Studied by 0 people
British Psychological Society (BPS)
Creators of Ethical Guidelines of Research (in 1978)
Ethical Guidelines in Research
1) Informed consent
2) Deception
3) Debriefing
4) Withdrawl
5) Confidentiality
6) Protection from harm
Elaboration on
1) Informed consent
Participation must be voluntary
Elaboration on
2) Deception
Do not reveal true aims of experiment as it might skew results
Elaboration on
3) Debriefing
After the study, participants must be fully informed about the nature of the study
Elaboration on
4) Withdrawal
Participants are free to leave whenever they want
Elaboration on
5) Confidentiality
Personal data stays confidential within a research agreement
Elaboration on
6) Protection from harm
No mental or physical harm must be caused
NON-EXPERIMENTAL methods of research in psychology
1) Observation
2) Interviews
3) Case studies
4) Questionnaires/Surveys
5) Correlation
EXPERIMENTAL methods of research in psychology
1) Laboratory experiment
2) Field
3) Natural/Quasi
Definition of QUANTITATIVE data
Has a numerical basis and is therefore easy to analyze statistically
Mainly gathered from experimental methods and closed questions in questionnaires and interviews
A concise way of presenting a lot of data
Advantages of QUANTITATIVE data
Easy to analyse
Can pick out specific info. from a large range of people
There are lots of participants - checking for reliability is easier
Info. can be transferred into visual representations such as graphs/tables
Can be collected easily (closed questions)
Disadvantages of QUANTITATIVE data
Lacks detail of rich qualitative data
Requires knowledge to read graphs and tables
Needs strict definitions of categories on tables
Lacks detail of questions that may not be asked (can’t ask questions beyond the ones specified - barely any elaborations can occur)
There is no background information as such
Definition of QUALITATIVE data
Non-numerical data (words) expressing ideas, feelings, thoughts and attitudes of a participant
This is very rich, detailed information mainly gathered from asking open questions
Advantages of QUALITATIVE data
Detailed rich information
Greater quantity of information to process
Can cover a wide range of topics/issues
Easier to understand
Can include anecdotal information
Disadvantages of QUALITATIVE data
Relies on participant’s memory (which might not be the greatest)
They might be exaggerated responses - questioning the validity of the data collected
It may be difficult for the interviewer to maintain impartiality (due to subjectivity and bias)
Not easily analysed
Time consuming to collect and interpret the data
A lot of effort is needed overall to collect/interpret data in this manner
Definition of Observation (non-experimental method of research)
Behaviour is observed in its natural setting, and all variables are left free to vary (go about as they please) as the investigator does not interfere or manipulate variables in any way
Advantages of Observation (non-experimental method of research)
Highly ecologically valid as presents a realistic picture of spontaneous behaviour
Avoids demand characteristics and evaluation apprehension
Can be used to establish possible relationships when studying behaviour for the first time, then go on to do lab-based experimentation
Useful for studying participants (children, animals, mentally-ill) and topics (funerals, weddings) which would not work well in a laboratory
Disadvantages of Observation (non-experimental method of research)
Ethical problems: deception, invasion of privacy, informed consent
No control over extraneous variables: eg. heat, distractions - therefore cause and effect cannot be stated
Observer bias
Inter-rater reliability could potential be hard to maintain (overarching biases/prejudices)
Ecological Validity
A measure of how an experiment’s performance predicts behaviour(s) in real-life settings
Demand characteristics
Occurs when participants act differently simply because they know that they are in a study
They may try to guess the aims of the study and act accordingly
Evaluation apprehension
A human tendency to try to look better or have the fear of being evaluated
Extraneous variables
Other variables which may control the study
They are undesirable as they can influence the relationship between the independent and dependent variables
Inter-rater reliability
The degree of agreement among multiple raters/evaluators when assessing the same thing
Determines the consistency of measurements when different people evaluate or measure the same subject/object
If there is a higher inter-rater reliability, the assessment is consistent and reproducible across different raters
If there is a lower inter-rater reliability, the assessment is not consistent and can’t be reproduced across different raters
Participant Covert (PC - type of observation)
Example: Undercover policeman
Advantage: Natural behaviours observed closely
Disadvantage: Ethics (no informed consent - violation in psychology ethical guidelines)
Participant Overt (PO - type of observation)
Example: Play school assessments, driving test
Advantage: Ethically sound (meets all Psychology experiment guidelines)
Disadvantage: Behaviour might not be completely natural - the results might be biased (implication of evaluation apprehension since the participants may want to look good in terms of the results they provide)
Non-participant covert (NPC - type of observation)
Example: CCTV
Advantage: Natural behaviours observed at a distance
