AQA A Level Psychology

STATISTICS & MATHEMATICS

WorkbooK
Example Answers
contents
Title
1 Measures of central tendency and dispersion 3

2 Calculating percentages 6

3 Normal and skewed distributions 10

4 Use of mathematical symbols 13

5 Decimals, fractions and ratios 14

6 Probability, significance and type I/type II errors 17

7 Interpreting data on graphs and tables 20

8 Plotting appropriate graphs 23

9 Choosing the correct test 26

10 Interpreting the results of statistical tests using statistical tables and critical values 30

11 Calculating and interpreting the sign test 34

Scenario 1: Are males more likely to pass their driving test first time than
12 37
females?
Scenario 2: Is there a correlation between the number of hours worked in a week
13 40
and how stressed someone feels?
Scenario 3: Is there a difference in spatial ability between physics and
14 42
psychology students?
This booklet consists of a series of activities and scenarios with questions to help you to prepare
for your Psychology A level examinations. At least 10% of the total marks will come from
questions that are testing your mathematical skills, and some of these questions could require
you to do calculations. Therefore, the more that you practise answering these types of question
the better!

Exam Hints: This booklet also includes exam hints and tips provided by our team and
from the examiner reports. It is important to pay careful attention to these hints and tips
as they will really help you to maximise your exam performance.

Answers: You can download example answers for all the questions in this Statistics &
Mathematics Workbook from your my tutor2u account. We would advise that you attempt
to answer each question before looking at the model answers!

2 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

Activity 1
Measures of central tendency & dispersion
Exam Hints: Remember that the measures of central tendency are the mean, median
and mode, while the measures of dispersion are the range and standard deviation. You
only need to know how to calculate the mean, median and mode, and the range. In the
exam it is important to show your workings when asked to calculate something.

Briefly explain how to calculate each of the following for a set of scores:

Measure of central
How to calculate
tendency / dispersion

Add all the scores together and divide by the total number of
Mean scores.

Put the scores in order; the median is the middle score. If

Median there is an even number of scores, then add together the two
middle scores and then divide by two.

Mode Find the score or scores that appear most frequently.

Range Subtract the lowest score from the highest score.

Calculate the mean, median, mode and range for the following sets of data.

Data Mean Median Mode Range

1,3,3,4,5,8,9,9,10
1, 4, 9, 3, 10, 52/9 = 10 - 1
A Middle score = 5 3&9
5, 8, 9, 3 5.77 =9

19,20,21,22,26,27,28,52
22, 26, 20, 52, 215/8 22+ 26 = 48/2 = 24 52 – 19
B None
28, 19, 27, 21 = 26.88 Middle score = 243 = 33

27,28, 29, 33, 36, 39,

41, 29, 39, 44, 443/12 39,40,41,42,44,45 45-27
C 36, 40, 27, 42, 39
= 36.92 Middle score = 39 = 18
45, 28, 39, 33

www.tutor2u.net 3
You may be asked which is the most appropriate measure of central tendency and to justify your
choice. You could also be asked to explain why a particular measure of central tendency is not
appropriate. For example:

QUESTION 1
Which is the most appropriate measure of central tendency to use for data set B and why? [2 marks]

Answer: The median would be most appropriate as it is no affected by extreme scores

(e.g. the score of 52) as would happen if the mean was used. The mode is not
appropriate for this data as each score only appears once.

QUESTION 2
Identify the most appropriate measure of central tendency to use for set C and explain why it would be the most
appropriate? [3 marks]

Answer: The mean would be the most appropriate as there are no extreme scores and it
is the most sensitive and informative because it uses information from all of the
scores. The median only uses information from the middle two scores and the mode
is only the score that appears most frequently.

QUESTION 3
Making reference to the data shown above, explain why the mode is often not useful as a measure of central
tendency. [2 marks]

Answer: The mode is often not useful because there may not be one (as in set B) or
there may be more than one mode (as in set A). Or it may fall towards the bottom or
the top of the data set (as in set C).

4 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 4
Making reference to the data shown above, explain one limitation of using the range as a measure of dispersion.
[3 marks]

Answer: One limitation of using the range as a measure of dispersion is that it does
not take account of all of the scores, only the lowest and the highest. In this case it
only takes into account the score of 19 and the score of 52 and does not show that
the rest of the scores are all in the 20s.

Although you do not need to be able to calculate the standard deviation, you need to
understand what it shows us about dispersion in a data set and how to interpret what differences
in standard deviations mean. Try answering these questions:

QUESTION 5
Explain what is meant by the standard deviation. [2 marks]

Answer: It is a measure of dispersion, that tells us how the scores are spread around
the mean. A large standard deviation means the scores are widely spread around
the mean, while a small standard deviation means the scores are tightly clustered
around the mean.

QUESTION 6
Explain one limitation of using the standard deviation as a measure of dispersion. [2 marks]

Answer: One limitation is that because the calculation of the SD is based on the mean,
it can be distorted by a single extreme value. This means it will not give a true
indication of how the scores are dispersed.

See activity 7 (page 20-22) for questions regarding interpreting differences in measures of central
tendency and dispersion, including standard deviations.

www.tutor2u.net 5
activity 2
Calculating percentages
Exam Hints: When asked to calculate something in the exam, like a percentage, you
should always show your workings out. That way if you are using the correct formula,
but make an error with the final number, you will still get some credit.

Here is some data from a study looking to see whether there were any differences in attachment
style between infants aged 12-18 months, where the mother was the primary caregiver and
where the father was the main caregiver. Each infant was assessed with their primary caregiver
using the Strange Situation.

Securely attached Insecure avoidant Insecure resistant

Mother primary caregiver 42 9 7
Father primary caregiver 22 4 5

Answers for questions 1 to 6 below should be calculated to 3 significant figures.

Exam Hints: Do not confuse 3 significant figures (counting from the first non-zero digit
for three digits, and then rounding the last digit) with 3 decimal places (three digits after
the decimal place).

