Chapter 1: Outline

Need for Psychological Science

Hindsight bias and judgmental overconfidence show that we cannot rely on intuition and
common sense Critical thin!ing must also be employed in order to perceive facts from
nonsense
Hind sight "ias
#he thought that once a person find out the outcome$ that the person !new the outcome
all along and could have predicted it
Proves that we need psychological research
Common sense describes what has happened more easily than predict what will happen
Overconfidence
#hin!ing is limited
%ven when you are 1&&' sure about something$ self prediction may change up to 1(' of
the time
)hen prediction becomes wrong$ individual*s attempt the +, was close* e-cuse
Often leads to the overestimation of our potential
S!epticism and humility must be added to come bac! to reality
Scientific .ttitude
Sometimes refutes s!eptics
"eing s!eptical but not cynical$ open but not gullible
S!eptical testing could separate the real facts
/e0uires s!epticism and humility
Need to reject our own ideas
Copernicus and Newton$ are e-amples of people who used the scientific attitude
Critical #hin!ing e-amines assumptions$ discerns hidden values$ evaluates evidence and
assess*s conclusion without blindly accepting arguments and conclusions

Scientific 1ethod
Scientific #heory e-plains through principles that organi2es and predicts behaviors or
events
"y organi2ing isolated facts$ theory simplifies things
Connection of observed dots$ we may discover a pattern
Hypothesis is testable prediction$ often started by a theory
/esearch allows one to test or reject the theory
Operational 3efinition is a statement of the procedures used to define research variables
/eplication when a procedure could be repeated with different participants in different
situations to see whether the same result is found
. theory is successful if it lin!s and organi2es observed facts and if it implies hypothesis
that offer testable predictions and practical applications
Case Study
.n observation techni0ue in which one person is studied in depth in hope of revealing
universal principles
Case studies can suggest hypotheses for further study
4nrepresentative information can lead to mista!en judgments and false conclusions
.necdotal stories may elevate the truth
3oes not wor! to find the general truths that cover individual cases
Survey
5oo!s at many cases in less depth
.s!s individuals about their behaviors and opinions
.nswers depend on wording
1ay use random sampling
/andom Sampling
6alse consensus effect is the tendency to overestimate the e-tent to which others share
our beliefs and behaviors
Populations includes all the cases in a group from which samples may be drawn from a
study
/andom sampling represents a populations because each member has an e0ual chance of
inclusion
7ery large samples may be more reliable
"asis of generali2ing is from representative sample of cases
Naturalistic Observation
Observing and recording behavior in naturally occurring situations without trying to
manipulate and control the situation
3oes not e-plain behavior$ it describes it
/obert 5evine and .ra Noren2ayan 8 compared pace of life
Can be used with correlation research
Correlation
)hen two traits seem to accompany each other$ they correlate
Correlation coefficient is a statistical measure of a relationship
/eveals how closely two things vary together and thus how well wither one predicts the
other
Scatter plots show how closely related the traits are associated with
%ach point plots the value of the two variables
Positive correlation means that two variables seem to rise or fall together
. negative correlation could mean inverse as well as negative relationships
,nverse means that while one variable is increasing$ the other is decreasing
. wea! correlation means that there is little or no relationship9 the coefficient is then 2ero
Correlation indicates the possibility of a cause9effect relationship but does not prove the
cause
,llusory Correlations
. perceived non e-istent relationship
%-plains superstitions$ beliefs
1ore li!ely to notice and remember the occurrence of two events in se0uence
Can be easily deceived by estimating that there is a correlation amidst random events
Perceiving order in random events
/andom se0uences often don*t loo! random
%-ample9 6lipping coins 8 (&' of getting each side
Strea!s may be found in any random data
Patterns or se0uences occur naturally in random data
%-ploring Cause and %ffect
%-periment is a research method in which an investigator manipulates one or more
factors to observe the effect on some behavior or mental process "y random assignment
of participants$ the e-perimenter aims to control other relevant factors
,ndependent 7ariable is being manipulated
3ependent 7ariable is the effect being observed
