Psychology

Research Methods
Summer Work

Name …………………………………………...

1
 What you need to know:
The following comes directly from the AQA Psychology Specification. You can find all of this information on the AQA
website

Research Methods

Students should demonstrate knowledge and understanding of the following research methods,
scientific processes and techniques of data handling and analysis, be familiar with their use and be
aware of their strengths and limitations:

•• Experimental method. Types of experiment, laboratory and field experiments; natural and
quasiexperiments.
•• Observational techniques. Types of observation: naturalistic and controlled observation; covert
and overt observation; participant and non-participant observation.
•• Self-report techniques. Questionnaires; interviews, structured and unstructured.
•• Correlations. Analysis of the relationship between co-variables. The difference between
correlations and experiments.

Scientific processes

•• Aims: stating aims, the difference between aims and hypotheses.

•• Hypotheses: directional and non-directional.
•• Sampling: the difference between population and sample; sampling techniques including:
random,
systematic, stratified, opportunity and volunteer; implications of sampling techniques, including
bias and generalisation.
•• Pilot studies and the aims of piloting.
•• Experimental designs: repeated measures, independent groups, matched pairs.
•• Observational design: behavioural categories; event sampling; time sampling.
•• Questionnaire construction, including use of open and closed questions; design of interviews.
•• Variables: manipulation and control of variables, including independent, dependent, extraneous,
confounding; operationalisation of variables.
•• Control: random allocation and counterbalancing, randomisation and standardisation.
•• Demand characteristics and investigator effects.
•• Ethics, including the role of the British Psychological Society’s code of ethics; ethical issues in the
design and conduct of psychological studies; dealing with ethical issues in research.
•• The role of peer review in the scientific process.
•• The implications of psychological research for the economy.

Data handling and analysis

•• Quantitative and qualitative data; the distinction between qualitative and quantitative data
collection techniques.
•• Primary and secondary data, including meta-analysis.
•• Descriptive statistics: measures of central tendency – mean, median, mode; calculation of mean,
median and mode; measures of dispersion; range and standard deviation; calculation of range;
calculation of percentages; positive, negative and zero correlations.
•• Presentation and display of quantitative data: graphs, tables, scattergrams, bar charts.
•• Distributions: normal and skewed distributions; characteristics of normal and skewed
distributions.
•• Introduction to statistical testing; the sign test.

2
Quantitative and qualitative research methods and data
Research methods are the ways that psychologists investigate a theory. Different methods will be appropriate for
different topics/theories/situations. In general, we class these methods as Quantitative and Qualitative, which
refers to both methodology and data.

What does the phrase; “It’s quality not quantity that matters” mean?

From this you can work out what the terms quantitative and qualitative mean.

Quantitative =

Qualitative =

Quantitative Data Qualitative Data

 Easier to analyse than Qualitative Data due to  Complexity of human behaviour is represented
the use of descriptive and inferential statistics  Gains insight into the thoughts and feelings of
individuals – gives the reason for behaviour, not
merely the frequency of it
 Rich detail is collected – data which is large in
volume and depth, which means conclusions
may be more valid

 Can over simplify complex behaviour  It is more difficult to draw conclusions and
 Can lack detail and lead to assumptions of identify patterns as a large amount of data is
meanings of and reasons for behaviour usually collected.
 The analysis is subjective – it is open to
interpretation. Two people could read the same
interview and take 2 different meanings from it.

3
 Quantitative Research Methods

Experiments
What is an experiment? We hear the word very often, but may not have a definitive idea of what it is. In Psychology
an experiment is a research method which enables a researcher to manipulate the situation a person is in and see
what effect it has on a person in order to test a theory to see if it is correct.

Aim
When conducting research, a Psychologist first needs to have an idea about what they would like to investigate and
why. This is referred to as the aim. For example, a Psychologist might state the following aim:

“The aim of this research is to investigate whether the dedication to and time spent on studies has an
effect on the grades a person achieves at AS level. This is due to the fact that the more time one spends
on a particular subject, the more they rehearse this information, therefore the stronger the memory will
be for this subject.”

