Transcript Methodology - Kennedy HS

Methodology
Quantitative Methods
Qualitative Methods
What is the difference?
• Quantitative Methods (Experiment)
– Can you quantify it?
– Narrow concepts to conduct statistical analysis
– Experiments
• All about numbers after manipulation
• Qualitative
– Quality in data (interviews, surveys)
• Thoughts, feeling and perceptions
• Eclectic Approach
– Quantitative and Qualitative approaches
Quantitative Research
• Experimental Studies with Controls
• Uses scientific method
– Hypothesis
– Independent Variable
– Dependent Variable
Psychology Experiment
(Mozart Effect)
• Variable
– IV: conditions altered or varied by experimenter. Suspected causes
for difference in behavior
• Music
– DV: measures results. Shows the effects that the IV have on
behavior.
• Math Score
– Extraneous Variables: other conditions the researcher wishes to
prevent
• Groups
– Experimental Group: exposure to IV
• Music
– Control Group: NO IV but all other variables
• No music
Make you own hypothesis…
• A hypothesis contains the following items
– Variable being altered by experimenter
– Behavior prediction
• Example
– Gum chewing during a math test will increase test
scores.
– If a participant holds a heavier weight prior to weight
prediction of another item, the participant will predict
that the item weighs more.
Experimental Control
• Random Assignment
– Equal chance of being in control or experimental
group
– Balances differences in the two groups
• How do you limit extraneous variables
– Make all conditions the same for each group
EXCEPT the IV
Placebo Effects
• Definition: inactive substance given in place of
a drug in research.
– Treat a patient by suggestion
• Effect: changes in behavior due to expectations
• Controlling Placebo Effects
– Single-blind: patient does not know
– Double-blind: patient and experimenter do not
know
• Example: 50% of the effectiveness from antidepressants are from placebo
Experimenter Effects
• Single-Blind Study
– May contain bias
• Experimenter Effects
– Unintended changes in participant behavior due
to cues inadvertently given by the experimenter
• Double-blind study
• Limits bias and effect
Qualitative Research
• Descriptive Research
– Examples
•
•
•
•
•
Naturalistic Observation
Correlational studies
Case Study
Survey Method
Meta analysis (Theoretical Foundation)
Naturalistic Observation
• Observing behavior in a natural setting
– Typical environment where an animal/person lives
– Jane Goodall in Tanzania
• Ape uses a tool
• Same as a Zoo?
– NO
• Limitations
– Observer Effect: changes in behavior caused by an
observer
– Observer Bias: observers see what they want to see
– Anthropomorphic Fallacy:attributing human thoughts,
feelings and motives to animals
Correlational Studies
• Correlated: linked together in an orderly
way
– Study: finds a degree of relationship between
two traits, behaviors or events
• Correlation Coefficients
– Degree of strength
• +1.00 and –1.00
– Positive: +1.00 Perfect relationship
• Correlation does not show Causation!!!!
Clinical Method
• Case Study: in-depth focus on a single
subject
– Natural clinical tests provide data
• Answers interesting questions
– Ethical guidelines don’t matter
• Limitation
– Little or no control is possible
– No control group
– Single case may be misleading
Survey Method
• Can you ask everyone in the world questions
pertaining to your research?
– NO
• Definition: public polling used to answer psychological
questions
• Sampling
– Representative sample: small group representing a large group
– Population: people belonging to a group
• Limitations
– Using psych students okay?
– MUST have representative samples
– Courtesy Bias: give “polite” or socially desirable answers
Meta-Analysis
• More than a 100 studies are done on
the same key concept
– So how do you get through all the data?
