Research Methods Glossary Flashcards
Independent Variable
The variable the psychologist manipulates.
Dependent Variable
The variable which is measured by the psychologist.
Extraneous Variable
Variables in a study that are not being measured or manipulated by the researcher but affect the results (DV) of ALL participants’ behaviour equally.
Confounding Variable
Variables in a study that are not being measured or manipulated by the researcher, that affect SOME participants’ behaviours but not others, having negative consequences for validity.
Operationalisation
Giving a precise definition of the behaviour being manipulated/observed/measured (IV/DV)
Aim
A broad statement of the purpose of the research.
Experimental/Alternate Hypothesis
A testable statement that a piece of research attempts to support or reject.
Null Hypothesis
A hypothesis that suggests there will be no difference/relationship between variables, any that does occur does so by chance
Directional Hypothesis
A hypothesis that suggests there will be a difference/relationship between variables, where the IV will affect the DV in one specific direction/outcome
Non-directional Hypothesis
A hypothesis that suggests there will be a difference/relationship between variables, the IV will affect the DV, but does not state a specific direction/outcome for results
Location: Laboratory
A location of research where scientific research and measurement can be taken using specialist/large equipment. Conditions and variables are controlled, and procedures are standardised.
Location: Field
A location of research outside of the laboratory, in a natural setting e.g. school, hospital, workplace etc. It has lower levels of control over confounding variables than the laboratory, but can still utilise a standardised procedure.
Location: Online
A location of research where participants are accessed via the internet/social networks/mobile apps etc. Often involves questionnaires but can also be experimental, correlational etc.
Lab experiment
A research method where cause and effect is measured in a controlled/contrived environment, through the control and manipulation of key variables, and where the participants are randomly allocated to experimental/control groups.
Field Experiment
A research method where cause and effect is measured in a natural setting, through the control and manipulation of key variables, and where the participants are randomly allocated to experimental/control groups.
Quasi Experiment
A research where the experimenter has NOT deliberately manipulated the IV and participants are NOT randomly allocated. Methods within this category include natural experiments and difference studies.
Natural Experiment
A research method where the IV arises naturally, although the DV can still be measured in a laboratory or any other location of the researcher’s choice. This method would be used when it is unethical to directly manipulate the IV.
Target Population
The group of individuals that a researcher is interested in studying
Sampling Frame
A group/population that is identified when it is unrealistic to study the whole target population
Random Sampling
A sampling technique where participants are selected from the sampling frame, where everyone has an equal chance of being selected. E.g. Names are pulled out of a hat, or a computer is used to randomly select participants.
Opportunity Sampling
A sampling technique where participants are selected at the researcher’s convenience without knowing any details about the sample in advance e.g. picking people who were there at the time, in your specific location.
Systematic Sampling
A sampling technique where every nth person on a list is selected by the researcher e.g. every 3rd house on a street, or 5th person on a register.
Stratified Sampling
A sampling technique where the target group is divided into subgroups, e.g. by sex, and then the participants are selected randomly from each subgroup.
Quota Sampling
A sampling technique where the target population is divided into subgroups, e.g. by sex, and the participants are chosen from each subgroup at the convenience of the researcher.
Self-selected Sampling
A sampling technique where participants volunteer (select themselves) for research e.g. they come forward/respond to the psychologist after reading an advertisement in a newspaper or on a notice board.
Snowball Sampling
A sampling technique where participants are initially recruited by the psychologist and then those participants recruit further participants from people they know, therefore the sample group appears to ‘snowball’.
Event Sampling
Where participants are observed by the psychologist, who records a specific behaviour (event) each time it occurs to create a total score.
Time Sampling
Where the psychologist observes and records behaviour (such as a score) at specific time intervals, e.g. every 15 minutes, and then creates an average score for each participant being observed.
Independent groups design
An experimental design where participants take part in only one experimental condition.
Repeated Measures design
An experimental design where participants take part in both the control and experimental conditions.
Matched Pairs design
A form of independent groups design where the experimental and control participants are deliberately similar e.g. there is a balance between gender and IQ levels in each group/condition.
Participant Observation
A research method where the researcher takes on the role of a participant whilst observing other participants’ behaviour around them.
Non-Participant Observation
A research method where the researcher watches and records participants’ behaviour without interfering in any way (from a distance).
Overt Observation
A research method where those being observed are aware of the fact.
Covert Observation
A research method where the researcher is “undercover”; the participants are unaware that they are being observed
Naturalistic observation
A research method where where behaviour is observed in real-world settings.
