Research Methods

Chapter 20 : Planning and Conducting Psychological Research

Scientific Research  Rooted in observation,
experimentation and knowledge
acquisition through a process of
objective reasoning and logic, testing a
hypothesis
 Empirical and measurable data is
gathered and replication possible
through the scientific method/
experiments
 Empirical evidence or data collected
allows researchers to draw valid
conclusions based on experiments

Non-Scientific Research  Ways of obtaining knowledge and

finding out why involves things like
common sense, philosophy and gut
feelings
 Not based on science and are often
labelled ‘pseudoscience’ to indicate
their lack of scientific foundation
 explain and even predict human
behaviour
 Example: in astrology movement of
stars predicts personality/ behaviour
Experimental  Involves variables that investigate a
cause and effect relationship, between
two or more variables that is whether a
change in one thing has an impact on
another
 For example: one might be interested
in finding out if eating a lot of pizza
(one variable) causes a change in
person’s weight
Non-experimental  Involves variables that are not
manipulated by the researcher and
instead are studied as they exist
 Involve observation: the collection of
data by carefully watching and
recording the behaviours of others
 For example: observation can be used
in classrooms if you want to conduct
research into children’s interactions
during collaborative learning activities
Sample A group of participants selected from a
population of interest
 It is important that participants in a
 sample are representative of the
population so the findings can be
confidentially applied to the population
 If the sample is biased and not
representative then the findings cannot
be applied to the population
 Used in experiments, observational
studies, interviews
Population A group of people belonging to a particular
category (have similar characteristics).
- In research it is a large group of interest
from which a sample would be drawn
Difference between population and sample Researchers refer to the large group of interest
to whom their findings apply to as the
population. The sample is a group which is a
smaller subset of the population. It is important
that the sample accurately reflects the
population of study at interest.
 For example: if a psychologist wanted
to investigate the relationship between
violence on TV and aggressive
playground behaviour in Western
Australia children aged to eight to 11
years. It would be impossible to study
all children in WA of this age so the
researcher has to obtain a sample of
children to represent the larger group.
The Scientific Method  Involves a systematic approach to the
planning, conducting and reporting of
research usually in the form of
experiments

Why do we use non-experimental design  Many variables of interest in social

methods psychology cannot be manipulated
because they are naturally occurring
and are generally unchangeable
variables
- Examples: personal characteristics such
as gender, socioeconomic status and
learning styles
 In some cases it is unethical to
randomly assign individuals to different
treatment conditions
- Example: You cannot observe the
effects of smoking by randomly
assigning individuals to a smoking and
non-smoking group for a given number
of years. Therefore the only ethical way
to investigate the potential effects of
smoking would be to identify a group of
 current smokers and a group of non-
current smokers and compare their
differences in current state of health.
Other variables should also be taken to
consideration
Examples of Non-Experimental (descriptive)  Case Studies
research methods  Surveys
 Archival Research
 Correlational Studies
Case Studies  An in depth, direct behavioural
observation of a single person,
individual or situation which can then
be used to related to the population
 Robson (2011) ‘ provide opportunity
the opportunity for detailed knowledge
about a single case or a number of
related cases’
 No fixed experimental design or any
hypothesis that guide the research
 Involve a single case or situation
(individual or group of interest)
 The case is studied in context and uses
a variety of data collection techniques
like observation and interview
 For example: the study of Phineas Gage
allowed us to learn the relationship
between parts of the brain and
behaviour
Surveys  Gather information about individuals,
can be conducted fact to face, over the
telephone or via the internet with tools
such as survey monkey can be written
or interview
 They provide quick answers about
people’s attitudes to and preferences
for various goods and services and are
often used to provide feedback to
manufacturers
 May focus on factual information about
the individual or it aim to collect the
opinions of the survey takers
 Long psychological surveys take time to
prepare and implement because they
involve careful planning and decision
making, well organised implementation
and coding and analysis
 Disadvantages to surveys: a low
response rate (not many people
complete or are prepared) and data can
be affected by the predisposition or
mode of respondent
  Examples: questionnaires with fixed
choice answers, 0-10 point Likert Scale
Steps in carrying out a small scale survey which 1. Developing the research question, the
produces valid and meaningful information study design (sample selection and pre
testing and main study) and the initial
draft of questionnaire = 30 days
2. Informal testing of draft questionnaire
= 5 days
3. Revising draft questionnaire= 3 days
4. Pre-testing the revised draft using
interviews= 3 days
5. Revising the questionnaire again
(revision of study design and main
study sample ) = 3 days
6. Carrying out the main data collection
interview= approx. 30 but depends on
sample size, telephone or via internet
and travelling time between
7. Coding the data and preparing data
files = 10 days
8. Analysing the data and writing the
report= depends on the size of data
and size of report
Archival Research  Study of old existing data which is
gathered, compared and analysed to
answer the intended research
questions
 Usually written material found in places
like libraries, universities, courts
 Often used by historians and writers
who reanalyse records to better
understand the events that happened
in the past
 Not as widespread in psychology
 Advantages: cheaper to use than collect
new data
 Disadvantage: time needed to find
appropriate data sets and obtain
permission to use them
 Examples: public documents and
official records held by hospitals or
private documents such as diaries and
letters
Correlational Studies  Look at the relationships between
different variables such as whether
boys and girls do better at maths in
school
 Correlation does not mean cause and
effect
 Research process in the same for
experimental studies (question and
 idea of the variables and what the
relationship is)
 Used in cross-sectional studies looking
at a number of variables at a single
point in time or used in prediction
times such as academic scores from
one time to another
Behavioural Variables (not independent or  CS investigate the relationship between
dependent) in Correlational Studies two or more variables that involve
assessing the degree and type of
relationship between these variables
- Often conducted when experimental
ones are inappropriate
- Example: investigating the relationship
between low birth weight and
academic performance in school, it
would be impractical to vary infants
birth weight so existing information is
used to explore the relationship
 Behaviour Variables are thus often
those that pre-exist and cannot be
carried out as an IV
 Relationships are then described as a
correlation coefficient which indicates
the strength of the relationship
- Score of +1= very strong positive
correlation
- -1 strong negative relationship
 Positive Correlation = high scores on
variable are associated with high scores
of the other variable
 Negative correlation= score is high on
variable and low on the other
 Example: correlation between the
amount of lead in a pencil and time
writing but the type of lead and
pressure by writer may also effect the
amount left

