Research Methods Flashcards
theory
an explanation for behaviour, tested using objective research methods
aim
a general statement explaining the purpose of a study
independent variable (IV)
the variable that the researcher manipulates
dependent variable (DV)
the variable being measured
operationalisation
making variables clearly defined and measurable
alternative hypothesis
statement of relationship beteeen variables
null hypothesis
a statement of no relationship between the variables
cause and effect
the only thing that should cause a change in the IV is the DV
extraneous variables (EVs)
unwanted variables that could affect the DV
then the change in the DV is due to EV and not IV
instructions to participants
you should give the same information about the study to all participants
standardised procedures
using the exact same methods and procedures for participants in a research study
aims to control EVs
randomisation
using chance (e.g. tossing a coin) to control effects of bias when designing a study
target population
group of people being studied
sample chosen from target population
spample should represent target population for making generalisations
sampling methods aim to avoid bias
random sampling
each person has equal chance of selection
numbers of target populatiin in hat / random generator
evaluation: no bias as everyone has an equal chance of selection
takes time as need list of all members of target population
opportunity sampling
selecting people available
evaluation: quick and cheap
only represents the population from which it was drawn
systematic sampling
selecting every nth person from list of target population
evaluation: avoids researcher bias
may end up with an unrepresentative sample
stratified sampling
selecting participants in proportion to frequency in target population
evaluation: most representative method
very time-consuming to sort sub-groups
ethical issues
conflict between participants’ rights and well-being and the need to gain valuable results
informed consent
participants should be told the purpose of research and that they can leave at any time
deception
participants should not be lied to or misled about aims
mild deception can be justified
privacy
participants have a right to control information about themselves
confidentiality
personal data must be protected and respected
BPS guidelines
a code of conduct all professional psychologists in the UK must follow
dealing with informed consent
participants (or their guardians) sign a form that tells them what is expected
dealing with deception and protection from harm
participants have a full debrief to explain the true aims, reduce distress
dealing with privacy and confidentiality
participants should be anonymous (given numbers or referred to by initials)
reliability
a measure of consistency
quantitative methods - in terms of reliability
tend to be the most reliable
laboratory experiments - controlled and easy to repeat exactly
interviews / questionnairs - same person should answer same questions in same way. closed questions likely to be more reliable
observations - one observer should produce same observations if repeated, or two observers (interobserver reliability)
qualitative methods - in terms of reliability
less reliable
case studies and unstructured interviews are difficult to repeat in the same way
validity
relates to whether a result is a true reflection of ‘real-world’ behaviour
sampling methods - in terms of validity
sample may not represent target population
representativeness low in opportunity sampling and high in stratified sampling
experimental designs
repeated measures - order effects challenge validity, overcome by counterbalancing
independent groups - participant variables challenge validity, overcome by random allocation
quantitative methods - in terms of validity
laboratory experiments - task, setting, participant awareness challenge validity. high control
field experiments - task and control challenge validity. more natural
methods producing numerical data (e.g. questionnaires) lack validity as they reduce behaviour to a score
qualitative methods - in terms of validity
case studies have greater validity as they give deeper insight into behaviour
difficult to analyse, which reduces validity
correlations
correlations show how things are linked together, associations
co-variables
correlations are quantitative - continuous, numerical data
scatter diagram
a special graph used to plot correlational data. one co-variable on the x-axis and the other on the y-axis.
a dot is placed where they meet
types of correlation
positive - as one co-variable increases, the other increases
negative - as one co-variable increases, the other decreases
zero - no relationship between co-variables
correlations - evaluation points
strengths:
- good starting point for research
- can be used to investugate curvilinear relationships, so many uses
weaknesses:
- don’t show cause and effect
- no control of EVs, so conclusion drawn may be wrong
interviews
face to face, real-time contact, though also on phone/text
structured interviews
interviewer reads list of pre-prepared questions
follow-up questions may be prepared as well
unstructured interviews
some questions prepard before
new questions created depending on what interviewee says
semi-structured interviews
some questions decided before but follow-up questions emerge
interviews - evaluation points
strengths:
- produce a lot of information
- insight gained into thoughts and feelings
weaknesses:
- data can be difficult to analyse
- people may feel uncomfortable talking face to face
questionnaires
prepared list of questions which can be answered in writing, over the phone, internet, etc.