%valuating #herapies
3ouble blind procedures9 when an e-periment is being performed without the !nowledge
of either the participant nor the administer
Placebo %ffect is researching the effect of e-pectations$ could be used to see whether a
medicine is needed or not
%-perimental Condition is the condition of an e-periment that e-poses participants to the
treatment
Control condition is the condition of an e-periment that contrasts with the e-perimental
condition and serves as a comparison for evaluating the effect of the treatment
/andom assignment is assigning participants to by chance$ minimi2es e-isting
differences
,ndependent and 3ependent 7ariables
#he ,ndependent 7ariable is the value being manipulated within a given e-periment
#he dependent variable is the effect produced by manipulating the independent
variable
. variable is anything that can vary
/andom assignment is preferred so that all other variables could be e0ual and not
favor one thing
Statistical /easoning
"y throwing out a big round number$ one could easily mista!e reality 5i!e when we
compare percentages and what they actually mean .n e-ample given in the te-tboo! is9
:1&' of people are lesbians or gay men Or is it ; to <' as suggested by various
surveys=>
3escribing 3ata
One important thing while reading graphs is not to forget to read scale and figure out the
range #he same graph could loo! different if given different scales or a different range
1easures of Central #endency
. single score that represents a whole set of scores is also !nown as the measure of
central tendency
1ode is the most fre0uent number repeated within a given set of numbers
1edian is the midpoint
1ean is the average of the numbers ?ou can find this out by adding up all the given
numbers and then dividing it by the number of numbers you have #he mean could be
e-aggerated or diluted based on a few e-treme scores 5i!e for e-ample$ the mean of <$ @$
A and ;&1 is (<($ even though there are not many numbers above (<( since the
information is s!ewed
1easures of 7ariation
7ariation describes how similar or different scores could be )hen a person has low
variability$ we may predict his ne-t move$ since there is not much of a chance that he will
change his ne-t move
#he range is determined based on how far the lowest and highest scores are
#he standard of deviation is a measure of how much numbers may deviate from one
another ,t shows how much individual numbers vary from the mean ,t is more
informative than just the mean
)hen is an observed difference reliable=
#hree principles that shall ma!e an observed difference more reliable includes
1 /epresentative samples are better than biased samples
; 5ess9variable observations are more reliable that those that are more variable
Bconsistent scores are more reliable than those with more variationC
< )hen a study includes more cases$ it is more reliable than having a study with
fewer cases
)hen is a difference significant=
)hen there is a big difference among an average$ that difference is !nown as the
statistical significance ,t also means that this difference was not due to chance variations
and that the mean is reliable
Statistical difference also shows the +li!elihood* that the result will happen by chance
Can 5ibratory %-periments illuminate %veryday 5ife=
#he main idea is that +the resulting principles9not the specific findings will help everyday
behaviors*
3ouglas 1oo! claims that the e-periment*s purpose is not re9create the e-act behaviors
of everyday life but to test theoretical principles
3oes "ehavior 3epend on One*s Culture
Culture is the shared ides and behaviors that one generation passes on to the ne-t
,t does matter since it influences our perspective on certain topics
Heritance does play a role though$ since dysle-ia can be seen more in certain races than
others$ the variation of languages are influenced by our cultures and society and so does
the way we live our lives$ the choices that we ma!e
3oes "ehavior vary by gender=
Dender does play a role in behavior
)omen are more li!ely to build relationships and carry on a conversation and men are
more li!ely to relate through advice #his information shows us a way to communicate
differently with each gender ,t is a !nown fact that men are less li!ely to tal! about
personal feeling as much as woman$ such differences$ re0uire different approaches
)hy do Psychologists study animals=
Human physiology resembles that of many other animals
.nimal e-periments could lead to curesEvaccines for humans
,s it ethical to e-periment on animals=
/esearchers defend themselves by stating that the number of animals they use for
research is only 1' of the number of animals !illed for food
Psychologists for the %thical #reatment of animals want researchers to use naturalistic