Seven Steps in Psychological Research 1. Identify a research issue or problem to

investigate: decision about the topic or
question to study and explore, might
test a theory, develop a theory or find
answers to a problem
2. Formulate a hypothesis: Making a
prediction about the relationship
between the two variables that can be
tested this can be made based on
previous work or a knowledgeable
hunch. The hypothesis is what you test
in and experiment
 3. Choose a research design and method:
Process/es used to collect data can be
experiment, observation, correlation,
survey, interview
4. Collect the data: Gathering evidence
through observation, surveys, recording
psychological responses or examining
archival files
5. Examine the evidence: use of statistical
techniques to analyse and summarise
the numerical data obtained
6. Interpret the results: The meaning of
findings. Is hypothesis supported?
Question answered? Findings apply to
whom? Suggestions for further
research?
7. Communicate your findings: Scientists
write reports/journals or talking such as
conferences or feedback to interested
parties so the theoretical implications
and practical applications of research
are known to working in the same area
as well as the general public and other
interested parties
Representative Sampling  Sample selection must reflect the
characteristics of the population at
interest and are being studied
- Studies with children samples must
reflect issues that may be important in
the interpretation of results such as
age, gender, school attended and
family structure
 Variables taken into account when
determine the type of sample and size
required if all variables are to be
studied
 RS the same as or equivalent to the
population from which it is drawn
Random Sampling  Ensures every member of the study
population has an equal chance of
being selected to participate in the
study to establish an unbiased sample
- Electoral Roll: identify all people in a
geographical area and names are
selected from the roll using a lottery
procedure that often involves a
numerical sequence
- A sequence of random numbers can be
generated and applied to a list to
obtain a sample
 Not easy to obtain so much research is
 limited in its ability to be generalised to
the population
Stratified Sampling  Important to ensure that a particular
group in a population of interest
appears in proportion representative of
that population
 Divides population to be sampled into
distinct subgroups or strata then
selecting a separate sample from each
stratum in the same proportions as
they occur in the target population
 Example: IQ, sex,religion, cultural
background, education
Convenience Sampling • Quick and easy way of selecting
participants
• Selected based on accessibility.
• E.g. at University, so many hours of
participation need to be completed
during undergraduate to be eligible for
4th year.
• E.g. local school, supermarket etc etc

Independent Variable  The independent variable is the

variable which is systematically
manipulated or changes in some way
by the experimenter in order to
determine its effect (on the dependent
variable).
 It is changed so that the experimenter
can see whether it affects another
variable and what those effects are.
 Can cause changes that might occur to
the second variable

Dependent Variable  variable which is measured and is the

variable expected to change as a result
of manipulation of the independent
variable.
 May or may not change as a result of
the manipulation of IV.
 Effect caused by the changes in IV
 Usually a numerical value
 For example: varying the presentation
time of an object on a screen to which
a person may respond quickly as
possible may lead to changes in speed
of reaction
- IV: variation in the time of object
presentation
- DV: speed of reacting (milliseconds)
- Give info about how quickly people
 react to changes in environment and
have implications for driving behaviour

