OverviewCritical thinking is a practice aimed at logically evaluating arguments based on evidence. It is exactly what social scientists, and sociologists more specifically, do when they engage in the research process. But it is not always clear what constitutes “good” critical thinking and how students, scholars, and everyday folks alike can engage in the fullest critical thinking. A key element of working in the social sciences is communicating your knowledge and insights to others, which is often done through written communication. The video above gives a brief overview of a reflection essay, which is the goal of this assignment. A more detailed tutorial is available in the resources section below and explains in detail how to tackle this type of writing assignment–consider it a must-watch!ScenarioYou are a writer for Thinking About Thinking, a website focused on critical thinking and issues in social research. As their content expert on critical thinking, your editor has asked you to read and reflect on the latest book, Is That True? by Joel Best. They want you to create an APA-style reflection essay describing the main ideas in the book, relating and analyzing the main ideas in light of your personal experience, course content, and/or current events/issues, and reflecting on how your beliefs and/or behaviors have been affected by the book. Ethics in Research
🙢
Ethical Principles
🙢
🙣 Research is governed by three ethical principles:
🙢 Respect for persons: Treating persons as autonomous
agents and protecting those with diminished
autonomy
🙢 Beneficence: Minimizing possible harms and
maximizing benefits
🙢 Justice: Distributing benefits and risks of research
fairly
The American Sociological
Association (ASA) Guidelines
🙢
🙣 Research should cause no harm to subjects.
🙣 Participation in research should be voluntary,
and therefore subjects must give their informed
consent to participate in the research.
🙣 Researchers should fully disclose their identity.
🙣 Anonymity or confidentiality must be
maintained for individual research participants
unless it is voluntarily and explicitly waived.
🙣 Benefits from a research project should
outweigh any foreseeable risks.
No Harm to Subjects
🙢
🙣Does it mean that subjects should not be at all
harmed psychologically as well as
physically?
🙣That they should feel no anxiety or distress
whatever during the study, or only after their
involvement ends?
🙣Should the possibility of any harm, no matter
how remote, deter research?
Informed Consent
🙢
🙣 Consent must be given by persons who are competent to
consent.
🙢 Whom might not be included in this group?
🙣 Consent must be voluntary.
🙢 Does this include research on prisoners? How about
research on students? Might there be an element of
coercion?
🙣 Participants must be fully informed about the nature of
the research.
🙢 What if explaining the research affects the spontaneity of
the subjects behavior or answers? Can you conduct covert
or deceptive research?
🙣 Not all harm is foreseeable and then can be disclosed to
participants.
Participants’ Privacy
🙢
🙣 Procedures must be created to minimize the risk of
access by unauthorized persons.
🙣 Laws allow research records to be subpoenaed
🙢 May require reporting child abuse.
🙣 Confidentiality does not apply to observation in
public places and information available in public
records.
🙢 Social media and digital technologies blur lines between
public and private behavior.
– Is behavior on websites such as Facebook and Twitter
public or private?
Regulating Ethics
🙢
🙣Federal regulations require that every
institution that seeks federal funding
for biomedical or behavioral research
on human subjects have an
institutional review board (IRB) that
reviews research proposals.
Paradigms
PURPOSE OF PARADIGMS
◻
Paradigm—A model or frame of reference through
which to observe and understand.
◻
“Patterns happen.”
◻
Theories seek to provide logical explanations.
1.
2.
3.
Theories prevent our being taken in by flukes.
Theories makes sense of observed patterns.
Theories shape and direct research efforts.
LEVELS OF ANALYSIS
🞆
Macrotheory – A theory aimed at understanding the
“big picture” of institutions, whole societies, and the
interactions among societies.
⚫ Examples:
class struggles, international relations, and
interrelations between social institutions
🞆
Microtheory – A theory aimed at understanding
social life on the intimate level of individuals and
their interactions.
⚫ Examples:
dating behavior, jury deliberations, studentfaculty interactions
THREE MAIN PARADIGMS
CRITICAL RACE THEORY
How does this
compare to
what you’ve
heard about
critical race
theory in
recent news?
FEMINIST THEORY
Watch the
first minute
and a half
to learn
about
feminist
theory
TWO LOGICAL SYSTEMS REVISITED
TRADITIONAL SCIENTIFIC METHOD
(DEDUCTIVE)
Conceptualization
and Measurement
Observing the social world
⦿ Everyday
observations are haphazard,
inconsistent, and filled with bias
⦿ Scientific
Measurement – Careful,
deliberate observations of the real world
for the purpose of describing objects
and events in terms of the attributes
composing the variable.
