Definitions and Concepts for AQA Psychology AS-level

Topic 7: Research Methods

Research Methods

Controlled observation: A type of observation where participants are observed in a lab,

increasing control and reliability but decreasing ecological validity.

Correlation: The extent to which two variables are associated.

Covert observation: A type of observation where the observer is hidden and therefore
participants do not know they are being observed. While this does reduce demand
characteristics, it can raise ethical issues around consent.

Experiment: A type of investigation wherein a hypothesis is tested by manipulation of an

independent variable, in order to view its effect on the dependent variable.

Field experiment: A type of experiment that is conducted in a real life setting, which reduces
the amount of control over extraneous variables, however the ecological validity is improved.

Interviews: A self-report technique wherein participants are asked questions by an interviewer,

which allows for flexibility in the information gathered.

Laboratory experiment: A type of experiment that is conducted in a highly controlled

environment, allowing control over extraneous variables at the expense of ecological validity.

Natural experiment: A type of experiment in which an independent variable that already exists
is tested in its natural environment, greatly reducing the control of extraneous variables. This
type of experiment allows for investigation of variables that cannot ethically be created.

Naturalistic observation: A type of observation where participants are observed in their natural
environment, increasing ecological validity but decreasing the amount of control over
extraneous variables.

Non-participant observation: A type of observation where the observer does not join the
group under observation. This allows for higher objectivity but reduces the insight that could be
achieved with a participant observation.

Observation: A type of data collection wherein participants’ behaviour is observed.

This work by PMT Education is licensed under https://bit.ly/pmt-cc

https://bit.ly/pmt-edu-cc CC BY-NC-ND 4.0

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
Overt observation: A type of observation where the participants know they are under
observation. This preserves informed consent but may increase demand characteristics.

Participant observation: A type of observation where the observer joins the group under
observation. This form of observation yields highly valuable qualitative information with insight,
but can reduce the objectivity of the researcher.

Quasi experiment: See “Natural experiment”

Questionnaires: A self-report technique wherein participants answer pre-decided questions, in

the form of paper or electronically. This form of self-report allows for anonymity.

Structured interviews: A form of interview wherein questions are pre-set beforehand, with no
flexibility. Usually, these consist of closed questions, and allows for replicability.

Unstructured interviews: A form of interview wherein questions may be set beforehand, but
the interviewer is allowed flexibility in the form of a natural conversation.

Scientific Processes

Aims: The objective or purpose of the experiment.

Bias: An inclination to a certain position or thought. For example, in hostile attribution bias,
hostility or negativity is more likely to be assumed from a neutral face.

Behavioural categories: An observational technique wherein participants' possible behaviours

are separated into more specific components. This allows for operationalisation of the
behaviour. For example, splitting aggression into categories of “swearing” and “punching”.

Closed questions: A type of question that can only be answered with a limited number of
answers, usually a “yes” or “no”.

Confounding variable: A type of extraneous variable that is related to the independent variable
in the experiment. For example, if you were testing the effects of anxiety on memory recall, the
relative levels of sensitivity to anxiety-inducing stimuli would be a confounding variable.

Control variable: Any variables that are kept constant through the experiment to prevent their
effects on the dependent variable.

Counterbalancing: To make half of the participant sample experience the different conditions of
the experiment in one order, and the other half of the participants complete it in the opposite
order.

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
Demand characteristics: Changes in the participants’ behaviour to comply with the hypothesis
of the researcher.

Dependent variable: The variable that changes in response to manipulation of the researcher,
that is being measured for the experiment. For example, if you were testing the effects of
anxiety on memory recall, memory recall would be the dependent variable.

Directional hypothesis: A hypothesis that specifies the direction of the relationship of the
experiment e.g. coffee will have an effect on the reaction time of participants.

Event sampling: An observational technique wherein an observer records every time a

particular behaviour or “event” occurs, usually in the form of a tally chart.

Extraneous variable: Variables other than the independent variable that have an effect on the
dependent variable. For example, if you were testing the effects of anxiety on memory recall, the
intelligence levels of the participants could be an extraneous variable.

Generalisation: To attribute information from a sample to the rest of the population.

Hypothesis: The prediction of the outcome of the experiment.

Independent groups: An experimental design wherein different participants are involved in

different conditions of the experiment. For example, using two different groups of people to test
the effect of music on memory recall, with one group memorising during music playing and the
other in silence.