Operational Hypothesis  Is a testable prediction that explains

exactly how the variables will be
measured and manipulated as well as
the populations from which the sample
is drawn. Therefore an operational
hypothesis is Workable, Testable and
Repeatable as it contains:
- Variables (IV and DV)
- Sample
- Measurement tool
 Guide research and predicted answers
to a proposed research question based
on a theory that provides basis for
predicted answer
 We test it by way of experiment can
never be fully proved as something may
occur to intervene that affects action of
IV
 A single negative effect can disapprove
a hypothesis
 Both cases refine experimental method
before discarding

Hypothesis  A general research prediction about the

direct interaction between the IV and
the DV and the population from which
the sample is drawn. That is, in the
presence of different levels of the IV,
does the DV increase or decrease?
 NULL hypothesis = predict that there
will be NO difference between
populations. IV doesn’t impact DV
 Experimental/ Alternative Hypothesis:
predicts that change or difference will
occur as a result of the manipulation of
IV

Controlled Variables  variables that are the same for both

the control and experimental groups so
that the changes observed in an
experiment can be explained by the IV,
nothing else.
 Example: same size computer screen

Uncontrolled Variables  those that don’t need to be controlled

(can stay random) as they would not
have any impact on the results
  For example: different eye colours of
participants do not have an effect on
the results
Extraneous Variables  example: red bull, drugs, coke, stress
which all elevate heart rate)
 is a variable other than the IV that can
influence the results (DV).
 Extraneous variables are unwanted
and should be identified and
controlled if possible

Experimental Group  The group that is exposed to the IV

(the experimental condition) in an
experiment.
 The purpose of the experimental
group is to determine the effect of the
IV on the participant’s response.

Controlled Group  The group that is not exposed to the IV

(control condition), but treated exactly
the same as the experimental group in
all other respects.
 It provides a standard against the
behaviour of the experimental group
can be compared in order to assess if
the IV has had an effect or caused a
change on DV
 Needs to be as similar as possible in
characteristics with the experimental
group (such as age)
 Groups are treated the same except for
the application of IV to experimental
group
 Sometimes the same people act as both
the experimental and control groups =
experimental and control condition and
the results are compared under the
two conditions
Experimental Group The group that is exposed to the IV (the
experimental condition) in an experiment. The
purpose of the experimental group is to
determine the effect of the IV on the
participant’s response.

Placebo  Defined as a dummy pill or harmless

substance given if It were a treatment
 Substance with no known medical
effects as it is a neutral substance =
sugar pill
 Sometimes given to a control group
that is blind to its status and thinks it is
 receiving treatment
 For example: in medical research two
groups were each given pill;
experimental to the researchers the
other is the placebo
 Used to control the effects of
participant expectation

Placebo Effect • Refers to the phenomenon in which

some people experience some type of
benefit after the administration of a
placebo.
• Participant expectation is a form of
extraneous variable
- Can potentially affect both the changes
in the DV and outcome of IV
 Why does the placebo effect occur?
- Personal characteristics, self-
presentation or desire to look good/
perform appropriately and them
demand of what it expected from the
experimenter
- Expectations of the patient can play an
important role in the placebo effect;
the more a person expects the
treatment to work, the more likely they
are to exhibit a placebo response.
- E.g. some patients in a medical study
administered a placebo while other
participants receive the actual
treatment. This is to determine
whether or not the treatment has an
actual effect. If participants taking the
actual drug demonstrate a significant
improvement over those taking the
placebo, the study can help support the
claim for the drug's effectiveness.
-

Single Blind Procedures • When the participants do not know

whether they have been assigned to
the control or experimental group
• Placebos are used and participants
unaware if they have real treatment or
placebo
• Reduces the effects that participants
expectations may have on the results

The Experimenter Effect  Experimenter’s own personal variables

as well as his or her expectations and
 behaviours that may bias results due to
inaccurate observation, recording or
interpretation of data or there may be
simple bias in the way the experiment
is presented to participants
 The experimenters behaviour, body
language, verbal cues and prompts,
preferential treatment of one group
over another etc may also influence the
results of the study!

Double Blind Procedure • Experimenter effect!! To eliminate and

increase reliability and validity of
results = Use Double Blind procedure …
• Participants and experimenters do not
know which participants have been
allocated to the control and
experimental groups.
• This involves a 3rd party to be privy to
which group is receiving the
experimental treatment, however they
are not directly involved with the
participants in any way and cannot
have any influence over them.
• OR computer may also be used.