observation while testing rather than libratory manipulation
%thics vary from culture to culture
Some say that the benefits out way the costs
Scott Plous noted that each individual has a list of animals they are more concerned
about
.P. advocates minimal pain and stress placed upon animals
%-perimenting on people
#he .P. states that in order to e-periment on people$ the following guidelines must be
followed:
1 Obtain the informed consent of potential participants
; protect them from harm and discomfort
< treat information about individual participants confidentially
@ fully e-plain the research after the e-periment
7alue of Fudgments
"ias definitely plays a role in how an individual may interpret results and study it
Plays a role in which goals we would li!e to reach
Psychology9 3angerous=
+power to deceive as well as enlighten*
.ddresses common problems$ meant to enlighten
The Scientific Attitude
Scientific approach that is s!eptical and open9minded
#o shift away from illusions to reality$ one must use Smart thin!ing or critical thin!ing:
thin!ing that does not blindly accept things$ but approaches with s!epticism and
e-amines the evidence carefullyG .s! how did they !now$ on guts and instinct= .re the
evidence biased=
However$ must remember to have humility as too e-treme would be stubbornness

The Limits of Intuition and Common Sense
,ntuition often ends up nowhere
#end to use a lot hindsight bias: tendency to believe that one would have !nown it after
the results are shownG
Seems li!e common senseG #he answer was right there and loo! how obvious it was
%-perience it usually when loo!ing bac! on historyG eg Dlen Clar! and the fast ferries
Humans tend to be overconfident$ thin! we !now more than we actually do Bprobably
result of self9serving biasC
Hindsight causes us to be overconfident as we believe we would have pic!ed the answer
when the results are in front of us

The Scientific Method
Scientific theory: e-planation using set of principles to organiseEpredict observations
No matter how good theory sounds$ must put it to test
1ust imply testable prediction H hypothesis
"eware of bias when testing
Dood e-periment can be replicated: the e-periment can be repeated and would yield
constant resultsG done with a different group of people or by a different person ending
with constant results
#heory useful if:
1 effectively organises range of observations
; implies clear predictions
Case study: research method where one person is studied in depth to find universal
principles Bthings that apply to allC
3rawbac! is that the individual being studied could be atypical$ results not universally
contained
Survey: research method to get the self9reported attitudesEbehaviours of people
5oo!s at cases less depth and wording of 0uestion affects the response given
BframingC#end to hang around group similar to us so using them as study is wrong
6alse consensus effect: tendency to overestimate other*s agreement with usG eg
7egetarians believe larger amount of pop is vegetarian than meat9eaters
Population: all the cases in the group being studied
#o ma!e a good sample$ use random sampling: sample that gives each case a good
chance of being studied to ensure results within range
Naturalistic observation: observing and recording behaviour in natural settings with any
control on situation
5i!e case study I survey$ doesn*t e-plain behaviour
)hen finding a trait that accompanies another$ not resulting effect$ but correlation: the
way ; factors vary together and how well one predicts the other
Positive correlation: direct relationship where factors increase or decrease together
Negative correlation: inverse relationship where one factor goes up while one goes down
3oes not e-plain cause$ simply show relationship between factors
,llusory correlation: perceiving correlation when none e-istG Notice random coincidences
as not random$ rather as correlated

Experiment
#o isolate cause I effect$ conduct e-periments
%-perimental condition: condition that e-poses subjects to treatment
Control condition: condition that serves as a comparison to see effects of treatment on
e-perimental condition subjects
4se random assignment: assigning subjects to e-perimentalEcontrol groups randomly to
ensure no bias
,ndependent variable: e-perimental factor being manipulated and studied Bby itself$ alone$
no need to depend on somethingC J -9a-is
3ependent variable: e-perimental factor that depends on independent variable and
changes in response to it J y9 a-is
Placebo: an inert substanceEcondition that maybe administered instead of a presumed
active agent
3ouble9blind procedure: procedure in which the e-perimenter and the subject noth donKt
!now which treatment is given