Disadvantage: Ethics (no informed consent - violation in psychology ethical guidelines)
Non-participant overt (NPO - type of observation)
Example: Inspectors watching lessons, “Big Brother“, Love Island TV show
Advantage: Ethically sound
Disadvantage: Behaviour can be affected by the presence of unknown observers - creating bias (evaluation apprehension)
Basic equipment for a Psychology experiment
Data collection table (digital/paper)
Disguises (if participant covert)
Paper/Pen/Clipboard
Ethical issue that needs to be considered when conducting an observation (Analysis of observational data)
Informed consent should be considered because the people should have a sense of comfort, and should know they are being used for an experiment
How to deal with having not considered “informed consent“
In a Participant covert/non-participant covert, the best time to deal with not having considered informed consent is when the data has been collected, but not processed - we can let the participants understand that their data is being used (also ensuring confidentiality/withdrawal)
Why Ecological validity should be considered when conducting an observation
Describes how participants act, but in real-life
Deception should be implemented to ensure the data is not biased (otherwise participants could become conscious/evaluation apprehension)
To maintain ecological validity, participant covert and non-participant covert would be suitable styles of observation
Operational Definition
‘What’ and ‘How’ to observe and identify the objective
This is done by creating a set of fixed criteria that enhances inter-rater reliability
With an example of observing ‘Violent behaviour’ - give an example for the ‘What’ and the ‘How’
What - Is there any visible/open wounds
How - Weapons, number of times physical contact occurs (punches), falls incurred
In the case of the ‘Hows’, maintaining a data table to tally the amount of times the actions occur (comparing the results of the other evaluators of the same experiment to reach a consensus of how many times each action occured - inter rater reliability)
Definition of Interviews
More personal, face-to-face situation where discussion takes place between interviewer and interviewee (there are 3 types of interviews)
General Advantages of Interviews
Good way to get a idea about general opinions about a topic/topics, within the population
General Disadvantages of Interviews
Time consuming - the interviewer has to hear a bunch of people talk
Costly - taking time out of people’s schedules needs to be compensated and resources for the interviews cost money too
Definition for Unstructured (type of interview)
Interviewee is free to discuss anything they want
Interviewer guides discussions and encourages discussion
Open question with the word ‘How’ used
“How are you feeling today?“
Example of Unstructured (type of interview)
Going to therapy/counselling where you get to talk about anything you want, and the therapist is there to guide the conversation
Advantages of Unstructured (type of interview)
Greater validity as participant is likely to tell the truth
Information that is rich and full is gained
The interviewee feels relaxed (i.e. has a sense of comfort)
The interviewer does not have to stick to any topic
Disadvantages of Unstructured (type of interview)
Difficult to analyse
Strongly influenced by characteristics of interviewer
eg. racist, sexist, bias for an ethnicity, level of attractiveness
Links into demand characteristics
Definition for Semi-structured (type of interview)
Interviewer has prepared open-ended questions
Interviewee is allowed to expand on these and go on relevant tangents
Interviewer has control over the whole discussion
Examples for Semi-structured (type of interview)
TV talk shows (James Corden, Jimmy Fallon)
Podcasts (the kind where special guests come on to talk about a specific topic)
Advantages for Semi-structured (type of interview)
Data analysis is relatively easier
Very useful for collecting large amounts of data
Interviewers need less training
Quick to administer
Disadvantages for Semi-structured (type of interview)
Participants react to formality and pick up on demand characteristics
Cannot follow up that easily on interesting points
Definition for Structured (type of interview)
Interviewer asks same set of standard questions to all interviewees in the same fixed order
Interviewer has control
Examples for Structured (type of interview)
Job interviews
English oral exams
Advantages for Structured (type of interview)
Data analysis is more simpler, more systematic
High reliability
Disadvantages for Structured (type of interview)
Less flexible as set questions must be asked
Interviewee feels less relaxed
Interviewer has to stick to the topic
The Aim when:
Observing the amount of rough and tumble play in a children’s playground
To investigate whether there is a lot of rough and tumble play in a playground
Closed questions would be asked since we want to get an idea about the AMOUNT of injuries
The Aim when:
Having a questionnaire for underage drinkers
To investigate whether underage drinkers like drinking
Semi-structured interviews would occur as it directs the conversation and gives room for elaboration
The Aim when:
An interview is conducted with a person that is suffering from Depression
To investigate the effect of depression on a person
Unstructured interviews would be posed as it helps the interviewee feel like they have a space to unload their thoughts and feelings
Considerations taken into account when an Observation is taking place
