Research methods Flashcards
What is the definition of ethics?
The moral principles that govern a person’s behaviour or how they conduct an activity.
When do ethical issues arise?
when there is a conflict between the rights and dignity of participants and the goals and the outcomes of research
What is the British Psychological Society?
a document which instructs psychologists what is and isn’t acceptable when dealing with human participants in research
What are the BPS guidelines for research?
Can- confidentiality
Do - deception
Can’t - consent
Do - debreif
With - withdraw
Participants - protection from harm
What is the role of the BPS code of ethics?
-protect participants, patients and clients
- helps maintain and promote professional standards
- provides a framework within which psychologists work
- guides decisions about appropriate and acceptable conduct
What 4 areas are the guidelines organsied into?
respect
compentency
responsibility
intergrity
What is compentency?
Psychologists should only give advice if they are qualified to do so.
What is intergrity?
psychologists should be honest and accurate while maintaining professional boundaries and avoid exploitation.
How can you deal with ethical issues?
follow ethical guidelines (BPS) - they regulary update them
ethics comitees
What is the role of ethics comitee?
they scrutinize research and proposals to ensure that the rights of all participants are fully respected
How can you deal with consent issues?
participants are asked formally to indicate their agreement to participate
What is presumptive consent?
taking a large random sample from the population to be studied and introduce them to the research design with deception. If they agree/disagree, they represent the views of each population group.
What is Prior general consent?
people involved are told they may be deceived about the true purpose of the study. Those who agree will be participants.
How can you deal with deception issues?
it must be approved by an ethics comitee = cost vs benefits
participants should be fully debreifed after the study and offered the oppotunity to withhold their data ( especially if retrospective consent was used)
role-playing
What is retrospective consent?
participants are used for their consent (during debriefing) having already taken part in the study. They may not have been aware of their participation.
How can you deal with withdrawal issues?
participants should be informed at the beginning of a study that they have the right to withdraw and allow this if they so wish.
They should be told of their right to withdraw their data after the investiagtion.
How can you deal with issues with protection from harm?
participant must require counselling
stop the study
avoid any risks greater than everyday life
What is an aim?
it is a general statement that describes the purpose of an investiagtion. It was developed from a theory and it should be clear and precise.
‘To investigate’
What is a hypotheses?
a statement that is made at the start of a study and clearly describes the relationship between the variables as stated by the theory.
What is a direction hypotheses?
has a clear difference that is anticipated between the two conditions
What is a non-directional hypotheses?
simply states that there is a difference between the conditions or groups of people
What is an alternative hypotheses?
states that there is a relationship between two variables being studied. It is what you would expect from the experiment.
What is a null hypotheses?
states that there is a relationship between the two variables. It is what you wouldnt expect from the experiment
What does falsifiability mean?
it is the logical possiblity that a hypotheses or theory can be proven to be false by an observation or experiment. Hypotheses should never be proven to be true, it should only be subjected to research attempts to prove them false.
What is quantitive data?
a measurement. numerical form
What is qualitative data?
opinions, statements, decriptions
What is the independant variable?
some aspect of the experiment that is manipulated by the researcher or changes naturally
What is the dependant variable?
what is measured by the researcher
any affect on the DV should be caused by the change in the IV
What is operationalisation?
clearly defining variables in terms of how they can be measures
all groups results must be recored in the same format
What is an extraneous variable?
any variable other than the IV that may affect the DV if it is not controlled. It does not vary systematically with the IV
What is a confounding variable?
a type of extraneous variable that varies systematically with the IV. you can not tell if the change in the DV is due to the change in IV or the CO-V
What are demand characteristics?
any cue from the researcher or the research itself that may be interpreted by the participants as revealing the purpose of the inestigation = participants could change their behaviour ( over-perform/ under-perform).
Participant reactivity is a significant EV that is difficlut to control.