QUESTION 1
What percentage of the sample had their father as the primary caregiver? [2 marks]

Answer:
42 + 9 + 7 = 58 mothers
22 + 4 + 5 = 31 fathers
Total = 58 + 31 = 89
31/89 x 100 = 34.8%

6 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 2
What percentage of infants with a mother as primary caregiver were securely attached? [2 marks]

Answer: 42 + 9 + 7 = 58 mothers as primary caregiver

42/58 x 100 = 72.4%

QUESTION 3
What percentage of infants with a father as primary caregiver were assessed as insecure resistant? [2 marks]

Answer: 22 + 4 + 5 = 31 fathers as primary caregivers

5/31 x 100 = 16.1%

QUESTION 4
What percentage of infants overall were assessed as being insecure avoidant? [2 marks]

Answer: 42 + 9 + 7 = 58 mothers
22 + 4 + 5 = 31 fathers
Total = 58 + 31 = 89
Total insecure avoidant = 9 + 4 = 13

www.tutor2u.net 7
You may also be asked to calculate a percentage increase or a percentage decrease. Read
through this study and then answer the questions that follow. Again, it is important to show your
calculations.

Exam Hints: Here is a reminder of how to calculate a percentage increase and a

percentage decrease:

Percentage increase Percentage decrease

1. First work out the difference (increase) 1. First work out the difference (decrease)
between the two numbers you are comparing. between the two numbers you are comparing.

2. Increase = new (larger) number – 2. Decrease = original (larger) number –

original (smaller) number new (smaller) number

3. Then divide the increase by the original 3. Then divide the decrease by the original
number and multiply the answer by 100. number and multiply the answer by 100.

For example, to find the percentage increase For example, to find the percentage decrease
from a score of 4 to a score of 9 from a score of 17 to a score of 14
1. Increase = 9 – 4 = 5 1. Decrease = 17-14 = 3
2. 5/4 = 1.25 2. 3/17 = 0.17647
3. 1.25 x 100 = 125% 3. 0.17647 x 100 = 17.65%

8 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

A study was carried out to look at the effects of listening to music on the time taken to complete
5 simple anagrams. One group completed the anagrams while listening to classical music,
a second group completed the anagrams while listening to rock music. The control group
completed the anagrams in silence.

The mean times for each group were as follows:

Classical music Rock Music Control group

Mean (in seconds) 68 82 63

QUESTION 5
Calculate the percentage increase in the mean time it took participants to solve the 5 anagrams when listening
to classical music, compared to silence (control group). Show your calculations. Give your answer to three
significant figures. [4 marks]

Answer: Increase = new number (Classical time) – original number (silence time)
68 – 63 = 5
Divide the increase by the original number and multiply by 100
5/63 x 100 = 7.94%

QUESTION 6
Calculate the percentage decrease in the mean time it took participants to solve the 5 anagrams when listening
to rock music, compared to classical music. Show your calculations. Give your answer to three significant
figures. [4 marks]

Answer: Decrease = Original number (rock music) – new number (classical music)
82 – 68 = 14
Divide the decrease by the original number and multiply by 100
14/82 x 100 = 17.1

www.tutor2u.net 9
Activity 3
Normal and skewed distributions
Exam Hints: Do not confuse distributions (normal and skewed) with measures of
dispersion (range and standard deviation). The specification requires you to know about
characteristics of normal and skewed distributions.

Here are the average scores from a study looking at performance on a memory test in a cold
room, a hot room and a room at normal temperature. The three different groups of participants
were shown pictures of 20 objects for one second at a time. They were then given a distractor
task before being asked to recall as many as they could in 2 minutes.

Measure of central Group C - Normal

Group A - Hot room Group B - Cold room
tendency room temperature

Mean 13.25 10.75 12.11

Median 12.5 12 12.5

Mode 10 13 12

Draw a distribution curve to represent each of these sets of data. Label the mean, median and mode.

Graph A

Mode

Median

Mean

0 1 2 3 4 5 6 7 8 9 1 0

10 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

 Graph B
Mode

Median

Mean

0 1 2 3 4 5 6 7 8 9 1 0

Graph C Mean
Median
Mode

0 1 2 3 4 5 6 7 8 9 1 0

QUESTION 1
Using the data in the table, explain how the distribution of scores in Group A differs from the distribution of
scores in Group C. [4 marks]

Answer: Group A’s scores suggest a positively skewed distribution as the mean
(13.25) is higher than both the median (12.5) and the mode (10). However, Group
C’s scores suggest a normal distribution as the mean (12.1), median (12.5) and
mode (12) are almost the same.

www.tutor2u.net 11
QUESTION 2
Using the data in the table, explain how the distribution of scores in Group B differs from the distribution of
scores in Group C. [4 marks]

Answer: Group B’s scores suggest a negatively skewed distribution as the mean
(10.75) is lower than both the median (12) and the mode (13). Whereas, group C’s
scores suggest a normal distribution as the mean (12.1), median (12.5) and mode
(12) are almost the same.

A study was conducted to see whether there was a difference in the time taken to complete a
maze, depending on whether the participants completed the task in silence or whilst listening to
music. Here is a distribution curve for the group who completed the maze whilst listening to music.

Mode 35

Median39

Mean 45

QUESTION 3
What does this distribution tell us about the scores of the participants completing the maze while listening to
music? [2 marks]

Answer: This distribution tells us that the scores are positively skewed. This means
that a small number of participants took a long time to complete the maze which led
to the mean score being higher than the median (middle score) and the mode (score
that occurred most frequently).

12 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

Activity 4
Use of mathematical symbols
Exam Hints: Symbols may form part of a statement that you are asked to interpret and
you need to ensure that you show what the symbol means in your answer.

These are the symbols and some of the abbreviations included in the AQA specification. Identify what each
means:

Symbol or Symbol or
What it means What it means
abbreviation abbreviation
> Greater than ∝ Proportional to

>> Much greater than P Probability

< Less than df Degrees of freedom

<< Much less than N Number of participants

= Equals X2 Chi-square

~ Approximately

Symbols can be included in questions. Answer the following questions:

An investigation was carried out to see whether participants recalled more short words than long
words. The same participants took part in both conditions and their scores were as follows:

1 2 3 4 5 6 7 8 9 10

Condition A (short words) 9 10 14 8 11 7 16 6 8 12

Condition B (long words) 8 7 5 6 8 5 10 6 7 9

Question Answer
1. In condition A, how many participants scored < 9? 4

2. In condition B, how many participants scored > 7? 4

3. In this study, N = 10

The probability of the results occurring

4. Explain in words what this means: p = 0.05.
by chance equals 0.05

The probability of the results occurring

5. Explain in words what this means: p < 0.01.
by chance is less than 0.01

www.tutor2u.net 13
Activity 5
Decimals, fractions & ratios
Part 1: Decimals
Exam Hints: You could be asked to convert a percentage to a decimal, which involves
moving the percentage sign and moving the decimal point two places to the left. For
example, to convert 62% to a decimal, you would remove the decimal point two places to
before the 6, making the answer 0.62.

