Extraneous variables Independent variable Key words

The IV is a variable (factor) that is directly

These are factors within an experiment that may interfere with the IV and DV. For manipulated by the researcher. For
example when testing the influence of age on memory anything that interferes with the example when testing the influence of age Independent variable – manipulated
effect of age on memory would be classed as an extraneous variable. For example, on memory, age is the IV because it is what Dependent variable – measured
noise, reading ability of participants, light in the room etc. There are 2 types of you are changing in the conditions of the
extraneous variable: situational or participant. Operationalise – Making the IV and
experiment.
DV specific to the study
Extraneous variables – Factors which
Dependent variable could influence the results of a study
The DV is a variable (factor) that is directly Confounding variables – An
Confounding variables Situational variables extraneous variable that will effect
measured by the researcher. For example
If you don’t control extraneous variables These are a type of extraneous variable when testing the influence of age on the results
which may effect the cause and effect memory, the DV is the memory as that is Situational variable – An extraneous
they become confounding variables. This
relationship between the IV and the DV. what you are measuring. This may be done variable from the environment
means that they will or have effected the
These are factors from the environment through the number of words recalled.
results. Confound means damage, so they Participant variable - An extraneous
have damaged the results. such as noise, light levels, temperature etc.
An example may be ‘order effects’ this is a variable from the person
situational variable which means that the You must remember……. Demand characteristics – When the
order in which participants did the ‘The IV is manipulated and the DV is participants behave in a way they
Demand characteristics experiment effected the outcome, for measured!’ think is expected of them
example if you are doing an eating test and Extraneous variables may effect the results Order effects – When the order of
These are a type of situational variable in you eat a bowl of chips and then are asked of an experiment, confounding variables the study effects the results
which the participant will change their to eat a bowl of cereal second to see what will effect the results.
behaviour to meet the needs of the study, fills you up better, by the time you eat the
for example they may have deciphered the cereal you are already full from the chips so Participant variables
true aim of the study and behaviour in a it’s a bad experiment!
way they think the experimenter wants These are factors associated with the participants that may interfere with the results. This
them to behave. This makes the results less could include IQ, personality type, memory ability, life experiences etc. The researcher only
valid. This can be made worse from needs to consider what participant variables are likely to influence the findings of the study,
Investigator effects as this is when the Standardised procedure e.g. driving ability is unlikely to effect the results of a memory tests but if the memory test is
researcher may unitentially give the reading and remembering words then reading ability will influence the results. To counter act
The procedure of the study is kept the
participant clues about how to behave e.g. this the experimenter should screen all participants before and ensure they have the same or
same across all conditions to reduce
nod when they give the correct answer. similar reading ages.
extraneous variables.
 Hypothesis Null hypothesis Key words
Every study has an experimental hypothesis A null hypothesis is a prediction that the result will find no or Hypothesis – A predicted outcome of a study
and a null hypothesis. Hypothesis are little effect. They usually start with ‘There will be no significant Null hypothesis – A prediction that there will be no difference or
different from a research aim because the difference between’ and end with ‘Any difference found will relationship between the IV and DV
aim shows the areas of interest for the study be due to chance’. Experimental hypothesis – A hypothesis used in an experiment
whereas the hypothesis states the predicted
that suggests that there will be a relationship between the IV and
outcome of a study.
the DV and the IV will cause an effect on the DV
PRECISE AND TESTABLE STATEMENT OF THE Experimental (Alternative) hypothesis
RELATIONSHIP BETWEEN TWO VARIABLES. Directional hypothesis – An experimental hypothesis that suggests
A hypothesis is a statement about what is the direction of the results, e.g. an increase or decrease
Just to add to confusion an experimental hypothesis is
being tested and involves things that are sometimes called an alternate or an alternative hypothesis. If Non – directional hypothesis – An experimental hypothesis that
measurable such as the IV and the DV. it is an experiment (such as lab or field) it is an experimental suggests there will be a difference but doesn’t state what the
For example, there will be a significant hypothesis, It is called an alternate when it is a survey or difference will be.
difference between younger and older interview (not an experiment). This is when we state the
people in performance of a memory test. expected outcome of a study, where there will be a difference.
If the study is a correlation it does not look This hypothesis is usually written based on a theory.
for a difference instead it looks for a
‘relationship’ and the word relationship
should be used in the hypothesis. A directional (One tailed) hypothesis
This is a type of experimental hypothesis that clearly states
the predicted direction of the experiment, for example there
will be a significant increase in the number of words recalled
A non - directional (Two tailed) from psychology students then art students. This is directional
hypothesis because of the word ‘increase’ which states the direction.

