Research Methods Flashcards
What are the five ethical issues?
Right to withdraw
Anonymity
Protection from harm (psychological and physical)
Informed consent
Deception
What is the right to withdraw?
Participants should be told at the beginning of the experiment that they have the right to withdraw at any time - before, during or after the experiment - without needing to give a reason
What are the limitations with the right to withdraw?
Participants may feel pressure to stay in the experiment even when they have been told they can leave
Pressure can increase if they are being paid by the psychologist to take part
What is anonymity?
Researchers should not record the name of any participants - they should give anonymity codes instead (numbers or false names)
No information should be published that could make someone recognisable
What are the limitations of anonymity?
In some unique cases, it can be hard to keep the participant anonymous - especially in case studies
Some participants reveal themselves as the participants
What is protection from harm?
Participants should leave the experiment in the same or better state than when they started the experiment
Avoid any risks greater than everyday life and stop the study if harm suspected
What are the limitations of protection from harm?
Psychologists cannot predict all types of harm that could occur
Short term harm is sometimes seen as acceptable
What is informed consent?
Participants should know everything about the study before and during the experiment
Participants are asked to sign a document which shows that they agree to take part in an experiment, which should include the nature of the study (task, duration)
What are the two types of consent?
Presumptive consent
Prior general consent
What is presumptive consent?
Speak to people who are similar to the participants (same age/sex/location) if they would take part in the experiment
If they say yes, assume real participants would say yes
What is prior general consent?
Ask them to give consent to multiple studies which includes your real study
Ask them if they consent to being lied to
What are the limitations of informed consent?
Can lead to demand characteristics
Presumptive consent doesn’t always work as a solution
What is deception?
Need for deception should be approved by the ethical committee, weighing up benefits of the study to the costs of the participants
Participants should be fully debriefed after the study to explain true nature, where they should be able to discuss any concerns that they have
What is a debrief?
Speak to the participants after the study and let them know the true aim of the study
What are the limitations of deception/debrief?
Debriefs don’t stop the lying from occurring
Some participants may still feel upset by the deception
When are consent forms given?
At the beginning
What is the format of a consent form?
Aim:
Conditions: task, instructions, duration, other groups
Ethical issues: mainly right to withdraw and anonymity
Sign:
Print name:
Date:
When is a debrief form given?
At the end
What is the format of a debrief form?
Aim:
Conditions: task, instructions, duration, other groups
Thank them for taking part:
Remind them of ethical issues:
What are demand characteristics?
Participants guessing the aim of the study, causing them to change their behaviours
What is mundane realism?
How similar the experiment is to real-life
How “realistic” the task is
What are investigator effects?
Researcher bias
Unconsciously the investigator could manipulate the experiment to get the results they want
What is a single blind experiment?
Participants don’t know what conditions they are in/if there are more conditions
What is a double-blind experiment?
Both researcher and participant don’t know the aim of the study
What does operationalised mean?
Specific
Measurable
Detailed
Objective
Comprehensive
Mutually exclusive
What is an independent variable?
Variable you change
What is a dependent variable?
Variable you measure
What are control variables?
Things you keep the same
What are extraneous variables?
Procedure variables
Any variable that has affected the DV which is not the IV
What are some examples of extraneous variables?
Location
Time
Noise levels
What are confounding variables?
Participant variables
Anything different about the participant which affects the DV that is not the IV
What are some examples of confounding variables?
Age
Eye sight
Personality
What are the four types of experiment?
Laboratory
Quasi
Natural
Field
What is a laboratory experiment?
Highly controlled environment
Artificial task
Unrealistic setting
(Not always in a laboratory)
What are the advantages of laboratory experiments?
High control of extraneous variables
Establish cause and effect
Replication is possible
What are the disadvantages of laboratory experiments?
Low ecological validity
Higher demand characteristics
What are quasi experiments?
IV is an existing difference between people such as age/sex/personality
Majority of time conducted in lab like conditions
What are the advantages of quasi experiments?
High control of extraneous variables
Establish cause and effect
Replication is possible
What are the disadvantages of quasi experiments?
Confounding variables
What are natural experiments?
IV is natural
Realistic task
Typically realistic setting (but can be in a lab)
What are the advantages of natural experiments?
Can test things that aren’t ethically possible to create in a lab
High ecological validity
What are the disadvantages of natural experiments?
Confounding variables
Rare opportunities
Difficult to generalise as unique
Loss of control of extraneous variables
What are field experiments?
