Research methods Flashcards
Why do psychologists carry out research
to find answers to questions they ask and collect data for real implication
What is the difference between primary and secondary data
-Primary research = the information the psychologist collects for the purpose of their aims
-Secondary research = pieces of information that already exist and are collected by others
What is the difference between quantitative and qualitative data
qualitative data = harder to analyze but gives rich detailed information that explains complex human behaviour
-quantitative data –> easy to put into a computer and analyse
What are the stages of the research project
1) Developing the aims of the research
2) Pilot studies
3) selecting samples + experimental design
4) collecting the data
5) analysing the data
6) Evaluating the project
What occurs during developing the aims of research
outline intentions and purpose of research, predicting what you will find
What occurs during pilot studies
decide on your research method and do a small scale pilot to check if it will work
What occurs during selecting samples and experimental design
select the group of people you will study and how you will group participants
What occurs during collecting, analysing and evaluating data
4) collecting the data —> go out and collect data using research method chosen
5) analysing the data –> place data into tables/graphs and apply statistical tests to find patterns
6) Evaluating the project –> research conclusions, strengths + limitations, suggest practical application, future research
What are the aims of pilot studies
small scale trial run of the investigation, check the procedures/material, allow researcher to make modifications if necessary
Why should psychologists use a pilot study
-check for errors/identify potential issues
-save money/time in the long run
-ensure research method is appropriate and not ambiguous
- check to see if research is ethical
Define the experimental method
involves manipulating one variable to determine if these changes cause changes in another variable. This method relies on high levels of control to test a hypothesis
What is a research aim
a general statement about what the researcher intends to investigate (purpose of research study
define a variable
things in the investigation that are manipulated, measured and controlled,
what is the independent variable
an aspect of the research situation that is manipulated
what is the dependent variable
the variable in the experimental situation that is measured by the researcher and should be caused by the independent variable.
what is operationalise
to be precise and clear about what is being manipulated and measured (testable and repeatable)
what is a hypothesis
statement of what you believe is true
what researcher predicts will happen
what are the steps to forming a hypothesis
1) possible hypothesis
2) 2 independent variables and another possible hypothesis
3) Operationalising the independent variables
4) operationalize the dependent variable
5) fully operationalised hypothesis
what is the difference between a directional and non directional hypothesis
Directional hypothesis –> Participants who….will…compared to participants who
Non-directional hypothesis –> There will be a difference in…between…and…
when would each hypothesis be used
A directional hypothesis is used if there is an agreeable outcome whereas a non-directional hypothesis is used if the previous research is contradictory/there is limited research.
what are the 4 types of experiments
lab experiment
field experiment
natural experiment
quasi experiment
what are lab experiments
an experiment that takes place in a controlled environment within which the researcher manipulates the IV and records the effect of the DV whilst maintaining strict control of extraneous variables e.g bobo doll
what are field experiments
an experiment that takes place in a natural setting within which the researcher manipulates the IV and records the effect on the DV – some variables cannot be controlled due to unpredictability of a public setting
what are natural experiments
an experiment where the change is not brought by the researcher but would have happened even if the researcher had not been there. Participants cannot be randomly allocated to experimental groups as they are already pre set environments e.g capture people in daily life
what are quasi experiments
a study that is almost an experiment but lacks key ingredients, the IV has not been determined by anyone and the variables simply exist e.g being older or younger. E.g adoption rates
what are strengths and limitations of lab experiments
Strengths of Lab experiments –> controlled environment; we can easily see changes and what caused it. Replicable and extraneous variables are removed
Limitations of Lab experiments –> lacks ecological validity; the ability to repeat the experiment in a real world setting cannot apply to everyday life
what are strengths and limitations of field experiments
Strengths of field experiments –> more realistic to daily life –> done in an environment that is more natural which means that results produced may be more valid and authentic especially if participants are unaware of the experiment
Limitations of field experiments –> loss of control over extraneous variables and lacks ethics due to concerns around consent/privacy
what are strengths and limitations of natural experiments
Strengths of the natural experiment –> provides opportunities for research; high external validity because they involve the sway of real life issues and problems as they happen
Limitations of natural experiment –> participants may not be randomly allocated to experimental decisions which means researcher may be less sure whether the IV affected the DV
what are strengths and limitations of the quasi experiment
Strengths of the quasi experiment –> controlled conditions which share strengths of lab experiments
Limitations of quasi experiment –> cannot randomly allocate ppts to conditons and therefore may be confounding variables
what are the 3 types of experimental design
independent groups
repeated measures design
matched pair designs
what are independent groups
different participants are used in each condition of the IV. This means that each condition of the experiment includes a different group of participants which should be done by random allocation. 2 groups
what are repeated measures design
the same participants take place in each condition of the IV
what are the matched pairs design
pairs of participants are matched depending on variables. One of the pair goes to control group and the other goes to the experimental group
what are advantages and disadvantages of independent groups
Advantages –> avoids order effects e.g fatigue as people participate in one condition only
Disadvantages –> differences between participants may affect results e.g variations in age
Random allocation deals with limitations (participant variables) through computer generator and anonymised name draws
what are advantages and disadvantages of repeated measures design
Advantages –> participant varibales are reduced and controlled, easier to conduct due to fewer participants needed