What are we measuring?
⦿ Conceptualization
– The process through
which we specify what we mean when
we use particular terms in research.
⦿ Uses
› Deductive—translate abstract into testable
hypotheses involving specific variables
› Inductive—make sense of related
observations
What are we measuring?
Concepts are fuzzy notions of things in the
real world
⦿
›
›
Race
Self-esteem
An attribute is a characteristic or quality of
something (ex: female, old, student).
⦿ A variable is a logical set of attributes (ex:
gender, age).
⦿
›
›
The attributes composing it should be
exhaustive.
Attributes must be mutually exclusive.
What are we measuring?
⦿
Selecting Variables
›
›
›
›
Theories
Utility of variables in previous research
Constraints and opportunities in research
setting
Relevance of variables to analysis at hand
How will we measure?
⦿ Operationalization
is the process to
devise operations that actually measure
intended concepts
› Goal🡪
achieve measurement validity
1. Define concepts
2. Choose variables , considering purpose of
research
3. Identify construct indicators
4. Feasibility
Levels of Measurement
⦿ The
mathematical precision with which
the values of a variable can be expressed
› The nominal level of measurement, which is
qualitative, has no mathematical interpretation
› The quantitative levels of measurement—ordinal,
interval, and ratio—are progressively more precise
mathematically.
⦿ Level
of measurement is not inherent in
variable, but is chosen by the researcher.
Levels of Measurement–NOIR
Reliability and Validity
Increasing Reliability
🞑 Test-Retest Method
■ Make the same measurement more than
once.
🞑 Established Measures
🞑 Already tested for reliability
🞑 Reliability of Research Workers
🞑 Inter-observer reliability
research design
Three Purposes of Research
• Exploration
•
•
Get a better understanding
Test feasibility of more extensive study
• Description
•
Detail phenomenon
• Explanation
•
Answer “why” questions
The Logic of Nomothetic
Explanation
• Goal: to find a few factors that
can account for many of the
variations in a given
phenomenon, but not necessarily
all
Causation vs. Correlation
Three Criteria for Nomothetic
Causality
1. The variables must be correlated
■
Correlation – An empirical relationship between two variables
such that changes in one are associated with changes in the
other, or particular attributes in one are associated with
particular attributes in the other.
2. The cause takes place before the effect (time order)
3. The variables are non-spurious
■
Spurious Relationship – A coincidental statistical correlation
between two variables shown to be caused by some third
variable
Spurious Relationship
Example
Necessary and Sufficient Causes
• A necessary cause represents a condition that
must be present for the effect to follow.
• A sufficient cause represents a condition that, if
present, guarantees the effect in question.
• Most satisfying outcomes in research include
both necessary and sufficient causes.
Units of Analysis
• Individuals :
information about
individuals is
aggregated and
generalized
• Organizations :
information about
formal social
organizations is
aggregated and
generalized
• Groups : information
about a collective is
aggregated and
generalized
• Social Interactions :
information about what
goes on between
individuals, groups or
organizations
• Social Artifacts :
information from a
product of social beings
is aggregated and
generalized
Ecological Fallacy in 3 Min
The Time Dimension
• Cross-Sectional Study – A study based on
observations representing a single point in time, a
cross section of a population.
The Time Dimension
• Longitudinal Study – A study design involving the
collection of data at different points in time.
Conceptualization &
Operationalization
Levels of Measurement
Why does this matter?
OK to compute….
Nominal
Ordinal
Interval
Ratio
frequency distribution.
Yes
Yes
Yes
Yes
median and percentiles.
No
Yes
Yes
Yes
add or subtract.
No
No
Yes
Yes
mean, standard deviation,
standard error of the mean.
No
No
Yes
Yes
ratio, or coefficient of variation.
No
No
No
Yes
Reliability & Validity
Reliability & Validity
Experiments
Experiments in General
● Experiments are good for studies
involving limited (and well-defined)
concepts, hypothesis testing, and
explanatory research.
◆
Goal: To make generalizable causal
inferences
● How is this accomplished??