Hypotheses
Once an aim has been established, depending on the research method being used, the Psychologist may then create
an experimental hypothesis. A hypothesis is;
 A clear statement – it should be waffle free and NOT include any explanation
 A prediction – it should suggest what the results will find
 Testable – it should be precise – words like “better” SHOULD NOT BE USED! What is better?

For example, taking the aim from above the following hypothesis could be created:

“The grade which people get in Psychology AS is affected by the amount of time spent completing
extended study”

Is this a good hypothesis? Is it:

 A clear statement?
 A prediction?
 Testable?

Hopefully the researcher will then find that their hypothesis is correct, and will be able to accept it as such.

4
 Activity 1 – Hypotheses
Create a suitable hypothesis for the following aims:
1. An investigation into the effects of caffeine on reaction time.

2. An investigation into whether listening to Mozart affects memory

3. An investigation into whether gender affects performance in school.

Variables
When a hypothesis is created a Psychologist is usually attempting to identify a causal relationship between two
factors, which we call variables. The way in which this is done is to change the variable we think is the cause, and
measure whether this has an effect on the other variable.

In psychology we talk about these variables as the independent (IV) and dependent variables (DV).

IV =

DV =

For example, if we think that the grade you get in Psychology AS is affected by the amount of time you spend doing
extended study, what would the IV and the DV be?

IV:

DV:

In the examination you will need to be able to identify / create IVs and the DVs for a given scenario.

Whenever you do this, you must check that you have gotten these variables the correct way around. The way to do
this is to put your answers into the following sentence:

(IV) affects (DV)

5
So, if we use the example from earlier, that the grade you get in Psychology AS is affected by the amount of time you
spend doing extended study, which of the following sentences makes sense:

1) The grade you get in Psychology AS affects how much time you spend doing extended study
OR
2) How much time you spend doing extended study affects the grade you get in Psychology AS

Variables must be Operationalised – i.e. taken from being an abstract concept and made into something you can
change (IV) or measure (DV)

For example, if you take the aim “To investigate the effect of caffeine on Short term Memory”

IV – Caffeine
DV – Short Term Memory

BUT these are not precise or testable, therefore they have not been operationalised.

How can we vary Caffeine to make it something you can change?

How could we make short term memory something you could measure?

Activity 2 – Variables
Identify the IV and DV from the following examples. Make sure you check you have them the right way round by using
the sentence “IV affects DV”, and make sure they are Operationalised

1. Listening to music whilst completing extended study leads to worse grades than completing extended study in
silence.
IV:

DV:

2. If you drink more than 3 caffeine based drinks a day you are more likely to be stressed.
IV:

DV:

3. People who are more attractive have an increased chance of being successful in job interviews
IV:

6
 DV:

4. You will become more intelligent if you study Psychology A level.

IV:

DV:

5. More intelligent people choose to study Psychology at A level than any other subject.
IV:

DV:

Directional vs Non-directional Hypotheses

Hypotheses can be directional (also known as one-tailed) or non-

directional (also known as two-tailed). This refers to the prediction of
the relationship between the IV and DV.
 If the hypothesis tells you what effect the IV has on the DV (will
it make it higher, lower, faster, slower etc) than it is called
directional, because you can predict the direction of the effect of
the IV on the DV.
 If the hypothesis just suggests that there is going to be a
difference due to the IV, but doesn’t tell you how the IV will affect
it, it is a non-directional hypothesis.

Directional – One tailed  Non-directional –Two tailed  or 

Predicts the way the results will turn out Just says there will be a difference between groups
Predicts which group will do better Do not know which will do better
Could draw a rough graph from this prediction Could not draw a graph from the hypothesis
Uses phrases such as; Uses phrases such as;
 Faster than...  A difference between...
 Higher than...
 Greater than...
 More than...

