Myers Modules 2,3 =
Module 2: Research Strategies: How Psychologists Ask and Answer Questions
How Everyday Thinking Can Lead to Wrong Conclusions
Common Sense Misconceptions:
One’s common sense, based on numerous casual observations, can sometimes be incorrect.
Casual observations may not always provide accurate information (leading to flawed conclusions)
Familiarity Breeds Belief:
Statements often seem like common-sense truths because they're frequently repeated.
Mere repetition, regardless of accuracy, makes information easier to process and remember, leading to a perception that its true
Scientific Impact of Easy-to-Remember Misconceptions:
Easily memorable misconceptions, like "Vitamin C prevents the common cold," can overshadow scientifically proven facts.
These misconceptions, due to their simplicity and repetition, may gain more traction despite lacking empirical evidence.
What are the three common flaws in common sense thinking
Hindsight bias
Overconfidence
Perceiving the order for random events
HOWEVER scientific inquiry can help us sift reality from illusion.
What is Hindsight Bias
known as the "I-knew-it-all-along phenomenon” (an illusion in a way)
the tendency to believe that the outcome of a situation was more foreseeable or obvious than it truly was
Such errors in people’s recollections and explanations show why we need psychological research
common sense describes, after the fact, better than it predicts what will happen (make prediction afterward)
What is Overconfidence
We humans tend to think we know more than we do (tend to be more confident than correct)
“We are not good at predicting social behavior”: Psychologist Philip Tetlock collected more than 27,000 expert predictions of world events. Findings: Predictions made with 80 percent confidence were right less than 40%
What are “superforecasters.”? (what do they avoid?)
Only about 2 percent of people can majorly predict social behavior
Phillip Tetlock calls them “superforecasters.”
Superforecasters avoid overconfidence.
Faced with a difficult prediction, they gather facts, balance arguments, and settle on an answer.
Perceiving Order in Random Events
We all have an eagerness to make sense of our world (an unpredictable world is unsettling)
Even in random data, we often find patterns, because—here’s a curious fact of life—random sequences often don’t look random.
Example: Hot and cold streaks in basketball look like a pattern, it really isn’t (can’t tell which next one will be a miss or hit)
Post Truth World Definition (Facts vs Beliefs)
describing a modern culture where people’s emotions and personal beliefs often override their acceptance of objective facts.
opinion overrides fact
feeling overides reason
bias overrides evidence
Political Bias and Challenges in Perceiving Facts
Psychologist Peter Ditto discovered partisan bias in both liberals and conservatives, (nearly identical levels)
Political biases, evident across party lines,
Both Democrats and Republicans share concerns about the struggle to discern fact from fiction.
Political biases distort perceptions and hinder the ability to distinguish factual information, raising concerns about the foundation of democracy.
How can we build a real-truth world, and get rid of this post truth world?
by embracing a scientific mindset or scientific method
we must adopt the spirit of critical thinking
To accept everything is to be gullible; to deny everything is to be a cynic
The Scientific Method (known as what type of process, what are the 4 main steps)
needs to combine curiosity, skepticism, and humility.
a self-correcting process for evaluating ideas with observation and analysis.
welcomes new theories
Hypothesis is made
Puts theory to the test (research and observation)
Confirm, Reject or Advise (Researchers submit their work to a scientific journal. Peer reviewers provide evaluations. Journal editor decides whether the research deserves publication)
How do theories advance psychological science?
a theory explains behaviours or events by offering ideas that organize observations.
theory summarizes and simplifies.
principle that neatly summarizes a list of observations about the reaction
Our theories can bias our observations. How can one check on their own biases?
Psychologists ensure the reliability of their research by providing clear and precise descriptions of their procedures and concepts. This allows others to replicate their experiments and confirm their findings. Replication, or repeating experiments to get similar results, is crucial for confirming the validity of scientific findings. To improve reliability, researchers often use larger sample sizes, increasing the likelihood of replication. Additionally, psychologists are increasingly using preregistration, where they publicly share their study design, hypotheses, and planned analyses before conducting the research. This transparency prevents any later modifications to fit the data, ensuring the integrity of the scientific process.