This is a type of experimental hypothesis that

clearly states there will be a difference but
does not state what the difference will be. It Test yourself
still states that a difference or relationship
will be found, but does not specify what that Create a directional and a non – directional hypothesis as well
difference will be, for example: there will be as a null hypothesis for a study with the aim of seeing the
a significant difference in the number of effect of hair colour on IQ score.
words recalled from psychology students Challenge – How would you test this? Can you design an
then art students. experiment for it?
 Random sampling Stratified sampling
Truly random sampling only occurs when every member of the target population Involves dividing the target population into important subcategories (or strata) and then selecting
has an equal chance of being selected. Each individual is chosen entirely by chance members of these subcategorise in the proportion that they occur in the target population. For
and each member of the population has a known, but possibly non-equal, chance example, if a target population consisted of 75% women and 25% men, a sample of 20 should include
of being included in the sample. For example, putting names of every member of 15 women and 5 men. For example, suppose a farmer wishes to work out the average milk yield of
the target population into a hat and pulling a sample out (without looking). each cow type in his herd which consists of Ayrshire, Friesian, Galloway and Jersey cows. He could
Strengths divide up his herd into the four sub-groups and take samples from these.
Should generate a representative sample because each member had equal chance
of being selected, this makes it more generalisable. Strengths
This helps to control participant variables giving the study more validity Ensures that the sample is totally representative of the target population.

Weaknesses
Participants who are selected may not agree to take part which would make the Weaknesses
study less representative Time consuming and people can still refuse to take part, this can lead to a sample error which can lead
Not possible unless the target population is very small. to invalid conclusions being drawn.

Volunteer sampling
Opportunity sampling Target population and sample
Consist of those individuals who have consciously or unconsciously Simply involves selecting those subjects that are
determined their own involvement in society, in other words they around and available at the time, an effort may be A target population is the group of people that an
volunteer. For example, studies or passers by who become made to not be biased in selecting particular types investigation is concerned with or wishes the findings of the
involved in field studies ie, in bystander intervention studies. of subject. This may simply consist of choosing the study to apply to. Because often it is impossible to study the
first 20 students in your college canteen to fill in whole target population we take a sample. For example I cant
Strengths – your questionnaire. For example, university study every 15 year old in the UK but I can take a smaller
Involves minimal effort from the researcher psychologists may sample from their own students. sample of 40 different 15 year olds from different schools and
Ethical as participants have fully consented to the research study them. The sample chosen should represent the target
Quickest and simplest technique population and thus allow the findings of the study to
Weaknesses – generalise to all the target population.
Less representative because not everyone will see the advert or Can lead to biased sample as not everyone will be
A sample Is a selection of the target population that is directly
can respond available so not generalisable to the target
studied in an investigation.
Can lead to a bias sample as volunteers tend to be a type of person population.
Generalisability – This is the extent to which the results of a
that want to take part. This may mean the findings are unreliable. Unreliable as replicating the study may not find the
study represent the whole population not just the sample
same results.
used.
 Experimental and research Independent measures Repeated measures design Matched pairs design
designs design
When designing research, experimenters This involves splitting the This is when the same participants complete all This uses different people in each condition
must chose a research design. This describes participants in to groups and conditions of the study, for example all participants but tries to use similar participants to
how participants are used within the study testing each group separately on do both conditions; such as if an experiment was eliminate participant effects. This means
and in what conditions of the study are one condition. Sometimes this is testing the effect of time of day on a memory task, when they recruit participants they would try
tested. An experimental design is the name the only option, e.g. if testing the all participants would sit the test at 9am and then and recruit similar participants, e.g. they may
given to a research design when used in an effect of age you would have to the same participants would sit the same exam at match them all for age, IQ score, similar
experiment. There are 3 types of do 2 groups as one person cant 5pm and the researchers would compare the backgrounds etc.
experimental design. be 2 ages at the same time. scores to see if time of day had an effect.