Disadvantages –> order effects arise due to repeating 2 tasks, Acts as a confounding variable and performance varies
Counterbalancing –> half participants take part in condition A then B and reverse for other half (B then A)
what are advantages and disadvantages to the matched pairs design
Advantage –> order effects and demand characteristics are less of a problem due to one conditioned group per participant
Disadvantages –> hard to reduce participant variables and matching can be time consuming/expensive, impossible to match people exactly
Random allocation –> randomly assigned to conditions
Define extraneous variables
variables that are not the IV but could affect the DV if not controlled e.g time of day, noise distraction etc
name the 4 potential extraneous variables
demand characteristics
investigator effects
situational variables
participant variables
what are demand characteristics
changing your behaviour to how participants think the experimenter wants them to react
how can demand characteristics be controlled
deception (lying about true aims of the study),
single blind design or
double blind design
what are investigator effects
any clues from the researcher that encourage certain behaviours in the participant) e.g indirect investigator effects – the measurement of the variables may be carried out in a way which makes the desired outcome more likely
how can investigator effects be controlled
a double blind design, interrater reliability (researcher checks results) and standardized instructions (pre written script)
what are situational variables
features of research situation that may affect participants behaviour for example order effects
how can situational variables be controlled
standardized procedures and standardized instructions and counterbalance which eliminates order effects
what experimental designs include participant variables
independent groups
matched pairs design
how can you control participant variables
random allocation can be done/randomisation of biases
what happens if extraneous variables are failed to be controlled
confounding variables may occur that damages the internal validity of an experiment
define a population
a group of people who are the focus of the researchers interest from which a sample is drawn
define a sample
a group of people who take part in a research investigation, The sample is drawn from a target population and is presumed to be representative of that population
define sampling techniques
the method used to collect a sample
define a bias
when certain groups may be over of under represented within the sample selected (limits generalisation)
define generalisation
the extent to which findings and conclusions from a particular investigation can be broadly applied to the population. This is more possible if the sample of participants is representative of the population
what are the 5 sampling techniques
-random sampling
-stratified sampling
-systematic sampling
-volunteer sample
-opportunity sample
what is random sampling
all members of the target population have an equal chance of being selected e.g names randomly picked from a list
what is systematic sampling
chosen for selecting from a target group. Differs to random sampling as it does not give an equal chance of selection to each individual in the target group e.g every 4th person in a list is used
what is stratified sampling
composition of the sample reflects the proportions of people in certain groups within the population
select participants in groups from a specific target population
what is an opportunity sample
the researchers simply decide to select any participants who happen to be willing and available e.g someone passing by on a street
what is a volunteer sample
the participants choose to be part of the sample
what are the advantages an disadvantages of random sampling
Advantages –> free from researcher bias
Disadvantages –> difficult and time consuming to conduct as obtaining a list of the target population may be difficult and participants may not want to take part
what are the advantages and disadvantages from systematic sampling
Advantages –> assuming the list order has been randomised, this method offers an unbiased chance of gaining a representative sample
Disadvantages –> if the list has been assembled in any other way bias may be present
what are the advantages and disadvantages of stratified sampling
Advantages –> avoids researcher bias and produces a representative sample so generalisation is possible
Disadvantages –> stratification is not perfect, separate conditions must be met properly
what is the disadvantages and advantages of an opportunity sample
Strengths –> it is convenient – this method saves the researcher time and is less costly
Limitations –> researcher bias as has complete control over participants and the sample is unrepresentative of the target population as it is drawn from a specific areaw
what is an advantage and disadvantage of a volunteer sample
Advantages –> it is easy to collect the sample and requires minimal input from the researcher
Disadvantages –> volunterr bias
define ethical issues
arise when a conflict exits between the rights of participants in research studies and the goals of research to produce authentic, valid and worthwhile data
what are the 6 ethical issues
-confidentiality/privacy
-deception
-informed consent
-debrief
-right to withdraw
-protection from harm
what is confidentiality/privacy/anonymity
Anonymity means data is not identifiable by their name. Confidentiality refers to how the data is stored and who sees it. Privacy is about not listening to what a person is saying or looking at what they are doing. Privacy is a particular issue in observational studies
what is deception (ethical guidelines)
participants must be told more or less what they are doing so that they are not completely deceived. This could make them feel bad after the experiment
what is informed consent
participants may only participate if they have consented to take part before the study knowing all the information involved
what is debrief
participants must be told that they have the right to remove themselves or any of their data from the study. They also have the opportunity to talk to the experiments to know the true names of the study after the experiment
what is the right to withdraw
The participants are told that they can remove themselves from the study at any time. They are told how to withdraw and how to not be harmed when withdrawing
what is protection of participants
participants must be managed so they can exit the study unharmed –> counselling sessions can be provided by the researcher if necessary
what does ethics refer to a balance between
costs and benefits
who is responsible for the ethical guidelines in Britain
BPS (british psychological society)
what are 3 ways to deal with consent if it is impractical to get informed consent
-presumptive consent –> similar group of people are asked if the study is acceptable
-prior general consent –> participants give their permission to take part in a number of different studies involving on of deception