Convert the following percentages into decimals:

87% 32.5% 5% 10% 1% 22.4%

0.87 0.325 0.05 0.1 0.01 0.224

Exam Hints: It is also important to understand the difference between 3 significant

figures (counting from the first non-zero digit for three digit and then rounding the last
digit), and 3 decimal places (three digits after the decimal place).

All of these decimals have 3 or more decimal places. Convert them to 3 significant figures:

5.631 56.589 2.629 8.9156 5.2682 22.492

5.63 56.6 2.63 8.92 5.27 22.5

Part 2: fractions
Exam Hints: You could also be asked to convert a decimal to a fraction. If there is one
decimal place then you divide by 10, if there are two then divide by 100, if there are
three then divide by 1,000 and so on. Once you have your fraction then you need to
reduce it by finding the lowest common denominator (the largest number that can divide
evenly into both parts of the fraction). For example, if you were asked to convert 0.4 to a
fraction you would divide it by 10 which makes 4/10, this can then be reduced to 2/5.

Convert the following decimals to fractions:

0.12 0.8 0.02 0.06 0.001 0.005

12/100 = 2/100 = 6/100 = 5/1000 =
6/50 = 3/25
8/10 = 4/5 1/1000
1/50 3/50 1/200

14 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

Part 3: ratios
Exam Hints: You could be asked to calculate a ratio. Ratios can be part-to-whole or
part-to-part. For example, in a study of 40 participants, where 17 were male and 23 are
female, the part-to-whole ratio for males would be 40:17 and the part-to-part ratio for
males to females would be 17:23. Like fractions, ratios should be reduced by finding the
lowest common denominator.

Practise by working out the following ratios:

QUESTION 1
In a study where there are 5 males and 15 females, what is the ratio of males in the study (part-to-whole)?
[2 marks]

Answer: 5:20 = 1:4

QUESTION 2
In the same study, what is the ratio of males to females (part-to-part)? [2 marks]

Answer: 5:15 = 1:3

Here are the results of an observational study to see whether there are gender differences in road
crossing behaviour at traffic lights:

Wait for green light

Cross on red light
to cross
Males 46 26
Females 92 24

QUESTION 1
In this study, what is the ratio of males to females (part-to-part)? [2 marks]

Answer: Males = 46 + 26 = 72
Females = 92 + 24 = 116
72:116 = 36:58 = 18: 29

www.tutor2u.net 15
QUESTION 2
What is the ratio of those who cross on a red light to all the participants in the study? [2 marks]

Answer: Total cross on red light = 26 + 24 = 50

Total in study = 50 + 46 + 92 = 188
50:188 = 25:94

QUESTION 3
What is the ratio of those who cross on a red light to those who wait for a green light? [2 marks]

Answer: Total cross on a red light = 50

Total wait for a green light = 46 + 92 = 138
50:138 = 25:69

QUESTION 4
What is the ratio of males who wait for a green light to males who cross on a red light? [2 marks]

Answer: 46:26 = 23:13

NOTES:

16 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

Activity 6
Probabilities, significance & type I/type II errors
Part 1: probabilities
Exam Hints: Probability refers to the likelihood of something happening and is
represented by a number between 0 and 1, where 0 means it is impossible and 1 means
that it is certain to happen. It is abbreviated to p and either shown as a fraction or a
decimal. To convert a fraction to a decimal, you take the top number (numerator) and
divide it by the bottom number (denominator). For example, ¼ would be 1 divided by 4 =
0.25

Think about this by answering the following:

Fraction Decimal

What is the probability of throwing a number 6 on a die? 1/6 0.167

If there are 100 raffle tickets in a bowl, what is the probability 1/100 0.01
of pulling out number 88?

There are 5 pairs of socks in a drawer: one red, two blue, one
black and one white. If it is dark, what is the probability of 1/5 0.2
picking out a black pair?

What is the probability of pulling out a blue pair? 2/5 0.4

Part 2: significance
Exam Hints: When you calculate a statistical test, it tells you the probability (likelihood)
that the results have occurred due to chance, and from this you decide whether to
accept or reject the hypothesis. The conventional probability level in psychology is 0.05
or 5%.

Here are some other probability levels that you might see on a statistical table written as decimals. Convert them
to percentages:

0.1 0.01 0.025 0.02 0.001

10% 1% 2.5% 2% 0.1%

www.tutor2u.net 17
Part 3: type I & type II errors
Exam Hints: It is really important to get these the correct way around. A type I error is
a false positive as it is when you accept the research/alternative hypothesis when it
should have been rejected. A type II error is a false negative as it is when you reject the
research/alternative hypothesis when it should have been accepted.

Here are some questions:

A study was conducted to test the hypothesis: ‘Participants will rate themselves lower for anxiety
(on a scale of 1 to 10) after taking part in one hour of meditation per day for a month’. The
researcher chose the 0.05 level of probability to try to avoid making a Type I or a Type II error.

QUESTION 1
Distinguish between a Type I error and a Type II error in the context of this investigation. [4 marks]

Answer: A Type I error is a false positive, where you accept the research hypothesis
when it should have been rejected. In this investigation it would be stating that
there are lower levels of anxiety after taking part in the meditation, when there is no
significant difference.

A Type II error is a false negative, where you reject the research hypothesis when
it should have been accepted. In this investigation it would be stating there is
no difference in anxiety ratings before and after the meditation, when there is a
significant difference.

A study was conducted to see whether there was a correlation between the amount of exercise a
participant had taken part in the previous week, and their rating out of 10 for happiness.

QUESTION 2
With reference to Type I and Type II errors, explain why it is appropriate to use the 5% level of significance in this
investigation. [2 marks]

Answer: It is appropriate to use the 5% level of significance because it achieves a

balance between making a Type I error (saying there is a correlation between
the amount of exercise a person has done and their happiness rating, when there
is actually no significant relationship), or a Type II error (saying there is no
correlation between the amount of exercise a person has done and their happiness
rating, when there actually is a significant relationship).