IV is manipulated
More realistic setting
What are the advantages of field experiments?
Mundane realism
Higher ecological validity than lab
What are the disadvantages of field experiments?
Loss of control of extraneous variables
Ethical issues
What is a hypothesis?
Statement in which you predict what will happen
What is an aim?
Stating the general purpose of the experiment
What are the two types of hypothesis?
Directional
Non directional
What is a directional hypothesis?
Can predict which condition will perform better than the other
Used when there is previous research to state which direction its likely to go
What is a non-directional hypothesis?
Think there will be a difference between your conditions but are not sure which condition will perform better/worse
No previous research
How do you choose which type of hypothesis to write?
Previous research = directional
No previous research = non-directional
What is the structure of a non-directional test of difference hypothesis?
There will be a difference in DV for IV and IV
What is the structure of a directional test of difference hypothesis?
IV’s DV will be higher/lower than IV’s DV
What is the structure of a non-directional test of relationship hypothesis?
There will be a relationship between DV and DV
What is the structure of a directional test of relationship hypothesis?
There will be a positive/negative relationship between DV and DV
How do you write a hypothesis?
- decide if directional or non-directional (“previous research)
- test of difference or relationship (“relationship”)
- find variables (IV and DV)
- operationalise variables (look for table)
What are the four types of graph that are needed to know in the AQA spec?
Scatter graph
Bar graph
Histogram
Line graph
How do you draw a bar graph?
Categories (discrete data) on x axis (nominal data)
Frequency on y axis (continuous data)
Height of column represents frequency
Bars don’t touch (gaps inbetweeen)
How do you draw a scatter graph?
Continuous data on x and y axis
“Relationship” and 2 DVs
Participants NOT on axis
Crosses = participants
No lines of best fit
How do you draw a histogram?
Frequency on y axis
Continuous scale on x axis
No gaps between bars
How do you draw a line graph?
Continuous data on x and y axis
One participant’s data
What is the success criteria for graphs?
Title
x axis label - operationalise
y axis label - operationalise
Units/correct scale
Accurate plotting
Choose correct graph (6th mark)
What are the measures of central tendency?
Mean
Median
Mode
What data is the mean used for?
Interval data
How do you calculate the mean?
Add up all the numbers
Divide by how many numbers there are
What are the advantages of the mean?
Most sensitive/representative
What are the disadvantages of the mean?
Easily distorted by extreme values/anomalies
What data is the median used for?
Ordinal data
How do you calculate the median?
Put the numbers in order
Cross out one number from both sides until you reach the middle
What are the advantages of the median?
Extreme scores don’t affect it
What are the disadvantages of the median?
Less sensitive
What data is the mode used on?
Nominal data
How do you calculate the mode?
Most common number
What are the advantages of the mode?
Extreme scores don’t affect it
Quick
What are the disadvantages of the mode?
Less sensitive
What is a normal distribution?
Most people located in the middle of the curve
Allows mean, median and mode to have same midpoint (SIMILAR numbers)
Symmetrical
What is a positive skew distribution?
Most of the curve towards left side
Tail of the curve continuing positively
Right foot/happy whale
Is the mode the lowest or highest score in a positive distribution?
Lowest
What is a negative skew distribution?
Most of the curve towards right side
Tail of curve continuing negatively
Left foot/sad whale
Is the mode the lowest or highest score in a negative distribution?
Highest
How do you draw a distribution?
- label y axis frequency
- label x axis score/scale
- plot the mode as the highest point on y axis and label
- plot the median lower than the mode on y axis and label
- plot the mean lower than the median on the y axis and label
- sketch curve
How do you change from a positive skew to normal distribution?
Test is too difficult
As mode is a low score
Should make test easier
Should increase mode and make a normal distribution
How do you change from a negative skew to a normal distribution?
Test is too easy
As mode is a high score
Should make test harder
Should decrease mode and make a normal distribution
What are self-report techniques?
Questionnaire
Interview
What is a questionnaire?
A pre-set list of written questions which participants record their answers
What can a questionnaire be used to assess?
DV
What are the advantages of questionnaires?
Cost-effective
Large amounts of data quickly
Easier to analyse (can produce graphs and charts)
Less effort (can be completed without researcher being there)
What are the disadvantages of questionnaires?
Demand characteristics
Social desirability
Participants may not understand questions
What are the two types of questions?