7
 Activity 3 – Directional / Non-directional?
Identify the IV and DV from the following examples and then circle whether they are directional or non directional:

1. There is a difference in the speed that people react to red lights and green lights.

IV:

DV:

Directional / non-directional

2. Dogs will bite people in uniform more than they will bite people in civilian clothing

IV:

DV:

Directional / non-directional

3. Drivers under 50 will have a higher score on a hazard perception test than those over 50.

IV:

DV:

Directional / non-directional

Null Hypotheses
When designing research, Psychologists must consider the possibility that their predictions will be wrong. This means
that when they are creating their research hypothesis, they must also create a null hypothesis. This is basically the
researcher staying “if my research shows I am wrong about my hypothesis, what can I say?”

Null hypotheses are written in the format:

There will be no significant differences between (condition 1) and (condition 2) in terms of (DV). Any
difference will be due to chance.

OR

(IV) does not affect (DV).

So for hypothesis used earlier, that “The grade which people get in Psychology AS is affected by the amount of time
spent completing extended study”

A suitable Null Hypothesis would be:

The amount of time spent completing extended study will not affect the grade people get in
Psychology AS.

8
Remember: THE NULL HYPOTHESIS IS NOT THE OPPOSITE OF THE RESEARCH HYPOTHESIS – e.g. if you think women
remember more items than males, your null would not be males would remember more than females, but that there
would be NO DIFFERENCE between males and females.

Activity 4 – Null Hypotheses

Create suitable null hypotheses for the following examples:

1. Children who watch aggressive television will be more aggressive at play time than those who watch non-
aggressive television.

2. Listening to classical music whilst studying will affect the grades a person gets

3. Students will be more likely to suffer from symptoms of colds in the week before a major examination than in
the week before a school holiday.

4. People will remember more items from a list of 20 when they are shown pictures of the items than when the
list is presented in words.

Selection of Participants and Sampling Techniques

Once a Psychologists has identified their variables and operationalised them, they then need to select ppts to take
part in their research, as it is often impossible or impractical to include everyone that you would wish to.

To select who a researcher wants to include, they must first identify their target population. This is the group of
people who you would like to apply your conclusions to, so for example, it could be sixth form students in RGS, or it
could be everyone in the country. The aim is to obtain a sample which represents that group as a whole, therefore
although not everyone will be experimented on, we should have an idea about how everyone would react if they were
included.

There are many ways to obtain a sample of ppts, all of which have their own strengths and limitations

9
 Activity 5 – Sampling

Complete the table below, adding in an example of each sampling technique and a strength and weakness of each:

Technique Description Advantages Disadvantages

Random

Volunteer /

Opportunity

Stratified

Systematic

10
Experimental Design
Once you have obtained your ppts, you then need to decide how you wish to use them. This is referred to as
Experimental Design, and there are 3 main designs; independent groups, repeated measures and matched pairs /
mixed design.

Independent Groups

You take your sample and split it into two separate groups using random allocation, e.g. flip a coin to decide who goes
into which group.
Group 1 complete condition 1 whilst group 2 complete condition 2.
You then compare the results of the two groups to see the difference in their performance

Advantages Weaknesses
 Easy and quick to obtain and assign to separate  Participant variables may affect the results – see
conditions below
 Quick for ppts

Participant Variables – these are the things which make us unique from each other. The problem with these is that,
if we use random allocation then we could end up with an unequal balance of ppts in each group, e.g. all those with
perfect eyesight in one condition, which may then make it seem that condition 1 performed better than condition 2
or vice versa, when really it has nothing to do with which condition they are in.

Repeated Measures

You take your sample and make them all complete condition 1 and then all complete condition 2. You then compare
performance on condition 1 and 2 to see if there is a difference.

Advantages Weaknesses
 Eliminates participant variables, as the two  Order effects may affect the results – See below
groups are the same  More likely to lead to demand characteristics, as
ppt go through both conditions. – see single-
blind as a way to overcome this

Order effects – this is when performance in the second condition gets better due to practice (called the practice effect)
or worse due to tiredness, or lack of motivation (called the fatigue effect). This could make it appear that the
performance is better in one condition than the other, when really it is nothing to do with the condition itself, merely
that they have been through the process twice.