What is the ‘Please-U effect’?
when participants over-perfrom to please the researcher
What is the ‘Screw-U effect’?
when participants under-perfrom to mess up the results
What are investigator effects?
any effects by the investigator’s behaviour (conscious/ unconscious) on the research outcome (DV)
eg: design of study, selection of/ interaction with participants, materials used, instructions, leading questions
What is randomisation?
the use of chance methods to control for the effects of bias when designing materials and the order of experimental conditions
Why is randomisation used?
to minimise the effect of EV/ CO-V on the DV
reduces researcher unconscious bias
attempts to control investiagtor effects
What is standardisation?
using exactly the same formalised procedures and intructions for all participants in a research study. All participants should be subject to thhe same environment, information and experience.
What is objectivity?
the tendency to base judgements and intpretations on external data rather than on subjective factors (feelings, beliefs, experiences)
What are the qualities of a laboratory experiment?
highly controlled environment
IV manipulated
high control over EV and CO-Vs
random allocation of Participants
Dv measured in control environment
What are the strengths of a lab experiment?
highly controlled = limits EV and CO-vs = high internal validity and increases confidence in results
replication is possible and easy to do = validity and confidence
What are the weaknesses of a lab experiment?
lacks generalisability
demand characteristics
loss of ecological validity and mundane realism
artificial environment, not generalisable to the real world
What is a feild experiment?
the IV is manipulated in a natural way, it is conducted in a more everyday setting
the researcher goes to the participants usual environment
no control over EV and CO-V
no random allocation of participants
Dv measured in uncontrolled environment
What are the strengths of a feild experiment?
high mundane realism and ecological validity
real-life setting = results more valid and authentic
reduces demand characteristics, participants dont know they are being observed = validity
What are the weaknesses of a feild experiment?
lack of control of EV = real life setting makes it difficult to establish casue and effect
ethical issues with consent
What is a natural experiment?
an experiment where the change in IV is not brought about by the researcher but would of happened even if the researcher had not been present
What are the qualities of a natural experiment?
naturally changing IV
IV not manipulated
no control over EV/ CO-V
Participants not randomly allocated
DV measured in natural environment
can be tested in a lab
What are the strengths of a natural experiment?
can study sensitive issues that wouldn’t be ethical to manipulate
high external validity = real world problems and issues as they are happening
What are the weaknesses of natural experiment?
naturally occuring events happen rarely = reduces oppotuntites for research, limits generalisation
reduced likehood of desired behaviour if IV is not commonly ocurring
lack of control = reduces likehood of cause and effect
no randomisation to experimental conditions = EVs could be present
In lab setting = lacks realism and demand characteristics
What is a quasi experiment?
The IV is fixed and is an already exisiting difference between people
What are the qualities of a quasi experiment?
IV not manipulated
control over EV and CO-V
participants not randomly allocated
DV measured in control environment
What are the strengths of a quasi experiment?
carried out (often) uner controlled conditions
replication is easy to do
What are the weaknesses of quasi experiment?
no randomisation of participants to conditions
IV not deliberately changed by the reaseacher = can not claim the IV has a cause and effect on the DV
What is an experimental design?
the different ways which participants can be organised in relation to the experimental conditions
What is an independant groups design?
participants are allocated to different groups where each group represents one experiment condition
What are the strengths of independant groups design?
order effects do not effect the research
demand characteristics are minimised
the same test and materials can be used
What are the weaknesses of a independant group design?
subject variables differ = can become a CO-V unless controlled
worse statistical tests can be used = more variation between conditions
more subjects required = less economical
How can you deal with the limitations of independant group design?
randomisation with conditions to avoid differing subject variables
What is a repeated measures group design?
all participants take part in all conditions of the experiment
What are the strengths of a repeated measures design?
subject variables are the same in all conditions
better statitical tests can be used = less variation between conditions
fewer subjects required = more economical
What are the weaknesses of a repeated measures design?
order effects = errors
demand characteristics
more materials required if experiment is being repeated
How do you deal with the limitations of a repeated measures group design?
counterbalancing
What is counterbalancing?
an attempt of controlling the effects of order by alternating the order in which participants experience the different conditions in a repeated measure
What are matched pairs design?
pairs of participants are first matched on the same variables that may affect the DV. Each member of the pair will experience a different condition. It is an attempt to control EVs and CO-Vs of subject variables.