• Two Types
– Research = Meta Analysis
– Own Research = analyze in relation to
meta analysis
• Triangulation
– Definition - When studying human behavior,
more than one method or approach is used
– There are 4 different types of triangulation
• Data Triangulation
– Involves using different times, locations and
participants
• Observer Triangulation
– Using more than one observer in the study
• Methodological triangulation
– Using a variety of qualitative and quantitative
research methods
• Theory triangulation
– Using more than one perspective to understand
human behavior
Triangulation…(continued)
• Advantages
– Qualitative research
• broader inferences
• Broader indications
– Increases the likelihood of the use of
qualitative research
• multiple methods and perspectives
used
Triangulation…(continued)
• Disadvantages
– Too simplistic
– Confusing
• Too many approaches
• Too many methods
Interviews
•
These are the six types of interviews
1.
2.
3.
4.
5.
6.
Email and Telephone
One to one
Conversational
Structured
Semi-structured
Focus Groups
Problems With Methodology
When evaluating theory and research,
is the BIG FOUR!!!
Evolution
• In the beginning…
– Qualitative (introspection and
psychoanalysis)
• Behaviorism and the “black box”
– If you cannot observe it directly and
objectively, it could not be scientifically
studied.
– Reject all other perspectives
• Inadequate and unscientific
• Cognitivists… yes you can
• Behaviorist legacy
– Only numerical results were scientific
• BUT… what do those numbers mean?
– Interpretation
• Protocol Analysis: the way participants were
thinking during research
• New emphasis on human meaning and
experience
– Ethical issues
• Today… both methods are used and
accepted
Evaluating Research
1.
2.
3.
4.
5.
6.
Sampling
Random Sampling
Demand characteristics
Researcher bias/expectancy
Cultural bias
Gender Bias
Sampling
• Population
– Definition: set of research participants
– Representative sample of typical population
• Group size
– Small size may have distorted data
• Individual differences are more noticeable
– Large, good, thousands
Random Sampling
• Definition
– Everyone in the population has an equal chance
of being selected
• Opportunity sample… most widely used
– BUT…open to bias and distortion
• Sampling bias
– Undergraduate students represent general
population?
Demand Characteristics
• Validity issues
– Does it measure what it is supposed to?
• Humans are treated like subjects and the only thing
different is the variable
– Is it?
• Martin Orne (1962)
– People act differently in a psychological
experiment
• Overly cooperative
• Lacks ecological validity
+ cultural biases
Researcher Bias
• Self-fulfilling prophecy and
expectancy effect
– Pick up small, nonverbal cues
• Expectancy Effect in analyzing
data
Problems of Gender Research
• Begins with biased assumption
– Males are more intelligent
• Fails to questions assumption
• Asking questions based on assumption
• Misinterpretation and failure to question
when results contradict assumptions
• Sex  biology
• Gender  Culture (similar arguments)
What is culture?
• Culture, Ethnicity, Nationality?
• Defined as…
– Shared rules that govern the behavior of
members
– Set of values, beliefs and attitudes shared by
most
Cultural Bias
• Interpretation of data is/could be different
between the cultures
– WHY? The subjects interpretation of the
question/topic may be different
Types of Cultures
• Collectivist
– define themselves in terms of their relationships with
others
– more inclined to give up their individual needs
• conflict between their needs and group needs
– Emphasis on “in-group”
• Individualistic
– define themselves as autonomous entities independent
of a group
• People in this group tend to be more competitive
Problems with the study of Culture
1. Methods and Sampling
–
–
Questionnaires/surveys  language
representativeness and attitudes
2. Interpretation
–
–
Interpretation in light of culture
Example: Circumcision
3. Stereotyping
–
Pre-existing stereotypes impact interpretation
4. Reitification of culture
–
–
Avoid circular reasoning
STOP  Regard intangible processes (feelings) as literal objects
Ethical guidelines
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Do no harm
Accurately describe risks
Ensure voluntary participation
Minimize discomfort
Maintain confidentiality
Do no unnecessarily invade privacy
Use deception wisely
Debrief (remove misconceptions)
Provide results and interpretations (Debrief)
Treat participants with dignity and respect
Ethics #1
• Explain why the following research is
unethical.