Meta-Analysis
Psychological and medical science also harness the power of meta-analysis. Meta-analysis is a procedure for statistically synthesizing a body of scientific evidence. By combining the results of many studies, researchers avoid the problem of small samples and arrive at a bottom-line result.
A theory will be useful if it … (5)
1) organizes observations
2) implies predictions that anyone can use to check the theory or to derive practical applications.
3) stimulate further research (leads to a revised theory, its generative, same with hypothesis)
4) Must be falsifiable (same with hypothesis)
5) More simple (parsimony)
the more complicated, the more likely the theory will be wrong and wont stand
parsimony = tells us to choose the simplest scientific explanation that fits the evidence
covers lots of domains (covers, explains all data)
Theory = general. Hypothesis = specific
How can one test a hypotheses and refine theories (the 3 methods for study)
Descriptive methods (describes behaviours, often through case studies, surveys, or naturalistic observations)
Correlational methods (associate different factors)
Experimental methods (manipulate factors to discover effects)
How do psychologists use case studies, naturalistic observations, and surveys to observe and describe behavior
All part of the Descriptive Method for study
starting point of any science is description.
Observe and describe people, often drawing conclusions about why they think, feel, and act as they do, more objectively and systematically,
case studies (in-depth analyses of individuals or groups).
naturalistic observations (recording the natural behavior of many individuals).
surveys and interviews (asking people questions).
Case Studies
the oldest research method
examines one individual or group in depth in the hope of revealing things true of us all.
To find general truths, we must employ other research methods.
Naturalistic Observation
records responses in natural environments
naturalistic observation does not explain behavior. It describes it.
offers interesting snapshots of everyday life, without controlling for all the factors that may influence behavior.
What are the advantages and disadvantages of naturalistic observation?
Naturalistic observation does not explain behavior, and it does not control the factors that may influence behavior (just describes it)
the descriptions it provides can be revealing, expanding our understanding and paving the way for future studies.
Surveys (what is the challenge here?)
A survey looks at many cases, asking people to report their behavior or opinions
asking questions is tricky (People may shade their answers in a socially desirable direction)
answers depend on how questions are worded and how respondents are chosen.
Even small changes in the order or wording of questions can make a big difference
critical thinkers will reflect on how the phrasing of a question might affect people’s expressed opinions
Random Sampling
How one obtains a representative sample
random sample = every person in the entire population has an equal chance of being included in the sample group.
Large representative samples are better than small ones (However, you cannot compensate for an unrepresentative sample by simply adding more people)
Political pollsters sample voters in national election surveys just this way.
What is an unrepresentative sample, and how do researchers avoid it?
An unrepresentative sample is a group that does not represent the entire population being studied.
Random sampling helps researchers form a representative sample, because each member of the population has an equal chance of being included
What does it mean when we say two things are correlated, and what are positive and negative correlations?
correlation = A statistical measure that helps us figure out how closely two things are associated. Helps us see the world more clearly (how well either one predicts the other)
Perfect correlations rarely occur in the real world.
Positive correlation = if two sets of scores, tend to rise or fall together (r=+1.00)
Negative correlation = if two sets of scores relate inversely, one set going up as the other goes down (r=-1.00)
No relationship: r=0.00
Correlations and Scatter Plots (when it used)
When comparing things people might not detect a relationship them. The correlation can be clear if we display the data as a scatterplot. Each dot in a scatterplot represents the values of two variables.
We cant recognize a strong relationship when data are presented systematically, showing the need for statistical illumination.
What are illusory correlations
When we believe there is a relationship between two things, we are likely to notice and recall instances that confirm our belief.
can feed an illusion of control (chance events are subject to our personal control)
Ex: Gamblers (believe they can influence the roll of the dice)
What is regression toward the mean?