Reliability  Conclusions are consistent or stable

over time
 Reliability of a measure is the extent to
which you would get the same result if
the same measured were to be given to
the same individual again under the
same circumstance
 Example: taking the same IQ test
multiple times and get the same score
 Statistical measures of reliability are
often used in research studies and
reliability scores are used in training
manuals
Types of Reliability Internal Reliability
- Assesses the consistency of results across
items within a test.
- Tested via Split Half method - Score on one
half of the questionnaire should
match/correlate with score on the other half of
the questionnaire.
External Reliability
- Refers to the extent to which a measure varies
from one use to another.
- Tested via Test Re-test method - Scores
should be the same on questionnaires
 administered some time apart

Validity  Refers to the extent to which the

results reflect what the measurement
instrument designed/claims to measure
 Measures what it claims to measure so
accurate conclusions can be drawn
 Example: much of the debate in the
measurement of personality relates to
the validity of the tests designed to
measure it
Types of Validity - Face – If a test appears to measure
what it claims
- Construct – If a test or tool measures
the constructs (exact topic) that it was
designed to measure
- Concurrent – If the test
correlates/compares with a previously
validated measure/test.
- Predictive - Is the extent to which a
score on a scale or test predicts scores
on some criterion measure.

Cross Sectional Research  A research design that compares

groups of people at different ages at
one point in time.
 Used in developmental psychology to
overcome the problem of children
growing and developing and showing
changes in behaviour as they come
more socially, emotionally and mentally
skilled
 Method takes into account age related
developmental changes by comparing
children of different ages
 For example: Kohlberg’s study of
presenting dilemmas to children of
different ages were asked to respond to
the issue= older children more likely to
reason at a higher level than younger
children
 Limitations: in the above study we
cannot predict if the children in the
youngest age group will reason like the
children in the oldest age group when
they are that age
- Participants at each age group are
different people who come from
different cohorts
  Advantages: Method is frequently used
because it is quick and easy especially
when there are time and financial limits

Longitudinal Methods  A research design that follows the

development of a group of individuals
across their lifespan, testing them at
different points in time.
- The total time period may be very brief,
over a period of months but in
developmental research is it often over
a period of years
- Example: Ann Colby and her colleagues
followed the moral development of 36
males over 20 years
 Individuals were followed across time
we do not have to be concerned about
possible cohort effects
 Limitations:
- Expensive
- Researchers also run the risk of losing
participants as they move or die
- Practice effects can distort the findings
as the question of participants getting
better because they have carried out
the same task numerous times?
- Cross-generational problem: drawn
from one cohort and may have
different experiences at each point in
their lifespan from those children in
earlier of later generation

Longitudinal Sequential Design  Combines features of both the cross-

sectional and longitudinal design in an
attempt to overcome the limitations of
each
 Groups of participants are followed
over time but at each measurement
point a new group is added that is the
same age as the first group in the first
measurement point
 Allows us to look for changes in
individuals across time as in
longitudinal design
 Lets us look for age differences in
behaviour as in a cross sectional design
and it lets us check for cohort effects
 Limitations: complex and expensive and
requires a well-organised research
team to track the data

Qualitative Data  Techniques that use words or texts to
describe or summarise what people
feel, think, say or do in relation to
particular entities to reflect their
attitudes.
 Cannot be analysed statistically
 Include: observations, recordings,
notes, self-reports, INTERVIEWS

3 Types of Interviews  Interviews are the most common

method when it comes to collecting
qualitative data usually involves a
researcher asking a series of questions
and receiving answers to these
questions
 STRUCTURED INTERVIEW :
- Involves pre-determined questions in a
pre-set order with fixed wording
 SEMI-STRUCTURED INTERVIEW
- Involves pre-determined questions but
the order of asking them can be varied
by the researcher depending on the
circumstances and the answers being
elicited
- Wording of the questions can be varied
and explanations given if needed
- Questions can be omitted or even new
ones added if required
 UNSTRUCTURED INTERVIEW:
- Involves a conversation around a
researcher’s general area of interest
- It is informal and casual but often in
depth
Qualitative Interviews are suitable in the  Study focuses on the meaning of a
following circumstances particular phenomenon to the
participants
 Prospective processes are to be
introduced for example: in an
organisation perceptions of possible
change are needed
 Study wishes to obtain historical
perspectives on a phenomenon
 Explanatory work is required prior to an
experimental study to ascertain the
validity of research
 Qualitative data are required to
validate the results of a quantitative
study and help elucidate interpretation
of the results
Data DATA - information that is collected in
research.
 Data can take different forms. It can be
expressed as written descriptions or in
numerical values.
Two types:
1. Qualitative
2. Quantitative (subjective and objective)

