a level psychology research methods key terms + definitions

variables

factors that may be involved in people's decisions and behaviours.

theory

proposed explanation for a psychological phenomenon that has been observed

hypothesis

a prediction based on a theory (testable statement)

method

how information is gathered/obtained

operational definition

stating exactly in practical terms what we mean when writing a hypothesis

independent variable

the variable that might cause the change in another (what the experimenter manipulates/changes on purpose)

dependent variable

the variable that may be influenced/affected by the alteration of the IV (what the experimenter measures)

experiment

a method which includes an IV & DV, and the only method that can enable you to deduce cause + effect

alternative hypothesis

predicts that change in the IV will result in significant changes in DV

non-directional hypothesis

predicts a change in the IV will lead to a change in the DV, but does not predict the direction of the change in the DV

directional hypothesis

predicts that a change in the IV will lead to a change in the DV in a particular stated direction

null hypothesis

change in the IV will NOT result in a significant change in the DV; and that any small changes that do occur will be due to chance

field experiment

research studies where there is a clearly identifiable IV and DV, but the research is conducted outside of laboratory conditions

natural experiment

research studies where there is a clearly identifiable and operationally defined IV and DV. The IV cannot be manipulated

laboratory experiment

an experiment carried out in a controlled setting, where the researcher purposefully manipulates the level of an IV in each of the conditions, in order to measure its effect on the DV

quasi-experiment

a study that has some elements of experimentation, but doesn't meet the criteria needed to actually be a proper experiment.

experimental condition

any condition in which the IV is present

control condition

a condition in which the IV is absent

extraneous variable

any variable, other than the IV, which is capable of affecting the DV

confounding variable

uncontrolled extraneous variables which occur in a study

population

all people/animals who possess a particular characteristic

sample

a group of participants that are selected from the population

target population

the population where the sample is selected from

representative sample

a sample which has the same characteristics as the target population

random sampling

selecting the sample in such a way that every member of the target population has an equal chance of being selected

stratified sampling

the best way of selecting a representative sample and should be used if the target population contains subsets which differ from one another in important ways

systematic sampling

this involves selecting participants at fixed intervals. the researcher will decide on this interval before selecting the participants, so as to gain the appropriate number of participants from the target population

volunteer sampling/self-selected sampling

the sample is made up of people who offer to take part in a study, often by responding to an advertisement

opportunity sampling

this involves selecting a sample from people who are willing + able to take part in an experiment when asked to do so

ethnocentric

centred around one particular culture, most commonly white, western, middle-class culture

androcentric

centred around males

gynocentric

centred around females

experimental design

the way in which participants are allocated to the conditions of the experiment

repeated measures design

each participant takes part in all conditions of the experiment

sensitive data

data that can be analysed to a more clearer level

order effect

occurs when a participant's score in whichever condition they experienced second is influenced by the fact they've already experienced the first condition

practice effect

a type of order effect where a participant's performance in the 2nd condition is better than it would otherwise have been, because they have already experienced the first condition

fatigue effect

a type of order effect where a participant's performance in the 2nd condition is worse because they have already experienced the first condition. (may include boredom effect)

counterbalancing

making sure each condition of an experiment follows, and is followed by every other condition an equal number of times. this is a way to limit order effects.

constant error

where all the results from one condition are unfairly better/worse and thus affects the conclusion

independent groups/measures design

most common + simplest design. each participant is allocated to 1 condition only, on a random basis

matched participants design

each participant is paired with another participant who is very similar to them on the basis of important subject extraneous variables for the particular experiment. when participants have been put into pairs, one member of each pair is randomly allocated to each condition. therefore, each participant only completes 1 condition

participant reactivity

where participants may alter their behaviour, because they know they are being investigated, and because of the situation they are in

demand characteristics

the cues which may appear to reveal the experimental hypothesis, and so suggest to the participant what result the experimenter hopes to obtain

expectancy effects

when the participant acts in the way they think the experimenter wants them to act, NOT in the way they would act if they didn't know they were being observed. alternatively, the participant may wish to 'prove the researcher wrong', and thus purposefully act in a way contrary to what they think is expected

single blind technique

the participants do not know who is in the experimental and who is in the control, making it more difficult to 'guess' the hypothesis

placebo

an inert substance/condition which prevents participants from realising they are in the control group

double blind technique

where neither the participant, nor the person collecting the data, knows which participant is in each condition

standardised instructions

where the study is scripted, so that each participant in one condition is treated in exactly the same way, whenever they are taking part in the research. this can help to reduce the effects of participant reactivity and experimenter effects.

subject extraneous variables

characteristics of the participant/subject which are capable of affecting the DV. they will therefore differ from study to study. also known as participant variables.

