AQA Psychology: Research Methods Core
Concepts
Aims & Hypotheses
Aims are a statement of what the researcher intends to find out through a study. When using the
scientific method to achieve an aim, we follow a very simple process:
Make observations
State expectations/formulate a hypothesis
Design an experiment
Collect Results
Conclude
Within the experimental process there are several important concepts it is worth defining from
the start. These are:
The Dependent Variable (DV): The dependent variable is what you measure in an experiment, and
must be operationalised to ensure it can be tested. This essentially means the variable being
measured must be specific.
The Independent Variable (IV): This is what you change, and defines the conditions of your
experiment. Each condition is described as a ‘level’ of the IV. You are looking to see whether
changing the IV has any effect on the DV.
Experiment: A method in which we investigate natural phenomena to draw causal conclusions
about the way things work. This is because an IV is deliberately manipulated in order to see the
effect on the DV.
Extraneous Variables: Variables that do not vary systematically with the IV, and can change what
happens to the dependent variable. These are nuisance variables that affect the validity of
conclusions and must be controlled.
Standardised Procedures: Very important in experiments to ensure that participants encounter
the same set of procedures in order to be able to repeat a study. Furthermore, changes in
procedure may impact the DV and alter the effect of the IV.
Hypotheses
Although research questions and aims state the purpose of a study, there are usually too
imprecise to test. Hypotheses are clear statements of what is being tested, the research
hypothesis being proposed at the beginning of a piece of research and outlining what is expected
to be found.
Null Hypothesis
This is assumed to be true until results are obtained, and may be rejected if data supports the
research hypothesis. Frequently, a null hypothesis is a prediction that there will be no relationship,
or that any correlation is due to chance. In this case, the research hypothesis would be that the
variables are linked, but you can be precise about such a link using directional hypotheses.
, Directional/One-tailed Hypothesis
If a hypothesis states which direction a relationship will be found, it is a directional prediction and
thus hypothesis. This are often used when previous research findings suggest how the results may
go.
Non-directional/Two-tailed Hypothesis
Non-directional hypothesis predicts a difference but without specifying the exact direction, and
are often used when there is little previous research
or when previous findings are mixed or inconclusive.
Validity & Control
Validity, in essence, refers to the extent to which the results of an experiment accurately reflect
behaviour. By controlling variables (holding them constant or regulating them), researchers may
prevent non-independent variables affecting validity. In other words, validity describes whether
the study successfully isolated and measured the effect of the IV on the DV.
This brings in the concept of confounding variables. These are similar to, and a special type of,
extraneous variables, in that both can alter the DV and ‘confound’ the effect on the IV. However,
unlike extraneous variables confounding variables vary systematically with the IV. These are more
challenging as they vary with the IV, which essentially means they accompany the IV and change
when the IV changes. Consider the following example.
One of the most common types of confounding occurs when an experimenter does not or cannot
randomly assign participants to groups, and some type of individual difference (e.g., ability,
extroversion, shyness, height, weight) act as a confounding variable. For example, any experiment
that involves a comparison of men and women is inherently plagued with confounding variables,
the most commonly cited of which is that the social environment for males and females is very
different. Extraneous variables occur when we cannot be sure the variable varies with the IV. With
gender, the variable always varies with the IV, because gender is fixed, so this is confounding.
However, control must be considered against real-life, in that the latter does not involve strict
control of variables in an artificial setting. Mundane realism refers to the extent to which a study
mirrors the real world, and to what degree the experiences and tasks involved reflect real
experiences and tasks. The whole point of trying to ensure realism is to make sure results are
generalisable to the real world, that the study has relevance beyond the conditions of the study.
These terms relate to the next concepts of internal and external validity.
Internal and External Validity
Internal validity concerns whether an observed effect on the DV was due to the manipulation of
the IV, i.e. making sure the research tests for the intended variables and includes enough
mundane realism to be successfully testing the IV. To gain internal validity researchers must design
an experiment carefully and control as many variables as they can.
External validity is affected by internal, as you cannot generalise the results of a study with low
internal validity to the real world. This concerns the ability to generalise to different contexts
(ecological), different populations (population) or different times (historical).