◆ Random assignment
Experiments in General
● Processes of designing and conducting
experimental research
◆
Formulate experimental conditions and
procedures
• Ensure IV shows clear and systematic effects on DV
◆
Reducing imprecision in measurement associated
with
• Sampling
• DV measurement
• Experimental procedures
◆
Controlling extraneous variables
• Enhance plausibility of IV causing changes in DV
Concepts
● Concepts
◆ IV and DV (X & O; cause & effect,
respectively)
◆ Pre and Posttest (1 and 2, respectively)
◆ Experimental (top case) and Control
(bottom case)
◆ Placebo
◆ Double-blind
Classic Experimental Design
● R: Experimental O1
● R: Control
O1
X
O2
O2
Selecting Ss (cont)
● Still, E and C groups should be comparable
● Techniques for accomplishing this are:
◆
Probability sampling
• If samples resemble population, they will resemble each other
◆
Randomization
• Are people that answer an ad for experimental subjects like those
that don’t? (between group differences)
• Are these Ss r: assigned to E and C conditions at least like each
other? (within group differences)
◆
Matching
• Match E and C groups for each “relevant” variable(s)
– Ex. gender, ADHD diagnosis
Quasi Experimental Designs
Internal Validity
● Internal validity occurs when extraneous
variables have been controlled and the
observed effect can be attributed to the
IV.
◆
Goal: isolate IV and DV to determine
causal relationship
Threats to Internal Validity
●
●
●
●
Testing – “OFL” on eye exam
History – Events happen
Instrumentation – pre- to posttest
Selection—Group difference
●
●
●
●
Maturation – older/wiser; tired/bored
Experimental Mortality – drops out are different
Statistical Regression—extreme cases disappear or are minimal
Selection-maturation interaction—combo of threats interact to confound
treatment effect
●
Diffusion of experimental effect—treatment may diffuse to control group
who seek desirable treatment
Rivalry (Compensatory) – try extra hard
Equalization of treatments (Compensatory)—Even Steven
Ambiguous temporal precedence—timing is all wrong
Demoralization – give up
●
●
●
●
External Validity
● External Validity
relates to the
generalizability
of study results.
◆
Goal: To make
generalized
inferences
about a larger
population
and/or multiple
settings based
on experimental
results.
Natural Experiments
● Naturally occurring independent
variables can’t be studied in a lab
◆ Example
• Hurricane (IV) and effect it has on people that
do and do not have homes or other property
destroyed (DV)
• COVID-19 pandemic (IV) and mental health of
[insert population]
Evaluating Experiments
Strengths
● Great for isolation of
experimental variable
● Fewer Ss…which
leads to easier
replication
● Less time
● Less money
Weaknesses
● Artificiality
SURVEY RESEARCH
Topics Appropriate for Survey
Research
◻
◻
Descriptive, exploratory, and explanatory
Surveys are most appropriate when
researchers
🞑
Want to collect large samples
🞑
Want to collect original data
🞑
Are interested in measuring attitudes and/or orientations
◻ Any decision about the best survey design
for any particular study must take into
account the unique features and goals of the
study.
Designing a Survey
What types of
questions are
appropriate for
questionnaires?
What are the dos
and don’ts for
creating “good”
questions?
What are the parts
of a
questionnaire?
Questionnaire Design
Process
Good and Bad Survey
Design
Great
examples
that will help
you
complete
this week’s
assignment!
Survey Administration Evolution
Interview Surveys
◻
◻
Interview – A data-collection encounter
in which one person (interviewer) asks
questions of another (respondent).
This is different than a qualitative
interview!
🞑
🞑
Interviewer must administer every
questionnaire exactly the same to avoid
introducing bias.
Interviewer does not add words or ask followup questions.
Interview Surveys–Part 2
◻
◻
◻
◻
◻
Interviewers solicit higher response
rates (80-85%) than mail surveys.
Interviews minimize “don’t know” and
“no answer.”
Interviewers serve as a guard against
confusion.
Interviewers can observe respondents
while completing the questionnaire.