-retrospective consent –> participants are asked for their consent after the study
what are two ways of dealing with confidentiality
-anonymity
-reminding participants that their data will be protected
define overt observation
Participants behaviour is watched and recorded with their knowledge and informed consent
what are strengths and limitations of overt observation
Strengths: more ethically acceptable
Limitations: participants may change their behaviour If they know they are being watched
define covert observation
participants behaviour is watched and recorded without their consent and knowledge
what are the strengths and limitations of covert observation
Strengths: removes the problem of participant reactivity so increased validity
Limitations: ethics of privacy are questioned
define participant observation
researcher becomes a member of the group whose behaviour they are recording and watching
what are the strengths and limitations of participant observation
Strengths: increased insight and increased validity of finding
Limitations: going native (loosing objectivity of the experiment)
define non-participant observation
the researcher remains outside of the group whose behaviour they are watching
what are the strengths and limitations of non-participant observation
Strengths: Less danger of going native as a physical distance is kept
Limitations: loose valuable insight (too far removed)
define naturalistic observation (unstructured)
provides qualitative data - no structure and records all behaviour in a descriptive way
define structured observation
researcher has various systems to organise the recordings of observations e.g behaviour categories or time scanning
what are the strengths and limitations of structured observation
Strengths: more efficient, easier to interpret, put in graphs and make comparisons
Limitations: reduced to few characteristics
what are the two types of observational structures
behavioural categories and sampling procedures
what are the 3 requirements for behavioural categories
Operationalise –> breaking the behaviour into a set of components
Be objective –> no inferences to be made about the behaviour. Simply just recording
Be mutually exclusive –> no overlapping, you should not have to mark 2 categories at the same time
what are the two sampling procedures (un/structured observation)
Event sampling –> recording the number of times something happens
Time sampling –> recording behaviours in a given time frame e.g what someone is doing every 30 minutes
what are the strengths and limitations of event sampling
Strengths –> faster to produce and easier to analyse
Limitations –> too many observations happen at once so may be difficult to record. If the event is complex, specific details may be missed
what are the strengths and limitations of time sampling
Strengths –> reduces the number of observations to be made. Fixates on set periods of time so is more focused and reduces chances of fatigue
Limitations –> some behaviours will be missed and therefore observation may not be representative
What are the three levels of measurement
nominal
ordinal
interval
what is nominal data
(data in separate categories) e.g attachment types – used for graphs etc
what is ordinal data
(Data placed in a specific order or scale) e.g organizing people from least to most aggressive – highly subjective and used for calculating range
what is interval data
(Data measured using units of equal interval) e.g number of correct answers –> used for standard deviation
what are the measures of central tendency
mean
median
mode
what is mean and its strengths/limitations
average number – calculate by adding all data items and divide by number of data items. Used for interval data
Strengths –> most sensitive measure as utilises all values
Limitations –> can be distorted by extreme values so unrepresentative
what is median and its strengths/limitations
middle value in data set – all items are arranged in order and middle number is left – used in interval and ordinal data
Strengths –> not affected by extremes
Limitations –> not sensitive as exact values are not reflected
what is mode and its strengths/limitations
Mode –> most common value – used in all three (nominal, ordinal and interval)
Strengths –> unaffected by extreme values
Limitations –> not always useful (more than one common choice)
what are two measures of dispersion
range
standard deviations
what is range and its strengths/limitations
difference in set of data - highest value – smallest value. Useful for data which is finding the same mean but has different ranges.
Strengths —> easy to calculate
Limitations –> affected by extreme values which might skew data
what is standard deviation and its strengths/limitations
Standard deviation –> provides information about the distribution of scores around the mean e.g how spread out the scores are from the mean – used to measure the average distance between each data item
Strengths –> precise measure as it takes all exact values
Limitations –> may hide some of the characteristics of data set
–> the smaller the standard deviation the more narrow the range and the larger the population the more normal distribution.
what is normal distribution
bell curve shape
mean,median and mode are all in the middle
what is negative skew
-majority of participants score high so tail on self is longer than right. mode is more than mean/median
what is positive skew
-majority of participants score low so tail on right side is longer. Mode is less then mean/median
what are 3 types of graphs
bar charts
histograms
scatter graphs
what are bar charts
height of each bar represents the frequency. The categories are x axis and the frequency is the y axis. Bar charts are non continuous data e.g nominal data and a space is left between each bar to indicate this.
what is a histogram
The x axis must be continual so no categories and there should be no gaps between bars. Used for interval and ordinal data
what is a scatter graph
used for correlation analysis. Provides initial indication if there is a relationship between two variables
what is a co-variable
variables investigated within a correlation e.g height. plotted by a scatter graph
what is a correlation co-efficient
value that states the strength of the correlation
what is a perfect positive correlation
+1.0
what is a perfect negative correlation
-1.0
what is a positive correlation
if one thing increases so does the other
what is a negative correlation
if one thing increases, the other either increases or decreases
what is the difference between a directional and non-directional correlation hypothesis
Directional —> There will be a relationship between…As…increases…this increases showing a positive correlation
Non-directional –> There will be a relationship between…As…increases…this either increases or decreases.
what are the strengths and limitations of correlations
Strengths of correlations –> can be used as starting points to assess patterns between co-variables before committing to an experimental study, quick to carry out, secondary data can be used
Limitations of correlations –> difficult to establish cause and effect relationship. Correlations can be misinterpretated.
