(BRB) hypothesis, variables and sampling (pp1)) Flashcards
describe the order of the scientific method
- a phenomenon or behaviour is noticed
- theories are developed to explain or describe this
- hypothesis (testable predictions) or research questions are written
- studies and procedures are designed to test these predictions and/or answer these questions
- systematic and controlled data collection takes place
- analysis and interpretation of the data collected reaching conclusions
what is one of the main goals of psychological research
to describe and explain human behaviour
what are the 5 techniques for collecting/ analysing data
- experiments
- self-report methods
- observations
- correlations
- case studies
define a variable
- aspects of the world that are observable and measurable
define an experiment
- measure the effect of one variable on another, looking for a difference between two conditions
define a correlation
- describe a relationship between two variables
define an independent variable
- the variable which is manipulated by the experimenter
define a dependant variable
the outcome which is measured by an experimenter
what does the use of experiments in psychology allow researchers to establish
- cause and effect links between two variables
what do the IV and DV do in an experiment
- they are used to write an alternate hypothesis, the IV is predicted to have a measurable effect on the DV
what is an alternate hypothesis
- a prediction in the form of a testable statement
describe the nature of variables in correlations (co-variables)
- the researcher is attempting to measure the nature and association between 2 variables which are known as co-variables
what does a correlational study allow us to establish
they allow us to establish the strength (strong or weak) and direction (positive or negative) of the association between these two variables
what is the definition of operationalising variables
- if we operationalise a variable, we say in precise terms how we will measure or manipulate it
give the definition for validity
- refers to the extent to which a study has measured what it claims to measure
- it is therefore very important that variables are operationalised in an appropriate manner
how do you operationalise variables without being bias
- operationalise in an objective manner, meaning collecting data without allowing personal interpretation
what do you do when operationalising variables in a correlation
- operationalise the variables carefully as correlations require qualitative data
- not an experiment so you must say how you would operationalise variable one and two
what are the ways of collecting data for the 2 variables in a correlational study
- give participants a test under controlled conditions
- ask participants to self rate themselves
- gather data on participants from data bases
- observe participants behaviour in a structured quantitative manner
- give participants a questionnaire that results in quantitative data
what is an extraneous variable in an experiment
- any variables that might have an effect on the DV apart from the IV
what is a confounding variable in an experiment
- uncontrolled extraneous variables that have had an effect on results
how can confounding variables affect the reliability of an experiment
- when confounding variables interfere with the variables that we are studying they can reduce the reliability
how can uncontrolled extraneous variables affect the validity of an experiment
- uncontrolled extraneous variables (confounding variables) are also a threat to the validity of the research as they influence how accurately we can measure the DV
what is a situational variable
- variables in the research situation itself can have an effect on the dependant variable
how can you prevent the difference in the room conditions where the participants are tested affect the outcome
- standardise the procedure
- standardise instructions
what is standardisation
- the materials used usually have to be the exactly the same for all participants so they experience the same conditions
- this controls a possible source for confounding variables
what are pilot studies
- a way to test materials
- a smaller version of the actual study and allows identification of problems and modification of materials
- e.g wording of questions in a questionnaire
what are participant variables
variables that come from the individual other than the IV that have an effect on there personal performance
what are order effects
- the order in which participants carry out tasks that has an effect on there performance
- e.g AB,BA
what is counterbalancing
- often used to control order effects
- half the participants complete the task on one order then the other half complete it in the reverse order
what are demand characteristics
- all of the cues that a participant might receive that indicate the purpose of the research
what are researcher effects ?
-if a researcher knows the aim of the study and an expected outcome, they may struggle to maintain objective and influence the results whether they mean it or not to reflect there expectations
what is an alternate hypothesis
a prediction in the form of a testable statement
when is alternate hypothesis used ?
in either an experiment or a correlational study
what is an experimental hypothesis
predicts a cause and effect relationship between the IV and the DV
what is the definition of significance
- the results are statistically meaningful
- the results could NOT have been due to chance
what is a correlational hypothesis
- predicts a consistent association between 2 co-variables
what can you noy talk about when a hypothesis predicts a correlation between 2 variables
- a cause and effect relationship
what is directional hypothesis (one tailed)
- predicts the direction the difference or correlation will be
what is non-directional hypothesis (two tailed)
- predicts only that there will be a difference or correlation thus allowing for an outcome in either direction
what is null hypothesis
- predicts that any difference between the two groups researched will be due to chance
what does the null hypothesis predict in a correlational study
- that any correlation found is not significant (due to chance)