Open-ended
Closed
What are closed questions?
Respondents have limited choices
What type of data do closed questions produce?
Quantitative
What are some types of closed questions?
Likert scale
Rating scales
Fixed choice options
What is a Likert scale?
Respondent indicates agreement with a statement using a scale of usually five points
What is a rating scale?
Gets respondents to identify value that represents their strength of feeling on a particular topic
What is a fixed choice option?
Includes list of possible answers
Respondents required to pick all that apply to them
What are the advantages of closed questions?
Easier to analyse and compare
What are the disadvantages of closed questions?
Forces people to pick (limited choices)
Lacks depth and detail
What are open-ended questions?
Does not have a fixed answer
Respondents provide own answers expressed in words
What type of data do open-ended questions produce?
Qualitative
What are the advantages of open-ended questions?
Rich in depth and detail
What are the disadvantages of open-ended questions?
Difficult to analyse
What are interviews?
Face-to-face or virtual interaction between an interviewer and interviewee
Done individually or in a group
What are structured interviews?
Made up of a pre-determined set of questions that are asked in a fixed order
What are the advantages of structured interviews?
Easy to replicate
Reduces bias
What are the disadvantages of structured interviews?
Not able to understand the insight of answers or gather detail
What are unstructured interviews?
There are no set questions, just a general topic or aim
What are the advantages of unstructured interviews?
More flexibility
Gain insight/detail
What are the disadvantages of unstructured interviews?
Difficult to analyse and replicate
How should interviews be designed?
Most involve interview schedule (list of questions to cover)
Should be standardised for each participant to try and reduce interview bias
Conducted in single or group settings
What are the different observations?
Naturalistic
Controlled
Overt
Covert
Participant
Non-participant
What is a naturalistic observation?
Setting or context where target behaviour is usually seen
What are the advantages of naturalistic observations?
High ecological validity
What are the disadvantages of naturalistic observations?
Lack of control of variables (EV)
Hard to replicate
What are controlled observations?
Manipulate the situation or behaviour
What are the advantages of controlled observations?
High control of variables (EV)
Easier to replicate
What are the disadvantages of controlled observations?
Low ecological validity
What is an overt observation?
Participants know they are being watched
What are the advantages of overt observations?
Ethical
What are the disadvantages of overt observations?
Demand characteristics
What is a covert observation?
Participants don’t know they are being watched
What are the advantages of covert observations?
Lower risk of demand characteristics
What are the disadvantages of covert observations?
Unethical - don’t give consent
What is a participant observation?
Observer becomes part of the group they are observing
What are the advantages of participant observations?
Increased insight of the meaning of behaviours
What are the disadvantages of participant observations?
Lose objectivity
May forget information
What is a non-participant observation?
Observer remains separate from participants
What are the advantages of non-participant observations?
Objective distance from participants
Can write down information as it happens
What are the disadvantages of non-participant observations?
Lose insight
What is an unstructured observation?
Records all information without a system
Mainly used in a new situation to decide what behaviours should be focused on
What is a structured observation?
Record information through behavioural categories and sampling procedures
What are behavioural categories?
A behavioural checklist of target behaviours which you then tally
Must be operationalised
What is event sampling?
Counting number of times a behaviour occurs
What is time sampling?
Recording behaviour within a pre-established time frame (every 30 seconds), typically focused on one person’s behaviour
What is the method of time sampling?
- record child’s behaviour at set time intervals during a time period, eg every 20 seconds
- at the specified time intervals, tick/make one or more categories from the behavioural checklist according to the behaviour
- give an example of a behaviour
- behaviours observed in between the time sampling frames should be ignored and should not be recorded
What are the advantages of time sampling?
Easier to manage as there are likely many behaviours occurring throughout set observation time
Most appropriate for recording ongoing behaviours that are likely to occur in a social setting
What is a pilot study?
Small scale trial run of the experiment before the real experiment
What is the purpose of a pilot study?
Check for any potential problems which they can then modify the design or procedure to correct
What does a pilot study for an experiment check for?
Timing
Instructions are clear and understood
Ask participants their opinion on the experiment for feedback/improvements
Check materials
What does a pilot study for an interview/questionnaire check for?
Test out a few of their questions to check for understanding
Instructions are clear and understood
Ask participants their opinion on the interview/questionnaire for feedback/improvements
What does a pilot study for an observation check for?