what is the peer review process
1) scientists conduct research
2) scientists write their results
3) journal editor receives an article and sends it out for peer review
4) peer reviewers read the article and provide feedback to the editor
5) editor may send reviewer comments to scientists who may revise their article and sends it back. If does not meet standards it will be rejected
6) if an article meets the peer standards it is published in a journal
typically how many peer reviewers are there
1-3
what are peer reviewers
other psychologists who are experts in the same field of research check work for validity
what do peer reviewers consider
-influence of extraneous variables
-whether the data collected is appropriate and accurately recorded
-whether the research method was suitable for the aims of the study
-interpretation of finding/valid claims
who is cyril burtt
example of a psychologist who has been discredited for tampering with the numbers in his research - controversial as he helped created the 11+ exams for schools
what are research papers
academic writing that has been published
what are purposes of peer review
-to know which research is worthwhile and hence funding can be allocated to it
-to validate the relevance and quality of research (prevent fraudulent)
-to suggest possible improvements or amendments
what are limitations of peer review
- anonymity is a problem – affects objectivity of viewers
-publication bias involved in peer review – negative reports not published
-It can be difficult to find an expert
-Any research that opposes mainstream theories tends to be suppressed – rate of change is slowed down
-Fraudulent research can be long lasting
define validity
refers to how well a test or a piece of research measures what it says it
measures
what is a null hypothesis
There will be no difference in the number of words spoken in one minute by participants who drink energy drinks and water. Any difference will be due to chance.
what are statistical tests used to determine
whether a difference or an association is statistically significant – more than could happen by chance. This is then used to decide whether we should accept or reject the null hypothesis
what are the 3 factors to decide whether a test should be used
–> Difference or correlation
—> Experimental design –> independent groups (unrelated), repeated measures (related), matched pairs (related) - not relevant to correlations
—> Levels of measurement –> ordinal (unsafe data as lacks precision and is a subjective scale), nominal (discrete data as each item can only appear in one category), interval (most precise and necessary for parametric tests)
what are parametric tests
parametric tests are much more robust than other tests, they are able to detect significance within data sets that non-parametric tests cannot.
what are some examples of parametric tests
related t-test
unrelated t-test
Pearson’s R
what is the criteria that must be met for parametric tests
1) data must be interval
2) Data should have been drawn from a population thus showing a normal distribution
3) should be homogeneity of variance (scores in each condition should have similar dispersion which can be determined by standard deviations in each condition – related design)
nominal data
mode and no measure of dispersion
ordinal data
median and range
interval data
mean and standard deviation
what are the three steps to the simon cowell thingy
-test of difference/association
-related or unrelated design if is a test of difference
-level of measurement
what is a nominal related design test
sign test
what is an ordinal related design test
Wilcoxon
what is an interval related design test
related t-test
what is an unrelated nominal test
chi-squared
what is an unrelated ordinal test
Mann-whitney
what is an unrelated interval test
unrelated t-test
what is a nominal test of association
chi-squared
what is an ordinal test of association
spearman’s rho
what is an interval test of association
pearson’s r
what is the usual level of significance in psychology
0.05 or 5% p ≤ 0.05
what does 0.05 or 5% p ≤ 0.05 mean
the probability of the observed/measured effect happening by chance is equal to or less than 5%. This means that in research where they claim to have found a significant difference/correlation there is still a 5% chance it was a fluke. Psychologists can never be 100% certain of their results as they have not tested everyone in the world in every conditions
how do you use the table of critical values
1) one or two tailed test –> one tailed is a directional hypothesis and two tailed is a non-directional hypothesis
2) number of participants (usually appears as the N value on the table or DF)
3) Level significance –> known as the P value. p<0.05 is the standard level
what is the rule of R
some statistical tests require the calculated value to be equal to or more than the critical value
what statistical tests have a calculated value more than/equal to (unsignificant)
-chi-squared
-spearman’s rho
-related t-test
-unrelated t-test
-pearson’s r
what statistical tests have a calculated value less than/equal to
-sign test
-wilcoxon
-mann whitney
what are the statistical tests
-sign test
-wilcoxon
-mann whitney
-chi-squared
-spearman’s rho
-related t-test
-unrelated t-test
-pearson’s R
what is a Type I error
null hypothesis is rejected and the alternative hypothesis is accepted when in reality the null hypothesis is true. This is an optimistic error or a false positive. When retesting should use 1%
what is a Type II error
The null hypothesis is accepted when it should have been rejected. This is a pessimistic error or a negative error. 10% should be used when retesting
how do you calculate the sign test
1) state the hypothesis (null or alternate)
2) If alternate hypothesis check if it is one or two tailed
3) work our values which increased, decreased or stayed the same
4) calculate total participants by subtracting the amount of people who stayed the same
5) Calculate the observed value (lower value of what either increased or decreased)
6) Find the critical value of S from the table provided (level significance will be 0.05
what is needed for the data to be significant (alternate hypothesis to be accepted)
observed value should be less than or equal to the critical value
why would a study lack validity
- the method of measurement doesn’t measure what it claims to measure.