Controlled observation
A research method where researchers conduct observations of participants in a controlled environment.
Structured observation
A research method where researchers observe behaviour in a systematic way (behavioural categories/tally)
Unstructured observation
A research method where behaviour is observed more informally (no checklist)
Self-report
A method which involves a participant reporting on their own thoughts and feelings though methods such as interviews or questionnaires.
Questionnaire
A research method that includes a list of written questions, which generate closed and/or open answers. These can be used in person, on-line or through other methods
Structured interview
A research method that includes standardised questions (like in a questionnaire), which are known as an interview schedule, and are usually asked face to face.
Semi-structured interview
A research method which involves asking participants questions, usually face to face. These can be in the form of an interview schedule, but could also include follow up questions to expand on answers of the questions asked.
Unstructured interview
A research method which involves asking participants questions, usually face to face. This may start with an opening question, but the interviewer will pose further questions based on the response of the interviewee
Open Question
Questions that require elaboration from the participant
Closed Question
Questions with a fixed response
Content analysis
A research method which involves exploration of behaviour to see what categories, codes or themes emerge, and tallying each time material fits a theme.
Case study
A method which involves an in-depth investigation of a phenomenon, which uses a descriptive analysis of a person, group or event. It is a holistic study through one or more methodologies that is usually longitudinal.
Correlation
A research method which involves a comparing two continuous variables (co-variables) to see if there is an association/relationship between them.
Co-variables
Co-variables indicate two or more quantities being measured in a correlation by the researcher that may or may not vary with each other.
Quantitative Data
A type of data that can be measured numerically by the psychologist, so that statistical analysis can be completed
Qualitative Data
A type of data that can be observed, but not measured numerically. It usually takes the form of words, thoughts and feelings, and is difficult to analyse
Primary sources
Information sources/data that is directly collected by the researcher first-hand
Secondary sources
Information sources/data that have not been directly collected /created by the researcher
Cross-sectional studies
A method which involves comparing one group of participants, representing a cross-section of society, against another at the same point in time.
Longitudinal studies
A method which involves conducting research over a long period of time in order to observe long-term effects of X on a specific behaviour. It may utilise a range of other methodologies such as case studies, interviews etc.
Nominal data
The level of measurement that shows categories of data represented by frequencies. The data sets have no relative numerical value
Ordinal Data
The level of measurement where data can be placed into ascending or descending order, but the intervals between data not necessarily equal e.g. the times for first, second and third in a race.
Interval Data
The level of measurement that has equal numerical intervals between scores, with no absolute/true zero e.g. temperature. The interval between 1 and 2 degrees is the same as between 21 and 22 degrees,
Ratio Data
The level of measurement that has equal intervals between scores and has an absolute or true zero point e.g. speed (mph)
Probability value
A numerical value that gives an indication of the likelihood that results are due to a real difference/correlation and not due to chance e.g. in psychology we accept a probability value of 95%, where results are due to chance in 5% of cases.
Significance level
A numerical value that is usually expressed in value including two decimal places. This level tells you the margin of error that could occur in your results e.g. 0.05 suggests that there is a 5% possibility that results are due to chance and not the difference/correlation between variables.
Observed/Calculated value
The numerical value that is created as a result of inferential statistical analysis of your data. This will be compared to the critical values for the test to calculate the level of significance.
Critical value
The tabulated numerical values that have been assigned to a particular inferential statistical test. It is compared to the observed value for your set of data to calculate significance.
Frequency tables
A chart/table which shows the rate of occurrence (frequency) for a number of measured values/categories.
Bar chart
A diagram that represents frequencies of non-continuous data.
Line graph
A diagram that shows a linear representation of frequencies of data.
Histogram
A diagram that represents the distribution of frequencies for discrete or continuous data.
Pie chart
A diagram that represents data proportionately, as part of a whole picture of responses.
Scattergraph
A diagram that represents a relationship/correlation between two or more co-variables.
Normal distribution
A type of distribution where the mean, median and mode are equal. This can be referred to as a bell curve.
Negatively skewed distribution
A type of skewed distribution, where the mode is greater than the mean.
Positively skewed distribution
A type of skewed distribution, where the mode is less than the mean.
Measure of central tendency
A form of estimation of a mid-point/average in a set of data.
Mean
The average that is shown by all scores in the data set when they are divided by n.
Median
The mid-point in a set of data that has been placed in order.
Mode
The most common value within a set of data.
Measure of dispersion
A measure that shows the spread of data, whether it is tightly clustered or has a broader spread.
Range
A value which shows the spread of data, representing the difference between the lowest and highest scores.