Concepts
Aims & Hypotheses
Aims are a statement of what the researcher intends to find out through a study. When using the
scientific method to achieve an aim, we follow a very simple process:
Make observations
State expectations/formulate a hypothesis
Design an experiment
Collect Results
Conclude
Within the experimental process there are several important concepts it is worth defining from
the start. These are:
The Dependent Variable (DV): The dependent variable is what you measure in an experiment, and
must be operationalised to ensure it can be tested. This essentially means the variable being
measured must be specific.
The Independent Variable (IV): This is what you change, and defines the conditions of your
experiment. Each condition is described as a ‘level’ of the IV. You are looking to see whether
changing the IV has any effect on the DV.
Experiment: A method in which we investigate natural phenomena to draw causal conclusions
about the way things work. This is because an IV is deliberately manipulated in order to see the
effect on the DV.
Extraneous Variables: Variables that do not vary systematically with the IV, and can change what
happens to the dependent variable. These are nuisance variables that affect the validity of
conclusions and must be controlled.
Standardised Procedures: Very important in experiments to ensure that participants encounter
the same set of procedures in order to be able to repeat a study. Furthermore, changes in
procedure may impact the DV and alter the effect of the IV.
Hypotheses
Although research questions and aims state the purpose of a study, there are usually too
imprecise to test. Hypotheses are clear statements of what is being tested, the research
hypothesis being proposed at the beginning of a piece of research and outlining what is expected
to be found.
Null Hypothesis
This is assumed to be true until results are obtained, and may be rejected if data supports the
research hypothesis. Frequently, a null hypothesis is a prediction that there will be no relationship,
or that any correlation is due to chance. In this case, the research hypothesis would be that the
variables are linked, but you can be precise about such a link using directional hypotheses.
, Directional/One-tailed Hypothesis
If a hypothesis states which direction a relationship will be found, it is a directional prediction and
thus hypothesis. This are often used when previous research findings suggest how the results may
go.
Non-directional/Two-tailed Hypothesis
Non-directional hypothesis predicts a difference but without specifying the exact direction, and
are often used when there is little previous research
or when previous findings are mixed or inconclusive.
Validity & Control
Validity, in essence, refers to the extent to which the results of an experiment accurately reflect
behaviour. By controlling variables (holding them constant or regulating them), researchers may
prevent non-independent variables affecting validity. In other words, validity describes whether
the study successfully isolated and measured the effect of the IV on the DV.
This brings in the concept of confounding variables. These are similar to, and a special type of,
extraneous variables, in that both can alter the DV and ‘confound’ the effect on the IV. However,
unlike extraneous variables confounding variables vary systematically with the IV. These are more
challenging as they vary with the IV, which essentially means they accompany the IV and change
when the IV changes. Consider the following example.
One of the most common types of confounding occurs when an experimenter does not or cannot
randomly assign participants to groups, and some type of individual difference (e.g., ability,
extroversion, shyness, height, weight) act as a confounding variable. For example, any experiment
that involves a comparison of men and women is inherently plagued with confounding variables,
the most commonly cited of which is that the social environment for males and females is very
different. Extraneous variables occur when we cannot be sure the variable varies with the IV. With
gender, the variable always varies with the IV, because gender is fixed, so this is confounding.
However, control must be considered against real-life, in that the latter does not involve strict
control of variables in an artificial setting. Mundane realism refers to the extent to which a study
mirrors the real world, and to what degree the experiences and tasks involved reflect real
experiences and tasks. The whole point of trying to ensure realism is to make sure results are
generalisable to the real world, that the study has relevance beyond the conditions of the study.
These terms relate to the next concepts of internal and external validity.
Internal and External Validity
Internal validity concerns whether an observed effect on the DV was due to the manipulation of
the IV, i.e. making sure the research tests for the intended variables and includes enough
mundane realism to be successfully testing the IV. To gain internal validity researchers must design
an experiment carefully and control as many variables as they can.
External validity is affected by internal, as you cannot generalise the results of a study with low
internal validity to the real world. This concerns the ability to generalise to different contexts
(ecological), different populations (population) or different times (historical).