18 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 3
Identify whether you would be more likely to make a Type I or a Type II error if you chose to use the 0.1 level of
significance. Justify your answer. [2 marks]

Answer: You would be more likely to make a Type I error if you chose to use the
0.1 level. This is because it means there is a 10% probability that the results have
occurred by chance, meaning you are more likely to accept the research hypothesis
when it should have been rejected.

QUESTION 4
Identify whether you would be more likely to make a Type I or a Type II error if you chose to use the 0.01 level of
significance. Justify your answer. [2 marks]

Answer: You would be more likely to make a Type II error if you chose to use the 0.01
level of significance. This is because it is too strict, meaning you are more likely to
reject the research hypothesis when it should have been accepted.

NOTES:

www.tutor2u.net 19
Activity 7
Interpreting data in graphs & tables
Exam Hints: When you are asked to explain information in a graph or a table of results, it
is important to refer to as much of the data as you can. For example, if it is a table, then
try to use all of the rows and all of the columns.

A study was conducted to see if there was a correlation between the number of hours sleep a
participant said they had the previous night and their score out of 20 on a memory test. The
results can be seen in the graph below:

QUESTION 1
What is the name given to this type of graph? [1 mark]

Answer: Scattergram

QUESTION 2
Explain what the graph shows about the correlation between the two co-variables. [3 marks]

Answer: It shows that there is a fairly strong positive correlation between the number
of hours sleep the previous night and the number of items correctly recalled on the
memory test as the more hours sleep participants had the previous night, the higher
they scored on the test. However, this was not the case for all participants as there is
an outlier who had lots of sleep but obtained a low score on the memory test.

20 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

A study was conducted to see whether there are gender differences in helping behaviour. A
psychology student dropped some books while passing other students in the university corridor
and then noted the gender of the person and whether they helped them to pick up the books or
not. The results were as follows:

Males Females
Helped pick up books 12 22
Did not help pick up books 30 62

QUESTION 1
Explain what the data in the table shows about gender differences in helping behaviour. [4 marks]

Answer: The data in the table suggests that there are no gender differences in helping
behaviour. This is because 12 out of 42 males helped which is about 25%, and 22
out of 84 females helped which is also about 25%. Similarly, there are no gender
differences in those that did not help as 30 out of 42 males did not help (about
75%), and 62 out of 84 females did not help (about 75%).

A study was conducted to see whether time of day has an effect on concentration. Half of a class
of psychology students were given a task where they were timed to see how long it took them
to find 6 differences between two cartoon pictures at 10am in the morning. The other half were
given the same task at 3pm in the afternoon. The results were as follows:

Time taken in seconds to find 6 differences

Morning (10am) Afternoon (3pm)
Mean 346 385
Standard deviation 7.52 12.16

QUESTION 1
What do the mean and standard deviation values in the table suggest about the effect of time of day on
concentration? [4 marks]

Answer: The mean values suggest that people are able to concentrate better in the
morning. This is because the students who completed the task in the afternoon took
longer to find the 6 errors (385 seconds) than the students who completed the task
in the morning (346 seconds). The standard deviation values suggest there is a
greater spread of scores for the students who completed the task in the afternoon
than those who completed it in the morning, because the standard deviation in the
afternoon (12.16) is larger than the standard deviation in the morning (7.52).

www.tutor2u.net 21
A study was conducted to see whether the age of a child affects food preferences. Children aged
5 to 6, or 9 to 10, were asked whether they preferred bananas or apples and the results were
plotted on a graph.

QUESTION 1
What is the name given to this type of graph? [1 mark]

Answer: Bar chart

QUESTION 2
Explain what the graph shows about the effect of the age of the child on food preferences. [2 marks]

Answer: The graph shows that more children aged 5-6 said they preferred bananas to
apples and that more children aged 9-10 said that they preferred apples to bananas.

22 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

Activity 8
Plotting appropriate graphs
Exam Hints: When asked to draw an appropriate graph it is very important to include a
title, to label both axes, and to make sure it is the correct type of graph for the data. You
also need to plot the data accurately.

A study was conducted to see whether there are gender differences in helping behaviour. A
psychology student dropped some books while passing other students in the university corridor
and then noted the gender of the person and whether they helped them to pick up the books or
not. The results were as follows:

Males Females

Helped pick up books 12 22

Did not help pick up books 30 62

QUESTION 1
Sketch a bar chart to represent the data in this table. [4 marks]

A bar chart to show if there are any gender differences in helping a student
pick up dropped books
70

60

50
Number of people who helped

40
males

30 Females

20

10

0
Helped Did not help

www.tutor2u.net 23
QUESTION 2
Explain why a histogram would not be an appropriate choice of graphical display for the data in this table.
[2 marks]

Answer: A histogram would not be appropriate as it can only be used with interval
data, where data is measured on a continuous scale with fixed units and equal
distance between the points. The data in this table is nominal or frequency count
data and so a bar chart is the appropriate choice.

A study was conducted to see if there

was a correlation between the rating a Participant Stress rating Score
student gave for how stressed they felt 1 9 9
(between 1 and 10) and their score on a 2 7 12
psychology test (between 1 and 20). The
results are shown in the table below: 3 6 15
4 5 14
QUESTION 1
Sketch a suitable graphical display to 5 2 16
represent the data in this table. [4 marks] 6 7 15
7 1 12
8 6 18
9 3 16
10 5 19

Scattergram to show the relationship between a student's rating for how

stressed the feel and their score on a psychology test
20

18

16

14
Score on the psychology test

12

10

0
0 1 2 3 4 5 6 7 8 9 10
Stress rating

24 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 2
Explain what the graph shows about the correlation between how stressed a student feels and their score on the
psychology test. [2 marks]

Answer: The graph shows a fairly weak negative correlation between how stressed a
student feels and their score on the psychology test.