Strengths Weaknesses Controlling problems Key words

There are no order effects as

Order Effects: An effect that can occur when a repeated measures
Independent Any differences between Participants can be
measures participants only take part in one conditions could be due to randomly allocated design is employed. If the participants always complete one condition
design condition so cannot get better through individual differences in to each condition first, by the time they get to the second condition they may
practice, or under-performance due to participants, for example one
fatigue, or change their behaviour due experience order effects, such as practice, boredom and fatigue. This
group could do better on recall
to demand characteristics.
because they are more
could then affect their performance in the second condition.
It allows task variables to be controlled
for example participants can be given motivated or more intelligent. Demand characteristics - refers to an experimental artefact where
the same word list in each condition so participants form an interpretation of the experiment's purpose and
that this does not become a unconsciously change their behaviour accordingly
confounding variable.
Experimenter variables - The experimenter effect is a term used to
Any differences between conditions
describe subtle cues or signals from an experimenter that affect the
Repeated Order effects (e.g. practice Order effects need to
measures are likely to be due to changes in the effect, fatigue effect, be controlled using performance of participants in studies. The cues may be unconscious
design IV and not due to participant variables. recognising demand counterbalancing or nonverbal cues, such as muscular tension or gestures. They may be
Fewer participants need to be characteristics) as participants randomisation.
recruited, as they are used twice vocal cues, such as tone of voice.
take part in all conditions.
Counterbalancing - Important control when using repeated measures
as it reduces ‘carry over’ effects. Half PPs do condition A then B,
Matched Fair comparisons can be made It is time-consuming to match
pairs design between the groups as they are all participants and not all
Other half do condition B then A
equally matched – participant characteristics can be equally Participant variables – An extraneous variable from the participant
variables don’t interfere with the matched. that will interfere with the results.
results.
 Reliability Validity Key words
Quantitative data – Numerical data, often collected through closed questions
Reliability means to repeat and refers to the Validity refers to whether a study measures what it is Qualitative data – Rich and detailed data often gathered through open questions.
consistency of a result. If something is reliable supposed to measure. For example if a study is Researcher bias – when a researcher interprets the outcome of a study according
it means if you repeated it you could expect measuring obedience it is important that the measures their own view (this is also what subjective means)
to get the same results. If findings are reliable actually measure obedience (this is known as internal Triangulation – When more than one measure is taken for a behaviour to validate
we can trust them and know they are not a validity). If the sample the study uses is representative the findings.
one off. Reliability can be increased if a study of the target population and we can generalise the Objective – Not often to bias. The opposite of subjective. Quantitative data is
has a standardised procedure. results then we would say the study has external often objective.
validity. Validity – Whether something measures what it is supposed to measure
Reliable – If you replicate it you would get the same results.

Quantitative measures
This is numerical data. It is a scientific form of data. Often quantitative data is
collected through closed questions from questionnaires and interviews.
Quantitative methods are designed to gather facts and measure behaviour that
can be applied to the target population so the data is generalisable and has
external validity. This is more objective research method.

Qualitative measures
This is rich, in-depth and detailed data that often comes from open questions
that allow participants to give extended responses. This type of data is more valid
as it gives a much clearer picture of the thought, feelings and beliefs of the
participants but is less reliable as it is harder to replicate and get the same
answers.