• The Monster Study was a stuttering experiment on 22 orphan
children in Davenport, Iowa, in 1939 conducted by Wendell
Johnson at the University of Iowa. Johnson chose one of his
graduate students, Mary Tudor, to conduct the experiment and he
supervised her research. After placing the children in control and
experimental groups, Tudor gave positive speech therapy to half
of the children, praising the fluency of their speech, and negative
speech therapy to the other half, belittling the children for every
speech imperfection and telling them they were stutterers. Many
of the normal speaking orphan children who received negative
therapy in the experiment suffered negative psychological effects
and some retained speech problems during the course of their
life. Dubbed “The Monster Study” by some of Johnson’s peers
who were horrified that he would experiment on orphan children
to prove a theory, the experiment was kept hidden for fear
Johnson’s reputation would be tarnished in the wake of human
experiments conducted by the Nazis during World War II. The
University of Iowa publicly apologized for the Monster Study in
2001.
Ethics #2
• South Africa’s apartheid army forced white lesbian and gay soldiers to
undergo ’sex-change’ operations in the 1970’s and the 1980’s, and
submitted many to chemical castration, electric shock, and other
unethical medical experiments. Although the exact number is not
known, former apartheid army surgeons estimate that as many as 900
forced ’sexual reassignment’ operations may have been performed
between 1971 and 1989 at military hospitals, as part of a top-secret
program to root out homosexuality from the service.
• Army psychiatrists aided by chaplains aggressively ferreted out
suspected homosexuals from the armed forces, sending them discretely
to military psychiatric units, chiefly ward 22 of 1 Military Hospital at
Voortrekkerhoogte, near Pretoria. Those who could not be ‘cured’ with
drugs, aversion shock therapy, hormone treatment, and other radical
‘psychiatric’ means were chemically castrated or given sex-change
operations.
• Although several cases of lesbian soldiers abused have been
documented so far—including one botched sex-change operation—
most of the victims appear to have been young, 16 to 24-year-old white
males drafted into the apartheid army.
• Dr. Aubrey Levin (the head of the study) is now Clinical Professor in
the Department of Psychiatry (Forensic Division) at the University of
Calgary’s Medical School. He is also in private practice, as a member
in good standing of the College of Physicians and Surgeons of Alberta.
Ethics #2
• While animal experimentation can be incredibly helpful in
understanding man, and developing life saving drugs, there have
been experiments which go well beyond the realms of ethics. The
monkey drug trials of 1969 were one such case. In this
experiment, a large group of monkeys and rats were trained to
inject themselves with an assortment of drugs, including
morphine, alcohol, codeine, cocaine, and amphetamines. Once the
animals were capable of self-injecting, they were left to their own
devices with a large supply of each drug.
• The animals were so disturbed (as one would expect) that some
tried so hard to escape that they broke their arms in the process.
The monkeys taking cocaine suffered convulsions and in some
cases tore off their own fingers (possible as a consequence of
hallucinations), one monkey taking amphetamines tore all of the
fur from his arm and abdomen, and in the case of cocaine and
morphine combined, death would occur within 2 weeks.
• The point of the experiment was simply to understand the effects
of addiction and drug use; a point which, I think, most rational
and ethical people would know did not require such horrendous
treatment of animals.
Ethics #3
• In 1924, Carney Landis, a psychology graduate at the University of
Minnesota developed an experiment to determine whether different
emotions create facial expressions specific to that emotion. The aim of this
experiment was to see if all people have a common expression when feeling
disgust, shock, joy, and so on.
• Most of the participants in the experiment were students. They were taken to
a lab and their faces were painted with black lines, in order to study the
movements of their facial muscles. They were then exposed to a variety of
stimuli designed to create a strong reaction. As each person reacted, they
were photographed by Landis. The subjects were made to smell ammonia,
to look at pornography, and to put their hands into a bucket of frogs. But the
controversy around this study was the final part of the test.
• Participants were shown a live rat and given instructions to behead it. While
all the participants were repelled by the idea, fully one third did it. The
situation was made worse by the fact that most of the students had no idea
how to perform this operation in a humane manner and the animals were
forced to experience great suffering. For the one third who refused to
perform the decapitation, Landis would pick up the knife and cut the
animals head off for them.
