hypothesis | A statement predicting a difference, correlation or association. | It receives support when the null hypothesis is rejected. |
Reject the null hypothesis | Conclude that the result is statistically significant. | If the null hypothesis is true, this decision is a Type I error. |
Retain the null hypothesis | Conclude that the result is not statistically significant. | If the null hypothesis is false, this decision is a Type II error. |
Significance level | The probability threshold used to determine statistical significance. | Changing it affects the likelihood of Type I and Type II errors. |
Statistical significance | A judgement that a result meets the chosen probability criterion. | A significant result leads to rejection of the null hypothesis. |
\(p\leq0.05\) | A significance criterion allowing a probability of no more than \(5\%\) under the null hypothesis. | It is less stringent than \(p\leq0.01\). |
\(p\leq0.01\) | A significance criterion allowing a probability of no more than \(1\%\) under the null hypothesis. | It reduces Type I risk but increases Type II risk. |
Stringent criterion | A stricter threshold requiring stronger evidence before the null hypothesis is rejected. | A more stringent level reduces Type I errors. |
Chance variation | Random variation that may produce an apparent pattern in a sample. | A Type I error occurs when a chance result is treated as genuine. |
Hints from the Examiner Reports 💡
No lesson-specific examiner guidance was identified in the provided reports.
Common Mistakes ⚠️
Mistake: Defining a Type I error as retaining the null hypothesis.
Why this is incorrect:
A Type I error involves rejecting a null hypothesis that is actually true.
How to improve:
Remember:
$$\text{Type I}=\text{Reject a true null hypothesis}$$
Mistake: Defining a Type II error as rejecting the null hypothesis.
Why this is incorrect:
A Type II error involves retaining a null hypothesis that is actually false.
How to improve:
Remember:
$$\text{Type II}=\text{Retain a false null hypothesis}$$
Mistake: Saying that every significant result is a Type I error.
Why this is incorrect:
A significant result is only a Type I error if the null hypothesis is actually true.
If the null hypothesis is false, rejecting it is the correct decision.
How to improve:
Include both the researcher’s decision and the true situation in your explanation.
Mistake: Saying that every non-significant result is a Type II error.
Why this is incorrect:
A non-significant result is only a Type II error if the null hypothesis is actually false.
If the null hypothesis is true, retaining it is correct.
How to improve:
Check whether a genuine effect exists before naming the error.
Mistake: Calling a Type I error a false negative.
Why this is incorrect:
A Type I error is a false positive because the researcher incorrectly concludes that an effect exists.
How to improve:
Use:
$$\text{Type I}=\text{False positive}$$
$$\text{Type II}=\text{False negative}$$
Mistake: Saying that a stricter significance level increases Type I errors.
Why this is incorrect:
A stricter level makes the null hypothesis harder to reject, reducing the likelihood of rejecting a true null hypothesis.
How to improve:
Remember:
$$p\leq0.01\rightarrow\text{Lower Type I risk}$$
Mistake: Saying that a stricter significance level reduces both errors.
Why this is incorrect:
Making the criterion stricter reduces the risk of a Type I error but generally increases the risk of a Type II error, all other factors being equal.
How to improve:
Describe the trade-off between the two errors.
Mistake: Assuming that \(p\leq0.01\) is less strict because the number is smaller.
Why this is incorrect:
The smaller probability sets a more demanding criterion. The result must be less likely under the null hypothesis before it is considered significant.
How to improve:
Remember that \(0.01\) permits only a \(1\%\) probability criterion, compared with \(5\%\) at \(0.05\).
Mistake: Identifying an error without referring to the null hypothesis.
Why this is incorrect:
Type I and Type II errors are defined by whether a true or false null hypothesis is rejected or retained.
How to improve:
Use the full definitions rather than saying only that the researcher was wrong.
Mistake: Claiming that the researcher always knows whether an error occurred.
Why this is incorrect:
The statistical test provides a probability-based decision. The true population situation is not usually known with certainty.
How to improve:
Refer to the possibility of an error unless the question provides evidence about reality.
Mistake: Confusing a Type II error with a small effect.
Why this is incorrect:
A Type II error is a particular incorrect decision about a false null hypothesis. It is not simply a description of the size of the effect.
How to improve:
State whether the null hypothesis was retained and whether it was actually false.
Exam-Style Questions ✍️
Question 1
Define a Type I error.[2 marks]
Question 2
Define a Type II error.[2 marks]
Question 3
Explain one difference between a Type I error and a Type II error.[2 marks]
Question 4
A researcher concludes that a new revision technique significantly improves memory performance.
In reality, the revision technique has no genuine effect.
Identify the type of error made by the researcher. Explain your answer.[3 marks]
Question 5
A psychologist concludes that a relaxation programme does not significantly affect anxiety.
In reality, the programme genuinely reduces anxiety.
Identify the type of error made by the psychologist. Explain your answer.[3 marks]
Question 6
Complete the table by identifying whether the researcher has made a Type I error, a Type II error or a correct decision.
Reality | Researcher’s decision | Outcome |
The null hypothesis is true | Reject the null hypothesis | |
The null hypothesis is true | Retain the null hypothesis | |
The null hypothesis is false | Reject the null hypothesis | |
The null hypothesis is false | Retain the null hypothesis |
[4 marks]
Question 7
Explain how changing the significance level from:
$$p\leq0.05$$
to:
$$p\leq0.01$$
affects the likelihood of:
a) a Type I error;[2 marks]
b) a Type II error.[2 marks]
Question 8
A statistical test produces:
$$p=0.03$$
a) State whether the result is significant at:
$$p\leq0.05$$
[1 mark]
b) State whether the result is significant at:
$$p\leq0.01$$
[1 mark]
c) Explain how the different decisions illustrate the relationship between significance levels and Type I and Type II errors.[4 marks]
Question 9
A student writes:
“Using \(p\leq0.01\) eliminates both Type I and Type II errors because it is a more accurate significance level.”
Explain why the student is incorrect.[4 marks]
Question 10
A psychologist investigates whether sleep deprivation affects concentration.
The result is not statistically significant, so the null hypothesis is retained.
Explain:
a) the circumstances under which this decision would be correct;[2 marks]
b) the circumstances under which this decision would be a Type II error.[2 marks]
Question 11
A researcher uses a significance level of:
$$p\leq0.05$$
Another researcher recommends using:
$$p\leq0.01$$
Explain one advantage and one disadvantage of adopting the stricter significance level. Refer to Type I and Type II errors in your answer.[4 marks]
Question 12
A psychologist rejects the null hypothesis after finding a statistically significant correlation between stress and sleep quality.
Explain why a Type I error is possible. Your answer should refer to:
the researcher’s statistical decision;
the true status of the null hypothesis;
what the error would mean in the context of the study.
[4 marks]
Comments