Primary / Secondary data Ecological validity

Primary data is data the researcher has gathered for a specific piece of research – original research. Secondary Is the measure like a real life/natural situation? If an experiment is
data is data a researcher uses that was gathered previously for a different purpose. A problem with secondary conducted in a laboratory it may be argued that it is NOT ecologically
data is that the original purpose of the first study may not quite suit the study that is reusing the data. But it can valid. If a self report has only used closed questions or Likert Scales it
be cheaper to use existing data. may be argued that it is NOT ecologically valid. If an observation is
structured it can be argued that it is NOT ecologically valid
 Ethics Informed consent Deception Protection from harm
ETHICS are standards of Participants must give their consent to take part in research Deception means that information is The BPS ethical guideline states that
conduct that distinguish and this consent must be ‘informed’. This means that withheld from participants; they are misled ‘Investigators have a primary
between right and wrong, information must be made available on which to base a about the purpose of the study and what responsibility to protect participants
good and bad, justices and decision to participate or not. Participants should be told will happen during it. According to the BPS from physical and mental harm during
injustice. what they are letting themselves in for. Only then are they guideline ‘Intentional deception of the the investigation’.
in a position to give informed consent. participants should be avoided whenever But, no investigation is risk free; the
The primary aim of psychology To study participants without consent would be ethically possible’. In particular deception is guiding principle is that risks should be no
must be to improve the quality acceptable so long as what happens to the participants unacceptable if it leads to ‘discomfort anger greater than the risks participants are
of human life and to do this it could just as likely happen to them in everyday life. E.g. or objections from the participants’ when exposed to in their normal lifestyles.
is necessary to carry out observation in naturalistic setting such as bus queues. the deception is revealed after the research Participants should be encouraged to
research with human People in bus queues may be observed by anyone. In the has been complete’. contact the investigator after the research
participants. Research case of young people under 16, consent should also be if they have any worries or concerns. The
psychologists have a duty to obtained from their parents. But, the BPS accepts that sometimes investigator has a responsibility to detect
respect the rights and dignity deception is unavoidable. In such cases the and remove any consequences of the
of all participants. A major problem with informed consent is the possibility researcher must: research.
This means that they must that the researcher will ‘give the game away’ and thus Make sure that alternative
follow certain moral principles influence participants’ behaviour. As a result, a case can be procedures which avoid deception
and rules of conduct, which made for withholding information. are not available Confidentiality
are designed to protect both Consult with colleagues about how Protection from harm involves
participants and the participants might be affected by the confidentiality. Participants may be asked
reputation of psychology. The Debrief deception personal questions. They must be told
professional organisation that Reveal the deception to the that there is no need to answer these
governs psychology in Britain is Debriefing involves telling it all after the investigation is participants immediately after the questions and, if they do that their
the British Psychological complete. Researchers should discuss the aims of the research has been completed. answers will be treated in confidence.
Society (BPS). They have research with the participants, making sure they understand
Right to withdraw That is, they should remain anonymous.
produced a list of ethical how they have contributed to meeting those aims. Any
Invasion of privacy can result in unease or
guidelines that all practising deception is disclosed, explained and justified. Attempts Participants should have the right to distress. The BPS states that unless
psychologists must follow. are made to undo any negative effects of the research. withdraw from an investigation at any time. people have given their consent, they
Codes of conduct for research state that all participants have This is true even if they are being paid to should only be observed in situations
An ethical issue is any a right to a debriefing session after the investigation has take part. They should be told this at the where they would expect to be observed
situation that repeatedly gives been completed. But, participants may be reluctant to start of the research. No attempt should be by strangers. This limits observational
rise to an ethical dilemma. express negative feelings during debriefing due to made to encourage or persuade them to research to public places.
embarrassment or not wanting to upset the researcher remain.
 Experiments Lab experiments Field experiments
Experiments look for This type of experiment is conducted in Field experiments are done in the everyday (i.e. real life) environment
the effect that a well-controlled environment (not of the participants. The experimenter still manipulates the
manipulated variables necessarily a laboratory), where independent variable, but in a real-life setting (so cannot really control
(independent variables, accurate measurements are possible. extraneous variables).
or IVs) have on The researcher decides where the Strength: Behaviour in a field experiment is more likely to reflect real
measured variables experiment will take place, at what life because of its natural setting, i.e. higher ecological validity than a
(dependent variables, or time, with which participants, in what lab experiment.
DVs), i.e. causal effects. circumstances and using a standardized Strength: There is less likelihood of demand characteristics affecting Reliability and validity
An experiment is an procedure. Participants are randomly the results, as participants may not know they are being studied. This The more controlled an experiment (lab
investigation in which a allocated to each independent variable occurs when the study is covert. experiments) is the more reliable it is. This
hypothesis is group. Limitation: There is less control over extraneous variables that might is because the standardised procedure
scientifically tested. In An example is Milgram’s study of bias the results. This makes it difficult for another researcher to means that it is more likely to be replicable.
an experiment, an obedience. replicate the study in exactly the same way. However, the more controlled an
independent variable Strength: It is easier to replicate (i.e. experiment is the more it lacks validity. This
(the cause) is copy) a laboratory experiment. This is is because these extra controls often mean
manipulated and the because a standardized procedure is Natural experiments that participants know they are being
dependent variable (the used. studied and may change their behaviour
effect) is measured; any Strength: They allow for precise control Natural experiments are conducted in the everyday (i.e. real life)
accordingly. If they decipher the aim of the
extraneous variables are of extraneous and independent environment of the participants, but here the experimenter has no
experiment they may demonstrate demand
controlled. variables. This allows a cause and effect control over the IV as it occurs naturally in real life.
characteristics where they change their
relationship to be established. For example, Charlton’s St Helena study.
behaviour based on what they think is