• The consequences of the study were actually more important for their
evidence that people are willing to do almost anything when asked in a
situation like this. The study did not prove that humans have a common set
of unique facial expressions.
Ethics #4
• In 1965, psychologists Mark Seligman and Steve Maier conducted an
experiment in which three groups of dogs were placed in harnesses.
Dogs from group one were released after a certain amount of time,
with no harm done. Dogs from group two were paired up and leashed
together, and one from each pair was given electrical shocks that could
be ended by pressing a lever. Dogs from group three were also paired
up and leashed together, one receiving shocks, but the shocks didn’t
end when the lever was pressed. Shocks came randomly and seemed
inevitable, which caused “learned helplessness,” the dogs assuming
that nothing could be done about the shocks. The dogs in group three
ended up displaying symptoms of clinical depression.
• Later, group three dogs were placed in a box with by themselves. They
were again shocked, but they could easily end the shocks by jumping
out of the box. These dogs simply “gave up,” again displaying learned
helplessness. The image above is a healthy pet dog in a science lab, not
an animal used in experimentation.
Ethics #5
• In 1965, a baby boy was born in Canada named David Reimer. At eight
months old, he was brought in for a standard procedure: circumcision.
Unfortunately, during the process his penis was burned off. This was due to
the physicians using an electrocautery needle instead of a standard scalpel.
When the parents visited psychologist John Money, he suggested a simple
solution to a very complicated problem: a sex change. His parents were
distraught about the situation, but they eventually agreed to the procedure.
They didn’t know that the doctor’s true intentions were to prove that nurture,
not nature, determined gender identity. For his own selfish gain, he decided
to use David as his own private case study.
• David, now Brenda, had a constructed vagina and was given hormonal
supplements. Dr. Money called the experiment a success, neglecting to report
the negative effects of Brenda’s surgery. She acted very much like a
stereotypical boy and had conflicting and confusing feelings about an array
of topics. Worst of all, her parents did not inform her of the horrific accident
as an infant. This caused a devastating tremor through the family. Brenda’s
mother was suicidal, her father was alcoholic, and her brother was severely
depressed.
• Finally, Brenda’s parents gave her the news of her true gender when she was
fourteen years old. Brenda decided to become David again, stopped taking
estrogen, and had a penis reconstructed. Dr. Money reported no further
results beyond insisting that the experiment had been a success, leaving out
many details of David’s obvious struggle with gender identity. At the age of
38, David committed suicide.
Higher Level
Qualitative Research
What is Qualitative Research
Methodology?
• Qualitative research is one of the two major
approaches to research
• Relies on reasons behind various aspects
of behavior
– investigates the why and how of decision
making
– Compared to what, where, and when of
quantitative research
• Triangulation
– Definition - When studying human behavior,
more than one method or approach is used
– There are 4 different types of triangulation
• Data Triangulation
– Involves using different times, locations and
participants
• Observer Triangulation
– Using more than one observer in the study
• Methodological triangulation
– Using a variety of qualitative and quantitative
research methods
• Theory triangulation
– Using more than one perspective to understand
human behavior
Triangulation…(continued)
• Advantages
– Qualitative research
• broader inferences
• Broader indications
– Increases the likelihood of the use of
qualitative research
• multiple methods and perspectives
used
Triangulation…(continued)
• Disadvantages
– Too simplistic
– Confusing
• Too many approaches
• Too many methods
Interviews
•
These are the six types of interviews
1.
2.
3.
4.
5.
6.
