research methods Flashcards
what makes something a science?
paradigms, theory construction and hypotheses, falsifiability, replicability, objectivity
what is a paradigm?
a clear, distinct concept accepted by the majority of people in a scientific field
what did philosopher Kuhn say?
that a paradigm makes something a science
we could argue that some paradigms or psychology are…
that the brain influences/ causes behaviour, lab experiments are required to prove theories
what is a paradigm shift?
when a group of researchers begin to question an accepted paradigm, and there is too much evidence to ignore, so the paradigm changes
give an example of a paradigm shift
the earth is flat -> the earth is round
what does science test?
theories
what is a theory?
a set of general laws or principles which can explain events or behaviours
theory construction occurs through
gathering evidence via direct observation
the role of hypothesis testing is…
ensuring it is possible to make clear and precise predictions based off the theory
evidence supporting the theory leads to it being
strengthened
evidence refuting the theory leads to it being
revised, revisited or rejected
the steps of creating a theory are…
- formulate a question
- construct hypothesis
- research and observe
- test and experiment
- analyse the results and conclude
has psychology has a paradigm shift?
it can be argued yes- approaches from psychodynamic to behavioural to cognitive
why might psychology not be seen as a science?
too many internal arguments, no paradigm
an example of where psychology met theory construction…
Bandura BoBo doll study, 1961
an example of where psychology did not meet theory construction…
Freud’s psychodynamic approach
who said that falsifiability made something a science and when?
Karl Popper, 1930s
what does falsifiability mean?
a scientific theory must have the ability to be proven wrong
nothing can be proven, as we cannot test everyone in the population. however, it is stronger as
it hasn’t been proven false yet. it increases in strength the longer that it is around without being disproven.
what is a pseudoscience
a science which lacks falsifiability
which element of psychology is a pseudoscience
psychodynamic approach, can’t be proven wrong as involves unconscious mind
replication helps to improve…
validity
if we are to trust a scientific theory, its findings must be shown to be…
repeatable across various contexts and circumstances
replicating findings shows us to what extent they can be…
generalised
scientists need to investigate in an - manner
objective
objectivity is…
something based on facts rather than opinion, which is true for everyone
empirical methods emphasise the importance of…
data collection based on direct, sensory experience
a theory cannot claim to be scientific unless it has been…
empirically tested and verified
aims are… that describe…
general statements, the purpose of an investigation
the independent variable is what we
manipulate
the dependent variable is what we
measure
the experimental group tests the…
independent variable
the compare group allows us to
compare the groups
a hypothesis is…
a statement that is made at the start of a study that states the relationship between variables
a hypothesis is not a question, it is a
testable statement
the 3 types of hypothesis are…
- directional (one tailed)
- non directional (two tailed)
- null hypotheses
a directional hypothesis states
a particular outcome that the researcher is expecting specific to which group will score higher
a non directional hypothesis states
that there will be a difference between the groups, without saying which will be higher
a null hypothesis states that
there will be no difference between the two conditions
when do we use directional hypotheses
when findings of previous research suggest a particular outcome
when do we use non directional hypotheses
when there are no previous findings or previous findings are contradictory
in a good experiment, the independent variable…
affects the dependent variable
variables other than the independent variable which may affect the dependent variable must be…
controlled or removed
situational means that the variable is
in the environment, eg lighting, noise, distractions in the room.
participant means that the variable us
in the people, for example age, gender, or previous experience
extraneous variables affect
both groups in the study
extraneous variables are usually easy/hard to control
easy
extraneous variables may cause
difference in the results of all participants, however don’t always ruin the study as we can still compare the groups
confounding variables affect
only one condition of the independent variable
confounding variables are a big problem, as they mean that
we can’t tell whether the DV has been caused by the IV, or because of the confounding variables
demand characteristics are
cues which allow participants to guess the aims of a study, causing them to behave artificially
demand characteristics are - variable
an extraneous
demand characteristics can either cause participants to act to - the experimenter, or to -
please, sabotage
investigator effects are
any unwanted influence of the investigator on the outcome of the research
investigator effects can be… variables
extraneous or confounding
randomisation is the use of.. to…
chance, reduce investigator effect
standardisation means that
to the extent possible, all participants should be subjected to the same environment and information
experimental designs refers to
the way in which participants are used in experiments
there are three experimental designs which are
- independent groups
- repeated measures
- matched pairs
independent groups are where
participants are put into 2 groups, and each group experiences a different condition
independent groups strengths:
- no order effects
- less chance of demand characteristics as participants only experience one condition
independent groups weaknesses:
- participant variables make it very hard to definitely say that the DV was caused by the IV, not participant variables
- less economical as more participants needed
repeated measures are where
one group of participants experience both conditions, and the same person is compared to themselves
repeated measures weaknesses:
- order effects can occur
- demand characteristics are more likely to occur as participants can guess the aims of the study more easily
order effects are where
the 2nd test is changed either by boredom or by practise.