Ecological validity
Operational definition
Demand characteristics
Inter-rater reliability
Extraneous variables
Evaluation apprehension
Considerations taken into account when a Questionnaire/survey is taking place
Ecological validity
Operational definition
Demand characteristics
Inter-rater reliability
Extraneous variables
Evaluation apprehension
Social desirability bias
Pilot study
Open and closed questions
Considerations taken into account when an Interview is taking place
Ecological validity
Operational definition
Demand characteristics
Inter-rater reliability
Extraneous variables
Evaluation apprehension
Social desirability bias
Pilot study
Open and closed questions
Interviewer effects
Structure (un-structured, semi-structured)
Deductive approach
Begins with a theory, developing hypotheses from the theory, and then collecting and analysing data to test the hypothesis
Essentially testing a new theory
It is a psychological process that helps people make decisions and solve problems
Example: All dogs have ears → golden retrievers are dogs → all golden retrievers can hear
Inductive approach
A form of reasoning in which inferences and general principles are drawn from specific observations and cases
Tests an existing theory
Example: If after every time you eat a sweet food you get a cavity → reasoning would lead to the thought that sweet food ALWAYS gives cavities
Content analysis
Allows for the quantification of qualitative data by measuring frequencies of categories and themes
Has two different distinctions
Manifest content: surface data accessible to the researcher
Latent content: underlying data and meaning (not quantifiable, only inferred)
Not tied to any theory or conceptual framework
Example: Quantifying the amount of times the same word has been used in a literary essay
Definition of Case study
Researcher observes the behaviour (feelings, attitude, IQ, blood pressure) of an individual, or group of people for a long period of time (6 months at least), about the topic under investigation
Advantages of Case studies
Rich data (in-depth) in a natural setting of the participant(s) - hereby expressing ecological validity
Disadvantages of Case studies
Time consuming
Attrition rates (people don’t show up eg. death)
Can be subjective overall
Examples of Case studies
3 identical strangers (triplets separated at birth, had no idea they were triplets, etc)
Clive Wearings (Severe case of amnesia - retrograde, meaning he lost all memories of his past and is unable to form new ones)
Phineas Gage (A metal rod went through his head, but he still was able to perform significant cognitive functions)
Definition of a Questionnaire/survey
Involves asking participants questions about their attitudes, opinions, behaviour, and intentions
Involves written answers recorded by the participant themselves, but can also be oral answers to researcher’s questions
Researcher does not need to be present when the questionnaire is administered (can be by post, telephone, internet, etc)
Advantages of a Questionnaire/survey
Large amount of data an be collected at relatively little cost to analyse trends
Can collect sensitive data not available through observation, etc (eg. info about private life)
Participants more likely to tell the truth as it is anonymous
Faster way of collecting data compared to interviews
Disadvantages of a Questionnaire/survey
Only suitable for some participants as there is a need for the ability to read
i.e. not suitable for young children, elderly, illiterate, learning difficulties, different cultures/languages
Relies on participant’s honesty - there might be social desirability bias
Limited choices answers may exclude information
Answers may be misinterpreted by researchers - researcher bias, designer bias
Potential questionnaire design issues
For a questionnaire we have to consider the following and limit anything that may come in between it during the survey:
Simplicity
Speed
Ease of analysis
Leading questions
Technical terms
Open and closed questions
Invasion of privacy
Order of questions
A pilot study must be conducted prior to the actual experiment in order to limit factors that could impact the above mentioned considerations for a good questionnaire/survey
Pilot study
Small trial versions of proposed studies to test the effectiveness and make improvements
Investigator defects
Occurs when a researcher intentionally or unintentionally influences the outcome of any research they are conducting
Definition of Triangulation
Using multiple research methods, theories, and investigators to study a particular topic - helps in refining ideas and finding new ways of understanding/interpreting data
Definition for Data triangulation
The process of checking data by comparing results obtained from multiple sources (used in qualitative and quantitative research)
Advantages of Data triangulation
Helps ensure the data is accurate
Reveals unique findings
Disadvantages of Data triangulation
Increased time
Conflicts may occur due to biases and theoretical frameworks
Misinterpretation of data
Definition of Investigator triangulation
Involving multiple researchers in collecting/analysing data
Advantages of Investigator triangulation
More than one investigator decreases chances of biases
Inter-rater reliability
Increased expertise brought in
Disadvantages of Investigator triangulation
Other researchers may have different views on the data collected, jeopardizing findings
More disruptive
Definition of Theory triangulation