Email and Telephone
One to one
Conversational
Structured
Semi-structured
Focus Groups
Email and Telephone Interviews
• Phone or Email (websites too)
–Advantages
• high response rates
– some people like talking over the phone
» Problem: sample bias
– some enjoy filling out the Email Questionnaires
• Sample research Can be hugely Diverse
• Lower tendency for the participant to give
desirable responses
• Reduction of risk of the researcher being physically
attacked
Email and Telephone Interviews
Disadvantages
–Rapport (the one to one level)
• Difficult, if not impossible, to achieve
–Literacy issues with email responses
–The lack of visual cues
• may cause misinterpretation
Email and Telephone Interviews
• So although this type of
interview may yield
quantitative results, it still
has much qualitative
attributes and results
Focus Groups Interviews
• Start
– It originated in market research in the 1920’s
• USE
– political parties: Most often
• assess the likely response to proposed policies
– Social Research now
Focus Groups Interviews
• Advantages
– Efficient way of generating substantial amount of data
– Participants tend to enjoy the experience
– Inexpensive and flexible
• Disadvantages
– Lack of Confidentiality
– Lack of information provided by participants
– Conformity
Focus Groups Interviews
• Overall, not the best;
– Emphasis on quantitative
• Designed for qualitative
– Impact of environment
One to One Interviews
• Addresses the concerns of focus group situations
• The interviewer tries to build a relationship
• Advantages
– More willing to give valuable personal information
– More data, as only one person is responding
– Allows a relationship between the participant and the
interviewer
• this may yield better results in the future for follow up
interviews
One to One Interviews
• Disadvantages
– Time consuming
– Difficult to conduct well
– Researcher needs to keep their
subjectivity
• Reflect on bias
• Reflect/report lack of subjectivity
One to One Interviews
• Difficult to conduct well
• Large amount of qualitative
data
–rapport
Conversational Interviews
• Unstructured Interviews
• Advantages
– Interviewed without knowing they are in
a study
• ethical implications/considerations
– Participant feel relaxed
• respond freely to a specific set of questions
Conversational Interviews
• Disadvantages
– Not standardized
– participant can easily take control
– researcher needs proper training or experience
• But…
– This method enables a good amount of
qualitative results
Structured Interviews
• Patterned interview
– Very straight forward
• The interviewer has a standard set of
questions that are asked of all
candidates
• Advantages
– Analysis:
• Easier to compare participants
• less time consuming
Structured Interviews
• Disadvantages
– No in depth qualitative data
• limited to a structured response
• Doesn’t engage participants on a
personal level
– Little if any rapport
• participants respond to structured
questions
Semi- Structured Interviews
• One to one
• Aims to obtain feed back or explore an issue
• cover a range of pre-set topics
– expecting for the interviewee to answer freely
• Advantages
– Answers many of the criticism of structured interviews
– Allows them to consider feelings and concerns
• Disadvantages
– Limits on what is asked
– Not fully conversational
– Allows for subjectivity and bias
Semi- Structured Interviews
• involves the interviewees feelings
• doesn’t limit the range of possible
answers
• strong approach to qualitative
research methodology!
Transcribing
• Interview is recorded electronically or on
tape
– and transcribed focusing only on the words
• Advantages
– Quicker and easier than post modern
• Disadvantages
– Doesn’t consider non verbal cues
• such a tone of voice, pauses, and speech rate.