order effects are - variables
confounding
repeated measures strengths:
- more economical
- no participant variables
matched pairs are where participants complete - condition, in -
one, pairs matched together based on a variable relevant to the study
in matched pairs, pairs are split and - into groups
randomly allocated
matched pairs strengths:
- less demand characteristics as participants only experience one condition
- no order effects
- far less problems with participant variables
matched pairs weaknesses:
- very time consuming
participants that form part of research studies are selected through the process of…
sampling
a population is a large group of people that
a researcher is interested in studying
a target population is a
subset of a larger population
a researcher will select a small group from their target population to study- this is the
sample
the sample needs to be - of the target population
representative
if the sample is not representative of the target population, then the findings of the study cannot be
generalised
the 5 types of sampling are…
volunteer, opportunity, random, systematic and stratified
in volunteer sampling, participants are selected through
self selection
an example of volunteer sampling is
Asch’s study
strength of volunteer samples
easy and not time consuming
weakness of volunteer sample
volunteer bias- study attracts a certain type of person who is willing to participate in a study, and findings may not be generalisable
in opportunity sampling, participants are selected through
the researchers asking whoever is around them at the time of the study
opportunity sampling strength
easy and not time consuming
2 weaknesses of opportunity sampling
- all participants drawn from the same place, may not be representative, generalisation issues
- researcher bias can occur
participants in random sampling are selected through the process of
a complete list of all members of the target population is created, members are assigned a number and then numbers are drawn at random
strength of random sampling is
completely free from researcher bias
3 weaknesses of random sampling is
- difficult
- time consuming
- can still create an unrepresentative sample
a sampling frame is
a list of members in the target population organised in a relevant format
in systematic sampling, participants are selected
as every nth member in a sampling frame, eg every 3rd person on the register
strength of systematic sampling is
no researcher bias
a weakness of systematic sampling is
sample may still be unrepresentative
strata are
sub groups in a target population
in stratified sampling the sample reflects the
proportions of strata within the target population
the process of stratifies sampling is
identifying the strata, work out the proportion of the target population they make, participants for each stratum are selected with random sampling
2 strengths of stratifies sampling are
- no researcher bias
- produces a representative sample which can be generalised confidently
a weakness of stratified sampling is
- very difficult and time consuming
the 5 ethical issues in psychology are
informed consent, deception, privacy, protection from harm and confidentiality
to participate in studies under 16s need
parental consent
informed consent means that people have the right to
know what they are consenting to before they consent to the study
informed consent is a problem in psychology as it leads to
demand characteristics and artificial behaviour
retrospective consent is where
the researchers get consent after the study is done
presumptive consent is where
researchers ask a similar group of people if they would consent to the study
deception involves
lying to or withholding information fro participants
active deception is where
researchers lie about the study
passive deception is where
information about the study is withheld from participants
in psychology deception means that participants have the right to
know what they are doing in the study
psychologists have to use some deception to avoid
demand characteristics
psychologists can only deceive
when it is absolutely necessary for the study
if participants are deceived, they have he right to 2 things
- a debrief after the study
2. their data to be destroyed if they wish
confidentiality is a legal right from the
data protection act
in studies, any identifying data must be
protected and not published
to maintain confidentiality, researchers must
use a coding system to avoid using participant names, avoid recording any personal information of participants
protection from harm means that participants have the
right
not to come to any physical or psychological harm during a study
participants should leave in the same
state that they arrived in
to follow protection from harm psychologists should
- conduct a pilot study to identify risks
- make participants aware of their right to leave at any point
the limitations of protection from harm are
that we can’t always identify the risks of a study
privacy means that people have a right not to
be observed where they wouldn’t expect it
the 3 ways to respect privacy are…
- respect if participants don’t want to answer sensitive questions
- observe people in public places only
- get informed consent
a limitation of privacy is
participants may behave artificially if they know they are being observed
what are the 4 types of experiment
lab experiment, field experiment, natural experiment and quasi experiment
lab experiments are conducted in
a highly controlled environment
2 strengths of lab experiments are
- control over extraneous variables increasing internal validity