Quantitative Data - Numerical information about the

quantity or amount of what is being
studied.
- The use of numerical data makes it
easier to summarise and interpret
information and therefore it can be
analysed statistically.
- Can be either:
a. Subjective Data
b. Objective Data

Objective Quantitative Methods  the results are based on facts and do

not influenced by personal
feelings/opinions
 Provides actual data and is not bias or
controlled directly by participant.
 Example: brain waves, heart rate, body
temperature and conductivity of the
skin
Subjective Quantitative Methods • Rely on self reported measures
provided by the participants which are
all based on personal
feelings/opinions
Rely on opinion and the participant
being honest
• eg. Checklists, rating scales
• Limitations:
- Relies on people being honest!
- Often not possible to gather subjective
data:
- People with a disability
- Young children

Rating Scales  Regarded as versions of questionnaires

or interviews
 Provide a scale on which an individual’s
standing on an issue can be measured
and they are commonly used for
attitude measurement
 Commonly have a range of statements
typically used to which a person rates
 their attitude
 The statements/ items are carefully
composed so they are related to the
attitude being measured and they also
need to have the assignment of
numbers to particular answers so that
meaningful conclusions can be reached.
 Most Common: Likert Scale: where a
pool of items or statements are related
to the issue your are researching and
reflect both positive and negative views
on the issue. The responses can be
categorised through:
o 1=strongly disagree,
o 2=disagree,
o 3=neutral,
o 4=agree,
o 5=strongly agree

Role of the Experimenter  In experiments researchers tend to

work with aggregated data: results
represent the group average rather
than the individual scores
- Non- experimental studies: report
averages or percentages. Group
averages enable researchers to identify
trends and patterns that help describe
and predict behaviour although they do
not capture individual complexities
- Single Case studies: are an exception as
they report quantitative data or the
behaviour of a single person
 Experiments assume that the
experimenter will be objective to
ensure that the researcher has no
effect on the behaviour being observed
or recorded on the results
- To counter the experimenter effect it is
common practice for researchers to
acknowledge any expectations and
potential biases and to put in place
strategies to minimise these as far as
possible through physical and emotion
distance from the study = this is easier
in experimental studies than in
qualitative interview studies.

Participants Rights  Privacy

 Anonymity
 Confidentiality
  Voluntary Participation
 Withdrawal Rights
Privacy  The right of protection from unwanted
intrusion by the government or other
people into one’s personal affairs
 Refers to the collection, storage and
sharing of personal information
 Privacy Laws in Australia protect any
personal information given to a medical
practitioner in the course of seeking
treatment to maintain privacy =
Commonwealth Act of 1988
 Light Touch: privacy legislation is not
controlling and deals mainly with
collection, storage and use of personal
information and a person’s right to
amend the factual details if they are
incorrect
Anonymity  The protection of people’s identity
through not disclosing their name or
now knowing it
Example: Survey Collection: the people
complete the questionnaire often do not put
their name or any other identifying information
on them.
Confidentiality  Resides in the relationship between a
professional (such as psychologist) and
their patient and refers to the degree
of secrecy attached to the information
given by the patient of client
- Based on trust and is a cornerstone
legal and ethical concept in the
establishment and maintenance of a
therapeutic relationship
- Based on the understanding that
information given by the patient will
not be disclosed to anyone unless the
patient consents to the disclosure and
under exceptional and usually legal
circumstances
- Similar provision applies to data
collection particularly on health
conditions where there is the potential
to identify participants
Informed Consent  Those taking part in any study should
know why it is being carried out and
what they will be expected to do so the
nature and purpose of the experiment
 If explaining the true purpose before
the study is carried out defeats the
purpose of the study then the
 researcher must make sure participants
do not experience any distress and they
must be fully informed of the purpose
of the study after it is completed
 Must be appropriately documented
 Participants who cannot give informed
consent because they are too young
(U18) or do not have the intellectual
ability to understand what they are
being told then consent must be
obtained from those legally responsible
for them.
Voluntary Participation  Participants must agree to participate
in research studies of their own accord
 Participants must not be forced into an
experiment or be coerced into
consenting to participate through
bribery, offers or threats such as failure
on a course
Withdrawal Rights  Participants must be informed that
they have the right to withdraw from a
study at any time and for any reason
(with out questioning), during the
research if they wish to at any time
without penalty and will suffer no
disadvantage and do not give a reason