Limitation: The artificiality of the Strength: Behaviour in a natural experiment is more likely to reflect
expected of them. Equally the artificial
Next steps setting may produce unnatural real life because of its natural setting, i.e. very high ecological validity.
environment is likely to effect behaviour
behaviour that does not reflect real life, Strength: There is less likelihood of demand characteristics affecting
Research the following too as the setting isn't natural. This means
i.e. low ecological validity. This means it the results, as participants may not know they are being studied.
studies and identify and the study will lack ecological validity – as it
would not be possible to generalize the Strength: Can be used in situations in which it would be ethically
justify which method of isn't like real life. Field and natural
findings to a real life setting. unacceptable to manipulate the independent variable, e.g. researching
experiment they used experiments have more validity because
Limitation: Demand characteristics or stress
and why this was the the setting is more natural for participants
experimenter effects may bias the Limitation: They may be more expensive and time consuming than lab
most appropriate: so behaviour is more normal, however
results and become confounding experiments.
• Milgram (1963) because of this they cant have controls in
variables. Limitation: There is no control over extraneous variables that might
• Zimbardo (1973) place which means they lack reliability and
bias the results. This makes it difficult for another researcher to
• Peterson and we are less likely to gain a cause and effect
replicate the study in exactly the same way.
Peterson (1959) relationship between the IV and the DV.
 Ways to gather Types of question Surveys Questionnaires
data Set of questions that are designed to investigate a particular topic. Good way
Both questionnaires and Surveys are a commonly used research to get a large volume of data. Can ask open and closed questions.
interviews can contain open or method in social psychology. A survey can be Questionnaires involve asking people what they think about a topic of
Different methods
closed questions. An open thought of as an umbrella term for a number interest. Questionnaires have to be designed carefully. They ask for personal
can be used to
question is one that can be of different research designs including data e.g. age, gender and background.
investigate behaviour
answered in any way the questionnaires and interviews. Because questionnaires involve a written format there is no flexibility about
The
participant chooses. It yields These methods investigate specific research the questions. Space can be left for the participants to write comments but
non-experimental
qualitative data- data that questions by gathering self-report data. otherwise set questions are answered. Questions are most likely to be
method is still a
consists of words that describe the Self report data is data given about closed and may make use of a Likert-type scale.
scientific methods
participant’s views. A closed themselves You can use any format you like depending on your aim.
(as the information
can be measured) question limits the responses that More straightforward questions come first, in depth ones follow.
and it can aim for can be made e.g. yes or no. It Remember! Personal questions should come at the end as they take more time, so the
objectivity (as it is yields quantitative data- data that respondent doesn’t get bored.
can be reduced to numbers or Interviews tend to give data with more Questionnaires should not be too long.
fact based)
quantities. validity. However they can involve A pilot survey should be carried out. This means…
subjectivity and are hard to repeat. It is hard Questions should be set up in a way that allows a range of responses to be
STRENGTH WEAKNESS
to test for reliability. gathered to avoid response bias.
Qualitative data Represents the true complexities of More difficult to analyse so that Negatives should be avoided e.g. asking if someone is not a racist person.
human behaviour and gains access conclusions are difficult to draw Questionnaires are reliable and less likely to STRENGTHS
is information that cannot be to thoughts and feelings that about the topic hypothesis
counted for example, about cannot be assessed using other involve subjectivity. However, they tend to the same questions are asked to all participants using the same
how you feel or think. It comes
typically from asking open
methods Replication is more difficult
because the results are detailed
be less valid, as any open questions may be standardised procedure this means that there is little variation in
questions to which the answers More valid because the feelings, and descriptive so reliability may missed or answered briefly. how people are asked the information, this means that data us
are not limited by a set of emotions and details of the be lower
choices or a scale, whereas situation are being taken into realistic and valid and uninfluenced by the researcher.
closed questions generate consideration
quantitative data directly. By using set procedures the questionnaires can be easily replicated
as the same one can be used again therefore ensuring reliability.
Easier to analyse because the data Reduces information about
Subjectivity / Objectivity WEAKNESSES
Quantitative data is in number format, so graphs and people to over-simplified
analysis is more straight forward statistics and so important We are SUBJECTIVE when we consider administering questionnaires can be difficult and this may mean
is numerical information for and conclusions can be drawn information may be missed out,
example, about your age, how easily ie what and how people feel. something from our own perspective and we other variables like location and others present could influence what
many hours do you work per
week, how highly you rate a TV Reliable because the results can be Lower in ecological validity are OBJECTIVE when we see what is really the respondent will fill in and ultimately bias the results
programme. It is the data that compared and replicated if because the data is numerical in
there, unaltered by our own biases. Both questionnaires often have restricted questions which means that the
represents how much, how far necessary and similar results are format and doesn’t tell us about
and how long etc. there are of likely. the descriptive situation, questionnaires and interviews are subjective results could be invalid, closed questions may not offer enough
something ie, behaviour is emotions or feelings of the
measured in terms of numbers research because the data analysis involves options and open questions may restricted length, which leads to
and quantities.
TIP! interpretation from the researcher. validity problems.
REMEMBER… Replicate = Reliable
 Interviews Evaluation of Interviews
Researchers can ask closed questions during an interview but you STRENGTHS
can probe the participant to find out what lies behind superficial - interviews enable a large amount of data to be collected which is descriptive and may give a better picture of what
attitudes. is going on in real life so are valid to what is being studied
Unstructured interviews are most likely to give qualitative data. - interviews give access to information which is not available through direct observation, such as what individuals
Here the questions are open and the structure of the interview is think and feel about certain topics which again makes it a more valid method.
flexible. There is a research question, which the interview is based - interviews can gather a lot of information which can produce results which give insight into areas which may not
around, but things are left quite unspecified to see what emerges have been thought of for example, if you are looking at people attitudes towards hard drug users you may not have
during the interview. thought of in your survey those who are ill and use prescribed drugs long-term.