open and closed questions
open questions tend to produce qualitative data
closed questions have a fixed range of answers, e.g. rating scale, yes/no
experiments
look at a measurable change in the DV (quantitative), caused by a change to the IV
laboratory experiments
experimenter has high control over what happens
takes place in a laboratory
laboratory experiments - evaluation points
strengths:
- EVs can be controlled, so cause and effect established
- use of standardised procedures permits replication, can demonstrate reliability
weaknesses:
- behaviour in a lab ‘less normal’, so difficult to generalise
- participants may change their behaviour because aware of being watched
field experiments
take place in a natural setting
IV manipulated by experimenter
field experiments - evaluation points
strengths:
- more realistic than lab experiments as in a natural environment
- can use standardised procedures so some control
weaknesses:
- may lose control of EVs so difficult to show cause and effect
- ethical issues because participants not aware of study
natural experiments
take place in a natural or lab setting
IV is not changed by the experimenter, it varies naturally
natural experiments - evaluation points
strengths:
- may have high validity because real-world variables
- can standardise so some control over EVs
weaknesses:
- few opportunities to do this kind of research as behaviours may be rare
- may be EVs because participants not randomly allocated to conditions
experimental designs
experimental designs are the different ways can be organised in relation to IVs / conditions of the experiment
independent groups
different group of participants for each level of the IV (condition)
control and experimental groups
independent groups - evaluation points
strength:
- order effects are not a problem
weaknesses:
- different participants in each group
- participant variables can act as EVs
independent groups - dealing with problems
dealing witn participant variables - allocation to conditions:
participant differences can be dealt with by using chance or systematic method to allocate participants to conditions
repreated measures
all participants take part in all levels of the IV (conditions)
repeated measures - evaluation points
strengths:
- no participant variables
- fewer participants needed, so less expensive
weakness:
- order effects reduce validity, e.g. practice effect
repeated measures - dealing with problems
dealing with order effects - counterbalancing:
half participants do conditions in one order, other half do opposite order
matched pairs
participants tested on variables relevant to the study. participants then matched and one member of each pair goes in each condition.
matched pairs - evaluation points
strength:
- no order effects (fewer participant variables)
weaknesses:
- takes time to match participants
- doesn’t control all participant variables
case studies
an in-depth investigation of an individual, group, event, or institution
a qualitative method (case studies)
collects information about people’s experiences in words.
may include quantitative data, e.g. IQ scores.
longitudinal (case studies)
often carried out over a long period to see how behaviour changes.
may also collect retrospective case history.
case studies - evaluation points
strengths:
- research lacks specific aims so researcher more open-minded
- best way of studying rare behaviours
weaknesses:
- focus on one individual or event, so often can’t be generalised
- subjective interpretation of events
observation
a researcher watches or listens to participants, and records data
observations - natural vs. controlled
natural - record behaviour where it would normally occur
controlled: researcher manipulates aspects of environment
observations - covert versus overt
covert - participants not aware behaviour is being recorded
overt - told in advance
observations - categories of behaviour
target behaviour broken into separate observable categories
observations - interobserver reliability
two observers should produce the same record of behaviour
researchers watch at the same time, and correlate data
observations - evaluation points
strengths:
- greater validity because based on what people do
- real-life behaviour when participants not aware of being observed
weaknesses:
- ethical issues as can’t gain consent if observing in a public place
- observer bias - observer’s expectations can affect validity
quantitative data (including evaluation)
quantities (numbers) but can measure thoughts / feelings
evaluation points:
- easy to analyse and draw conclusions
- lacks depth, not reflecting real-world complexity
qualitative data (including evaluation)
data in words but can be turned into numbers by counting themes
evaluation points:
- more depth and detail
- difficult to analuse and summarise
primary data (including evaluation)
data that has been obtained first hand
evaluation points:
- suits the aims of research so more useful
- it takes time and effort to collect
secondary data (including evaluation)
second hand data from other studies or government statistics
evaluation points:
- easy and convenient to use, saving expense
- it may not fit what the researcher is investigating
descriptive statistics
express numbers in a way that shows the overall pattern
e.g. mean, median, mode, and range
range (including evaluation)
spread of data
arrange data in order and subtract lowest from highest score
evaluation points:
- easy to calculate
- can be distorted by extreme values
mean (including evaluation)
mathematical average
add up all scores and divide by the number of scores
evaluation points:
- uses all the data, so most sensitive measure
- can be distorted by extreme values
median (including evaluation)
middle value
data put in order from lowest to highest
evaluation points:
- not effect by extreme scores
- less sensitive than the mean to variation in values
mode (including evaluation)
most common score
evaluation points:
- very easy to calculate
- can be unrepresentative
scatter diagrams
to display correlation
one co-variable on x-axis and the other on y-axis. a dot is placed where co-variables meet
frequency tables
frequency means the number of times it occurs
frequency tables are a systematic way to organise data in rows and columns
frequency diagrams
histogram - continuous categories, no spaces between bars
bar chart - bars can be in any order
normal distribution - symmetrical spread forms a bell shape with mean, median, and mode at peak
decimals
any number written with a decimal
position represents value
fractions
reduce to simplest form
ratios
a way to express fractions
8:2 can be reduced to 4:1
percentages
fractions out of 100
finding the arithmetic mean
add all the scores and divide by number of scores
standard form
a mathematical shorthand to represent very large or small numbers
e.g. 3.23 x 10^6 is 3,230,000, and 3.23 x 10^-5 is 0.0000323
significant figures
simplifying a number to a certain number of places
e.g. 32,462 to 2 sf is 32,000 and 0.003256 to 2 sf is 0.0033
estimate results
a rough calculation