Deception in Research  In some circumstances researchers may

use deception in an experiment where
they hide the real reason for research
in order to reduce the likelihood that
participants will behave differently
 Ethics committees ensure that
deception is only used when necessary
and that participants will not be
stressed/distressed or embarrassed by
deception
Debriefing  At the end of a study the researcher
must inform participants about the true
nature, purpose results and conclusions
of the study to correct any misleading
ideas and impressions they have about
the study
 Researcher must also provide
information on the services available to
help treat any stress resulting from the
study
Professional Conduct  Any psychological, medical or scientific
research in Australia must comply with
the ‘National Statement on Ethical
 Conduct in Human Research 2007’
(Updated May 2015) and the
‘Australian Psychological Society’s Code
of Ethics’
 Both documents explain the
professional code of conduct required
of psychologists as they undertake
research in Australia
 Both documents make clear that
scientific research with humans in
beneficial and should not be of any
hard as the benefits should outweigh
any risks and therefore the welfare of
humans must not be compromised
 All institutions where research with
humans in undertaken have Human
Research Ethics Committees which
analyse research proposals and
approve all research involving humans
to ensure no harm is done
- Detailed applications submitted with
the benefit and risks need to be
justified before approval
- The same/different committee will also
deal with applications to conduct
research with animals which also
governed by various codes of ethics.

Ethics  Moral principles and codes of

behaviour that apply to all
psychologists regardless of which field
they work in and are important to
ensure the protection of the
participants and must be followed
 Different countries have different
customs and guidelines

CHAPTER 21: Processing and Evaluating Psychological

Research
Why Research/ Experiment  To discover if there is a casual
relationship between variables
 Does the IV cause a change in the DV?
 We need to conduct analysis of the
data draw conclusions about the
hypothesis
  Results and conclusions need to be
explained as to why were they
obtained and what do they mean?
Statistics  Researchers used statistics to analyse
quantitative data these are
mathematical procedures used in
psychological research to summarise,
describe and interpret results
 Descriptive Statistics: used to
summarise and describe results
- Allow numbers to be organised and
understood and represent
important aspects of data with a
single number
e.g: mean, mode, median
 Inferential Statistics : used to indicate
whether the results obtained are
meaningful/ significant and if the data
supports the hypothesis being tested
Data Analysis  Needs to be considered at the
experimental design stage so that
meaningful data are collected and
subsequent analysis are made as simple
as possible
Data Sets  Are usually organised into rows and
columns where the rows represent a
case and the columns are the variables
under different conditions of their
responses
 Each separated entry is a cell
containing data and the data set is a
matrix
 This can be done:
- Automatically if the experiment is
being conducted and data collected
via a computer
- Through the creation of a file that
can be imported through a
software program
- Or analysed by hand through a
simple manual entry
 In most cases before formal statistical
analyses, initial exploration of the data
is useful as it gives you a feel for the
data and what they represent
Methods of Displaying Quantitative Data  Tables: used to organise data in
columns and rows.
- Tables should always be numbered,
have a title, and each column and
row should be identified using a
sub-title.
  Graphs: also used to display data.
- Graphs should always have a title
and both axes be clearly labelled.
Units should be used where
appropriate.
 Diagrams: are images and should have
a title be labelled and drawn in pencil
 All of these visually summarise data
Frequency Table 1. List the values and organise them
from lowest to highest
2. On the other side right the amount
of times a particular value occurs
( the frequency)
Bar Graph  Bar charts/graphs are used to display
discrete data (NOT continuous)
 x-axis :
 data represented in distinct
categories
 bars are separate (don’t touch)
 y-axis: (height)
 the frequency of each category
 the bars can be either
horizontal or vertical.

Histogram  Shows the frequency of a particular

score (or range of scores) in a set of
data.
 X-axis:
 individual scores or intervals
 the bars touch
 data is continuous
 data is usually numerical
 Y-axis: (height)
 the frequency of each score
 Data for more than one group of
participants can be described on the
same histogram.

Frequency Polygon  Graphs the frequency of

scores/responses using a line
 X-axis:
 individual scores
 Y-axis:
 Frequency
 Dots are plotted at the intersection of
the X and Y axes to indicate individual
scores.
 Could graph:
 The amount of time you spend
 sleeping, being at school, watching tv,
doing homework, and other
Single points are placed for total values and a
line is drawn between the points
- Differs from line graph as it only
shows information about frequency
(how often something occurs)
- Advantage over histogram: several
sets of data can be used in one
graph as a legend can show what
different lines represent
Pie Chart  Pie graph: circular and shows the
proportions of values or scores for
different categories of data
 Is best used to compare different parts
of the same whole

Line Graph  Represents the relationship between

two variables in an experiment.
 X - axis :
 Has the IV plotted on it
 The value increases from left to
right.
 The variable on x-axis is
continuous
 Y - axis :
 Has the DV plotted on it