Instructions are clear and understood
Ask participants their opinion on the observation for feedback/improvements
Check behavioural categories/coding system
What are the experiment designs?
Matched pairs
Repeated measures
Independent groups
What are order effects?
Order tasks are done in affects the task
What is a matched pairs design?
Choose a characteristic to match ppts on
Example of characteristic
Test ppts on characteristic
Match ppts based on test score and then randomly allocate one of each pair to each condition
What are the advantages of matched pairs design?
Tries to reduce participant variables
No order effects
Less chance of demand characteristics
What are the disadvantages of matched pairs design?
Participants can never be matched exactly
Time consuming
What is the ‘fix’ for matched pairs?
Choose only one key characteristic
What is a repeated measures design?
All participants take part in both conditions of an experiment
Can be two different tests
Performance of each ppt compared to themselves
What are the advantages of repeated measures design?
No participant variables
Fewer participants needed
What are the disadvantages of repeated measures design?
Order effects
Higher risk of demand characteristics
What are the ‘fixes’ for repeated measures?
Counterbalancing
Match materials for difficulty
Does counterbalancing fix order effects?
NO - CHECKS FOR ORDER EFFECTS
What is counterbalancing?
Split ppts into two groups
First group does condition 1 then condition 2
Second group does condition 2 then condition 1
Compare scores for each condition - if scores different, order effects have occurred
What is an independent groups design?
Two or more groups of ppts who only take part in one condition of experiment
What are the advantages of independent groups design?
No order effects
Less risk of demand characteristics
What are the disadvantages of independent groups design?
Participant variables
What is the ‘fix’ for independent groups design?
Random allocation:
- give each participant a number
- place numbers into a hat/random number generator
- select ppts alternatively for each group
What are the different types of sampling?
Volunteer
Opportunity
Random
Systematic
Stratified
What is a target population?
Large group of people you are trying to understand/explain
What is a sample?
Smaller group of participants in the target population
What does representative mean?
Sample to have same characteristics as our target population
What is the volunteer sampling method?
Participants select themselves to take part in an experiment - typically put up an advert
What are the advantages of volunteer sampling?
Easy
Less time consuming
What are the disadvantages of volunteer sampling?
Volunteer bias - only certain people will volunteer (confident)
What is the method of opportunity sampling?
Psychologist chooses ppts who are willing to take part - passers by
What are the advantages of opportunity sampling?
Less time consuming
Easy
What are the disadvantages of opportunity sampling?
Biased sample - only from one area
Researcher bias - psychologists only ask people they “want” to take part
What is the method of random sampling?
Every member of the target population has an equal chance of being selected
- each person is given a number
- number goes into a hat
- select number of ppts needed
What are the advantages of random sampling?
No researcher bias
What are the disadvantages of random sampling?
Unrepresentative sample
Time consuming
What is the method of systematic sampling?
Every nth person of the target population is chosen to take part
What are the advantages of systematic sampling?
No researcher bias
What are the disadvantages of systematic sampling?
Unrepresentative sample
Requires time and effort
What is the method of stratified sampling?
Sample is mini version of target population
- choose sub group
- find out ratio in target population
- randomly select sample out of sub groups
- explain random sampling methods (number of people in sample per sub group)
What are the advantages of stratified sampling?
No researcher bias
Tries to make the sample representative
What are the disadvantages of stratified sampling?
Cannot create perfect representation of target population
Time-consuming
What is qualitative data?
Non-numerical data
Language - expressed in words, created by open-ended questions
What are the advantages of qualitative data?
Richness of detail
What are the disadvantages of qualitative data?
Difficult to analyse
What is quantitative data?
Numerical data
Numbers - statistics, age, test scores, created by closed questions
What are the advantages of quantitative data?
Easy to analyse
What are the disadvantages of quantitative data?
Lacks detail
What is primary data?
Collected first-hand by the psychologists for the purpose of their own research
What are the advantages of primary data?
More relevant to the experiment
Less peripheral information
More control
What are the disadvantages of primary data?
Time and effort
Could be biased
What is secondary data?
Previous research collected by someone else for the purpose of their research
What are the advantages of secondary data?
Quicker to collect
Easy and cheap
What are the disadvantages of secondary data?
Lack of control
Could be out-dated
What is meta-analysis?
Conclusion of several studies which have all completed the same or similar experiments
What are the advantages of meta-analysis?
Lots of data
Can see the general conclusion/norms
What are the disadvantages of meta-analysis?
