Research Methods Key Terms

3.2.3 Research Methods

independent variable (IV)

variable that is changed by the researcher to observe its effect in an experiment

dependent variable (DV)

the variable that is measured to see how it is affected by changes in the IV

aim

general statement about the purpose of the investigation

hypothesis

a precise, testable statement about the outcome of an investigation

directional hypothesis

a statement that predicts the outcome of the experiment, including the direction of the change (one-tailed hypothesis)

non-directional hypothesis

a statement that predicts a change but does not state the direction of the change (two-tailed hypothesis)

null hypothesis

a statement that predicts that there will be no change

operationalisation

defining variables to make them as clear and objective as possible

extraneous variable (EV)

any variable other than the IV that could influence the DV

confounding variable

type of EV that varies systematically and could affect the results

demand characteristics

any cue that may inform a participant as to the purpose of the experiment, changing their behaviour

investigator effects

any way that the investigator themself could affect the results; could be due to the experimental design, verbal / non-verbal cues or bias in interpretation of data

participant reactivity

any way in which the participant alters their behaviour as they are aware they are being observed / part of an experiment

randomisation

participants are randomly allocated to experimental conditions; eliminates bias and ensures each participant has an equal chance of being in any condition

standardisation

ensuring all procedures and instructions in an experiment are the same for all participants to maintain consistency (reduces demand characteristics and investigator effects)

ecological validity

the extent to which the results of a study can be applied on real-world settings or reflect real-life behaviour

internal validity

the extent to which an experiment accurately measures what it intends to

Hawthorne effect

a type of participant reactivity where an individual alters their behaviour as they know that they are being observed

social desirability bias

when participants alter their responses / behaviour to be viewed in a better light by others, usually aligning with societal norms

laboratory experiment

the researcher manipulates the IV in a controlled environment, using standardised procedure

field experiment

the researcher manipulates the IV in a real-world setting

natural experiment

the researcher does not manipulate the IV; change is caused by something or someone else

quasi experiment

the IV occurs naturally and cannot be changed (e.g, age or gender)

strengths of laboratory experiments

• high internal validity

• experiment can be repeated

• EVs can be controlled

weaknesses of laboratory experiments

• lacks generalisability

• low external validity

• demand characteristics may be present

• low mundane realism

strengths of field experiments

• high mundane realism due to natural environment

• high ecological validity

weaknesses of field experiments

• EVs are more likely

• harder to replicate experiment

strengths of natural experiments

• high ecological validity

• can conduct research that would otherwise be considered unethical

weaknesses of natural experiments

• lacks generalisability due to small sample size

• cannot randomly allocate participants

• low internal validity (difficult to prove IV cause the change in DV)

strengths of quasi experiments

• can be conducted in labs » DV can easily be measured

• replicable

weaknesses of quasi experiments

• cannot randomly allocate participants

• low internal validity (difficult to prove IV cause the change in DV)

sampling

selecting a subset of individuals from the larger population, to study and draw inferences about the entire population

sampling technique

method used to select people from the population

bias (sampling)

occurs when a certain group is under- or overrepresented within the sample group

generalisation

the extent to which findings from an investigation can be applied to the wider population

population

a group of people who are the focus of the researcher’s work, from which a smaller sample is selected

random sampling

every member of the population has the same chance of being chosen (e.g. assigning each participant a number then using a random number generator)

systematic sampling

selecting every n^th member of the population (e.g. every 5^th person on a register)

opportunity sampling

researcher selects anyone who is willing and available (e.g. waiting on a street)

volunteer sampling

participants select to participate themselves (e.g. answering a postal questionnaire)

stratified sampling

researcher divides the target population into subgroups (strata) based on key characteristics, then randomly selecting participants from each subgroup in proportion to their representation in the population

independent groups

each participant is allocated to a group and participates in one experimental condition » data from each group is independent of the other

repeated measures

each participant completes all conditions » data is related due to individual performance differences

matched pairs

participants are paired up based on variables relevant to the experiment; each half of the pair is randomly allocated to a condition

strengths of independent groups

• order effects cannot be observed

• less time-consuming to collect data as experiments can be carried out simultaneously

• participants less likely to guess aims

weaknesses of independent groups

• risk of participant variables affecting results, reducing internal validity

• difficult and time consuming to recruit different participants to each condition

ways to reduce the impact of participant variables in independent groups

• random allocation of participants to evenly distribute participant variables

• large sample size

strengths of repeated measures

• no participant variables » high internal validity

• fewer participants need to be recruited

order effect

how the sequence of conditions in a repeated measures design affects participant performance

weaknesses of repeated measures

• risk of order effects impacting the results, as participants may become better or worse at the task with experience

• high chance participants will figure out the study’s aim » increases demand characteristics

way to reduce order effects

counterbalancing: each group has a different order to complete the conditions in

strengths of matched pairs

• participants only complete one condition » demand characteristics reduced and no order effects

• risk of participant variables affecting results is reduced » increases internal validity

weaknesses of matched pairs

• expensive and time consuming to match up participants

• difficult to match up participants identically

pilot studies

small scale practice investigations carried out before the full experiment to identify potential issues with the design, method or analysis » allows researches to see if participants guessed the experimental aim (demand characteristics) and if the investigation will have significant results

control groups

group in an experiment treated like the experimental group, but does not receive the tested variable

benefits of having a control group

• allows for comparison » researcher can determine if the changes in the results (DV) are caused by the treatment (IV)

• increases internal validity of the experiment

placebo effect

psychological phenomenon where a person’s health or behaviour changes after taking a fake treatment

reasons why placebos are unethical

• participants believe they are receiving treatment, but are not

• could deprive participants of effective treatment, potentially causing harm

• violates ethical guidelines from BPS