Hindsight Bias: the tendency to believe$ after learning an outcome$ that one would have
foreseen it Balso !nown as the L, !new it all alongL phenomenonC
Critical Thinking: thin!ing that does not blindly accept arguments and conclusionsG
rather$ it e-amines assumptions$ discerns hidden values$ evaluates evidence$ and assesses
conclusions
Theor: an e-planation using an integrated set of principles that organi2es observations
and predicts behaviors or events
Hpothesis: a testable prediction$ often implied by a theory
!perational "efinition: a statement of the procedures BoperationsC used to define
research variablesG for e-ample$ human intelligence may be operationally defined as what
an intelligence test measures
#eplication: repeating the essence of a research study$ usually with different participants
in different situations$ to see whether the basic finding e-tends to other participants and
circumstances
Case Stud: an observation techni0ue in which one person is studied in depth in the hoe
of revealing universal principles
Sur$e: a techni0ue for ascertaining the self9reported attitudes or behavior of a particular
group$ usually by 0uestioning a representative$ random sample of the group
%opulation: all the cases in a group being studied$ from which samples may be drawn
Bnote: e-cept for national studies$ this does not refer to a countryKs whole populationC
#andom Sample: a sample that fairly represents a population because each member has
an e0ual chance of inclusion
&aturalistic !'ser$ation: observing and recording behavior in naturally occurring
situations without trying to control or manipulate the situation
Correlation: a measure of the e-tent to which two factors vary together$ and thus of ow
well either factor predicts the other
Correlation Coefficient: a statistical inde- of the relationship between two things Bfrom
91 to M1C
Scatterplot: a graphed cluster of dots$ each of which represents the values of two
variablesG the slop of the points suggests the direction of the relationship between the two
variablesG the amount of scatter suggests the strength of the correlations Blittle scatter H
high correlationC
Illusor Correlations: the perception of a relationship where none e-ists
Experiment: a research method in which an investigator manipulates on or more factors
Bindependent variablesC to observe the effect on some behavior or mental process Bthe
dependent variableCG by random assignment of participants$ the e-perimenter aims to
control other relevant factors
#andom Assignment: assigning participants to e-perimental and control groups by
chance$ thus minimi2ing pree-isting differences between those assigned to the different
groups
"ou'le('lind %rocedure: an e-perimental method where neither the participants nor the
research staff !now whether one person got the treatment or the placebo
%lace'o Effect: e-perimental results caused by e-pectation aloneG any effect on behavior
cased by the administration of an inert substance or condition$ which the recipient
assumes is an active agent
Experimental )roup: the group e-posed to the treatment Bto one version of the
independent variableC
Control )roup: the group that does not receive the treatmentG serves as a comparison for
evaluating the effect of the treatment
Independent *aria'le: the variable being manipulated and whose effect is being studied
Confounding *aria'le: a factor other than the independent variable that might produce
an effect in an e-periment
"ependent *aria'le: the outcome factorG the variable that may change in response to
manipulations of the independent variableG the variable being measured
Mode: most fre0uently occurring scoreBsC in a distribution
Mean: the average Badd scores and divide by the number of scores that there areC
Median: the middle score in a distributionG half the scores are above it and half are below
it
#ange: the difference between the highest and lowest scores in a distribution
Standard "e$iation: a computed measure of how much scores vary around the mean
score
&ormal Cur$e +normal distri'ution,: a symmetrical$ bell9shaped curve that describes
the distribution of many types of dataG most scores fall near the mean BAN' fall within
one standard deviation of it$ O(' fall within two$ and ON' fall within threeC
Statistical Significance: a statistical statement of how li!ely it is that an obtained result
occurred by chance
Culture: the enduring behaviors$ ideas$ attitudes$ and traditions shared by a group of
people and transmitted from one generation to the ne-t
Informed Consent: an ethical principle that research participants be told enough to
enable them to choose whether they wish to participate in an e-periment
"e'riefing: the post9e-perimental e-planation of a study$ including its purpose and any
deceptions$ to its participants