- more replicable, so further studies can increase validity
2 weaknesses of lab experiments are
- lack mundane reality, causing participants to act artificially and decreasing internal validity
- leads to more demand characteristics
in field experiments, the independent variable is
manipulated by the researchers in a natural setting
a strength of field experiments is
they have mundane realism, so are more authentic, increasing the validity of the studies
two weaknesses of field experiments are
- some control of extraneous variables lost, reducing validity as findings can’t be proven to be consistent
- ethical issues if participants don’t consent to the study
in natural experiments, reseachers take advantage of
naturally occurring independent variables
2 strengths of natural experiments are
- allow studies to take place which may otherwise be rejected
- high external validity as behaviour is not at all artificial
2 weaknesses of natural experiments are
- only occur very occasionally, lack utility
- participants not randomly allocated within independent groups, confounding variables can occur
quasi experiments use independent variables
which exist within the population
strength of quasi experiment
high control over extraneous variables, higher internal validity
weakness of quasi experiment
participants not randomly allocated, confounding variables can occur
quantitative data is expressed
numerically
quantitative data is gathered through
empirical observations
quantitative data strengths
- easy to analyse
- easy to compare
- more objective
quantitative data weaknesses
- narrower in detail, leading to less external validity
- may not represent daily life
qualitative data is expressed
verbally
qualitative data strengths
- rich in detail, leading to greater external validity
qualitative data weaknesses
- difficult to analyse
- creates subjectivity and can be open to researcher bias
primary data is collected by
the researcher specifically for the study
primary data strengths
- data specific to the research
- questions targeted to gather information needed
primary data weaknesses
- time consuming
- requires more planning and consideration
secondary data is collected by
other researchers before the study
secondary data strengths
- less expensive
- requires less planning and consideration
secondary data weaknesses
- variation is quality and accuracy of data
- data may not meet specific needs of researcher, challenging validity of conclusions
mean strengths
- quick and easy to calculate
- most sensitive measure, more representative of the whole data set
mean weakness
- highly affected by abnormal values
median strengths
- easy to calculate
- not affected by abnormal values
median weaknesses
- not very sensitive, so less representative of the whole data set
mode strengths
- easy to calculate
- has to be used with certain data sets
mode weaknesses
- not representative of whole data set
- can be multiple modes, not at all useful then
range weakness
-only accounts for extreme values, not representative of whole data set
define correlation
a relationship between two variables where changes in one variable go along with changes in the other
correlations have no
independent or dependent variables
a positive correlation means that
as one variable increases so does the other
a negative correlation means that
as one variable increases the other decreases
a correlation coefficient is a
mathematical value for a correlation
correlations coefficient is between
-1 and +1
strong correlations have values more extreme than
0.5 or -0.5
the closer to 0 the correlation coefficient is, the - the correlation is
weaker
a non directional correlational hypothesis says
that there will be a significant relationship between A and B
a directional correlational hypothesis says
that there will be a significant positive/ negative relationship between A and B
correlations demonstrate that
there may be a relationship between two sets of data
correlations do NOT demonstrate
that one of the factors causes the other
which 4 things can cause a correlation?
- A causes B
- B causes A
- external factor causes A and B
- the correlation is completely random and the factors are not related at all
2 strengths of correlations are
- they can be used to study variables that would be unethical to manipulate, avoiding ethical issues
- they are a good starting point for experiments, researchers can check if there is any link between factors before spending time and money researching further
an unstructured observation means that
the researcher records everything they see
unstructured observations are usually done with
small scale observations
structured observations are where
quantified observations using a predetermined list of behaviours and sampling methods are made
structured observations strengths
- makes recording data much easier and more systematic
2. produces quantitative data which is easier to analyse
structured observations weaknesses
- data is far less detailed compared to unstructured observations
unstructured observations strengths
- provide much more detailed data
unstructured observations weaknesses
- greater risk of observer bias, as the researcher doesn’t have objective categories to look for
which type of observation is better?
it depends on the type of research being conducted
behavioural categories are
clearly defined, precise target behaviours used to measure a wider concept
behavioural categories example of affection
smiling, kissing or hugging
behavioural categories must be
unambiguous, observable, self-evident, exclusive
