Research method Flashcards
What are the main ethical principles decided by the committee of ethics of the BPS?
Respect
Competence
Responsibility
Integrity
What is research ethics?
They are moral principles that act as a guideline to be followed by the researcher throughout the inception, completion and until publication of results.
Respect
Respect for the dignity of people
Recognise the inherent worth of all human beings.
All human beings are worthy of equal moral consideration
Competence
Value the continuing development of maintenance of a high standard of competence
Works within the recognised limits of your knowledge, skill, training, education , experience.
Responsibility
Value responsibilities to people, the general public, and to the profession and science of psychology.
Avoid harm
Prevent misuse or abuse of your contribution to the society
Integrity
Value honesty, probity, accuracy, clarity and fairness in interaction with all persons and peoples.
Seek to promote integrity in all facets of scientific and professional endeavours.
What does weighing the risk mean?
Minimise physical/ psychological harm, discomfort or stress
How much deception it involves
Justifying risks in terms of the scientific value of the results
What kind of approval is required prior to the experiment ?
Ethical approval
Unethical studies of the past
Experiments conducted by physicians in the concentration camps
during WW2
* Milgram study (early 1960s)
* Stanford Prison Experiment (1971)
* The U.S. Public Health Service Syphilis Study at Tuskegee (1932-1972)
Which unethical experiment led to the development of the Nuremberg Code of Ethics in Medical Research?
Harmful experiments conducted by Josef Mengele “Angel of Death” ( at least one of the main Nazi physician involved) on the inmates during WW2.
Nuremberg Code of Ethics in
Medical Research (1947)
Carefully weighing risks against benefits and the need for informed consent
Milgram Study (~1960s)
- A response to Nuremberg trials
- Advertised as a study on memory
- Learning under an authority figure
- 3 people in the room ( experimenter, learner and teacher)
- The teacher is the one who gets asked to shock the learner (the confederate)
What prods were made in Milgram studies to push the participants to continue the experiment?
- Please continue or please go on
- The experiment requires that you continue
- It is absolutely essential that you continue
- You have no other choice; you must go on
What percentage of Milgram studies participants continued?
65%.
Even when the shock ranged from 15V- 450 V
All participants were debriefed at the end
Stanford Prison Experiment (1971)
Conducted by Philip Zimbardo
- 24 participants male ( 12 prison guards and 12 prisoners)
- All participants signed the informed consent
( so agreed on some harassment, to some privacy and civil rights violated , minimal diet)
- Guards were asked to maintain law and order
- So they became abusive starting day 2
Planned duration - 14 days
The experiment lasted - 6 days only
BPS code guidelines for all
Written in vague language so can be applied to all members (including students, and clinicians).
Also says that where own “professional and ethical judgment” required use it
Syphilis Study at Tuskegee (1932- 1972)
“Untreated Syphilis in the Male Negro”
Conducted by U.S Public Health Service in rural Alabama
600 African-American men (399 in the experimental and 201 in the control
group)
* Reward: free medical care, meals, and burial insurance
- When the study started there was no cure available
Why syphilis study was unethical?
Participants’ informed consent was not collected
* Researchers told the men they were being treated for “bad blood”
* 1943 penicillin became treatment of choice for syphilis, but participants in
the study were not offered treatment
Which study resulted in the formation of Ethics Advisory Board?
Syphilis study
Ethics Advisory Board has three ethical principles (Belmont Report):
- Respect for persons
- Beneficence
- Justice
What are the 8 main things informed consent should state ?
Participants need to be aware what they are signing up for (Briefing):
* Aims of the project
* Time commitment
* Risks outlined
* Opportunity to withdraw at any time without any consequences
* Compliance with Data Protection Act
* Confidentially and anonymity conditions
* Planned debriefing
* Contact details of the researcher
Under 16 -> parent consent
When is deception a choice?