What are the strengths of a matched pairs design?
subject variables are controlled
better statistical tests used
order effects do not occur
demand characteristics are minimised
the same test can be used = less materials = more economical
What are the weaknesses of a matched pairs design?
subject variables can not be matched perfectly
matching participants is time-consuming and expensive
could of missed relevant subject variables that could have an effect on the DV
another test may need to be done to match participants
more subjects required = less economical
How can you deal with the limitations of a matched pairs design?
restrict the number of subject variables to make matching easier
What is random allocation?
an attempt to control subject variables in an independant groups design which ensures that each participant has the same chnace of being in one condition as any other
What is a population?
a group of induviduals that a particular researcher may be interested in studying
What is a target population?
a subset of the general population that the researcher may be interested in
Why should a sample be representative?
so that generalisation can be possible
What is a sample?
a group of people who take part in the research, it is drawn from the target population and is presumed representative of that subset of the population
What is generalisation?
the extent in which findings and conclusions from an investigation can be broadly applied to the population
What is a sampling technique?
the method used to select people from the population
Why is it difficult to produce a representative sample?
majority of samples contain some degree of bias
What is bias?
when certain groups are over-represented or under-represented within the sample selected = limits generalisation
What is a random sample?
all members of the target population have an equal chnace of being selected
What is the method of random sampling?
the names on the list are given a number
sample is generated through the use of the lottery method
(computer-based randomiser/ picking numbers out a hat)
What are the strengths of a random sample?
unbiased = EV and CO-V are eually divided between the two groups = more internal validity
equal chance of selection
removes researcher bias
What are the weaknesses of a random sample?
difficult and time-consuming to conduct
complete list of target population may be difficlut to obtain
does not guarentee a represntative sample
participants may refuse to take part
What is systematic sampling?
every nth member of the target population is selected
What is the systematic sampling method?
la sampling frame is produced:
list of people in target population organised in an order
randomly generates a number to become the nth term
every participant in that nth term pattern is selected
What are the strengths of a systematic sample?
avoids researcher bias
objective
fairly representative
consistent
What are the weaknesses of a systematic sample?
less effective
not everyone has an equal chance of being seleted
cant guarentee a representative sample
time-consuming
participants may refuse to take part
What is a stratified sample?
the sample reflects the participants of people in certain subgroups (strata) within the target population
What is the method of stratified sampling?
- researcher identifies the different strata that make up the population
- proportions needed for the sample to be representative are worked out
- the participants make up each stratum selected using random sampling
What are the strengths of a stratified sample?
representative = designed accurately to reflect the composition of the population
generalisation can occur
avoids researcher bias
What are the weaknesses of a stratified sample?
time consuming
the strata can not reflect all the ways that people are different = complete representation is not possible
What is an oppotunity sample?
where paticipants who are most convinient or most available are recruited for the study
What are the strengths of a opputunity sample?
covinient
less costly in terms of money and time
simple
What are the weaknesses of an opputunity sample?
unrepresentative = no generalisation
researcher bias
What is a volunteer sample?
involves participants selecting themselves to be part of the sample
researcher needs to advertise
What are the strengths of a volunteer sample?
minimal input from researcher
less time consuming
wide range of people available
people more willing to participate
What are the weaknesses of a volunteer sample?
volunteer bias
demand characteristics = over-perfrom
unrepresentative and bias
What is a pilot study?
a small-scale version of an investigation that takes place before the real investigation is conducted to identify any potential errors
What are the purposes of a pilot study?
see if the investiagtion runs smoothly
indentify potential issues
to question the full scale investigation
to see if experiment is worth doing
to determine number of participants needed
prevent major issues
What is a single-blind procedure?
participants do not know the aim of the study and the condition they are in = an attempt to control deman characteristics
What is a double-blind trial?
reseracher and participants do not know the aims of the investigation and whos in what condition. A third-party conducts the investigation without knowing its main purpose = controls demand characteristics and investigator effects
What is a control groups and conditions?