The structured interview involves a pre-set order of questions. This WEAKNESSES

means there is little opportunity for the researcher to follow up areas - in interviews people often don’t know what they feel or do, and therefore are forced to rely on “social
of interest. desirability”, meaning that they tend to answer a question in the way that seems most representative of “good”
behaviour. This produces a SOCIAL DESIRABILITY a form of bias and reduces the reliability of the results.
Semi-structured interviews involve a schedule of questions that - in interviews the analysis of the information can be subjective especially if one person is carrying out the research,
should be answered, but the researcher also has freedom to follow they may miss important information that others would pick up because of their personal opinions, ie RESEARCHER
up on responses. BIAS.
Interviews mainly gather qualitative data, and so are used when
in-depth data is required. There are likely to be some quantitative Correlations Evaluation of correlations
data e.g. age, or yes/no questions. The more structured an interview
This is a technique designed to look for relationships between • Allow us to investigate and test things that
is, the more likely it is to include quantitative data. Less structured
variables. Can use an observation or questionnaire to gather data we may not be able to test otherwise, for
interviews are more likely to generate qualitative data.
Looking for a relationship between co-variables. Plot a scatter example the correlation between smoking
gram to see. and lung cancer couldn’t be tested
Issues to consider- participants should see:
Positive correlation: each co variable increases together experimentally as it would be unethical to
The interview schedule (a set of questions and the time required).
Negative correlation: as one increase another decreases force participants to smoke to see if they
The chosen format for recording the interview.
get lung cancer!
The full transcript of the interview (and agree with what has been
• However, we are unable to get a cause and
recorded).
effect relationship, as we cant be sure that
the results are not down to other factors
Subjectivity and objectivity
interfering.
In all research, the researcher can cause bias. Social desirability,
• Cheap and less time consuming
demand characteristics and response bias can all affect interviews.
Researchers can cause bias by interpreting the results using their
own views (subjectivity). Objectivity is when there is no bias
affecting the results. Scientific studies must be objective.
 Observations Controlled so is more reliable
Structured Can be replicated lack validity as a constructed situation
Psychologists observe, they watch people’s
Can see all behaviours
behaviour and measure particular aspects in a
way that is as precise as possible Time and cost effective Demand characteristics
Usually have more than 1 observer to avoid bias Manipulate a situation so a behaviour can be seen
Behaviour can be videoed. When observers valid because they act normally most observations cant be replicated as the day/ time/
interpret the data in the same way there is If time sampling is used carefully with tallying and specific setting are important and one day will differ to the next
inter-observer reliability. Researchers must first
Naturalistic categories for behaviours and more than one observer it can Can be observer bias- experimenters see what the want to
decide what behaviours are to be noted E.g.
measure of aggression; kicking, punching etc . be reliable and replicable see
Naturalistic Observer may be too involved to record the data- not
Conducted in an everyday environment High ecological validity as in natural setting objective and have to rely on memory
where participant behave normally
Participant Difficult to replicate as it is hard to find an observer who
Controlled/structured
Conducted in a lab – may be set up No stranger present so act normally can also be a member of the group- grown up
Overt experimenter looking at teenage habits
Participants know they are being studied Data is valid
Covert are objective- stands back and observes Your presence is likely to affect their behaviour- not
Participants don’t know they are being Can record data easily natural
studied Non Can use time sampling when tallying which cant really do if Might miss behaviours if have to stand far away
Participant
participant participating Might misinterpret behaviours if not involved
Observer becomes part of the group that
they are observing
Non-participant
Observer takes a step back from the group
Exam tip Key terms
Decisions over how and what time period the
researchers will use It is important to specify which observation • Inter – rater reliability: when more than one observer codes
Time sampling; record the child every 5 minutes technique you are referring to in your behaviour and their results are compared to check for agreement.
for 25 minutes for example Rules need to be put answer. If you are criticising observations for • Observer bias: when an observer interprets the observed behaviour
in place before to ensure for standardisation. ethical reasons, these may not apply to all
observational techniques so you need to say