- there are extraneous or confounding variables that affect the measurement.
define empirical
a scientific approach based on gathering evidence through observation and experience.
what are the features of science
-empirical
-objective
-replicable
-falsifiable
-theory construction and deduction
-paradigm and paradigm shift
-inductive and deductive
what is objectivity
observation without bias
what is replicability
Being able to repeat a study to check the validity of the results across other contexts/circumstances
what is falsifability
scientific statements are capable of being proven wrong. Popper argues that this is a key criteron for a scientific discipline, he proposed the theory of falsification. Successful theories have been constantly falsified and not proven false yet. Sciences that cannot be proven are called pseudo-sciences e.g Freud’s oedipus complex. Theories which survive more falsify attempts are the strongest.
what is paradigm
A set of shared assumptions and agreed methods in a scientific discipline. Kuhn suggests that paradigm’s are what separates scientific disciplines from non-scientific disciplines. He also believes that a paradigm shift shows progress in science. Psychology that has too many conflicting ideas is referred to as pre-science
what is a paradigm shift
revolutionary changes in scientific assumption
define inductive
use evidence from observation to formulate theories
define deductive
conclusions are taken from inference, these are tested leading to theory adjustment
what is theory construction and deduction
a theory refers to a set of general principles and laws which can be used to explain specific events and behaviours. Theory construction takes place through gathering evidence from direct observation during investigations. Deduction refers to the process of deriving new hypothesis from an already existing theory
arguments for is psychology a science
Cognitive approach –> uses empirical methods in highly controlled lab conditions
Biological approach –> Objective scientific data used to support theories with high levels of control
-Eye witness testimony (Pardigm shift)
-scientific methods are in many research studies giving them scientific credibility
-findings positvely impact individuals
arguments for psychology is both a science and not
-nature vs nurture
-systematic desensitization and medicine (treat phobais)
-client centered therapies and antidepressants (treat depression)
arguments for psychology is not a science
Psychodynamic approach –> not objective
-Humanistic approach –> not replicable
-experiment interpretations can be subjective
-not all research is generalisable e.g case studies
what is the structure of determining the significance of a value
The result of the…(test) is significant/not significant. The calculated value is greater/less than the critical level value by…where n=….for a one/two tailed hypothesis with a p<0.05. The null hypothesis can be rejected/accepted and the alternate hypothesis is rejected/accepted.
Explain why stratified sampling might improve this study
1) more representative = generalisation
2) reduces researcher bias
explain what is meant by primary data
data collected first hand by the researcher for the purpose of their aims
explain one reason why mean would be the most appropriate measure of central tendency
1) most representative
2) most sensitive
discuss the purpose of counterbalancing
-reduces effect of order effects
-A-B B-A (intermintent turns)
-However does not eliminate order effects as 2 separate tasks are complete by each person
what are the steps to planning a practical research
1)write research question
2)write operationalised one tailed hypothesis
3)describe method and procedure –> method/design, IV, DV, sampling method, material used, ethics, timings, controls, how data is obtained
4)state how you will analyse data
5)describe advantage of experimental design
6)outline one way you would take ethical issues into account
7)discuss ecological validity
8)evaluate sampling method
9)identify one disadvantage and how this will be overcome
define reliability
how consistent the results are
define validity
the extent to which a research method measures what it is designed to measure
what are the types of validity
external
internal
population
ecological
temporal
what is external validity
related more to factors outside of the investigation such as generalising to other settings, other populations and other areas
what is internal validity
refers to whether the effects observed in an experiment are due to the manipulation of the IV and not some other factor
what is population validity
ability to reasonably generalise findings from a sample to a larger population
what is ecological validity
the extent to which findings can be generalised to other settings (real world)
what is temporal validity
The extent to which findings hold true over time (not outdated) –> Freud’s penis envy doesnt have temporal validity
what are the two ways to test validity
-Face validity –> whether a test, scale or measure appears to measure what its supposed to measure. This can be done through eyeballing the measuring instrument or passing it onto an expert
-Concurrent validity –> results obtained are very close to another recognized and well established test. Close agreement between two sets of data would indicate that the test has high concurrent validity
what factors reduce the validity of a study
demand characteristics
investigator effects
individual differences
how can you improve validity
-Demand characteristics and investigator effects can be controlled using a single or double blind technique.
-Individual differences can be controlled through experimental design. Using either a repeated measures or matched pairs design reduces the likelihood of individual differences affecting the results, as the same participants take part in each condition. However, if a repeated measures design is used; other factors such as fatigue can affect internal validity
-Designing a pilot study is often one of the best ways to check that everything in your actual experiment will run smoothly which will improve the validity of results. It helps the researcher to identify any practical problems with the research method, design and so on. Pilot studies can also check items on a questionnaire to make sure they are easy to answer and unambigious
-to improve external validity research should be replicated
how can you assess reliability
-Test-retest –> involves administering the same test or questionnaire to the same person on different occasions. To be reliable the results obtained should be the same or similar. The set of scores obtained would be correlated
-Inter-observer reliability –> relevant to observation research as researchers interpretation of events may be subjective or biased. There should be 2 observers to check the observers are applying behavioral categories in the same way
how can you improve reliability
-Experimental research –> using a control group, standardise procedures to minimise participant reactivity and investigator effects, The use of single or double blind procedures
-Questionnaires –> incorporate a lie scale to control social desirability bias and use anonymous data
-Observations –> covert observation + behavioural categories
-Qualitative methods –> triangulation and case studies/interviews
How do you answer a 12 or 9 marker in the research method section
Each paragraph must have the subheading of each bullet point provided
Only answer what is asked in the questions
KEY TERMS!!!