they are used for the purpose of comparison and provides a baseline measurement for the investigation
What are non-experimental methods?
characterised by the fact they have no IV that is manipulated and therefore no established cause and effect
What are the 3 examples of non-experimental methods?
observations
self-report techniques
correlations
What are self-report techniques?
questionaires
structured and unstructered interviews
What are correlations?
the analysis of the relationship between CO-Vs
What is a naturalistic observation?
the behaviour is observed in natural situations
no direct manipulation of variables and therefore no attempt to influence the behvaviour being studied
When can naturalistic observations be used?
in situations where intervention would be inappropriate
with non-human animals especially those that don’t adapt well to lab experiments
What are controlled observations?
when a researcher controls aspects of the situation it can be covert or overt
What does covert mean?
the ps are being observed are not aware they are being observed
What does overt mean?
ps are aware they are being observed
What does participant and non-participant mean in an observation?
participant = researcher joins the group
non-participant = the researcher remains seperate
What are the advantages of using observations?
preliminary research tool = careful use of naturalistic observation - identification of appropriate hypotheses for further reseach which may help save time being wasted on unrealistic experiments
validity = naturalistic observations, findings may apply outside lab conditions - greater external validity
What are the disadvantages of observations?
there is a lack of cause and effect = no IV and level of control of Co-Vs
observer effects = the presence of an observer may change ps behaviour
categorisation of data is difficult as behaviour is being studied
What are the ethical issues of observations?
privacy = it should be carried out in places where people would expect to be observed by strangers
confidentiality = if published, their data must be identifiable as theirs. if anonymity is not guaranteed, ps should be able to give their consent to the data being published
What are the problems with participant observations?
the observers presence will have an impact on the group’s behaviour
What is an untructured observation?
researchers write down everything they see - produces rich detail
it is suited to situations where there are few ps and it is small scale
What is a structured observation?
uses bahvioural categories or sampling techniques to record data
What is time sampling?
involves observing and recording what happens in a series of fixed time intervals
What is event sampling?
it involves observing and recording a complete event of interest each time it occurs
What are behavioural categories?
involves breaking down target behaviours into a checklist
the behaviours must be precisely defined, observable, objective and mutually exclusive
each time a specific behaviour is observed it is recorded as a tally
at the end of the observation = you count the no. of times each behaviour was recorded and make comparisons between ps
What is a peer review?
the assessment of scientific work by specialists in the field to ensure that any research that is to be published is high quality
What are the aims of a peer review?
- to validate the quality and relevance of the research
- to suggest ammendments and improvements
- to allocate research funding
- to prevent publication of fraudulant research
What are the advantages to using questionaires?
simplicity = minimal training, specialists are not required
quick = large amount of data can be gathered from a large no of ps, relatively cheaply and in a short period of time
What are the disadvantages to using questionaires?
question wording = ps can interpret them in diff ways, leading Qs - ps respond in a certain way
unreliable = ps may not answer Qs accurately - social desirability
unrepresentative sample = reponse rates are usually low - could be a biased sample
What are the ehtical issues with questionaires?
confidentiality = must be guarenteed anonymity
informed consent and right to withdraw = ps should be informed of the aim and right to withdraw at any time
What are closed questions?
researcher determines the range of possible answers by directing ps to respond in a certain way
What is a good thing about closed questions?
easy to quantify and analyse
What is an issue of closed questions?
lack realism due to forced choice of answers
What are open questions?
where the researcher does not restrict the range of answers available
What is a positive and negative of open questions?
+ = produce rich info
- = more difficult to analyse
What are the 8 things that need to be considered when designing a questionaire?
leading questions
ambiguity
loaded questions
jargon
double-barrelled questions
negatives
complex questions
impossible questions
What is a leading question?
the question encourages respondants to give a particular answer
What is ambiguity in questionaires?
questions can be interpreted in various ways should be avoided
What are loaded questions?
questions that contain emotive language which is likely to produce an emotional reaction in the respondant
Why should jargon be avoided in questionaires?
technical jargon like psychological terms can confuse people as they dont know what it means
What are double-barrelled questions?