according to their own view.
Behaviour may be analysed and broken down into
behaviour categories such as; aggression, which technique is associated with the
friendliness etc ethical issue.
 Case studies Standard form Decimals
Ratio
A case study is a study of an individual or small group of Standard form is a way of writing Decimal form refers to any number
people. It allows in depth and detailed data to be collected. Ratios are used to compare down large or small numbers in the base-ten system
Within this method, other methods can be used, such as quantities. without including all the digits Numbers are thought of as over 10
interviews, observations and questionnaires. The case being A ratio shows how much one thing for example… or in items of tenths
studied occurs naturally, and the psychologist has no control compares to another. For example the 10 x 10 x 10 x10 x 10 x 10 So the fraction ½ written as a
over the situation. A case study is used, as the researcher ratio of 2 : 3 means for every 2 the 10 to the power of 6 = 106 number ‘over’ 10 would be 5 over
cannot control any variables, he simply has to observe and first person gets the second person 3200 written in standard form ten or 5/10
study. would get 3 parts. would be 3.2 x 103 5/10 = 0.05
Case studies allow a researcher to investigate a topic in far
more detail than might be possible if they were trying to deal
with a large number of research participants. A central tool Significant Figures
which is used is the gathering of a case history. May include
open ended interviews. Secondary data (data from schools etc)
can be collected. May write a case study about a child having
problems at school to try and solve these issues.
STRENGTHS Rounding numbers
Data tends to be more valid, as it is in depth and focuses on
real experiences in a real situation
This is a valuable research method because it may be the only
way to gather rich, detailed, qualitative information in context
and with meaning for those concerned e.g. round 3268 to 1 sf
Sometimes it is the only way to be gathering data about a the first significant figure is a 3,
particular issue, for example, a child who has been deprived of which represents 300, so we need
parenting (you cannot set up an experiment to test such a You are likely to be asked to round to 2 to round to the nearest thousand.
thing) decimal places (2dp) to do this look at the The number next to the 3 is a 2
WEAKNESSES second number after the decimal place. which isnt ‘5 or more’ so we round
They lack generalisability because they are about one individual Underline it and circle the number next to down to 3000.
(or small group) only, so they are specific rather than in general it, if that number is ‘5 or more’ you would
The data gathered cannot be applied to any other case, round up. If not you would keep it the
individual or group, as the data is unique same. E.g 4.5678 would round to 4.57
They are hard to replicate, so cannot be tested for reliability because the 7 which is next to the 6 is ‘5 or
Data may be subjective, and therefore cannot be used to build more’.
a body of knowledge.
 Percentages Set A Strengths of this Weaknesses of
Measure 2, 2, 3, 5, 5, 7, 8 measure this measure
A percentage is a fraction of 100
To work out a percentage just work out the fraction and multiply by The Mean = average Adding the numbers up gives: Makes use of all Takes into
100 2 + 2 + 3 + 5 + 5 + 7 + 8 = 32 scores. This as account
Example: if 200 people were surveyed to find out how many thought To find the mean, you need to add up
good as it extreme
practice, genetic make-up or good schooling were the reasons for all the data, and then divide There are 7 values, so you divide the total
increased validity values…Why is
something having a good memory this total by the number of values in by 7: 32 ÷ 7 = 4.57...
as all scores are this bad?
A fourth option ‘do not know’ was added the data. So the mean is 4.57
included
To find the percentage for each option divide the answer number by
200 and multiply by 100.
The Median = middle The numbers in order: Easy to calculate Hard to
To find the median, you need to put 2 , 2 , 3 , (5) , 5 , 7 , 8 with a small data calculate with a
Inferential statistics the values in order, then find the set large data set.
Descriptive stats are used to summarise raw data. The whole set of The middle value is marked in
middle value. If there are two values Is easily
scores from a study. Before any analysis takes place. brackets, and it is 5.
in the middle then you find the mean affected by
The mean, median and mode are types of average. So the median is 5
of these two values. anomalies
The range gives a measure of the spread of a set of data.
The table to the side shows how to calculate the mean, median, The Mode = most frequent The data values: Easy with a small Doesn’t take
mode and range for a set of data.
The mode is the value which appears 2,2,3,5,5,7,8 data set into account all
the most often in the data. It is values
The values which appear most often are 2
possible to have more than one mode
and 5. They both appear more time than
if there is more than one value
any of the other data values.
which appears the most.
So the modes are 2 and 5