Lack of control
Publication bias - only look at positive research
What are the measures of dispersion?
Range
Standard deviation
What is the range?
Spread of the data
How is the range calculated?
highest value - lowest value
What are the advantages of the range?
Quicker than standard deviation
Easy to calculate
What are the disadvantages of the range?
Easily affected by extreme values
Not as representative/sensitive
What is the standard deviation?
Spread of data away from the mean
How is the standard deviation calculated?
Substitute the number into the formula
- work out the mean
- for each number, subtract mean and square result
- work out mean of squared differences
- take square root of that
What is a small standard deviation?
Data is close to the mean
What is a large standard deviation?
Data is far away from the mean
What are the advantages of standard deviation?
More representative/sensitive
What are the disadvantages of standard deviation?
Difficult to calculate
Time-consuming
What are the features of science?
Empirical methods
Replicability
Objectivity
Theory construction
Hypothesis testing
Paradigms
Falsifiability
What is empirical methods?
Conducting experiments/observations/self-report to gather evidence
Information tested using research, through direct observation or experiment, without bias
What topics show empirical methods?
Asch
Milgram
Loftus and Palmer
Behaviourist approach
What topics DON’T show empirical methods?
Humanistic approach
What is replicability?
Standardised procedures which meant that they could be repeated/replicated to assess consistency/reliability of the findings; this increased the validity of the conclusions drawn
Detailed report so that other psychologists can repeat the experiment in a different context and circumstance and see if you get consistent results so original experiment can be validated
What topics show replicability?
Milgram / Game of Death
What topics DON’T show replicability?
Asch / engineering students
Zimbardo / BBC
Strange Situation / Takahashi
Case studies
What is objectivity?
Keeping a critical distance from the experiment and having a factual/scientific DV
Data collected from the empirical research should be factual data that cannot be argued against and the experimenter must keep a critical distance from their research
What topics show objectivity?
Milgram
Double-blind procedure
What topics DON’T show objectivity?
Zimbardo
What is a theory?
Collection of general principles that explain observations and facts
What is theory construction?
Development of explanations/theories so that human behaviour can be predicted
Theories that help us understand and predict natural phenomena around us created by gathering evidence via direct observation
Inductive - theory then experiment
Deductive - experiment then theory
What topics show theory construction?
Explanations of conformity
What topics DON’T show theory construction?
Humanistic approach
What is hypothesis testing?
Manipulating an IV to measure the effect on the DV and keeping other (possible confounding) variables constant
What is a paradigm?
Shared assumptions between all psychologists
Psychology is classed as a pre-science as there are too many disagreements/approaches to be classed as a science
What is a paradigm shift?
The result of a scientific revolution - a significant change in the dominant unifying theory within scientific discipline
What is falsifiability?
Ability to disprove a theory/prove it wrong
What topics show falsifiability?
Behaviourist approach
What topics DON’T show falsifiability?
Psychodynamic approach
Humanistic approach
What is a peer review?
Before you are allowed to publish your research, must go through a peer review
What is the process of a peer review?
Where other psychologists/EXPERTS (psychology who work in a similar field) read through your research to see if it is allowed to be published
Done anonymously
What is research checked for in a peer review?
Validity
Significance
Originality
Method of the experiment - did they use controls, is the sample generalisable
Why are peer reviews done anonymously?
Reduce bias
Remove bribery
Is non-significant research published?
Yes
What are the outcomes of a peer review?
Yes - published straight away
Yes but you need to make some adjustments first
No - start again
What is the purpose of a peer review?
To ensure quality and relevant of research (methodology, data analysis etc)
To ensure accuracy of findings
To evaluate proposed designs (aims, quality and value of research) for research funding
What are the strengths of peer review?
Protects the quality of the research - minimises fraudulent “bad” research and protects the reputation of psychology
What are the limitations of peer review?
Publication bias - tends to publish only certain types of research and others may be ignored
What is THE table?
Nominal and relationship = Chi
Nominal and repeated measures = Sign
Nominal and independent groups = Chi
Ordinal and relationship = Spearman
Ordinal and repeated measures = Wilcoxon
Ordinal and independent groups = Mann-Whitney
Interval and relationship = Pearson
Interval and repeated measures = Related t-test
Interval and independent groups = Unrelated t-test
What are the levels of measurement?
Data
DV
Nominal, Ordinal, Interval
What is nominal data?
Category data
What is ordinal data?