Independent variable: The variable that is manipulated by the researcher to observe its effects
on the dependent variable. For example, if you were testing the effects of anxiety on memory
recall, anxiety would be the independent variable.

Investigation effects: Unconscious changes in the investigators behaviour to comply with the
hypothesis of the investigation.

Matched pairs: An experimental design wherein participants in different conditions of the

experiment are matched on certain variables to reduce the effect of participant variables. For
example, in the Bobo doll study, children were matched on scores of aggressiveness for each
condition.

Non-directional hypothesis: A hypothesis that does not specify the direction of the relationship
of the experiment e.g. coffee will change the reaction times of participants (whether it will
increase or decrease the times is not specified).

Open questions: A type of question that requires answers that are longer than “yes” or “no”.

Operationalisation (of variables): To clearly state and objectify a variable. For example,

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
instead of measuring “aggression”, researchers would convert it into observable categories like
“punching” and “kicking”.

Opportunity sampling: A sampling technique that involves obtaining a sample via anyone that
is available from the population at the time of collecting the sample. For example, handing
questionnaires out to people outside a shopping mall.

Peer review: The assessment of work by other people with similar levels of expertise in that
field, to provide an unbiased expert opinion of the quality of said work.

Pilot studies: Preliminary/trial studies carried out to ensure the clarity of the study itself. For
example, using a pilot questionnaire on a sample of people that give feedback on the clarity of
the questions.

Population: The group represented by a sample.

Random allocation: To allocate participants to separate conditions using some sort of

randomisation technique. For example, using a computer to randomly generate groups for
condition A and B.

Random sampling: A sampling technique that involves randomly generating participants from
the population by any randomisation technique. For example: random number generation from a
computer, picking names out of a hat.

Repeated measures: An experimental design wherein the same participants undergo all the
conditions of the experiment. For example, when testing the effect of coffee on reaction time, all
participants will be tested for reaction time with and without coffee.

Sample: A smaller group that aims to be representative of a population.

Standardisation: Keeping the experimental methods as identical as possible.

Stratified sampling: A sampling technique that involves establishing sub-groups (strata) within
the population investigated and picking participants to create a representative sample. For
example, if the population is ⅔ female and ⅓ male, the sample should also be ⅔ female and ⅓
male.

Systematic sampling: A sampling technique that involves establishing a method to pick

participants evenly distributed through the population. For example, picking every 10th
participant in a list of the entire population.

Time sampling: An observational technique wherein an observer only records specific

behaviours in specific time intervals. For example observing and recording the behaviour of
football fans at a stadium every 15 minutes for 30 seconds.

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
Volunteer sampling: A sampling technique that involves using participants that volunteer to
take part in the study, provided they meet the inclusion criteria. For example, putting a request
on an information board in your school and sending questionnaires to those that respond.

Data Handling and Analysis

Bar charts: A graphical representation of categorical data with numerical values.

Correlation: The extent to which two variables are associated.

Mean: A measure of the average of a data set that is calculated by adding all values together
and dividing by the number of values. This means it takes every single value into account,
including outliers.

Median: A measure of the average of a data set by determining the middle value in the data
set. This means it only takes into account the very middle value, ignoring the value of any
others.

Mode: A measure of the average of a data set by finding the most common value. This means it
only takes into account the number of values, and not the values themselves.

Negative correlation: When two sets of variables have a negative relationship i.e. when one
increases, the other decreases.

Normal distribution: A symmetric distribution of values around the mean, sometimes called
“the bell curve”.

Positive correlation: When two sets of variables have a positive relationship i.e. when one
increases the other increases.

Primary data: Data that has been collected first hand, by yourself.

Qualitative data: Non-numerical data, such as text, video, photographs or audio recordings.

Quantitative data: Data that is categorised by numerical values. For example, height, weight,
time.

Range: A measure of dispersion that is calculated by subtracting the smallest value from the
largest value.

Scattergrams: A graphical representation of the correlation between two variables.

Secondary data: Data that has been collected by someone else.

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc
Skewed distribution: An asymmetric distribution of values of values around the mean, which
can be positively or negatively skewed.

Standard deviation: A measure of dispersion that represents the average distance of values
from the mean. It is calculated by subtracting each value from the mean, squaring that
difference and finding the sum of all these squares. Then, dividing by the number of values and
finding the square root.

https://bit.ly/pmt-cc
https://bit.ly/pmt-edu https://bit.ly/pmt-cc