Transcribing
• This method is not one of the more
effective because it doesn’t consider
the emotions of the participant
– Needed for in-depth qualitative research
Post-Modern Transcription
• The researcher decides beforehand what
non-verbal cues he is going to record
– considers them later when transcribing
• Advantages
– Allows the full interview experience to be
researched rather than just the words
– Produces rich data and highly detailed analysis
Post-Modern (continued)
• Disadvantages
– Difficult to maintain reliability across researchers
– Time consuming
– Transcribing needs to occur right after the interview
• researchers perceptions are still fresh
– Distracts participants
– Every participant responds differently
• i.e. avoiding eye contact or talking too fast or using sarcasm
– Observable behavior is not always obvious
– Observable behavior does not imply the same in all
Verbal Protocol
• A record of what people say when they are asked
to think aloud as they perform a task
– i.e driving a police car in heavy traffic
• Advantages
– Very useful to describe what they are doing
– Described before it is forgotten
– On the spot analysis which provides information that in
an interview may be otherwise unobtainable
• Disadvantages
–
–
–
–
May not include vital information
i.e. embarrassment
They need training and practice
Multitasking may be conflicting
Questionnaires/Surveys
• Small Scale
– 200-300 people
• Advantages
– Straightforward
– Generalizations can be made
– Large amount of data and efficient
– Easiest way of collecting data for large groups
– Allows anonymity
Questionnaires/Surveys
• Disadvantages
– Affected by characteristics of respondents
• Memory, knowledge, experience, motivation, personality
– Respondents may not report accurately
– Low response rate
– Misunderstanding of the survey
– May not take it seriously
– Respondents may consider anonymity as an excuse to be less open
– This form of methodology may yield qualitative responses yet this
is limited by the content of what exactly is on that questionnaire or
survey
Questionnaires/Surveys
• Large Scale
– This takes a lot of time and processing
– They are useful and encompass a large number
of people and their views
– The same advantages and disadvantages apply
as in small scale
Likert Scale 1932
• This methodology is used by psychologist
to measure attitude
– Advantages
• Provides a universally known rating scale
– Disadvantages
• Cumbersome and time consuming
• This method is capable of yielding a good
amount of qualitative results
Participant Observation
• This means you have become a member of the
observed group
• Participant observation and qualitative techniques
can be scientific if the observer reports fully their
technique, this can also be applied to a variety of
other situations
– Advantages
• Allows researcher to blend in with the participants
• Allows researcher to subjectively experience what the
participants are experiencing
– Disadvantages
• Observer involvement may influence the group behavior
• Different observers experience and interpret things differently
(i.e. A football game)
Participant Observation
• This method give a huge amount of access
to qualitative results and techniques. This is
so because the observer blends in with the
group and experiences the exact same
things that the participants are experiencing
Non-Participant Observation
• To avoid the accusation of the researcher
influence on a group, one observes but does
not participate, this can be anything such as
a student taking notes in the library or an
audience at the opera
Non-Participant Observation
• Advantages
– Doesn’t impact behavior, yet still blends in
• Disadvantages
– The researcher does not get the subjective
experience as they are not “involved”
– Not participating may be odd to others and
depending on the situation it may be dangerous
to the researcher
About Psychology Methodology
• There are three major perspectives in
psychology research, they are:
– The nomothetic approach
– The ideographic approach
– The hermeneutic approach
Nomothetic Approach
• This approach is concerned with finding general
laws about human behavior and generalize this to
other humans in order to make predictions about
their behavior in given circumstances.
Psychologist involved in this type of research use
statistical methods in order to find the average in
variations within human behavior. They use
random and large samples and experiment is the
preferred method. The cognitive and behaviorist
perspectives are in favor of this approach
Ideographic Approach
• This is concerned with exploring uniqueness
– In other words, what makes a person an individual
(qualitative)
• More in depth studies
– Long term, flexible, and detailed
• Prefers to use case studies
• More general understandings
• Psychodynamic and humanistic approach on favor
of this
– Freud and Rogers
Hermeneutic Approach
• This is concerned with meaning
• Meaning occurs on a number of levels
– Conscious, unconscious, personal, cultural and
socio-political
• Researchers investigate how people
interpret their experience and how various
forms of symbolism are used to convey
meaning in human life
Quantitative vs. Qualitative
• The three different perspectives in psychology
have come about because methodology in
psychology has evolved
• In the earlier days of psychology the focus was
almost exclusively on the use of qualitative
methods
• Qualitative methods are mostly used by
behaviorist and cognitive psychologists
• Behaviorist claimed that anything that could not
be observed directly and objectively could not be
open to scientific investigation
– The mind thus wasn’t examined
– Favored method:controlled lab experiment
Quantitative vs. Qualitative
• Ethical issues in psychology helped show
participants not as material yet human beings
• It was recognized the qualitative methods were
often more suited to real world problems then
quantitative approaches, and this was later
recognized as a valid approach to data analysis
• Modern psychology is reaching a situation where
it is understood that both quantitative and
qualitative methods can be used in psychological
research