Ordered in some way (ranked, scale 1-10)
What is interval data?
Data on a scale with equal intervals
What experimental design does matched pairs count as when it comes to THE table?
Repeated measures
What is the order of the three decision makers in choosing an inferential test?
- data - level of measurement
- test of difference/relationship
- experimental design
How is a sign test completed?
- calculate where differences are + or - or =
- add up number of + and - separately
- smaller number is s value
- check number of participants ignoring = ppts
- check to see if asked to use specific significance level; if not use 0.05
- check number of participants value for correct significance level
- is critical value < calculate value = significant
What type of test does a directional hypothesis produce?
One-tailed
What type of tailed test does a non-directional hypothesis produce?
Two-tailed
How do you calculate the degrees of freedom?
(number of rows - 1) x (number of columns - 1)
How is an inferential test completed?
- cross out incorrect tailed test
- cross out incorrect significance level
- cross out incorrect number of participants/df
- left with one number which is critical value
How is significance written?
- the results were significant/non-significant
- because the calculated of … was more/less than the critical value of …
- for a one/two-tailed test, n/df = …, for a p < 0.05
- therefore we accept/reject the null hypothesis (accept = non-significant, reject = significant)
What are the different sections of a scientific report?
Abstract
Literature review
Method
Results
Discussion
References
How many words is a typical abstract?
100-300 words
What is an abstract?
First section of a report
Brief/summary of aims, methods, results and conclusions
What is the purpose of an abstract?
To allow the reader to decide if they want to read the full report
What is an introduction?
Literature review of previous research with similar aim/procedure
What is the purpose of an introduction?
Decision of what type of hypothesis to write
What is included in a method?
Sample
Ethics
Task
Instructions
Materials
What is the purpose of a method?
Allows for replicability
What is included in a results section?
Measures of central tendency/dispersion
Graphs
Inferential tests
What is the purpose of a results section?
Even insignificant results are published
What is included in a discussion?
Implications for society
WWW/EBI
What is the purpose of a discussion?
Peer review = should be published/funding
How do you reference a journal article?
surname, initial. (year) article title, italics journal name, volume(issue), page number.
How do you reference a book?
surname, initial. (year) italics book title, place of publication: publisher
What is the purpose of references?
Allows to give credit
Stops plagiarism
How do you write a design a study question?
Use subheadings to signpost
Signpost justification
Put a second justification for A*
Bring back around to scenario
Write 3/4 sentences per bullet point
What is internal validity?
Are you measuring what you set out to measure?
What is external validity?
Can you generalise your results?
What is ecological validity?
Can your results generalise to real life?
What studies show high ecological validity?
Natural studies
Yuille and Cutshall
What studies show low ecological validity?
Lab studies
Loftus and Palmer
What is temporal validity?
Can your results generalise to another time?
What studies show high temporal validity?
New research
Tulving et al
What studies show low temporal validity?
Old research
Asch
What are the problems with validity?
Retrospective data
Demand characteristics
Investigator effects
Social desirability bias
What are the different methods of assessing validity?
Face validity
Concurrent validity
What is face validity?
On the surface, are you measuring what you set out to measure
What is the method of face validity?
Ask an expert to check over method/results
Look for anomalies
What is concurrent validity?
Using an already established test to check your experiment’s results
What is the method of concurrent validity?
- compare your results to an already established test’s results (can say interview family/friends and see if they give same answer)
- check the agreement - should be 0.8
- use an inferential test to check significance - Spearman or Pearson. If significant - good validity
How can you improve population validity?
Use stratified sampling
How can you improve internal validity?
Control group
Standardised method
Lab experiment
How can you improve ecological validity?
Make more realistic
How can you improve the validity of a lab experiment?
Control groups
Standardisation
Single/double blind
How can you improve the validity of an observation?
Covert
Operationalise behavioural categories
How can you improve the validity of a questionnaire?
Lie scale
Anonymous
How can you improve the validity of qualitative research?
Interpretive validity - how their interpretation of the data matches the participants interpretation
Triangulation
What is reliability?
Consistency
What is internal reliability?
Is the method consistent?
What is external reliability?
Are the results consistent?
What are the different methods of assessing reliability?
Test-retest reliability
Inter-rater reliability
What is test-retest reliability?
The SAME test is given to the SAME ppts on two occasions to see if the SAME results are obtained
What is the method of test-retest reliability?