Order effects can be overcome by using counterbalancing. This is where although all ppts still complete all conditions,
you split your sample in half, and then make group 1 complete condition 1 then condition 2, whilst group 2 do the
opposite, i.e. do condition 2 then 1. This then cancels out any order effects, as if group 1 get worse in condition 2 due
to tiredness, so do group 2 in condition 1.

Note – there are some things that can never be repeated measures as they can’t be easily manipulated
by the experimenter – e.g. gender, or to a lesser extent age (although this could be done through a
longitudinal research method), nationality and so on.

11
Matched Pairs / Mixed design

This is where you match each ppt in group 1 with a similar person in group 2 based on criteria which are relevant to
the research – e.g. age in a test of memory. These must be relevant criteria though. There would be no point matching
people on gender for a memory test, unless there was a suggestion that gender affects memory.
Then group 1 completes condition 1, group 2 condition 2, and the results are compared.

Advantages Weaknesses
 Minimises participant variables, as the groups  May be difficult to match people on all relevant
of people are similar variables.
 No order effects, as ppts only go through one  May be some variables which are difficult to
condition. measure to therefore match people on.
 There will still be some participant variability.

Activity 6 – Experimental Design

Use the information above to complete explain how you would use each experimental design to test the following
hypothesis:

“Those who listen to Mozart whilst learning a list of 40 words will remember a greater number of words
when tested than those who learn the words in silence.”

Independent
Groups

Repeated
Measures

Matched Pairs
/ Mixed
Design

12
Controlling Extraneous Variables

Although in a hypothesis we vary the IV and see what effect it has on the IV, there are a wide range of other variables
which may affect the DV.

Think back to the example we have been using throughout that:

“The grade which people get in Psychology is affected by the amount of time spent completing extended study”

What variables other than the amount of time spent completing extended study could affect the grade people get in
Psychology Exam?

The other variables which may affect the DV are referred to as extraneous variables, and they cause a big problem for
researchers. If these are allowed to affect the DV, then our results may be corrupted, leading us to mistakenly either
reject our hypothesis when there may be a real causal link between then IV and DV, or to accept the hypothesis when
there is no relationship between the variables.

Attempting to reduce the effect of, or cancel out these is called controlling extraneous variables, and to do this, we
must try to keep things the same between participants. For example, if time of day that participants take part in our
research may affect our results, then we should keep the time of day the same for all participants. This is something
a researcher must consider during planning of research. This also keeps the research replicable.

To ensure that as many extraneous variables are controlled as possible, researchers must ensure that their method is
standardised, i.e. the same for all participants. To ensure this, they should create a detailed and strict guide to their
behaviour during the research, e.g. stating how long ppts have to complete certain tasks and containing a script of
what the researcher should say.

Demand Characteristics and Investigator Effects

The term demand characteristics was coined by Orne (1962). It refers to features of the experimental situation
which act as cues to help the participants to interpret what is expected of them – i.e. the difference in behaviour due
to the experimental setting.

A good example of this used by Orne was when he would approach people in the street and ask them to
do 10 star jumps for him. As one would expect, very few were willing. In the second condition of this
research, Orne first asked the people he approached whether they would be willing to take part in a
piece of research. If they said yes, which most did, he would then ask them to do 10 star jumps. Now,
virtually all people asked would complete the star jumps. This shows the unnatural behaviour we create
merely through telling someone they are taking part in research.

13
It may be that the demand characteristics lead the ppt to do what they think the researcher wants them to, or
indeed do the opposite of what is wanted of them, either way the effect is damaging to the integrity of the research,
as it is not measuring valid behaviour and lacks the realism of natural behaviour.

One way to overcome demand characteristics is to use a single-blind technique. This is where the ppt is not told
which condition they are in, so would find it difficult to work out the demands of the researcher.