be specific in details and comprehensive
what is descriptive analysis another word for
measures of central tendency and measures of dispersion
what is one limitation of controlled observation
lacks ecological validity
structure for answering the suitability of experimental designs
Which experimental design is used…..why it is appropriate….context of the study
explain why statistical testing is used in psychological research
-tells us whether the hypothesis should be accepted (alternate) or rejected (null)
-tells us if it is statistically significant
structure of 9/12 marker research methods
Title e.g same task for participants
standardisation is very important…..in this study…..to control this….it would also be necessary
9 and 12 marker
9 marker –> 4 points per paragraph
12 marker –> 3 points per paragraph
participant variables
individual differences e.g someone might be older or better at the task than others thus reducing validity
how many paragraphs do you write for an 8 marker
3 paragaphs
1 AO1
2 AO3
extended RM questions
-for 4 bullet points do 3 points per paragraph
-You can draw diagrams for RM questions
questionnaires vs interviews
-Questionnaires –> assess a persons thoughts or experiences through a number of different written questions
-Interviews –> involves a live encounter where a set of questions is asked by an interviewer to an interviewee to assess their thoughts or experiences
types of questionnaires
-open question (questions are phrased in a way that the participant is free to answer however they like – qualitative data) –> rich and useful for sensitive topics but is difficult to convert to statistical data and is harder to analyse
-closed question (questions which restrict you to a fixed number of responses – quantitative data e.g rating scales)
—> easy to analyse data and compare but lacks depth and detail
strengths of questionnaires
-cost-effective
-gathers large amounts of data quickly
-researcher does not need to be present
-easier to analyse
-responses can be anonymous
Limitations of questionnaires
-difficult to know target population
-takes a long time to design
-difficulty to assess validity due to biases
-participant bias is present e.g age
-response bias
types of interviews
-structured (involves a set of predetermined questions asked during the interview – waits for a suitable response) –> easily replicable, comparisons can easily be made but interview bias is present and social desirability bias
-unstructured (no pre-determined questions and instead it develops as the interview continues – more free flowing) –> lots of detailed data is collected and can be tailored to individuals due to greater insight but skilled interviewers are needed, interview bias, difficult to make comparisons, hard to analyse
what is a meta-analysis
A meta-analysis is where researchers combine the findings from multiple studies to draw an overall conclusion.
consistent results
smaller SD -> smaller range around the mean
why do psychologists normally use the 5% level of significance in their research
The 5% level is used as it strikes a balance between the risk of making the Type I and II errors
process and purpose of peer review
Process
* other psychologists check the research report before deciding whether it could be published
* independent scrutiny by other psychologists working in a similar field
* work is considered in terms of its validity, significance and originality
* assessment of the appropriateness of the methods and designs used
* reviewer can accept the manuscript as it is, accept with revisions, suggest the author
makes revisions and re-submits or reject without the possibility of re-submission
* editor makes the final decision whether to accept or reject the research report based on the
reviewers’ comments/recommendations
* research proposals are submitted to panel and assessed for merit.
Purposes
* to ensure quality and relevance of research, eg methodology, data analysis etc
* to ensure accuracy of findings
* to evaluate propose
replicable =
standardized
use median =
instead of mean if there is an anomolous result as it would skew/distotrt the data
strenghts of controlled observations
-standardized procedure
-less extraneous variables
-easy to distinguish cause and effect relationship
how to write non-directional hypothesis
there will be a difference between….
standard deviation
-measure of scores around the mean
-smaller SD = more narrow the range between the lowest and highest score
-large population = normal distribution
negative skew
mean is lower than median and mode
significant
calculated value is less than or equal to critical value
distribution of scores =
reference normal distribution, positive skew and negative skew
scatter vs bar vs histogram
scatter = correlation analysis (+/-), 2 co-variables, ordinal data
bar = nominal, discrete (non-continious), categorical
histogram = scale, interval, continious
t value
probability of observing results if the null hypothesis is correct
higher t value = lower probability
so null hypothesis is less likely to be correct
factors affecting size of t value
-difference between the means (bigger difference between 2 means = larger t value) = less likely null is correct
-dispersion of data (larger dispersion = smaller t value) = more likely null is correct
-sample size (larger sample size = larger t value)
6 parts to a report
abstract, introduction, method, results, discussion, references
abstract
-start of the report
-short summary
-briefly describes aspects of a study
introduction
-explains aims and hypothesis of the study
-explains what past research has found
method
-how variables were measured
-type of study
-exact procedure used in enough detail so the study can be replicated
results
-description of results obtained
-presentation of data
-description of how the data was analysed
discussion
-statement of whether results supported hypotheis
-interpretation of results
-whether results have been replicated from other studies
thematic analysis
-labels for data in thematic analysis are called codes
-then these codes are categorised into themes
-qualitative data
-data must be written up
content analysis
-category and tally table
thematic vs content analysis
-thematic = written data whilst content is any type of data
thematic = codes then themes but in content is it called coding units then counting this
thematic = report wiht interpretation of findings whilst content anlysis leaves a count
content analysis strength vs thematic
strength –> uses statistics to test hypothesis and can draw clearer conclusions e.g find mode then do chi suqared test whilst thematic analysis doesnt involve statistical tests
thematic = subjective partcipants experiences
content analysis
-Content analysis is a method used to analyse qualitative data (non-numerical data). In its most common form it is a technique that allows a researcher to take qualitative data and to transform it into quantitative data (numerical data). The technique can be used for data in many different formats, for example interview transcripts, film, and audio recordings.