- test SAME ppts using SAME test but later
- compare results of two tests
- check agreement - should be 0.8+
- use inferential test - Spearman or Pearson - and if significant, high reliability
What is inter-rater reliability?
The extent to which there is agreement between two or more observers involved in observations of behaviour
What is the method of inter-rater reliability?
- two psychologists discuss what they are going to analyse and how
- two psychologists observe/code/analyse same data but SEPARATELY (don’t talk)
- check agreement - should be 0.8+
- use inferential test - Spearman or Pearson - if significant, high reliability
How can you improve observational reliability?
Operationalising behavioural categories
Training
How can you improve self-report reliability?
Rewrite questions
Reduce ambiguity
Standardisation
How can you improve experimental reliability?
Standardise procedure of experiments
What is probability?
The likelihood of your results occurring by chance
What is the alternative (experimental) hypothesis?
A testable statement of a relationship (difference, correlation, association)
Directional or non-directional
What is a null hypothesis?
A prediction of no relationship or difference between variables studied
What does p<0.05 mean?
A LESS THAN 5% probability that any difference has occurred by chance
What does p<0.01 mean?
A LESS THAN 1% probability that any difference has occurred by chance
What is a type 1 error?
The null hypothesis has been WRONGLY REJECTED
False positive
What is a type 2 error?
The null hypothesis has been WRONGLY ACCEPTED
False negative
How do you choose which hypothesis to accept/reject?
SIGNIFICANT results
- ACCEPT the ALTERNATIVE hypothesis
- REJECT the NULL hypothesis
NOT SIGNIFICANT results
- REJECT the ALTERNATIVE hypothesis
- ACCEPT the NULL hypothesis
Why is 0.05 the default significance?
Balances the risk of having a type 1 and a type 2 error
0.01 is too narrow so you are more likely to make a type 1 or type 2 error
When is a significance level of 0.01 used?
When there is significant effect on human life
What is thematic analysis?
Produces themes, which are more descriptive than coding units
What is the type of data at the beginning and end of thematic analysis?
Start - QUALITATIVE data
End - QUALITATIVE data
What is the method of thematic analysis?
- read through several times
- find reoccurring themes and write them as quotes
- combine units in larger themes
- DO NOT CHANGE INTO QUANTITATIVE
- leave information as qualitative
What is content analysis?
A method of quantifying qualitative content via coding/categorisation
A form of (indirect) observation that examines artefacts/communications/forms of media that people produce
What is the type of data at the beginning and end of content analysis?
Start - QUALITATIVE data
End - QUANTITATIVE data
What is the method of content analysis?
- read through several times
- Have PRE-MADE categories previously chosen to look out for
- count the frequency of each category
- change the information into quantitative data
What are the strengths of content analysis?
Reliable way of analysing qualitative data as coding units not open to interpretation
Not time-consuming
Allows statistical analysis to be conducted
Many ethical issues may not apply if in public domain
What are the limitations content analysis?
Causality not established (only describes data)
Cannot extract deeper meaning or explanation
Communication studied out of context
Lack objectivity
What is a case study?
In-depth analysis of an individual, group, institution or event - typically ones who have had a unique experience
Typically longitudinal studies - conducted over a long period of time
How are case studies typically conducted?
Self-reports to gain a case history and produce qualitative
Tests (like IQ tests) to produce quantitative data
Self reports on family/friends to gain more detail
What are strengths of case studies?
Provide lots of data
Can investigate multiple factors at once
Able to update theories
What are the limitations of case studies?
Cannot generalise the results
Can be subjective data
Could have social desirability bias
What is a correlation?
Strength between two variables
What are the types of correlation?
Positive
Negative
Zero/no
What is a positive correlation?
As one variable increases so does the other
What is a negative correlation?
As one variable increases, the other decreases
What is zero/no correlation?
No relationship
What is the difference between an experiment and a correlation?
Experiment has an IV whereas correlations don’t
When should a correlation be drawn?
When they mention the word “relationship” in the scenario and there is no IV
What is a correlation co-efficient?
An estimation of strong the relationship of the two variables
What does a correlation co-efficient of 1 mean?
Positive correlation
What does a correlation co-efficient of 0?
No relationship
What does a correlation co-efficient of -1 mean?
Negative correlation
What are the advantages of correlations?
Useful preliminary tool to find out strength of relationship
Relatively quick and cheap to carry out
What are the disadvantages of correlations?
Cannot establish cause and effect
Could be other variables which affect the relationship