Costly and time consuming relative to
other methods of administering surveys
Comparison of the Different
Survey Methods
◻
Mailed Surveys
Cheaper and faster than face-to-face interviews
🞑 Low cost to mail
🞑 Requires small staff
🞑 Greater willingness to answer controversial items
🞑
◻
Interview Surveys
Fewer incomplete questionnaires
🞑 More effective for complicated questionnaires
🞑 Face-to-face is more intimate
🞑
◻
Telephone Surveys
Cheaper and more time efficient
🞑 Increasingly difficult when caller ID became widespread
🞑
◻
Online/Email Surveys
Lots of available software and websites for survey construction
🞑 Arguably the norm now
🞑
Strengths and Weaknesses of
Survey Research
◻
Strengths
🞑
🞑
🞑
🞑
◻
Useful in describing large populations
Make large samples possible
Surveys are flexible
Standardized questions
Weaknesses
🞑
🞑
🞑
🞑
🞑
Round pegs in square holes
Seldom deal with context of responses
Inflexible
Artificial
Weak on validity (but strong on reliability)
Common Problems and Alternatives
◻ Check out this article, specifically the table, to see examples
of common problems and better alternatives
1
Much
What Is Critical
as
politicians
Thinking?
endorse motherhood and
apple pie,
nearly
thinking.’Collegeprofessors
agree that they want their students to become critical
thinkers, but
so do teachers in lower
grades. I’ve heard first- and second-grade
teachers declare that teaching critical
thinking was one of their
most important jobs. Most educators are with
the program.2
But we can suspect that when
virtually everyone agrees that
everyone who teaches praises critical
something is good, they probably define it in different ways. The
word critical can take on many different
meanings. I recall one stu-
dent recoiling in horror when I spoke positively about critical think-
ing: “Oh, I don’t want to become a critical person!” Then there are
sociology professors who will boldly declare that they embrace
“critical race theory” or “critical animal studies” or .. .youget the
1dea. Used in this way, oritical usually signals that their approach is
aligned with some sort of liberal/progressive/radical/leftist political perspective. In effect, they use the word critical as a sort
of brand name to contrast their approach with rival schools of
thought that, they charge, support the status quo. While they may
assume that adopting a “critical” approach makes them critical
thinkers, that’s not whatIwill mean by “eritical thinking” here.
Rather, this bookviews critical thinking as a set oftool.
sfor evalu
case. We encounter claims all the time in conversatione gto be the
in what we
read, in the media, indeed, on pretty much every occasion
we co
ating claims. A claim is any statement asserting somethin
nect with other people, and we’ve all had to learn to internu
claims.We classify claims as being more or less convincina
, using
terms like fact or information toidentify claims that seem sound
d, and
terms
like
rumor or
fake to label claims that seem more duhios
Learning to make these distinctions starts early: a lot of parenting
ing
involves helping small children become better at evaluating thin.
ngs
they hear (“He’s only teasing,” *I’m really serious,” “That’s just a
story”). At some point, kids have to learn to distinguish betweenTy
programs and the content of commercials, and to understand that
advertisers’ claims may not be completely truthful. As we get older,
we learn that flattery and compliments might not reveal what others
actually are thinking, just as most ofus learn to discount rival claims
made during election campaigns. We learn to distinguish question-
able claims from others that seem more likely to be true.
The ability to think critically is important. Imagine a person
incapable of critical thinking: this would be someone so suggest
ble (and vulnerable) that he or she takes every commercial’s advice
to rush out and buy the product being advertised, and finds every
politician convincing. Obviously, few people are that weak. Tc
while becoming skeptical or suspicious of what we’re told bypE
ple who want to sell us something is a useful skill, it isn’t enou
d artiWe
constantly encounter claims in news stories, boOKS, a
posts,
in
CICs Trom radio, television, and online personalities; blog we to
podcasts, downloaded videos, and social media. How ae
e
Claims? Howcan we separate ones thatprobaoy
can be accepted as true from those that we should doudtr
12
WHAT IS CRITICAL
THINKING
People have different standards for making these
judgments.
One popular standard throughout
history has been to assume that
we already know what is true–that there is
some sacred book that
contains all the truths
need to know and that we can
simply
judge all claims by whether they are consistent with this holy writ.
we
Or that some great thinker-Aristotle,
say, or Confucius, or Marx
already explained how the world works, and we can evaluate
today’s claims in terms of how well they match those classic inter-
pretations. Assuming you already know what’s right and true can
be comforting, if only because it justifies ignoring those who hold
different views. Anyone who has
ever
gotten into
an
about religion knows that people who believe some sort
argument
of authori
tative doctrine are hard to budge.
This book presents critical thinking as a more modest, alternative approach for assessing claims. Instead of simply assuming that
we already know what is true, critical thinking requires that we
consider the possibility that our assumptions might be wrong. At
bottom, critical thinking is about evidence. Evidence is information
that can help us judge whether a claim is true. When we hear a
claim, we ought to evaluate the evidence for and against it. The
claim may be about something small and personal (“I love your
hair in that style”) or aimed at a much larger audience (today’s top
news story). It doesn’t matter. Thinking critically involves examin-
ing the evidence for a claim and deciding whether it is convincing.