However, an investigator may also affect the behaviour of the ppts. The investigator may unconsciously give more
or less encouragement, or more or less positive responses to the behaviour of ppts in one condition than the other.
It may be as simple as using a more animated tone of voice, or facial expression.

To overcome this, a double-blind technique may be used. This is where neither ppt or researcher know which
condition the ppt is in, therefore the experimenter cannot influence the outcome of the results.

Why might it be difficult to conduct double-blind techniques?

Reliability and Validity

These are two huge issues within Psychological research which you must make sure that you understand.

Think about the phrase “The dog ate my homework”

This is not a valid excuse as it is not true. However, it may be a reliable one because
a person might use it time and time again.

This demonstrates the difference between validity and reliability.

Another way to think of validity and reliability is using a dartboard analogy. If the goal of research is the bullseye,
then validity is how close we get to the bullseye, whilst reliability is how consistent we are, regardless of how close
we get to the bullseye.

14
Complete definitions of each in the space below:

Validity:

Reliability:

Why is it important to have validity in research? Why is it important to have reliability in research?

Validity can be further split into different types. The main 2 that we are interested in are;

Internal Validity External Validity

This is whether we have been able to establish cause This is whether our conclusions can be accurately
and effect in the research, or whether the results may applied to other people and other settings beyond the
be due to confounding or extraneous variables research setting. A related issue here is mundane
realism – how like real, everyday life was the task the
ppts completed?

15
Types of Experiment
Whilst all experiments involve the measurement of the DV after change of the IV, there are several different ways in
which may be conducted

Laboratory Experiments
A Laboratory should not be considered as a specific room, but instead a controlled environment. Anywhere could be
your laboratory, provided you are able to have a high level of control over the variables in that place. In a laboratory
the IV would be directly manipulated by the researcher. This high level of control is excellent in establishing cause
and effect, as any extraneous variables will be easily controlled, and precise manipulation and measurement of the
IV and DV are enabled, however, it is unlikely that a laboratory would yield research which is high in realism, as it is
so artificial and likely to cause demand characteristics.

Field experiments
The field simply refers to a natural habitat, so anywhere the ppt would have been at that time without the research
taking place. The researcher still manipulates the IV, but the environment is not controlled, which could lead to
extraneous variables having an effect on the DV. A field experiment is likely to have greater realism and lower
demand characteristics than a laboratory experiment.

Natural Experiments

The word natural refers only to the way in which the IV is varied. If it is something which the experimenter cannot
directly control, then this termed a natural experiment. A natural experiment could take place in a laboratory or in
the field, as deeming something a natural experiment has nothing to do with the setting in which the research will
be carried out. For example, if I were to compare the memory of students from grammar school and high school, I
could not randomly decide who should go to which school purely for the purposes of my research, but I could still
use a laboratory to investigate the effects.

A further type of experiment is known as a quasi experiment. Some will group quasi and natural experiments
together, although there is a subtle difference between these. In a quasi experiment the IV will be naturally
occurring, such as age or gender, whilst in a natural experiment, the IV could be manipulated by someone else, but
not the researcher, e.g. the Island of St Helena had TV introduced by television networks, but researchers were still
able to investigate the effect using a repeated measures design.

Complete the summary table on the next page to identify the advantages and weaknesses of these types of
experiments:

16
You need to know at least two advantages and two weaknesses of each method.

Method Description/use Advantages Weaknesses

Laboratory Experiment

Field Experiment

Natural Experiment

17
 This means you must make
it relevant to the example
Activity 7 – Types of Experiment given, not just repeat the
evaluative points from the
Answer these questions about each of these stimulus examples. table on the other page

1. What is the research method that has been used in this study?
2. Explain one advantage and one weakness of this method in the context of this study.

A study was conducted to investigate the effects of anxiety on performance. Participants were given a task to complete
in a set time. One set of the Ps was mildly stressed during the task. This was achieved by arranging for the researcher
to watch their performance closely. The other set of Ps were watched but in a friendly manner.