situational extraneous variables

characteristics of the study/situation/set-up which are capable of affecting the DV

standardisation

the way the researcher attempts to make sure that instructions, procedures and scoring for all the conditions of the experiment are made as standardised (i.e. the same) as possible

randomisation

deciding the order of presentatio of stimuli, materials, conditions etc randomly

pilot study

a method used to identify extraneous variables and prevent them from becoming confounding variables. a small-scale study conducted in the way you hope to complete the full research, keeping a record of any problems you encounter any any changes that may need to be made

right to withdraw

participants have the right to cease participation in the study at any time

debriefing

a part of the research process where the participants have the whole study explained to them in detail AFTER they have taken part. they can ask questions to the researchers, and will be asked if they are happy for their data to be used

case study

in-depth investigations/studies of a single person, group, event or community

subjective knowledge

information/data that is not factual - opinions, feelings etc

objective knowledge

measurable facts

content analysis

a process undertaken to examine the elements of a qualitative source, producing quantitative data

quantitative data

data to do with numbers

qualitative data

data to do with words

thematic analysis

similar to content analysis but DOES NOT produce quantitative data

interview

self-report technique conducted on a one-to-one basis by trained researchers asking participants questions

structured interview

this involves asking the same questions to all those being interviewed. information from this type of interview can then be directly compared across participants and frequency data can be obtained (i.e. the information gained can be quantitative and can be statistically analysed)

unstructured interview

this involves asking questions about some particular topic area, but each question will follow on from the previous answer and may build upon themes brought by the participants. therefore, no two sets of questions will be exactly alike and answers CANNOT be directly compared. answers from this type of interview however, can be very rich in detail and give a good insight into a participant's state of mind. data from this type of research is said to be qualitative and cannot be statistically analysed, but could be placed into categories or reported in the form of quotations.

semi-structured interview

some of the questions are set for all people, and some follow on from answers/issues raised by the participants.

survey/questionnaire

a research method consisting of a series of questions for the purpose of gathering information from respondents

correlation

a non-experimental method that examines whether there is a relationship between two variables. it uses sets of data that have NOT been manipulated by the researchers

positive correlation

this is when, as one variable increases, so does the other (i.e they both go in the same direction)

negative correlation

this is when, as one variable increases, the other decreases (i.e. they go in different directions)

controlled observation

an observation where the researchers set up the situation(s) and put the participants into this. they can also be non-experimental however, with no clear IV, but instead a series of events may take place in a stipulated order, so that different people's reactions to this sequence can be observed + compared

naturalistic observation

this is where the researcher has no influence over either the setting or the events that unfold - they simple observe + record what happens

behavioural category

when researchers must decide which specific behaviours should be examined. this will vary from study to study and should be determined by the theory they are investigating

event sampling

this is when the researcher records each and every event (that they are interested in), no matter when it happened. other things that happen are not recorded

time sampling

this is when the researcher records what is going on at specific times, no matter what it is. events that happen outside of this time period are not recorded

participant observation

this is where the observer actually becomes part of the group being observed

non-participant observation

this involves the researcher observing the participants from distance, and recording their observations

covert observation

a type of observation where the participants DO NOT KNOW they are being studied

overt observation

a type of observation where the participants know they are being studied.

tally chart/behavioural checklist

a method where the researcher counts the number of times a particular behaviour occurs

coding grid

this is a more sophisticated version of a tally chart, which allows the researcher to note not only how many times a particular behaviour occurred, but when it occurred also.

rating system

with this technique, the researcher will have previously decided on the aspects of the situation they wish to record, but this allows them to evaluate each aspect on a sliding scale (continuum)

diary description

when using this method, the researcher attempts to note down all the behaviours that they saw occurring, in sequence, and may distinguish between verbal and physical actions. (this may be written by participants themselves without the researcher present, when it is done this way, it is a method of self-report, not an observation)

taping

this may be done with just audio tape or with video tape/film

reliability

the consistency of a test/research procedure

test-retest reliability

this is a commonly used method of assessing the consistency of a test. when using this method, a researcher will administer the test/procedure to a group of participants and their scores are noted. the same test is then given to the same group of participants at a later date and their scores noted

split-half reliability/internal consistency

this method assesses how reliable a test is within itself (i.e. do all the items on a questionnaire measure the same ability?). this involves comparing participants' scores on half the questions/procedures, with their scores on the other half. the two sets of scores would be analysed using correlation. an internally reliable test would show a strong positive correlation between the two sets of scores

parallel-forms reliability

in some research studies, it may be necessary to design two sets of materials that aim to test the same ability, but which are not exactly the same test. this involves taking a large group of participants, none of whom are to be used in the actual study, and giving them both forms of the test and noting their scores on both forms. this would be analysed using correlation, and would hopefully show a strong positive correlation

inter-rater reliability

the degree of agreement between two or more assessors.

intra-rater reliability

the degree of agreement that one assessor has with themselves

validity

the truthfulness + accuracy of the information gathered by research

face validity

this involves simply 'looking at' the test to see if it appears to be measuring what it's supposed to measure 'on the face of it'.