A study was conducted to see to see whether caffeine

improves concentration. Half of the participants were Score in seconds Frequency
given two cups of strong coffee and timed to see how 20-29 1
long it took them to solve 3 simple anagrams. The 30-39 5
other half completed the same task, but were not given
coffee. Here are the scores for the caffeine condition: 40-49 4
QUESTION 1 50-59 9
Draw a suitable graphical display to represent 59-69 8
this data. Label your graph appropriately.
70-79 0
[4 marks]
89-90 1

Histogram to show the times taken by the caffeine group to solve 3

anagrams
10

7
Frequency (number of people)

0
20-29 30-39 40-49 50-59 59-69 70-79 89-90
Time taken in seconds

www.tutor2u.net 25
Activity 9
Choosing the correct test
Exam Hints: In order to be able to choose the correct test in the exam, you need to find
a flow chart or mnemonic that works for you. The three things that you need to consider
are: Does the study look for a difference or a relationship? What level of measurement
is the data (nominal, ordinal or interval)? Is the design related (repeated measures or
matched pairs), or unrelated (independent groups)?

Test of Difference Test of Association or

Related Design Unrelated Design Correlation
Sign test Chi-squared
Nominal Data Simon Cowell

At Least Wilcoxon Mann-Whitney U Spearman’s rho

Ordinal Data Wants More Singers
Related t-test Unrelated t-test
Pearson’s r (Parametric)
(Parametric) (Parametric)
Interval Data Praise
Receiving Unanimous
For each of the following studies, identify which statistical test would be appropriate and explain why. In the
examination, these types of questions are typically worth 3 marks: 1 mark for identifying the correct test and 2
marks for the justification. The justification should be written in the context of the investigation described in the
STEM.

QUESTION 1 An experiment was conducted to investigate the effect of word length on recall in
STM. In one condition, the participants were asked to recall short words and in another the same
participants were asked to recall long words.

Answer: Wilcoxon. The study is looking for a difference in the number of words
recalled and a repeated measures design has been used as the same participants are
asked to recall both the short words and the long words. The data (number of words
recalled) is at least ordinal as it can be argued that although the measuring scale is
continuous, we cannot be sure that there is equal distance between the points on the
scale as some words may be easier to recall than others.
The related t test would also be credited here.

26 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 2 A study was conducted to investigate whether there are gender differences in food
preferences. Males and females were asked to choose whether they prefer chocolate or crisps.

Answer: Chi-square test as the study is looking for a difference in food preferences
between males and females. Independent groups design has been used as the
participants were either male or female and the data collected is nominal as it
involves counting the number of participants in each category (e.g. females who
prefer chocolate).

QUESTION 3 A study to investigate the relationship between students’ ratings of their own memory
ability (out of 10) and their scores on a memory test.

Answer: Spearman Rho as the study is looking for a relationship/correlation between

students’ ratings of their memory ability and their scores on a memory test. In
addition, the students’ ratings of their own memory ability are ordinal level data,
because although there is a continuous scale of measurement, there is not equal
distance between the points.

QUESTION 4 An experiment to see whether alcohol leads to slower reaction time. Half of the
participants were given a drink containing 2 units of alcohol and the other half were given a drink
with no alcohol. They were then timed to see how many seconds it took them to click the mouse
when a red cross appeared on the computer screen.

Answer: Unrelated t test as the study is looking for a difference in reaction times.
An independent groups design has been used as participants only take part in one
condition and the time taken to click the mouse is interval level data because there
is a continuous scale of measurement with equal distance between the points.

www.tutor2u.net 27
QUESTION 5 A study to see if there is a correlation between the age of a child (in months) and the
time taken to complete a maze.

Answer: Pearson r test, as the study is looking for a relationship between the age of
a child and the time taken to complete a maze and the data for both co-variables
is interval. This is because both the age of the child and the time taken to complete
the maze have continuous scales of measurement with equal distance between the
points.

QUESTION 6 A study to investigate whether infants who are first born are more likely to be securely
attached than infants who have older siblings. The infants were assessed using the Strange
Situation and categorised as either being securely attached or insecurely attached.

Answer: Chi-square test as the study is looking for a difference in attachment

type between infants who are first born and infants who have older siblings. An
independent groups design has been used and the data collected is nominal as it
is collected by counting the number of infants in each category (e.g. first born
infants that are securely attached).

There are other ways that your knowledge of which statistical test is appropriate can be assessed. Here are
some examples below to work on:

QUESTION 7 A study to investigate whether smokers are more likely to give themselves a higher
rating for being impulsive than non-smokers.

The psychologist used a Mann-Whitney U test to analyse the data. With reference to levels of measurement,
explain why this is a suitable test to use for this data. [2 marks]

Answer: It is a suitable test to use because the level of measurement of the data is
ordinal level. This is because impulsivity is being measured using a rating scale
and although this scale will be continuous, there will not be equal distance between
points on the scale.

QUESTION 8 A study to investigate whether older people (35 to 70) are more likely to smoke than
younger people (16 to 34).

A chi-square test was used to analyse the data. With reference to levels of measurement, explain why this is a
suitable test to use for this data. [2 marks]

28 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

 Answer: It is a suitable test to use because the level of measurement of the data is
nominal level. This is because smoking behaviour is being measured by counting
the number of participants in each category; in other words, how many younger
people smoke or don’t smoke, and how many older people smoke and don’t smoke.

Exam Hints: In these types of questions (7 and 8), it is only necessary to refer to levels
of measurement, it is not necessary to include any other justification for using that
particular test.

QUESTION 9 A psychologist investigated whether room temperature affects learning. Half of the
participants learned and recalled the words in a warm room, and half learned and recalled them
in a cool room.

The psychologist used a Mann-Whitney U test to analyse this data. Explain why a Mann-Whitney U test is an
appropriate test to use and a Spearman’s rho test is not. [2 marks]

Answer: A Mann Whitney test is appropriate to use as the study is looking for a
difference (between the number of words recalled in a warm room or a cool room).
The Spearman’s rho test is not appropriate as the study is not looking for a
correlation.

A psychologist investigated whether hunger affects concentration. Participants were timed to

see how long it took them to find all the errors in a paragraph of text. Half of the participants did
this shortly after eating their lunch. The other half completed the task 5 hours after they had last
eaten.

QUESTION 10 The psychologist used an independent t-test to analyse the data. Explain why an independent
t-test is an appropriate test to use and a related t-test is not. [2 marks]

Answer: An independent t test is appropriate to use as the study has used an

independent groups design, as different participants completed the task before and
after eating. The related t test is not appropriate as the study has not used a repeated
measures or matched pairs design.