Research design:

Advantage:

Disadvantage:

Participants were asked to complete a revision quiz on the learning Platform as part of their extended Study. The final
question was not graded, but did ask participants to state whether they had listened to music whilst completing the
extended study.

Research design:

Advantage:

Disadvantage:

18
Correlations

If you wanted to design an experiment on whether low educational performance was related to truancy, it
would be difficult to manipulate the amount of truancy, i.e. to have two groups of participants, telling one
group to play truant and another group to attend school every day without fail. It is not ethical to
manipulate participants in this way.

Instead, the researcher can look for an association between the variables of educational performance and
truancy as they occur naturally.

Studies that are designed to look at associations of this kind are known as correlational designs, as we look
for a relationship between the two variables being studied. Unlike in an experiment, we do not need to
identify the IV and the DV, just the two covariables that we are looking for a relationship between.

Rather than stating an experimental hypothesis, a psychologist who is investigating whether there is a
correlation will create a correlative hypothesis, for example

“There will be a significant positive correlation between the number of hours per week a
student spends studying and the marks they obtain in their AS level examinations”

However, their disadvantage is that you can never be quite sure what caused the relationship between the
two variables, you can see that there is a relationship between them but we cannot tell whether variable 1
affects variable 2, vice versa, or even whether something we have not considered at all affects both
variable 1 and 2.

Thus, when a correlation is found there could actually be three possible explanations:
1) A causal relationship e.g. truancy caused low educational performance (or vice versa)
2) Both variables are affected by a third variable e.g. truancy and low educational performance are
caused by poor housing
3) The correlation occurs by chance – e.g. it is a coincidence that both truancy and low educational
performance occur together

For example, a perfect positive correlation is found between the number of storks in
Denmark and the number of babies born each month.
1) A causal explanation would be that the increase in storks causes an increase in
babies.
2) A third variable explanation would be that variations in the weather cause the
number of storks to vary. As it gets colder they fly away, as it gets warmer they
return. Babies are planned for the spring to give them a better chance of survival.
3) Or it could simply be a chance correlation.

From the point of view of science, the existence of a correlation between two variables can be very useful.
This is so because when a correlation exists, it is possible to predict one variable from information about
one or more other variables. The ability to make such predictions-to forecast future events from present
ones- is one important goal of science.

19
Strictly speaking a correlational study is not an experiment – cause and effect are not being predicted.
This is also true as in correlations, often neither variable is manipulated by the experimenter.

Many correlation techniques calculate a correlation coefficient, basically a measurement of the strength of
the correlation, a statistic which has a value on a scale between +1 and –1.

negative correlation Zero correlation positive correlation

(-1 = perfect negative) (+1 = perfect positive)
means that high scores on one There is no relationship means that high scores on one
variable are associated with between the two variables variable are associated with
low scores on another variable high scores on another variable

The results can be plotted on scattergrams, which always have a numerical scale for each axis, and should
be labelled clearly, with a title for the graph.

Activity 9 – Correlations
Describe the correlations between the variables plotted on these graphs:

Advantages Weaknesses
20
 Gives a numeric value to the strength of  it always raises questions of chicken and
a relationships between variables egg situations – does smoking cigarettes
 It can be used in natural settings as well makes one’s attitude to advertising to be
as in the laboratory; such research is more lenient, or does one’s lenient
known as field research. attitude to advertising cause one to
 The correlational method is often highly smoke more cigarettes?
efficient and can yield a large amount of  the relationship may be due to other
data relatively quickly. extraneous variables, for example height
 It can be extended to include many and IQ might be linked because diet
different variables at once. influences both.
 It can be used where experimental  The findings it yields are generally not
manipulation would be unethical or conclusive with respect to cause-and-
impossible. effect relationships. That is, the fact that
 It can indicate a trend which can then be two variables are correlated, even highly
used to guide experimental research correlated, does not guarantee that
there is a causal link between them-that
changes in one cause changes in the
other. Rather, in many cases, the fact
that two variables are correlated simply
reflects the fact that changes in both are
caused by a third variable.
 measurement of non-linear relationships
not possible (e.g. attention span and
time of day – relationship may be
positive early in day and negative late in
day – a non-linear relationship)

21
 Quantitative Data Analysis and presentation
Data is the results from research. Rather than presenting all of this in its raw form in your report we use
data analysis, descriptive and inferential statistics to summarise these results. This means that anybody
reading the report of the study will have a concise summary of the results and conclusions can be reached.

