Chapter 2 - Research Methods Flashcards
Why we need good research methods?
-protect against bias to ensure results are accurate and conclusions are true
-avoid errors
Prefrontal Lobotomy
-surgical procedure that severs fibres connecting the frontal lobe of the brain with the thalamus
-used to treat schizophrenia
-results based only on subjective clinical results
-didn’t conduct systematic research and assumed their observations were enough to verify their results
System 1 Thinking
-intuitive
-fast
-relies on feelings and gut reactions
-may be an assumption
-relies on heuristics
-prone to error
Heuristic
-mental shortcut or rule of thumb that helps us to streamline our thinking and make sense of our world
-based on past history
System 2 Thinking
-analytical
-slow
-relies on careful evaluation of evidence
-reflective
Scientific Method
-not a singular way of doing things
-it is a toolbox of skills that can be applied in specific ways
-permit us to test hypotheses
Surveys
-used to measure peoples opinions and attitudes
-a self-report measure
Questionnaire
-a self-report measure
-assess personality traits, mental illnesses, interests
Self-Report Measure Advantages
-easy to administer and gather large amount of data
-cost effective
-allows assessment of internal processes, thoughts, feelings that outside observers are not typically aware
Self-Report Measure Disadvantages
-the wording of the question can lead to different results
-assumes respondents have enough knowledge to report accurately
-assumes participants are honest
Response Sets
-tendency of research participants to distort their responses to questionnaire items
Malingering
-tendency to make ourselves appear psychologically disturbed
-aim of achieving a clear-cut personal goal
-may be trying to obtain financial compensation or escape military duty
Rating Data
-a self-report measure
-someone else is asked to comment on a person’s behaviour
-it is assumed they know the person well
Rating Data Advantages
-gets around malingering and response set bias
Rating Data Disadvantages
-halo effect
-horns effect
-susceptible to stereotypes
Halo Effect
the tendency for a high rating in one positive characteristic to spill over and enhance the ratings of other characteristics
Horns Effect
the tendency for a high rating in one negative characteristic to spill over and lower ratings of other characteristics
Random Selection
-procedure that ensures that every person in a population has an equal chance of being chosen to participate
-allows us to generalize our results
-studying fewer people broadly is better than studying more people narrowly
Generalizability
-a measure of how useful the results of a study are for a broader group of people or situations
-ie. if the study results are applicable to many people or situations = good generalizability
Reliability
-an evaluating measure
-the consistency of a measurement
-the study will measure the same way over and over again
Test-Retest Reliability
-reliable questionnaire yields same scores over time
-consistent results
-ie. stepping on a scale and measuring the same weight multiple times
Interrater Reliability
-the extent to which different people who conduct an interview or make observations agree on the characteristics they are measuring
Validity
-the extent to which a measure assesses what it purports to measure
-ie. lie detector polygraph is not valid
Reliability + Validity
-reliability is necessary for validity
-validity is not necessary for reliability
Replicability Crisis
-open and transparent science is needed to ensure findings are replicable and reproducible
-sparked the open science movement
Post Data Publicly
-a response to the replicability crisis
-data needs to be public to be replicated and peer reviewed
Conduct Replications
-a response to the replicability crisis
-replications of own and others work needs to be available
Preregister Research
-a response to the replicability crisis
-prevents forged numbers
-when you provide a blueprint it is unchangeable
Publish ALL Sound Science
-a response to the replicability crisis
-journals only publish fancy findings to gain interest
-publish all works including replications
-scientists want to be published and wont replicate if they know they won’t be selected
Less Emphasis on Single Study Findings
-a response to the replicability crisis
-more emphasis on replicated studies
3 Types of Research Methods
-descriptive
-correlational
-experimental
Naturalistic Observation
-descriptive research method
-observe behaviour naturally without trying to manipulate it
-ie. observing in the wild
Natural Observation Advantages
-high in external validity
-captures natural behaviour
Naturalistic Observation Disadvantages
-low internal validity
-doesn’t allow us to infer causation
-possible reactivity (people know they are being studied)
-possible observer bias
-no control over other variables
External Validity
-extent to which we can generalize findings to real-world settings
Internal Validity
-extent to which we can draw cause-and-effect inferences from a study
Case Studies
-descriptive research method
-an in depth analysis on an individual, group, or event over an extended time period
Existence Proof
-demonstration that a given psychological phenomenon can occur
Case Study Advantages
-can provide existence proofs
-allows study of rare phenomena
-good for hypotheses generation
-offers insights for later testing
Case Study Disadvantages
-typically anecdotal
-don’t allow us to infer causation
-generalization may be an issue
-possible observer bias
Correlational Designs
-research design that examines the extent to which to variables are associated/related
-allow us to generate predictions about the future
Correlate
-two things that relate to each other statistically rather than interpersonally
Advantages of Correlational Designs
-flexible and easier to conduct than experiments
Disadvantages
-cannot explain causation
Positive Correlation
-as the value of one variable changes, the other goes in the same direction
- + = perfect positive correlation
Zero Correlation
-the variables don’t go together at all
Negative Correlation
-as the value of one variable change, the other goes in the opposite direction
- -1.0 = perfect negative correlation