Using multiple theoretical models to study a particular phenomenon/issue (tests a competing hypothesis and applies alternate theoretical frameworks)
Advantages of Theory triangulation
Resolves data contradictions
Helps in understanding personal conclusions in other perspectives
Richer analysis
Disadvantages of Theory triangulation
May cause confusion if the theoretical frameworks are not clearly identified/made known
Definition for Methodological triangulation
Using multiple methods of data collection (combination of qualitative and quantitative methods) - can be experimental and non-experimental
Advantages for Methodological triangulation
Ensures objective analysis
Prevents researcher bias
Refines study by identifying gaps and incosistencies
Disadvantages for Methodological triangulation
Difficulty in combining statistical and descriptive data
Definition for Correlation
Tests whether a systematic relationship exists between two or more co-variables (hence why it’s called co-rrlations)
Example of Positive correlation
eg. shoe-size EU to forearm size in cm (in-class test)
Example of Negative correlation
eg. age of car in years to the value of the car in euros
Example of No correlation
eg. current math grade and height in cm
Correlation coefficient
The number which represents the direction and the strength of the relationship between two variables
Ranges between -1 to +1
Correlation is NOT causation
Correlations are only RELATIONSHIPS between two variables
Causation is when one variable predicts the result of the other
Advantages of Correlation
Can collect much info. from many subjects at one time
Can study a wide range of variables and their interrelations
Study variables that are not easily produced/ethical to produce in the laboratory (can study them naturally)
Easy to see a relationship in the data (represented in a graphical format)
Disadvantages of Correlation
Correlation does not indicate causation
No control over extraneous variables that might affect the correlation
Definition of Reliability
Refers to the ability if a test/observation/interview, measuring instrument or questionnaire to perform consistently (i.e. the same performance each time)
There are 3 methods to test for reliability
Test-Retest (test for reliability)
Performs consistently over-time - i.e. feet size and forearm size
Participants carry out the test and then after some time the same participants carry out the exact same test again
If the two sets of results are strongly/positively correlated, the test is externally reliable
There is a chance that people underperform as they might be bored
There is a chance they may learn the answers by being exposed to it a lot (practice) hence presenting a limitation of test-retest
Equivalent forms (test for reliability)
Performs consistently over time
To avoid waiting for participants to forget the details of test so that a retest can be done - a very similar equivalent second test can be used for the ‘retest’
Participants carry out the 1st test, and then after some time the participants carry out the equivalent 2nd test
Then the two sets of results are correlated
If the test is externally reliable, then there will be a strong positive correlation between the two sets of results (i.e. seeing the correlation between car-driving and motorbike driving tests - equivalent = the same!)
Split-halves (test for reliability)
All items in a test performing in the same way - the participants carry out the test and then the test is split into halves - odd and even numbers
Participants’ performance on the even numbers is then correlated with their performance on the odd numbers
If the test is internally reliable, then there will be a strong positive correlation between the two sets of results
Definition of Validity
A matter of whether the test or measuring implemented measures what it is supposed to → eg. a scale to test/weigh how heavy one is in kilograms
i.e. length of forearm as a test of IQ would be reliable (that is if you get the same result time after time)
However it would not be an accurate or true measurement of IQ - therefore it is not a valid test
Content or face validity (method for testing Validity)
Examine the test and see whether, on the face of it, it looks valid! (educated opinion)
Predictive validity (method for testing Validity)
Many tests are used to predict future behaviour (eg. IB score ought to predict who would be best on a high school level)
The predictive validity of IB points tests can be assessed by doing a correlation between the score on IB and later performance at university
If the IB tests a good predictive validity, then there will be a strong, positive correlation between the two sets of results
Difference between Reliability and Validity
Reliability looks for consistency
Validity looks for accuracy
Definition of an Independent variable
The variable that is changed/manipulated usually by the experimenter
Ideally it is the only thing that is different between the conditions
eg. use of words in a memory experiment
conditions: gender (male/female)
Definition of a Dependent variable
The variable that is measured by the experimenter, as a result of the manipulation
eg. number of words that are remembered
conditions: gender (male/female)
Variables must be operationalised
Variables must:
be stated in a form that can be measured
eg. not just memory, but number of words remembered
eg. not just stress, but stress score
Three types of experiments
Laboratory experiment
Field
Natural
Note 
Flashcard (26)