If there are no other effective procedures
* Strong scientific merit → Benefits need to outweigh the risks
* Debrief as soon as experiment/ study is completed
* Appropriate risk management and harm alleviation strategy
Deception
Leading participants to believe that something
other than the true independent variable is involved, or withholding information such that the reality of the investigative situation is distorted
Prevents undesirable behaviour
Disadvantage:
- Lowers public trust in Psychological research
- Goes against the integrity and responsibility criteria of the BPS code of Ethics
- They did not consent to it
Debriefing
Informing participants about the full nature
and rationale of the study they’ve experienced, and
attempting to reverse any negative influence.
Debriefing in a longitudinal study:
No possible or partial debrief after each session.
Dibbets & Schulte-Ostermann (2015)
Possible reasons for no debriefing :
Does not reflect real -life situation (no debriefing in real- life )
Dibbetts don’t conduct debriefing in the first session but to minimize the harm they recruit people not exposed to abuse; they lowered the intensity of the abuse scene in VR; a healthcare psychologist was made available for consultation.
Traditional (in-lab) data collection
1) Advertise
2) Participant would get in touch with you
3) Schedule participants to take part in the experiment on a particular day and time
4) Participant takes part
Traditional (outside lab ) data collection
e.g. surveys
1) Advertise
2) Participants would get in touch
3) Mail survey
4) Take part
5) Return survey
Advantages of traditional in-lab data collection
- Recruitment criteria
- Careful responding
- Participant completes the study
once
Disadvantages/ challenges of traditional in-lab data collection
Time
* Recruiting is an effortful process/ scheduling
* On the day
* No-shows/ participants may be late
* Participant money
Costs
* Printing costs
* Travel/ postage costs
Data management
* Data entry
* Scoring/ response scale
Advantages of online data collection
- Internet
- Computer technology
- Large pool of participants
globally accessible - Access to “hidden populations”
- Increase sample diversity (e.g.
more balanced ratio of genders) - Randomisation of items
- Time
- Recruiting/ scheduling
- More than one participant at the
same time - Costs
- No printing costs
- No travel/ postage costs
- Data management
- No data entry → reduces human
error & increases accuracy - Scoring/ response scale
Challenges of online data collection
- Building an experiment may require knowledge of programming skills if not should hire another employee
Choice of Software: open-source vs licence
- Hosting platform/ server – website from where experiment is
accessed - In-house (e.g. Experimentum) vs external (e.g. Gorilla, Pavlovia, OSWeb)
- Recruitment
- Email/ Social Media vs Sona vs Prolific/ Amazon Mechanical Turk (MTurk)
Cost vs Time trade-off
Why is there decline in data quality in online data collection?
- Experimenter absent → less control than in-lab
- Less committed
- Divided attention
- Higher drop-out rates
- Missing data = data point has been skipped/ missed
- Tech glitches
- Inattention
- Outliers = response values that are deemed “extreme” compared to
the distribution - Response patterns in questionnaires
Reason for multiple submissions in online data collection?
- Forgetfulness
- To gain incentives
- Distort results
Advantages of online data collection for participants
- Anonymity → more likely to answer truthfully to sensitive topics
- Convenience
- Provide information at their own pace
- Increased sense of control
- Incentives
*Methods of solving challenges in online data collection
stimulus presentation issues can be resolved by:
- conducting practice trials
- screen calibration
- Viewing distance
- Timing
Resolving multiple submission issue
Depends on the type of experimental set- up:
- Email/ Social Media vs
recruitment platforms
How to improve data quality issues (such as less commitment, divided attention and high drop out rates) ?
For less committed and divided attention:
* Recruitment services
* Approval rates (e.g. > 95%)
* Previous number of submissions
- Attention checks
- Objective criterion
For higher drop-out rates:
* Predetermine sample size
* Encourage participation →
incentives
How to resolve the missing value issue?
Decide whether the missing value is important in answering the specific research question or not.
If it is important to the RQ including that incomplete data in the results will reduce the validity
How to prevent any outliers from corrupting the results?
1) Researcher decides what is an outlier
2) Before you look at the data
3) Justification backed up by evidence (should be clear before excluding outliers! )
What is straightlining? How can we prevent this?
Giving the same answer for all questions.
A possible form of an outlier (although it may be valid).
- Reverse coding may help us decide whether the participants are straightlining or not therefore whether it is valid or not.
- Completion time (extend it?)
