Psych Research Methods exam 1 Flashcards
Three methods we answer questions
Intuition
Authority
Science
Method of Intuition
Draws conclusions on the basis of personal experience and judgment
Method of authority
adopts conclusions of trusted sources such as expert, friend, doctrine
Method of science
Draws conclusions on the basis of systematic observation
Why statistics are needed
Helps us make analyze (makes sense of) lots of information
Allows us to make conclusions about our selected population based on data from a small portion of it
Four goals for understanding behavior
- Describing behavior
- Predicting behavior
- Explaining behavior: Causes of behavior and why it works
- Influencing behavior: Control or change
Research Process
- Generate Research Question by observing the world
- Consult Theory
- Generate Hypothesis
- Generate Prediction by identifying variables
- Collect Data to Test Prediction by measuring variables
- Analyze Data by graph data/model
- Generalize results
Mertonian norms
Four norms of good scientific research were introduced by Robert K. Merton.
Describe what constitutes the ethos of modern science
Communism, Universalism, disinterestedness, and organized skepticism
Communism
Expresses the shared ownership of scientific discoveries and the need for scientists to share their discoveries publicly
Universalism
Idea anyone can do science regardless of race, nationality and gender
Everyone’s scientific claims should be scrutinized equally without discrimination
Disinterestedness
Expresses the idea that scientists should only work for the benefit of science
Organized scepticism
Expresses the idea that the acceptance of scientific work depends on assessments of the scientific contribution, objectivity and rigor
National Research Act
Passed in 1974
Established a commission to regulate human-subjects research
Required all research institutions that receive federal funding to establish Institutional Review Boards (IRBs) to supervise human-subject research
Belmont Report
Passed in 1979
Establishes ethical guidelines for human-subjects research
Guidelines integrated into the APA ethics code
The 3 principles of the Belmont Report is the
Beneficence
Justice
Respect for Persons
Institutional Review Boards (IRB)
Committees that evaluate the ethics of all research conducted at an institution
Researchers must obtain permission before beginning any study-even one with no risk to participants
Three key principles of Belmont Report
Beneficence
Justice
Respect for Persons
Justice
Benefits and risks should be distributed fairly
It is unfair when an individual or group carry a large share of risks of research without getting a proportionate share of the benefits
Formulations
(1) to each person an equal share
(2) to each person according to individual need
(3) to each person according to individual effort
(4) to each person according to societal contribution
(5) to each person according to merit
Beneficence
Persons are to be treated in a ethical manner not only be respecting their decisions and protecting them from harm, but also by making effort to secure well-being
- Do not harm
- Maximize benefits and minimize risks
Has the interests of research participants in mind
Respect for Persons
Should be treated as autonomous agents
Participants without full autonomy should be protected
Requires participants to be allowed to choose what shall or shall not happen to them. They should receive informed consent
Benefits of beneficence
Direct payment or other compensation
Education or learning
treatment or intervention benefits
Application of findings
Risks of beneficence
Physical discomfort or pain
Psychological stress
Loss of privacy
Confidential data in research
research can connect identities with data, but the public cannot
Beneficence: Researchers Responsibility
Assure no permanent physical damage or psychological harm
Protect privacy through anonymity or confidentiality
Make sure benefits outweigh risks
Repair any actual harm
Anonymous data in research
When researchers cannot connect participants identities with their data
Respect for persons: Informed consent
To fulfill Respect for Persons, obtain informed consent by providing all relevant information to each participant before they agree to participate
Relevant information includes:
General purpose, procedure, duration
Risks, benefits, and incentives
Assurance and limits of confidentiality
Right to decline or withdraw at any time
Address any immediate questions or concern
Researchers contact information
Justice: Participant selection
Should be guided by scientific consideration and concern for fairness
Should not concentrate on benefits on advantaged populations or risks on disadvantaged populations
Protected populations in research
Children and minors
People with psychological disorders or mental handicaps
Students and employees/subordinates
Prisoners
What to be mindful of when using protected populations in studies
- Be careful about risks
- If the participant is not fully capable of consent, obtain their consent, plus consent from a parent, guardian, or spouse
- Be careful to avoid coercion
Exemptions from Informed Consent
IRB can decide that studies don’t require informed consent if
- No more than minimal risks to participants
- Educational practice or assessment
- Job or organizational effectiveness
- Anonymous surveys, naturalistic observations, or archival research
Debriefing
Providing additional information to each participant after they complete study
Specific purpose, expected results
Address questions, concerns
Repair any damage psychologically or physically
How to learn about the results or conclusions
Deception
Involves withholding some information about a study’s purpose or procedure, or misleading participants
Throughly debrief participants, especially after deception
When is deception considered acceptable
Researchers fully disclose any greater-than-minimal risks
Researcher debriefs participants after deception
There is not a feasible alternative
Adam Frank’s 3 S’s of science
Spitballs
Supertankers
Stadiums
Spitballs
Individual research papers are spitballs
It basically means they’re not the last word on anything
Supertanker
Basically is the state of the study, all the spitballs have to line up on the same side to stop the supertanker and create a new scientific consensus
If individual research studies are on different sides of the supertanker, no consensus is going to be reached
Stadium
It refers to the community of people who interpret, process and decide on directions of the general scientific consensus and its studies that lead up to it
Peer review articles
Articles are written by experts and are reviewed by several other experts in the field before the article is published in the journal in order to ensure the article’s quality.