Exam Hints: In these types of questions (9 and 10), it is important to point out why one
test was appropriate (1 mark) and the other one was not (1 mark). 

www.tutor2u.net 29
Activity 10
Interpreting the results of statistical tests using
statistical tables & critical values
Exam Hints: If you are asked to interpret the results of a statistical test you need to
compare the calculated/observed value (the test result) to the appropriate critical value
from the statistical table. To be able to do this, you need to know whether the test is one-
tailed (directional hypothesis) or two-tailed (non-directional hypothesis) as this affects
which column in the table you look at. You will also need to know the sample size (N)
and for some tests df (degrees of freedom) also have to be calculated. For some tests
(chi-square, Spearman’s Rho, Pearson’s r, related t-test, unrelated t-test) the observed/
calculated value has to be greater than or equal to the critical value for the result to be
significant, while for the others (sign test, Mann-Whitney U and Wilcoxon) the calculated/
observed value has to be less than or equal to the critical value. You will always be told
this at the bottom of the table, but a simple rule is that any test with a letter R in the
name is greater than or equal to.

A psychologist conducted an observational study to test the hypothesis: ‘Males will be more likely
to walk under a ladder than females’. She placed a ladder over a footpath outside her office and
observed how many people walked round or under it and whether they were male or female. She
analysed the data collected using a chi-square test and obtained a calculated value of 2.981.

QUESTION 1
With reference to the critical values in Table 1 where df= 1, explain whether or not the calculated value of chi-
square is significant at the 5% level. [2 marks]
Table 1: Critical values for chi-square
Answer: Where df is 1, the critical
Level of significance for a one-tailed test
value for a one-tailed test
where p=0.05 is 2.71. As the 0.10 0.05 0.025 0.01 0.005 0.0005
calculated value of 2.981 is
Level of significance for a two-tailed test
greater than the critical value it
is significant at the 5% level. Df 0.20 0.10 0.05 0.025 0.01 0.001

1 1.64 2.71 3.84 5.02 6.64 10.83

2 3.22 4.61 5.99 7.38 9.21 13.82

3 4.64 6.25 7.82 9.35 11.35 16.27

Calculated/observed value has to be GREATER than or EQUAL to
the critical value for significance to be shown.

30 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

A psychology teacher wanted to see whether there was a relationship between the rating a
student gave themselves for how stressed they felt (between 1 and 10) and the score they
obtained in a psychology test. He was not sure whether those who were stressed would score
higher on the test or lower. After collecting the data from 10 students, he analysed it using the
Spearman Rho test and obtained a calculated value of rs= 0.582.

QUESTION 2
With reference to the critical values in Table 2, explain whether or not the calculated value of Spearman Rho is
significant at the 5% level. [2 marks]
Table 2: Critical values for Spearman Rho
Answer: Where N= 10 the critical
One-tailed test 0.05 0.01
value for a two-tailed test
where p=0.05 is 0.648. As the Two-tailed test 0.10 0.05
calculated value of 0.582 is less N= 7 .714 .786
than the critical value, it is not 8 .643 .738
significant at the 5% level.
9 .600 .700
Exam hint: This was a two-tailed 10 .564 .648
test as the psychology teacher was 11 .536 .618
unsure whether those who were
12 .503 .587
stressed would score higher or
Calculated/observed value of rho must be EQUAL to or
lower. GREATER than the critical value in this table for significance to
be shown.

A psychologist wanted to test the hypothesis: ‘There will be a difference in the number of
words correctly recalled whilst listening to music, compared to listening in silence’. 15 student
participants were tested using a repeated measures design and a Wilcoxon test was used to
analyse the results. The calculated value of the test was T = 27.

QUESTION 3
Using table 3 and the 0.05 level of significance, explain whether or not the psychologist was able to accept the
hypothesis. Justify your answer. [3 marks]
Table 3: Critical values for Wilcoxon
Answer: The psychologist will not
One-tailed test 0.05 0.025 0.01 be able to accept the hypothesis.
Two-tailed test 0.1 0.05 0.02 This is because the calculated
N = 13 21 17 12 value for a two-tailed test where
N= 15 and p= 0.05 is 27, and
14 25 21 15
this is greater than the critical
15 30 25 19 value of T which is 25.
16 35 29 23
17 41 34 27
Calculated/observed value of rho must be LESS THAN or
EQUAL TO the critical value in this table for significance to be
shown.

www.tutor2u.net 31
Some psychology students were interested in finding out whether there was a relationship
between the number of hours sleep the previous night and performance on a concentration task,
where the participants were timed to find out how many seconds it took for them to complete 3
simple anagrams. Their hypothesis was: ‘There will be a negative correlation between the number
of hours sleep the previous night and the time taken (in seconds) to solve 3 anagrams’. There
were 12 participants and when Pearson’s r was calculated, a negative correlation coefficient of
-0.581 was found. The degrees of freedom are calculated by N – 2.

QUESTION 4
Using table 4, explain whether or not the students were able to accept the hypothesis at the 0.05 level of
significance. Justify your answer. [3 marks]

Exam Hints: When you are comparing the calculated value to the critical value using a
correlation test, you ignore the minus sign if you have found a negative correlation.

Table 4: Critical values for Pearson’s r

Answer: The students will be able
Level of significance for a one-tailed test
to accept the hypothesis. This
is because the critical value for 0.10 0.05 0.025 0.01 0.005 0.0005
a one-tailed test where df = 10 Level of significance for a two-tailed test
and p = 0.05 is 0.497, meaning
that the calculated value of Df 0.20 0.10 0.05 0.025 0.01 0.001
0.581 is greater than the critical 10 0.398 0.497 0.576 0.658 0.708 0.823
value.
11 0.380 0.476 0.553 0.634 0.684 0.801

12 0.365 0.457 0.532 0.612 0.661 0.780

The calculated/observed value has to be GREATER than or
EQUAL to the critical value for significance to be shown.

32 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

 A psychologist wanted to investigate whether hunger affects concentration. 10 participants were
timed to see how many seconds it took them to identify 10 errors in a short passage of text, 4
hours after they had last eaten. Another 9 participants completed the same task 30 minutes after
they had last eaten. The psychologist used a directional hypothesis as previous research had
shown hunger reduces concentration. An unrelated t-test was used to analyse the data collected
and obtained a calculated value of 1.740. Df are worked out by adding together the number of
participants in each condition and subtracting 2.