Descriptive statistics

Descriptive stats allow research data to be described and presented. It is not helpful to the reader to be
given raw data of a study but it is important that they have a summary of that data. This may take the form
of:
o A table
o A graph
o Numerical average

Measures of central tendency

1. Mean – when all scores in a group are added together and the total is divided by the number of
scores.

Eg. the results from a test (marked out of 50)

36 39 21 18 32 30 = 176
176 / 6 = 36
Mean = 36

2. Median – this is the central value in a set of scores

after they have been put in rank order:

Eg. 95 109 121 130 140 Median = 121

If there is an even number of scores take the mean of the two central values:

Eg. 95 109 121 135 140 180

121 + 135 = 256 / 2 = 128
3. Mode - this is the most commonly occurring value in a set of scores:

Eg. 1 2 3 3 3 4 4 4 5 5 5 5 5 5 5 5 6 6 6 7

Mode = 5

This handy rhyme should help you remember which of the measures of central tendency is which:

Hey diddle diddle, the Median’s the middle.

You add and divide for the Mean

The Mode is the number most commonly found,

The Range is the difference between.

22
 Activity 10 – Means, medians and modes

Find the mean and median of the following.

1. The % scores in a Sociology exam:

52 64 58 41

2. The number of library books borrowed by students in one year:

14 9 6 12 18 9

3. The amount spent on weekly shopping (£):

45 84 52 38 42 66

Find the mode of the following:

4. The heights of the teachers at Wimbledon College:

5”9 5”3 5”8 6”2 5”8 5”1 6”4 5”2 5”3 5”3 5”9 5”7 5”9 5”10 5”7 6”0 5”9 5”7 5”1 5”9 5”6
5”10

5. The weekly pocket money given to Year 8 students (£):

4 6 2 5.50 3 8 4 5 4 1 7 4.50 10 4 6 3 4 2 5 7 4 7 4 3 2 8

6. The numbers of videos rented in one year:

14 15 25 12 14 18 10 0 28 25 14 18 7 4 19 14 10 30 19

Measures of dispersion

Measures of central tendency (mean/median/mode) are used to summarise sets of numbers giving a score
which is representative of the set. However in order to give a fuller picture, we need to know how spread
out (how dispersed) the scores are.

The Range – this is the difference between the highest and lowest value.
e.g. if your numbers are 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
The range = 16 - 7 = 9

Standard Deviation – The average amount all scores deviate from the mean. This is the most powerful
measure of dispersion. You will not have to work out standard deviation (hurrah!) but you need to know
how it is done.

To calculate the standard deviation:

23
  The difference (deviation) between each score and the mean of those scores is calculated and then
squared (to remove minus values).
 These squared deviations are then added up and their mean calculated to give a value known as the
variance.
 The square root of the variance gives the standard deviation of the scores.

Score Mean d d2
6 10 -4 16
8 10 -2 4
10 10 0 0
12 10 +2 4
14 10 +4 16
Total 40

Mean of d2 = 8 (= variance)
Square root of variance = standard deviation = 2.8

This means that the average distance away from the mean is 2.8.

The higher the value, of the standard deviation, the more spread out the data is.

Activity 11 – Measures of Dispersion

Find the range for the following set of scores:

1. Age at which Ps claim to have ‘been in love’

21 19 22 18 25 21

2. Amount spent on travel, per day (£)

8.5 9.5 17.5 12.5 14.5

3. Scores obtained in age-related reading test:

9.8 7.1 4.2 8.4 9.9

What would it mean if the following S.D were found?