The Range To find the range, you The data values: Easy to calculate Impacted by
first need to find the lowest and 2,2,3,5,5,7,8 and quick. extreme scores
highest values in the data. The range
The lowest value is 2 and the highest Shows us the
is found by subtracting the
value is 8. Subtracting the lowest from variance in the
lowest value from the highest value.
the highest gives: 8 - 2 = 6 scores.
So the range is 6
 Normal distribution Skewed distribution Outliers
If the mean, median and mode are not similar then a Outliers can cause skewed distributions (the mean is very
skewed distribution is produced. susceptible to outliers)
This is found when the mean, median and mode are A positive skewed distribution is caused by a high extreme set of
very similar or the same. scores, therefore a positive skew will contain more low scores than
The further the scores are from the mean the less often high scores. (The skew is caused by outlying positive scores)
they occur in a set of data. A negative distribution is caused by a low extreme of scores, and
The graph will be symmetrical like below. therefore will contain more high scores than low scores (the skew is
If the mean, median and mode are not similar then a caused by outlying low scores)
skewed distribution is produced.
A normal distribution is found if the mean, median and
mode for a set of data are very similar or exactly the Histogram
same. When data are normally distributed 50% of the
values are below the mean and 50% are above. The Histograms in psychology refer
majority of the scores are equally spread close to the to bar charts that show
mean on either side of it. The further the scores are continuous data and thus don’t
from the mean, the less often they occur in a set of have gaps in between the bars.
data. Many mathematical statistical tests can only be
carried out if data are normally distributed.
Scatter diagrams Bar charts
Only used for correlations Uses bars to describe categorical data
Shows a relationship between two variables As the data are discrete (not continuous)
We add a line of best fit by drawing a line there are gaps between the bars
that has half the scores above it and half
below.

Frequency
Don’t forget ‘frequency’ refers
to the ‘total’.