Diagrams Should:
 have a title
 be labelled
 be drawn in pencil

Measures of Central Tendency  Mean – Average: add all scores

together, and divide by n
 Median – The mid-point of a set of
ordered scores
 Mode – The most frequently occurring
score in a set of scores
 Range – the difference between
highest and lowest score and describes
the spread of values
- not used very often as it is a crude
amount and does not take clumps
of scores into account although it is
used in measures of dispersion

Measures of Dispersion  Frequency Distribution - Looks at the

results and shows how they are spread
across the possible results.
**Normal curve – display of the ideal
 distribution of variables, based on the
mean and standard deviation.
 Variance – the spread of variable
around the mean.
 Standard deviation – square root of
variance. AVERAGE spread of variable
around the mean.

Distribution  Normal Distribution – a bell shaped

curve
 Positively Skewed Distribution – scores
cluster around the lower end of the
graph
 Negatively Skewed Distribution –
scores cluster around the higher end of
the graph
 Bimodal Distribution – the data have
two modes

Variance  Describes spread of scores around the

mean. Lower variance = limited spread
 Calculated:
1. Subtract the mean from each score
to give each score’s deviation score
from the mean
2. Multiply each deviation score by its
self
3. Add up all the squared deviations
4. Divide the total by the number of
scores
 A variance of zero indicates no spread
 Not often used since we do not usually
think about the spread of scores in
terms of their squared value we usually
think about the scores as the values we
have been given

Standard Deviation  describes the spread of a group of

scores and represents the average
amount by which scores differ from the
mean
 Square root of the variance
 Calculated by:
1. X= each score and M = Mean
2. The score minus the mean (x-m)
3. This calculation is squared
4. All of these scores are added
together €
5. Divided by the number of scores
 (N)
 Used in psychological research as an
expression of the distribution of scores
Normal Distribution (Bell Curve)  Many psychological tests assume a
distribution where the majority of
scores are in the middle of the
distribution and fewer are at the
extremes
 Describes the data in a better
description then the range
 A normal distribution because of its
standard shape of a bell curve can
assume the mathematical properties
with:
- 34% of values fall 1 SD above and
34% of values fall 1 SD below the
mean
- 14% of values then fall between 1
and 2 SD’s above and below the
mean
- 2 % fall between 2 and 3 SD’s
above and below the mean
- There fore 68% of scores are
between 1SD above and below the
mean making a very useful
statistical property
Variability  Variability – The spread of scores
 Measure of Variability - indicates how
widely scores are spread around the
central point
 Low Variability – tightly packed spread
 High Variability – wide spread

Probability  PROBABILITY – indicates the likelihood

of an event occuring. Higher the
probability the more likely something
will occur. Always between 0-1

Statistical Significance  Refers to the significance of the

difference between two scores
 That is, whether we can attribute the
results to the IV or to merely chance
alone.
 Provides statistical proof of a causal
relationship, yes the IV does cause the
DV to change!
 Something which descriptive stats do
not do!
 Indicates whether results are due to
chance or occurred for a reason. The
closer a value is to the mean the more
 statistically significant it is. If there is a
greater that 5 % chance of a event
occurring at random then the results
are not statistically significant
 Statistical Significance is represented by
a “p value”
 If P ≤ 0.05 = statistically significant

P Values < is less than

> is greater than
≤ is less than or equal too
≥ is more than or equal too
 p ≤ 0.05 means that for the results of a
study to be statistically significant, the
probability that the results are due to
chance alone must be less than 5%.
- Results are statistically significant
- Results are highly likely to be due
to the IV
- Hypothesis accepted; conclusions
may be drawn.
 If the p value is > than 0.05 means the
probability that the results are due to
chance alone most be MORE than 5%
- The results are not statistically
significant and the results are likely
to be due to chance and not the IV.
- Therefore the hypothesis IS
rejected! No conclusions can be
drawn.