The researcher conducting a content analysis will use ‘coding units’ in their work. These units vary widely depending on the data used, but an example would be the number of positive or negative words used by a mother to describe her child’s behaviour or the number of swear words in a film
strengths and weaknesses of content analysis
Strengths of content analysis
It is a reliable way to analyse qualitative data as the coding units are not open to interpretation and so are applied in the same way over time and with different researchers
It is an easy technique to use and is not too time consuming
It allows a statistical analysis to be conducted if required as there is usually quantitative data as a result of the procedure
Weaknesses of content analysis
Causality cannot be established as it merely describes the data
As it only describes the data it cannot extract any deeper meaning or explanation for the data patterns arising.
what is a case study
in-depth study of a person or group of people. idiographic and individualistic
how could the researcher use thematic analysis
use the recordings to make a transcription of the interview
* use coding to initially analyse the transcripts
* review the transcriptions/codes looking for emergent themes/ideas that might
be linked to later aggressive behaviour, eg family violence, parental argument,
alcohol misuse.
thematic analysis
organising data according to themes or categories
positive correlation bc
as one co-variable increases the other co-variable increases
non-directional hypothesis for correlation study
There will be a relationship between….
controlled observation =
reduce extraneous variables
inter-observer reliability in controlled observations
the two observers’ tally charts would be compared to check for agreement/calculate the correlation
between the recordings of the two observers to determine the level of inter-observer reliability
* researchers generally accept +0.8 correlation as a reasonable degree of reliability.
features of an abstract in scientific report
- first section of a report (accept alternative wording)
- brief/summary/100-300 words (accept alternative wording)
- contains (a summary of) the aims, methods, results, and conclusions (accept alternative wording)
ecological validity =
events that occur in real world
external validity =
generalise to people, stimuli etc
features of science
- objectivity and the empirical method
- replicability and falsifiability
- theory construction and hypothesis testing
- paradigms and paradigm shifts.
Explain what it would mean if the results in the above study were significant at the
0.01 level.
there is only a 1% possibility that the difference in stress ratings between the two conditions is due to
chance (not due to the breaktime 2km run)
* there is ≤ 0.01 chance that the observed difference in the stress ratings reported is not a real
difference (resulting from doing or not doing the 2km run at breaktime).
Drawing bar chart
-x and y axis
-title
-key
explain how the data collected from the interview might have improved upon the data collected from content analysis
The data from the interview may have improved the data as it meant they had different kinds of data to compare to each other, they collected both qualitative and quantitative data therefore having a range to compare. Also the essays may be biased as the teenagers were writing them themselves and could pick and choose what they wanted to say whereas in the interview they were more on the spot and were asked specific questions that they had to answer.
suggest one possible modification to the design of the experiment and explain how this might improve validity
use a different sampling technique to avoid a self-selected sample and thus avoid volunteer bias
as that may make them more susceptible to demand characteristics.
change to independent groups design
The psychologist found that the difference in the number of nightmares reported in the two conditions was significant at p<0.05. Explain what is meant by ‘significant at p<0.05’ in the context of this experiment.
This means that the difference in the number of nightmares reported after watching horror films compared to romantic comedies is significant at 0.05 level. This means there is less than 5% (1 in 20) likelihood (probability) that the difference was due to chance/due to something other than the IV.
The psychologist used counterbalancing in this experiment. Explain why it was appropriate to use counterbalancing in this experiment.
order effects are likely to occur
if you saw the horror films in the first week you may still be thinking about them/have higher
levels of fear in the second week, resulting in more nightmares.
structure of a consent form
Format and style of consent form:
will require the participant’s agreement
could be written as a form that participants need to sign
could include space for the participant to sign the consent form
could include a space for the participant to write the date
could include a space for the participant to print their name.
Outline of what the experiment entails:
an explanation of the general purpose of the research
they would need to watch an assigned horror film every night before going to bed for 7 nights and
would need to watch an assigned romantic comedy every night before going to bed for 7 nights
a daily requirement to truthfully respond to a text message asking whether they had experienced
a nightmare
the two-week duration of the experiment
Outline how these researchers could have used a statistical test to establish concurrent validity of the mathematical reasoning ability tes
Concurrent validity would involve correlating the results on the maths test with results for the same group of people on an established maths reasoning test. A Spearman’s rho should be used for the two sets of test results and if the mathematical ability test is valid then there should be a significant positive correlation between the two sets of test scores at the 0.05 level.