Third-Variable Problem
-the reason correlational research doesn’t show causation
-while x and y might show correlation, there could be a third z variable that affects the two
Correlation Coefficient
-measures the strength of a correlation
-scale from -1.0 (negative) to + 1.0 (positive)
Scatterplot
-grouping of point on a two-dimensional graph (x, y) in which each dot represents a single person’s data
Scatterplot Negative Correlation
-the plot goes in a downward slope
-if the correlation isn’t a perfect negative (-1.0) that means that some students still did well and some students who drank non still did poorly
-ie. shows the more beers drank the worse students do on an exam
Scatterplot Zero Correlation
-no definite pointing up or down
-no association
Scatterplot Positive Correlation
-dots point in an upward slope
-ie. more psych classes attended, the better they do on their exam
Illusory Correlation
-perception of a statistical association between two variables where none exists
Experimental Design
-a research design characterized by random assignment of participants to conditions
-involves the manipulation of at least one variable
-allows researchers to establish as causal relationship between variables
Components of an Experiment
- Random assignment of a participants to conditions
- Manipulation of an independent variable
Random Assignment
-randomly sorting participants into groups
-cancels out preexisting differences between the two groups
-ensures the participant has an equal chance of being sorted into one of the two groups
Experimental Group
-in an experiment, the group of participants that receives the manipulation
Control Group
-in an experiment, the group of participants that doesn’t receive the manipulation
Independent Variable
-variable that an experimenter manipulates
Dependent Variable
-variable that an experimenter measures to see whether the manipulation has an effect
Between-subjects Design
-in an experiment, researchers assign different groups to the control or experimental condition
-group A gets drug, group B doesn’t
Within-subjects Design
-in an experiment, each participant acts as his or her own control
-they measure a behaviour before a variable is manipulated, and then after
Operational Definition
-a working definition of what a researcher is measuring
Confounds (extraneous/confounding variables)
-any variable that differs between experimental and control group
-may be responsible for the observed difference between the groups after manipulation
-researcher trie to control things that could affect the results that weren’t accounted for
-ensures there is only one manipulated variable
-ie. was the result do to the drug or financial status
Experimental Design Example
-Hypothesis: drug will increase a participants mood
-Independent variable: whether a subject gets the drug
-Dependent variable: mood of the participant
Placebo Effect
-improvement resulting from the mere expectation of improvement
Blind
-a way to control the placebo effect
-unaware of whether one is in the experimental or control group
Nocebo Effect
-harm resulting from the mere expectation of harm
-ie. voodoo dolls
Experimenter Expectancy Effect
-phenomenon in which researchers hypotheses lead them to unintentionally bias the outcome of a study
-usually in line with their expected hypothesis
-driven by confirmation bias
Double-Blind
-a way to protect against experimenter expectancy effect
-neither researcher nor participants are aware of who’s in the experimental or control group
Demand Characteristics
-cues that participants pick up from a study that allow them to generate guesses regarding the researchers hypotheses
-changes how they act based on their assumptions
Preventing Demand Characteristics
-researchers often disguise the true purpose of a study until after the study has been completed
Hawthorne Effect
-people’s knowledge that they are being studied changes their behaviour
Tuskegee Study
-an unethical study
-for 40 years (from 1932-1972) the US Public Health Service diagnosed black men in a poor community in Alabama with Syphilis
-They had no idea they had it and infected women and children
-Eventually died
-Treatment was available at the time
Belmont Report
-1979
-in response to Tuskegee study
-made guidelines about ethical research
Findings of Belmont Report
Research should be:
-allow people to make decisions about themselves
-be beneficent
-distribute benefits and risks equally to all participants
Informed Consent
-informing research participants of what is involved in a study before asking them to participate
Educating Participants
-protect them from harm
-know what they are getting into
-non-technical language
Freedom from Coercion
-you can’t bribe/coerce people to participate in a study
-especially if it’s harmful
Risk-benefit analysis
-determine the risks and benefits
-is it worth it?
-are the risks too high?
Statistics
-application of mathematic to describing and analyzing data
Descriptive Statistics
-numerical characterization that describe data
Central Tendency
-measure of the “central” scores in a data set
-where the group tends to cluster
Variability
-measure of how loosely or tightly bunched scores are
Mean
-average
-add up all numbers, divide by how many there are
Median
-middle score in a data set
-measure of central tendency
Mode
-most frequent score in a data set
-measure of central tendency
Range
-a measure of variability
-difference between the highest and lowest scores
Standard Deviation
-measure of variability
-average amount that an individual data point differs from the mean
Inferential Statistics
-mathematical methods that allow us to determine whether we can generalize findings from our sample to the full population
-allows to determine if results are likely to occur due to chance
Statistical Significance
-the probability that the findings are due to chance
-if they are statistically significant that means the results are unlikely to occur due to chance
Practical Significance
-determination of whether the finding has any real world importance
Peer Review
-process of quality control for research before it is published in an academic journal
-reviewers identify flaws that undermine the findings of a study