questions that contain 2 different options
What are the 3 types of interviews?
structured
semi-structured
unstructured
What is a structured interview?
all the questions are predetermined and the interviewer sticks firmly to their script
What are the advantages of structured interviews?
collect rich detail
easily repeated as the questions are fixed
What are the disadvantages of structured interviews?
interviewer effects = interviewer behaviour and attitude can influence answers from ps
social desirability
confidentiality can sometimes not be guarenteed
What is an unstructured interview?
the interviewer starts with a few questions but then created new questions in response to the ps answers
What are the advantages of unstructured interviews?
rich detail
allows for flexibility = issues can be explored in more depth by tailoring Qs
What are the issues with unstructured interviews?
interviewer effects
social desirability effect
confidentiality
What are semi-structured interviews?
combination of structured and unstructured interviews
What are the advatnages of semi-structured interviews?
flexibility
two-way communication = encourages conversational exchange = ps more comfortable
rich in detail
What are the issues with semi-structured interviews?
time-consuming = a lot of work to analyse them
less objective
unpredictable
confidentiality
What are the advantages of using questionaires over interviews?
cheaper
more reliable = less variation
less bias
anonymity and confidentiality
easier to analyse and conduct
What are the advantages of using interviews over questionaires?
more in depth answers and data
interviewer can clarify Qs for when ps dont understand
ps can express themselves freely
What are correlations?
they illustrate the strength and direction of an association between 2 co-variables
How are correlations analysed?
pair of scored obtained and plotted on a scatter diagram
What are the differences of correlations and experiments?
correlations are not looking for a difference but a relationship
No manipulation of IV in correlations = no C and E
What are the advantages of correlations?
detecting patterns = predictions can be made
eliminating C and E = no relationship means no C and E
large amounts of data = more rapidly analysed
What are the disadvantages of correlations?
no C and E = other variables involved?
cannot identify curvliniar relationships
What is secondary data?
data that has been collected by someone other than the person conducting the study
What is an issue with meta-analysis?
prone to publication bias = researcher may remove studies that are contradictory
What are the 2 types of decriptive stats?
measures of central tendancy
measures of dispersion
What is measures of dispersion?
tells you how widely spread the data is
What are the 3 measures of central tendancy?
mean
median
mode
What are the advantages of the mean?
takes all scores into account - representative
can be used in further stats tests
What is the disadvantage of the mean?
anomalies distort the mean - unrepresentative
What is the advantage of the mean?
unaffected by extreme scores - most appropriate with anomalies
used for skewed data
What is the disadvantage of the median?
doesnt take into account of all the data = less sensitive than the mean
What are the advantages of the mode?
unaffected by anomalies
useful when data is in categories
What is the disadvantage of the mode?
unstable = a small change in data can greately affect the mode
What are the 2 measures of dispersion?
the range and the SD
What is the advantage of the range?
easy and quick
What is the disadvantage of the range and the standard deviation?
they are very vulnerable to rouge scores - very high or very low scores
What is the standard deviation?
the average amount all scores deviate from the mean
What is the advantage of the standard deviation?
more accurate than the range = includes all data
more sensitive measure
What is the use of graphs?
a visual aid to help make sense of quantitative data obtained from psychological research = provide an overall pic to summarise the results
When are histograms used?
when representing continuous data on interval or ratio scales of measurement
When are bar charts used?
used when data is in categories (nominal or ordinal) or to illustrate average scores from different samples
When are frequency polygons used?
when you need to compare 2 or more frequency distributions
presenting continuous data on interval or ratio scales
When are scattergrams used?
used to present correlational data
How do you identify a normal distribution by a graph?
a symmetrical bell-shaped curve
How can you identify a normal distribution from a table?
the median, mean and mode are all equal
What % of data lies in 1 standard deviation?
68%
What % of data lies in 2 standard deviations?
95%
What is a positive skewed distribution?
most of the data lies to the left side of the graph
What is a negative skewed distribution?
most of the data lies to the right side of the graph
Why are medians used instead of the mean?
anomalies in data would distort the mean but the median includes all data even anomalies without being distorted