QUESTION 5
Using table 5, explain whether or not the students were able to accept the hypothesis at the 0.05 level of
significance. Justify your answer. [3 marks]
Table 5: Critical values for Unrelated t-test
Answer: The students will be able
One-tailed test 0.05 0.025
to accept the hypothesis. This
is because the critical value for Two-tailed test 0.10 0.05
a one-tailed test where df= 17 df = 16 1.746 2.120
(19 – 2) and p=0.05 is 1.740, 17 1.740 2.110
meaning that it is equal to the
calculated value of 1.740. 18 1.734 2.101
19 1.729 2.093
20 1.725 2.086
The calculated/observed value has to be GREATER than or
EQUAL to the critical value for significance to be shown.

NOTES:

www.tutor2u.net 33
Activity 11
Calculating and interpreting the sign test
Exam Hints: The sign test is the only statistical test that you can be asked to calculate
in the exam and should be used when data is nominal level, the study is looking for a
difference, and the design is related (repeated measures or matched pairs). However,
in the exam you may be asked to conduct a sign test on data that is either ordinal or
interval (e.g. scores).
Calculating a sign test involves working through a series of steps:

Decide whether there has been an increase (+), a decrease (-) or the response/
STEP 1 score has stayed the same (=).

Add up the pluses and minuses to work out the calculated/observed value (s),
STEP 2 which is the least frequent sign.

Compare the calculated/observed value to the critical value using Table 5. N is the
STEP 3 number of participants, but any responses that are the same are ignored.

Table 1: Critical values for the sign test Here are two research scenarios; read through
the information and then calculate the sign test
One-tailed test 0.05 0.025 0.01 and interpret the findings for each.
Two-tailed test 0.10 0.05 0.02

N=5 0

6 0 0

7 0 0 0

8 1 0 0

9 1 1 1

10 1 1 0

11 2 1 1

12 2 2 1

13 3 2 1

14 3 2 2

15 3 3 2
The observed/calculated value has to be EQUAL to or LESS
than the critical value for significance to be shown.

34 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

 SCENARIO A
A study was conducted to see whether memory is better in the morning or the afternoon (non-
directional hypothesis). 10 participants were shown pictures of 20 objects at 10am in the morning
and then asked to recall as many as they could. The same participants were then shown another
20 pictures at 3pm in the morning and asked to recall as many of the objects as they could. The
results are shown in the table below:

Number of objects Number of objects

Participant Step 1 (+ - =)
recalled at 10am recalled at 3pm
1 15 11 -
2 12 12 =
3 16 14 -
4 9 7 -
5 10 11 +
6 18 17 -
7 15 16 +
8 8 8 =
9 10 8 -
10 12 10 -

QUESTION 1
Calculate the value of s and show your workings. [2 mark]

Answer: Number of + = 2
Number of - = 6
Calculated value of S = 2

QUESTION 2
Where p= 0.05, explain whether or not there was a significant difference in the number of objects recalled in the
morning than in the afternoon. Use your answer to Q1 and Table 1. [2 marks]

Answer: There is no significant difference in the number of objects recalled in the

morning than in the afternoon as the critical value for a two-tailed test where N =
8, using the 0.05 level of significance is 0. As the calculated value of 2 is greater
than the critical value of 0, the difference is not significant.

www.tutor2u.net 35
 SCENARIO b
A study was conducted to see whether participants rated their anxiety levels to be lower after
taking part in 30 minutes of meditation once a day for a week (directional hypothesis). 12
participants were asked to rate their overall anxiety out of 10 before beginning the meditation
and again at the end of the week. The results are shown in the table below:

Anxiety rating before Anxiety rating after a

Participant Step 1 (+ - =)
beginning meditation week of meditation
1 9 6 -
2 8 5 -
3 10 10 =
4 6 4 -
5 10 8 -
6 9 8 -
7 5 4 -
8 9 6 -
9 10 8 -
10 7 6 -
11 7 9 +
12 8 4 -

QUESTION 1
Calculate the value of s and show your workings. [2 marks]

Answer: Number of + = 1
Number of - = 10
Calculated value of S = 1

QUESTION 2
Where p=0.05, explain whether or not there was a significant difference in the anxiety scores after the
meditation. Use your answer to Q1 and Table 1. [2 marks]

Answer: There is a significant difference in anxiety scores after the meditation as the
critical value for a one-tailed test where N = 11, using the 0.05 level of significance,
is 2. As the calculated value of 1 is less than the critical value of 2, the difference is
significant.

36 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

scenario 1
Are males more likely to pass their driving test first time
than females?
A psychologist wanted to test the hypothesis: ‘Males are more likely to pass their practical driving
test the first time than females’. He questioned a sample of university students who had passed
their driving test. The results are shown in Table 1:

Males Females

Passed practical driving test first time 69 59

Did not pass first time 63 73

QUESTION 1
Draw a suitable graphical display to represent this data. Label your graph appropriately. [4 marks]

Bar chart to show differences between males and females for passing
practical driving test first time or not
80

70

60
Frequency (number of people)

50

40

30

20

10

0
Passed test first time Did not pass first time

Males Females

www.tutor2u.net 37
QUESTION 2
What percentage of the overall sample passed their practical driving test first time? Show your calculations. Give
your answer to 3 significant figures. [3 marks]

Answer: Total passed first time = 69+59 = 128

Total in sample = 128 + 63 + 73 = 264
128/264 x 100 = 48.5

QUESTION 3
What percentage of males passed their practical driving test first time? Show your calculations. Show your
answer to 3 decimal places. [3 marks]

Answer: Total males = 69+63 = 132

69/132 x 100 = 52.273

QUESTION 4
What percentage of females passed their practical driving test first time? Show your calculations. Show your
answer to 2 decimal places. [3 marks]

Answer: Total females = 59 + 73 = 132

59/132 x 100 = 44.69%

38 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 5
The psychologist used a chi-square test to analyse this data. With reference to levels of measurement, explain
why a chi-square test is an appropriate test to use and a Mann-Whitney U test is not. [2 marks]

Answer: A chi-square test is appropriate to use as the study has collected nominal
data (the frequency of those who have passed their practical driving test the first
time or not). The Mann-Whitney test is not appropriate as it can only be used with
data that is at least ordinal.