Average results are more typical than extreme results
extraordinary happenings tend to be followed by more ordinary ones
Failure to recognize regression is the source of many superstitions and of some ineffective practices
regression toward the average can mislead us into feeling rewarded after criticizing others
When a fluctuating behavior returns to normal, fancy explanations for why it does so are often wrong. Its just regression toward the mean
What are the characteristics of experimentation that make it possible to isolate cause and effect?
create an experimental group (receive the treatment) and a control group (they do not)
To minimize any preexisting differences between the two groups, experimenters randomly assign people to each condition. (equalizes the two groups)
Experiments enable researchers to isolate the effects
What measures do researchers use to prevent the placebo effect from confusing their results? (2)
randomly assigns participants to an experimental group (which receives the real treatment) or to a control group (which receives a placebo)
A double-blind procedure prevents people’s beliefs and hopes from affecting the results, because neither the participants nor those collecting the data know who receives the placebo
VARIABLES definitions
Experimental factor or independent variable
This is the variable that is manipulated by the researcher and can vary independently of other factors. It is indeed a fundamental aspect of experimental design.
Confounding variables
These are indeed other factors that may potentially influence the results of the study, and controlling for them is essential to ensure the validity of the experiment. Random assignment is one method used to control for confounding variables.
Dependent variable
This variable is indeed measured or observed to assess the effects of the independent variable. It represents the outcome of the experiment and can vary depending on the manipulation of the independent variable.
Variable
anything that can vary (within the bounds of what is feasible and ethical to measure).
How would you know which research design to use? What are they?
Descriptive
observe and record behaviour
case studies, naturalistic observations, surveys
nothing manipulated
no control of variables (can be misleading)
Correlational
naturally occurring relationships (how one predicts others)
collect data from variables (no manipulation)
cant specify cause and effect, just showcase relationships
Experimental
explores cause and effect
manipulate one or more factors (independent variables)
use random assignment
may not generalize to other contexts
may not be ethical to manipulate certain variables
Are Off-the-top-of-the-head estimates reliable?
NO, Off-the-top-of-the-head estimates often misread reality and mislead
big, round numbers can easily be echoed and thus becomes public misinformation
Use critical thinking when presented with big, round, undocumented numbers.
Measures of Central Tendency (what does it help with)
helps to know about amount of variation in the data—how similar or diverse the scores are
Mean
most familiar / arithmetic average (total sum of scores divided by number of scores)
Mode
the most frequently occurring score(s) in a distribution
Median
midpoint/middle score (the 50th percentile)
What is the most useful standard for measuring how much scores deviate (differ) from one another?
it is the standard deviation
a computed measure of how much scores vary around the mean (average) score.
Normal Curve (whats the shape, most cases fall near the what?)
symmetrical, bell-shaped curve
normal distribution = Most cases fall near the mean (68 percent fall within one standard deviation) and fewer cases fall near extreme
Mostly Large numbers of data (Ex: height, intelligence score, aptitude tests, life expectancy, not income)
Correlation ≠ Causation (why or why not)
"Correlation does not imply causation"
The 2 types for experimenting on theories
Correlation means that two variables have a statistical relationship (association)
Causation means that one variable causes the other (controlled, independent, dependent)
Elements of Correlation (4)
Third Variables: (mystery variable not recorded)
The correlation between two variables might be influenced by a third variable that affects both. This is known as a confounding variable.
Direction and Strength
positive or negative
No Variables
Correlation does not provide information on variables
Calculates P value (probability value)
range 0 - 1
measures difference or variability
above = results not significant
below = results are significant
What are P values (Proof needs a P value score of what?)
Many psychological tests provide p-values
indicate the probability of the null hypothesis given the sample data
For most psychologists, proof beyond a reasonable doubt means the (p-value) of the null hypothesis is less than 5 percent (p < .05)
Scepticism
stance thats its almost impossible to prove anything, nothing is true
DOES NOT mean that they are just questioning or suspicious
cannot fully prove something (can only collect more evidence for it)
just haven’t disproven it yet
Prove means not falsifiable which means in itself that it can’t be scientific since falsifiable data is needed for scientific research
Habituation (adaption) VS Sensitization
Habituation (adaption)
Threshold changes — As habituation occurs, the threshold for responding to the stimulus increases.
sensory adaption takes place = responds less
decrease in response to a repeated or continuous stimulus over time.
Sensitization
opposite of habituation
an increase in responsiveness or sensitivity to a stimulus over time.
takes less stimulation to produce a response.
Increased Sensitivity