ratio data =
data which takes on number values for which we can tell how much bigger one value is from another
values cant go below zero
any variable whose data points can be counted up is
quantiative data
e.g number of ppl with green eyes in a list
frequency tables
1st column = category
2nd column = frequency column
discrete vs continious data
discrete = quantiative reduced to just certain numbers e.g nominal
Continuous data is quantitative data that’s not restricted to certain numbers. e.g interval
ordinal data = restricted number of values
drawing frequency graphs
y axis = frequency
structured interviews =
more reliable than unstructured interviews but unstructured interviews enable researchers to gain more information
problems with self report =
lack objecticity due to personal bias/social desirability bias
case study
an in depth investigation and description and analysis of an individual group, institution and event
production of qualitative data = analysis of unusual individuals or events
tend to take place over a long period of time (longitudinal)
content analysis
a research techhnique that enables the indirect study of behaviour by examining communications that other people produce
type of observational research
coding
stage of content analysis in which communication is studied by anlysing categories
initial stage of content analysis
quantitative data
categorise information into meaningful units
thematic analysis
inductive and qualitative approach to analysis that involves identifiying implict or explicit ideas within the data. Mostly after coding has done
strengths and limitations of case studies
strengths = rich, detailed insight
generate hypothesis e.g patient HM
limitations = inability to generalise and subjective interpretation = lack validity as it is prone to innaccuracy
strengths and limitations of content analysis
strengths = no ethical issues / high external validity = flexible data
limitations = indirect study = loss of information. Lack objectivity from the researcher
concurrent valididty
the extent to which a psychological measure relates to an existing similar measure
when can test retest be used
questionnaire, IQ tests, interviews
interobserver reliability
pilot studies
behavioural categories
can apply to other forms of observation such as content analysis or interviews
improving reliability
closed questions
correlation 0.80 + or minus
no leading questions in interviews through structured interviews
lab experiments
behavioural categories
internal validity
effects observed in an experiment are due to the manipulation of the independent variable
external validity
relates to more factors outside the investigation, generalising to other settings
ecological validity
generalise to everyday life
low mundane realism =
did not mirror everyday life
improving validity
-control group = easier to view impact of independent variable
-standardised procedures to minimse impact of particiapnt variables
-single and double blind
-lie scale = questionnaires
-triangulation
nominal data =
discrete
type 1 vs type 2 error
1 = false positive = accept alternate (significance level may be too high e.g 0.1%)
2 = false negative = accept null (signifiance level may be too low)
why is it better to conduct a case study over a questionnaire
case study = longitudinal = studied over time = more holistic
case studies involve several methods
case studies = qualitative rich detailed data
questionnaires = not descriptive (reductionist), social desiarability bias
STM and LTM =
functionally different
conversion
content analysis converts qualitative data to quantitative data
content analysis
-content analysis is a form of observation which analyses the communication between people e.g in films, TV, speeches etc
-aim is to summarise this communication in a systematic way so overall conclusions can be drawn
-mainly secondary data
coding categories
Coding categories are established prior to analysing the data (quantiative) to identify recurring themes and count up the number of times a particular word of phrase increases
Furrham and Farragher (2000) –> examined TV adverts to see how often men and women are depicted in professional roles. More men were shown as professional compared to women.
thematic analysis
Thematic analysis = a form of content analysis but the outcome is qualitative. The main process involves the identification of themes. A theme refers to a recurrent idea that keeps cropping up as part of the communication studied.
Matthews et al (2012) = analysed 1200 instances of graffiti from toilet walls. Graffiti was coded according to distinct categories such as sexual references, socio-political, enetertainment etc. Males composed significantly more sexual and name writing graffiti whilst females was romantic posed significantly more sexual
how to conduct content analysis
-pilot = research must begin by observing the material and familiarising themselves with
-coding = decide how the categories the material e.g counting the number of times a word of theme appears
-tally = counts up recurring themes
-analyse = data should be summarised
how to conduct thermatic analysis
How to conduct thematic analysis:
-familiarise yourseful with the data
-generate initial code
-search for themes
-conclusions
-produce the report
strengths of content analysis
-offers a method to analyse a variety of forms of data including media and self-report methods so that insights into cultural trends and experiences can be understood
-high external validity
-flexible in terms of quantiative and qualitative data
limitations of content analysis
-the identification of suitable themes and codes is subjective and decided by the researcher alone, meaning the conclusions lack objectivity
Describe an experiment to investigate whether execrise could increase feelings of happiness using ordinal level of measurement
experimental design = independent groups = reduce order effects that could impact aim and determine cause and effect relationship. Ppt variables msut be controlled using random allocation.
variables = 30mins exerecise, 30 mins no exercise. (be operationalised). DV = levels of happiness in a questionnaire on a scale from 1-10. (1= unhappy, 10 = very happy). Should be done at start of condition and range will be calculated.
Control = time of day etc
Questionnaire = done to obtain demographic information.
Data analysis = table split between conditon 1 and 2 and determine the change in happiness score for all ppts. Calculate the median from the table (draw table in exam)
infferential analysis = mann whitney = ordinal and independent groups.
if they ask you to write a repeated measures design within the controls write
order effects must be controlled and explain how counterbalancing would be used
different tasks may be used as same conditions but the tasks should be of the same difficulty
controls
=same tasks
-headphones for listenting task so dont affect other group
-time of day
-word puzzles = same or same difficulty
for example controls in a study o see f people remeber more wirtten or heard information
-ensure content of the story is the same for both conditions
-ensure same lenght of time is given to either hear or read the content provided
-and same amount of time to answer questions about the content to test memory.
be specific for extended RM questions when talking about sampling or sampling technique e.g
also mention why that sample would be approporate e.g for volunteer sample it would be appropirate to find left hand people as not everyone is left handed
40 students
aged 16-18
20 in each condition done by random allocation
selection using opporunity sample
outline of procedure
mention standardised instructions will be read out and then at the end be thanked and debriefed
test for musical ability for example
music teacher would rate strength on scale from 1-10
what to write in debrief
thank ppts
state aims
reminder of confidentiality and protection of data
state their right to withdraw
and ask if they have any questions
sampling most likely to be used in external environment
event sampling
when talking about behavourial categories
draw table
behaviour, tally, total