QUESTION 6
The psychologist analysed the data collected using a chi-square test and obtained a calculated value of 1.516.
With reference to the critical values in Table 2 where df= 1, explain whether or not the calculated value of chi-
square is significant at the 5% level. [2 marks]

Table 1: Critical values for chi-square

Answer: Where df is 1, the critical
value for a one-tailed test Level of significance for a one-tailed test
where p=0.05 is 2.71. As the 0.10 0.05 0.025 0.01 0.005 0.0005
calculated value of 1.516 is less
than the critical value it is not Level of significance for a two-tailed test
significant at the 5% level. Df 0.20 0.10 0.05 0.025 0.01 0.001

1 1.64 2.71 3.84 5.02 6.64 10.83

2 3.22 4.61 5.99 7.38 9.21 13.82

3 4.64 6.25 7.82 9.35 11.35 16.27

Calculated/observed value has to be GREATER than or EQUAL to
the critical value for significance to be shown.

NOTES:

www.tutor2u.net 39
scenario 2
Is there a correlation between the number of hours worked in
a week and how stressed someone feels?
A psychology student was Number of hours worked
interested in the relationship Participant no. Stress rating out of 10
in an average week
between work and stress. She
1 40 6
asked a sample of working
adults to estimate how 2 45 8
many hours they worked in 3 32 7
an average week and also
4 16 9
asked them to give a rating
of between 1 and 10 for how 5 48 9
stressed they felt, where 10 6 24 5
meant they felt extremely
stressed. The results are shown 7 32 3
in Table 1: 8 22 4
9 35 7
10 52 9

QUESTION 1
Draw a suitable graphical display to represent this data. Label your graph appropriately. [4 marks]

Scattergram to show the correlation between number of hours worked in an

average week and stress ratings out of 10
10

7
Stress rating out of 10

0
0 10 20 30 40 50 60
Number of hours worked in an average week

40 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 2
Explain what the data in the table and the graph shows about the correlation between the number of hours
participants work in an average week and their stress rating. [3 marks]

Answer: It shows that there is a fairly strong positive correlation between the number
of hours a participant works in an average week and their stress rating as the more
hours someone has worked, the higher the rating they give for how stressed they feel.
However, this was not the case for all participants as there is an outlier (participant
4) who only works 16 hours but has a stress score of 9.

QUESTION 3
Identify an appropriate statistical test for the psychologist to use to analyse the data. Justify your answer.
[3 marks]

Answer: Spearman Rho as the study is looking for a relationship/correlation between

the number of hours a participant works in an average week and their stress rating.
In addition, the stress ratings are ordinal level data because although there is a
continuous scale of measurement, there is not equal distance between the points.

QUESTION 4
The psychology student wants to avoid making a Type I error. Should she use a 0.05 level of significance or a
0.01 level of significance? Justify your answer. [3 marks]

Answer: She should use a 0.01 level of significance. This is because it means that
there is only a 1% probability that the results have occurred by chance, so it is less
likely she will make a Type I error (false positive) and say there is a correlation
between the between the number of hours a participant works in an average week and
their stress rating, when there is not a significant relationship.

www.tutor2u.net 41
scenario 3
Is there a difference in spatial ability between physics
students and psychology students?
A psychologist wanted to test the hypothesis: ‘There will be a difference in the time taken
(in seconds) by psychology and physics students to find their way through a simple maze.
Undergraduates studying each subject were shown the maze on a computer screen and timed
to see how long it took them to use the mouse to reach the end of the maze. The time taken (in
seconds) is shown in the table below:
Psychology students Physics students
Time taken in seconds Time taken in seconds
Participant no. Participant no.
to solve the maze to solve the maze
1 18 9 12
2 16 10 13
3 15 11 8
4 14 12 10
5 17 13 14
6 15 14 13
7 12 15 9
8 14 16 10
17 12

QUESTION 1
Calculate the mean time taken by the psychology students to complete the maze. Show your calculations. Give
your answer to 3 significant figures. [3 marks]

Answer: 18 +16+15+14+17+15+12+14 = 121

121/8 = 15.1

QUESTION 2
Calculate the mean time taken by the physics students to complete the maze. Show your calculations. Give your
answer to 3 significant figures. [3 marks]

Answer: 12+13+8+10+14+13+9+10+12 = 101

101/9 = 11.2

42 AQA A LEVEL PSYCHOLOGY STATISTICS & MATHEMATICS WORKBOOK

QUESTION 3
What might the psychologist conclude about differences in spatial abilities between psychology and physics
students, based on the mean times? [2 marks]

Answer: The mean times suggest that physics students have better spatial abilities
than psychology students. This is because the physics students took less time on
average to complete the maze (11.2 seconds) than the psychology students (15.1
seconds).

QUESTION 4
Identify the most appropriate measure of dispersion to use with this data. Justify your answer. [3 marks]

Answer: The standard deviation would be most appropriate as the data is interval level.
It is also more informative than the range as it tells you the spread of scores around
the mean. In addition, there are no extreme scores in either condition.

QUESTION 5
The psychologist used an unrelated t-test to analyse the data. Explain why this is an appropriate test to use in
this investigation. [3 marks]

Answer: It is an appropriate test to use because the investigation is looking for

a difference in spatial abilities between physics and psychology students.
An independent groups design has been used as the physics students and the
psychology students are different groups. In addition, the data collected is interval
data as it is the time taken in seconds so there is a continuous scale with equal
distance between points on the scale.

QUESTION 6
The researcher chose the 0.05 level of probability to try to avoid making a Type I or a Type II error. Distinguish
between a Type I error and a Type II error in the context of this investigation. [4 marks]

Answer: A Type I error is a false positive, where you accept the research hypothesis
when it should have been rejected. In this investigation, it would be stating that
there is a difference between physics and psychology students in the time taken to
complete a maze when there is no significant difference.

A Type II error is a false negative, where you reject the research hypothesis when
it should have been accepted. In this investigation it would be stating there is no
difference between physics and psychology students in the time taken to complete a
maze when there is a significant difference.

www.tutor2u.net 43
More Psychology revision and support at:
www.tutor2u.net/psychology

@tutor2uPsych ALevelPsychStudentGroup