The Concept of Statistical Significance and  In Normal Distribution we know that

Probability around 2% of the population for a
particular behaviour under examination
falls between 2 and 3 SDS above the
mean
- This results in a probability of 0.02
or 2% that the score will fall in their
region for a single person
- There is however a probability of
0.68 that the score of a single
person will fall between 1SD above
and 1SD below the mean = high
probability
 When testing for SS these are the
comparisons we make between the
distribution of scores and that of the
wider (assumed) population
- If More extreme results (at either
end of distribution) than those
 assumed by a normal distribution
or by chance than we conclude our
results are statically significant
 Also used to indicate whether
difference between a control and
experimental group is real that is due
to the independent variable and not by
chance
- Researchers use tests of statically
significance to compare the mean
scores of the groups and in general
accept that difference is due to the
independent variable when the
probability is due to a chance in 5
time or fewer in 100 repetition of
the study. This would be a result in
a pvalue of 0.05 (5%) that the
difference in scores was due to
chance
Correlation  Non-experimental research method
used to investigate the relationship
between 2 or more variables.
 It does not tell us if one causes the
other to occur but it tells us if there is a
relationship between the 2 variables.
 Measures of correlation are referred to
as correlation coefficients which
describe the strength and direction of
the co-relationship between two
variables

Correlation + and -  Positive correlation – Two variables

move in the same direction
 That is, they both increase or
both decrease
 For example, chocolate and Ms
Keating’s happiness
 Negative correlation – Two variables
move in opposite directions
 That is, one increases while the
other decreases, and vice versa
 For example, marking and Ms
Keating’s happiness
 The Correlation Coefficient shows the
strength and direction of the
relationship (between +1 and -1)

Correlation Coefficient  The closer the correlation coefficient is

to -1 or +1, the stronger negative or
positive the relationship is
 (respectively)
 The closer the correlation coefficient is
to 0, the weaker the relationship is:
- Strong positive relationship: r
between 0.75-0.99
- Moderate Positive= 0.5 and 0.74
- Weak positive = 0.26 and 0.49
- No relationship= -0.24 and +0.24
- Weak negative= -0.25 and -0.49
- Moderate Negative= -0.5 and -.74
- Strong Negative= -0.75 and -0.99

Scatter Plots  A scatter plot is a graph of scores

showing correlational data
 The spread of the dots indicates the
strength of the relationship.
 The closer the dots, the stronger the
relationship
 The slope of the dots indicates the
direction of the relationship.
 Sloping up shows a positive
relationship
 Sloping down shows a negative
relationship

Types of Correlation  Bivariate Correlation – when two

variables are considered
 Multivariate Correlation – when more
than two variables are considered

Validity .Validity – the degree to which the variables

measure what they are intended to measure,
and so lead to an accurate conclusion. For
something to be valid, it has to actually
measure what it claims to measure.
- For example, a test that claims to
measure self-esteem but actually
measures a liking for the colour blue is
not really valid.

Reliability  Reliability – the quality of a measure

relating to its relative consistency in
measurement. If something is reliable,
it means that the result we obtain is
repeatable.
- For example, if we took a test that
measured our IQ, for example, and the
first time you took it your IQ was 110
and the next time you took it your IQ
was 150, the test is probably unreliable.
Sources of Error and ways to Reduce 1. Chose the correct sampling method.
 Methods/Options
2. Use of placebo reduces the placebo effect
3. Double blind studies reduces experimental
bias
4. Randomized controlled trials

Sampling Methods  Random Allocation of participants in

groups - randomly allocate numbers to
participants = lotto draw ensures each
participant has an equal chance of
being selected to be in experimental or
control group and also makes all
possible combinations of people for a
particular sample size equally likely to
reduce differences between
experimental and control groups
 Stratified random sampling – a way of
selecting participants for a study by
dividing them into groups (stratas)
based on characteristics (eg. Sex –
males and females) and then randomly
allocating them
 Proportionate sampling – different
sample sizes based on representation
of groups in the population
 Disproportionate sampling –
oversampling of small/rare groups to
ensure there is a some representation
of them in a study
 Snowball sampling – researcher
identifies participants of interest for a
study. Participants are interviewed and
then pass on details of other people of
interest in the population. Used in
qualitative research
 These sample techniques reduce
sources of data error and reduce
likelihood of error when generalising
the results from sample to wide
population

Evaluating and Improving Research  Evaluation of Research at whatever

level can lead to improvement in
research weather it is in the design and
the variable included, analysis of data
or interpretation of results
Before any research is conducted a proposal
must be developed to get approval for the
research.
 Thought must go into considering prior
 research and then a plan to make
improvements for future research must
be made. When planning for future
research considering factors such as the
below is important:
- The aim and significance of the
research you want to conduct
- The research method and design that
will be used
- How the analysis and interpretation of
the data will occur
- The discussion implications and
applications of research
 Evaluating research systematically and
critically better research can be
developed thus ensuring psychology
remains rooted in scientific tradition
but also contributing to our knowledge
of how behaviour and mental processes
work and how to predict human
behaviour
Peer Review of Applications for Funding  Conducted by experts in the same field
of research and are familiar with the
theory, content, method, analysis and
interpretation of data
 People whom the researcher has
collaborated with or competitors
 Recognised way of getting one’s
research evaluated and funded or
published