These are the most scientifically valid articles
Timely peer reviewed article
Addresses current or pressing issue in fields, provides up to date information or research findings significant to the scientific community
Time bound article
schedule or timeline associated with the publication process and ensures that the article can be published at a reasonable time-frame
Levels of measurement
1.Categorical
2. Continuous
Categorical variable and types
Entities are divided into distinct categories
Binary variable
Nominal variable
Ordinal variable
Binary variable
Only two categories
e.g. dead or alive
Nominal variable
More than two categories
e.g. whether someone is an omnivore, vegetarian, vegan, fruitarian
Order does not matter
Ordinal variable
The same as a nominal variable but categories have logical order
e.g. whether people got a fail, a pass, a merit, or distinction on their exam
Order matters but not difference between values
Continuous variables
entities get a distinct score
Scale variables
Scale variables
a measurement variable — a variable that has a numeric value
Broken down into ratio variables and interval variables
Ratio variables
Have an absolute, true “zero” point
e.g. height weight
Makes it possible to compare measurements in terms of ratio
e.g had 2 cups of cereal while my wife had 1 cup
Interval variables
The value of “zero” does not indicate a total absence of the variable being measured
Difference between values are meaningful
Cohort study
a type of longitudinal study—an approach that follows research participants over a period of time (often many years)
Specifically recruits and follows participants who share similar characteristics like occupation or demography
Case-control studies
A study that compares two groups of people: those with the disease or condition under study (cases) and a very similar group of people who do not have the disease or condition (controls).
Relationships between variables
Describing relationships
Using relationships to predict behavior
Testing causal relationships
Using relationships to influence behavior
research designs used to test relationships between variables
- Experimental designs
- Correlational designs
Experimental design
Researcher directly manipulates one or more independent variables than observes effects on one or more dependent variables
Ex. Anxiety and test performance
Control group
Group of participants in the experimental design that is not manipulated
manipulation check
any means by which an experimenter evaluates the efficacy of an experimental variable,
verifies that the experimental variable is done as presented
Correlational design
Researcher measures two or more variables, but does not directly manipulate any of them
ex. Anxiety and test performance
Anxiety is the predictor variable
Test performance is the outcome variable
Casual relationship
Implies direct cause and effect relationship between the independent and dependent variable
The requirements are
Covariation
Temporal precedence
Alternative explanations
Easier for experimental studies to meet these requirements
Covariation
a statistical association between the cause and the effect.
Temporal precedence
Cause must occur before the effect
Alternative explanations
All plausible alternative explanations for covariation must be eliminated
Correlation Coefficient
Varies between -1 and +1
0 = no relationship
+-.1 = small effect
+- .3 = medium effect
+-.5 = large effect
in social studies
Confounding variable
Provides alternative explanation for a possible casual relationship
Benefits of correlational studies
- Easier to do
- Easier to observe behavior naturalistically, rather than needing to use an artificial lab setting
- Can study variables that are impossible or unethical to manipulate
What makes a good research study
High construct validity
High external validity
High internal validity
Construct validity
assesses whether the variables that you are testing for behave in a way to support your theory
Pertains to the measure that you are using
External validity
This means that the results of the study can be generalized to other people, settings, and operational definitions of the variables
Easier for correlational than experimental studies
Internal validity
Internal validity is the extent to which a research study establishes a trustworthy cause-and-effect relationship
This type of validity depends largely on the study’s procedures and how rigorously it is performed.
Usually easier for experimental than correlational
Ways to maximize construct validity
Use measures and manipulations that have been validated by previous research
How to ensure external validity
Use representative sample
Observe behavior in natural settings
Use realistic manipulations
Conduct multiple studies with different methods
John Stuart Mill
Created the 3 ways to identify a successful casual experiment
Covariation
Temporal Precedence
Alternative explantations
Hawthorne effect
people will modify their behavior simply because they are being observed.
Operational definition
The way a construct is measured in a study
ex. If you are measuring the absolute income hypothesis for money, an operational variable you’d be measuring is happiness
Qualitative Data
Information that cannot be counted, measured or easily expressed using numbers.
Quantitative data
data that can be counted or measured in numerical values
Operationalization
The way that a particular variable is measured or manipulated in a particular study
Two ways to quantify data
Categorical data
Continuous data
Internal validity in experiments
Used large samples for equal groups
Treat all groups identically except for the independent variable
Don’t communicate your hypothesis to the participants
What study design is high in internal validity
experimental studies
Robert K. Merton.
Created the 4 Mertonian Norms
