Doing research Flashcards
what is experiment
a method of collecting data which measures the effect of an IV on a DV by controlling other variables
what is interview
a method of collecting data that involves directly questioning people
what is questionnaire
a method of collecting data that involves people answering a series of pre-determined questions
what is case study
a method of collecting data that involves focusing on a small sample in detail
what is laboratory experiment
where an IV is manipulated in a controlled environment to test its effect on a DV
what is field experiment
where an IV is manipulated in a natural environment to test its effect on a DV
what is natural experiment
where an IV is not directly controlled by the experimenter but its effect on a DV is still tested
W and L of lab experiment
they have more control over extraneous experiments, more reliable than field experiments and easier to recreate
they have low ecological validity because of their artificial settings, they suffer from demand characteristics as participants are aware that they are in a lab experiment
W and L of natural experiment
experiment can be ethic as nothing is manipulated
difficult for researchers to match up the participants they are comparing if the IV is already in place
W and L of field experiment
higher ecological validity over laboratory experiments as the environment is authentic even if the IV is set up
less control over extraneous variables and affect the reliability of results
structured interview
an interview with pre-determined questions
unstructured interview
an interview where questions vary depending on the interviewee’s answers
W of interview
they can access people’s thoughts and feelings
they can clarify what questions or answers mean and increase chance of more valid responses
L of interview
people may lie or exaggerate when answering questions, which affects the validity of results
they rely on people being able to explain their thoughts and feelings
close questions
questions which have set responses to choose from
open questions
questions that have no fixed responses so participants can respond how they wish
rating scales
a way of answering a close question that requires the respondent to select a number to represent their response
W and L of close
they give quantitative data and easy to quantify participant’s responses and identify patterns
lack construct validity
W and L of open
more qualitative data, more in-depth responses, more valid results
data is difficult to analyse for trends
W and L of questionnaire
access people’s thoughts and feelings
possible to compare answers and look for reliable patterns and trends
misunderstand questions as they do them alone, unanswered or reliable answers
do not consider individuals, personal responses, affect validity
naturalistic observation
observing people in a real-life setting
lack of control over EV, less reliability
controlled observation
observing people in an environment that has been set up in some way
artificial, ecological validity
overt observation
observing people with their knowledge
behave differently, affect validity
covert observation
observing people without them knowing
cannot give consent, unethical
participant observation
observing people while also participating in their activities
can’t be objective, too involved
non-participant observation
observing people from a distance
miss important details, insight, not being involved
qualitative data
descriptive data
W and L of case studies
rich, in-depth qualitative data, high in validity as they go in detail
based on small samples, cases are rear, cannot represent, hard to generalise, subjective
quantitative data
numerical data
positive correlation
when two variables travel in the same direction
negative correlation
when two variables travel in opposite direction
zero correlation
when two variables show no relationship
correlation variable
a score that measures the strength and direction of the relationship between two co-variables
W of correlation
carry out statistical investigation of behaviours that cannot be experimented
high level of ecological validity as nothing is set up and manipulated compared to experiments
L of correlation
inform the researchers about the relationship between variable and does not tell them about a casual relationship
need to use quantitative data, lack of construct validity
reliability
a measure of how consistent or replicable something is
validity
a measure of how true or accurate something is
internal reliability
where a measure is consistent within itself
external reliability
where a measure is consistent beyond itself, on another occasion
inter-rater reliability
where two or more researchers agree on a set of results
ecological validity
how far a research setting mirrors real-life
unrealistic, artificial
construct validity
how far a variable is measured in relation to the whole concept
quantitative measures, narrow measure of subject
population validity
how far a sample represents the target population
a small bias sample
demand characteristics
cues from the procedure that suggest what the research is about
when participants know the aim, behave differently
observer effect
the effect on behaviour when people know they are being observed
behave differently as they know that they are being observed
social desirability
the pressure to respond to questions in a way that is expected and acceptable
they want to look good in the eyes of the researcher