4. Condition A – S.D = 5.2, Condition B – S.D = 1.2

5. Condition 1 – S.D = 0.2, Condition B – S.D = 0.25

For your exam you will need to understand how these things are calculated and what they mean,
but you will not be directly asked to calculate them.

However, if you are asked to identify findings from a set of data, being able to calculate the
means, medians or modes and the range would be useful.

24
Presentation of quantitative data

Tables

The first thing a psychologist will do with the results they have collected is create a table
of their results. In your exams you could be presented with a table and have to interpret it.

The first thing you need to know about a table is what experimental design was used. This should be
outlined in the blurb given before the graph. You must know this so that you can state findings from the
table easily.

Independent Groups

Condition A Condition B
4 7
6 6
3 9
9 6
2 2
5 8
3 9

Repeated Measures

Ppt Condition A Condition B

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

As you can see, in the IG table, the rows that people are in are irrelevant, whereas in the RMs table, the
rows people are in are very important, as this is one ppts performance in two different conditions.

25
Graphs
Graphs are used to display data in a form which is easy to read. You will be familiar with different types of
graphs but it is important to know when to use each type and what they mean.

Bar charts – these use bars which do not touch and can be used with a variety of types of data, and will
usually be used to show a specified measure of central tendency, such as the mean, for the conditions in a
piece of research.

Scattergrams / Scatterplots / Scattergraph – This is where two sets of continuous numerical data are
plotted against each other, where a line of best fit can then be used to represent the data and identify
whether there is a correlation between the two variables on the axes.

Graphs must have:

 Clearly labelled axes which include units / maximum values
 Both axes starting at zero – otherwise the data will be mis-represented
 A title which describes the graph
 The IV on the X axis and the DV on the Y axis – this rule does not apply for correlations

Activity 12 - Graphs

Produce an appropriate graph to display the following data:

1. Results of a study into how age affects types of play:

Play Age (years)

1 2 3
Solitary 16 8 4
Parallel 3 9 7
Co-operative 1 3 9

2. Scores from Ps who were asked to rate speakers on intelligence on a scale

0-5 (where 5 was very intelligent) when speakers had Northern and Southern English accents.

Rating Southern Northern

1 0 0
2 5 6
3 12 23
4 26 21
5 21 10

26
3. Results from a correlation study to see if practice on a driving test improves
performance:

Number of Points
attempts awarded
1 27
2 54
3 78
4 105
5 120
6 149

4. Ps who sleep more than 7 hours per night (on average) in one year will gain higher marks in the final
A’Level exam than Ps who sleep less than 7 hours per night (on average) in one year.

The results for this study are as follows:

P Hours of sleep Exam mark

(average)
1 8.5 73
2 5.8 52
3 5 35
4 6.1 61
5 7.4 66
6 6.9 70
7 7.4 65
8 6.6 56
9 7.5 71
10 8.9 79

27
28
29
 Key terms in Research Methods

You must be able to understand each of these terms in the context of a piece of research. Test yourself
as you go along and for revision.

1. Bar chart 19. Natural experiment

1. Confounding variables 20. Naturalistic observation

2. Correlational analysis 21. Negative correlation

3. Demand characteristics 22. Non-directional (two-tailed) hypothesis

4. Directional (one-tailed) hypothesis 23. Null hypothesis

5. DV (dependent variable) 24. Opportunity sampling

6. Experimental/alternative hypothesis 25. Pilot study

7. Experimental Design 26. Positive correlation

8. Field experiment 27. Qualitative data

9. Independent groups design 28. Quantitative data

10. Interview 29. Quasi-experiment

11. Investigator effects 30. Questionnaire survey

12. IV (independent variable) 31. Random sampling

13. Laboratory experiment 32. Range

14. Matched pairs (matched participants) 33. Reliability

design 34. Repeated measures design

15. Mean 35. Scattergraph / scattergram

16. Median 36. Standard deviation

17. Mode 40. Validity

18. Mundane realism

30