Standard deviation
A value which represents the amount of variation of results from the mean score.
<
Less than
>
Greater than
≥
Greater than or equal to
≤
Less than or equal to
Privacy
A person’s right to control the flow of information about themselves
Confidentiality
Third parties should not be able to trace information back to individual participants. This is usually achieved through providing anonymity e.g. using participant numbers not names.
Deception
Deliberately misleading or falsely informing participants about the nature of research.
Right to withdraw
Participants understanding that they can leave the study at any time – and remove their data once the study is over
Informed (valid) consent
Includes giving participants enough information (in a form they can understand) so that they can make an informed choice about whether they wish to participate.
Protection from harm
Participants should not experience negative physical or psychological effects beyond what would be normal in day to day life
Debrief
A method which aims to ensure participants are aware of the true nature of the study (overcoming any deception). It should return the participant back to their original state.
Ethics committee
The board that is in place to ensure that proposed research meets the standards of the current ethical codes of conduct and guidelines.
Ethical guidelines
Provide a set of moral principles that guide research from its inception through to completion and publication of results.
Peer review
A process that takes place before a study is published to ensure that the research is of a high quality, contributes to the field of research and is accurately presented. The process is carried out by experts in the related field of research.
Demand characteristics
A type of confounding variable where participants unconsciously work out the aim and act differently
Counterbalancing
A technique used to deal with order effects when using a repeated measures design. With counterbalancing, the participant sample is divided in half, with one half completing the two conditions in one order and the other half completing the conditions in the reverse order.
Randomisation
A way of controlling for the effects of extraneous/confounding variables. Allocating participants to tasks, selecting samples of participants, and so on, should be left to chance as far as possible, to reduce the investigator’s influence on a study.
Researcher bias
Where the researcher either directly or indirectly influences the results of a study, through the process of designing the study or through the way the research is conducted/analysed.
Social desirability bias
Where participants give the response that they think will show them in the best possible light. This may mean that they are not a true reflection of their real thoughts/feelings.
Single blind procedure
A specific research procedure in which the researchers (and those involved in the study) do not tell the participants what condition of the study they are in
Double blind procedure
A specific research procedure where neither the participants nor the experimenters know which conditions the participants are in
Standardised procedure
Keeping everything the same for all participants so that the investigation is fair.
Validity
Accuracy
Internal validity
The findings are accurate and the effects on the DV are caused by the IV. Therefore the study measures what it intends to measure (as confounding variables have been controlled and will not affect the results).
External validity
Whether the study paints a true picture of real life behaviours (e.g. if the tasks have mundane realism) and whether the findings would apply to different places, different times, or different people
Ecological validity
A measure of how test performance predicts behaviours in real-world settings. How far can the results be applied to everyday life
Temporal validity
Refers to the validity of the findings in relation to the progression of time. How far can the results be applied to today
Population validity
Whether the findings from your sample can be reasonably generalised to a larger group of people
Concurrent validity
Validating a measurement by comparing it with an established measurement that has known validity. If similar results occur on both tests, then this new test is valid. If not, then the new test would have to be redesigned and tested.
Construct validity
The most sophisticated test of validity as it looks at whether the overall results reflect the phenomena as a whole (external validity). Checking the existing definitions of the behaviour being studied and redesigning the test if it measures a different construct.
Content validity
This objectively checks the method of measuring behaviour is accurate and decides whether it is a fair test that achieves the aims of the study (internal validity). Ask an expert in that specific area of behaviour to check the test is valid.
Predictive validity
The degree to which a test accurately forecasts a future outcome on a more broadly related topic. Do the findings apply in different and more varied situations? E.g. Do those with high IQ score gain higher grades in exams?
Face validity
The least sophisticated measure of validity. This validity is simply whether the test appears to measure what it claims to, and hence is subjective. Tests where the purpose is clear, even to naïve respondents, are said to have this validity.
Reliability
Consistency
Internal reliability
The extent to which a test or measure is consistent within itself e.g. the use of a standardised instructions and procedure for all participants.
External reliability
The extent to which a test produces consistent results over several occasions.
Inter-rater reliability
Where two or more psychologists produce consistent results by using a standardised procedure, agreed coding system, or correlation of their data.
Test-retest reliability
Involves testing and retesting the same participants over time, with the same test, and comparing their scores. If the scores are the same the test has external reliability.
Split-half reliability
Involves splitting a pp’s test answers in half and seeing whether s/he got the same or similar scores on the two halves. If so, internal reliability is high; if not, it is low and individual questions would need to be redesigned.