When this book refers to “eritical thinking,” then, it wil mean ways
of weighing evidence and distinguishing between stronger evi-
dence and weaker evidence.
This sort of critical thinking has a history. It began to catch on
during the Enlightenment-the centuries-long movement refuting
the idea that all truth could be found in the Bible or Aristotle.
WHAT IS CRITICAL THINKING?
13
Instead, people
started
evaluating observable facts
collecting and
and information-i.e., evidence. For instance,
they used
to make observations of the planets and stars, and h.
convinced them that the earth revolved around the e
telescopes
at they saw
sun,*thus
tradicting theologians’ insistence that the earth was the
the universe. Later,
isms that
to
they used
microscopes to identify tiny
seemed to cause diseases, which led medical
reluctantly reject Aristotle s
imbalance of the
body’s
model
con
organ-
authorities
ofdisease being Caused
four humors.
These were tough
bates.
some theologians and physicians never stopped
resisting the new
ideas. But today, those pieces of evidence have won out: most ne
peo
ple accept that the earth orbits the sun, and that germs can lead to
disease. Yet we continue to argue about plenty of other things.
Most people now agree that evidence is important, even thoueh
they may disagree about what the evidence shows
Critical thinking in the sense of weighing evidence is a skill it
can be learned, and one gets better at it with practice. Perhaçs
you’re surprised that so many educators agree that teaching critcal
thinking is important. After all, your high school probably didn?
offer classes in critical thinking. You took classes in mathemanes
science, language or literature, and social studies or histor
Stil, your teachers probably thought that all those classes were
or
teaching critical thinking skills: math taught you to performnau
aysa
ematical reasoning, literature classes involved analyzing piay
ons ot
poetry, history encouraged you to assess different explaiau
teac
key events, and so on. Those lessons were designed to
and history,
Something about the substance of math, literature,
ure, and ni
they were also intended to make you a
one
more
critical
thinker,
who not only knew something about the subjects
14
WHAT IS
CRITICAL
THINKING
but
some”
estiot
but also could
apply
the
analytic
variety of topics and contexts.
Learning to
skills those lessons
taught to
think
a
critically is a major reason why there is a
strong correlation betweenlevel
of education and income: on average, high school graduates earn
more than those
who drop out;
with
some college make
people
more than those who
don’t go
beyond high school; those who graduate from
college earn a lot
more than those who don’t
receive a degree; and
people who on
to finish
graduate or professional degrees make more than go
graduates. Why should this be true? Lots of high school and college
college
classes
don’t
directly relevant to most jobs. But the subject
matter covered in those classes is less
important than acquiringthe
critical thinking skills students need to
succeed in college. A college graduate should have learned to read
thoughtfully enough to
comprehend difficult material, to locate information and evaluate
its quality, and to
develop, organize, and present their own reasoned arguments. By completing
coursework-doing the assigned
seem
reading, studying for tests, writing term papers, and so onstudents develop and use increasingly
sophisticated critical think-
ing skills. At bottom, it is those relatively rare and valuable skills
that qualify better-educated individuals for
higher-payingjobs.
In other words, while the term critical
thinking may seem vague,
abstract, or impractical, it is actually the key to education. Consider
a guestion sometimes posed to grade-school children: “There
are
125
sheep
and 5
dogs
in
a
flock. How old is the
shepherd?”3
Mathematics educators note that most children facing this question assume they must be being asked to produce a number, such as
25 (125 divided by s). After all, arithmetic students constantly confront “word problems” that require them to calculate the correct
WHAT IS CRITICAL THINKING?
5]
The Basics
2
Arguments and Assumptions
In this
book, the word argumnent simply means an attempt to per-
suade, a line of reasoning with one or more claims that lead to a
conclusion. An argument doesn’t need to be dramatic, or hostile. If
John says, “Since it is raining right now and we don’t want to get
wet, I think we should delay going outside until it stops,” he is
making an argument. The argument consists of grounds that provide basic information (it is
raining), wamantsthat justity drawing
some conclusion (we don’t want to get wet), and the conchusion
itself (therefore, we should stay inside and wait for the rain to
stop)
Critical thinking refers to evaluating or assessing an argument
to see whether it is convincing. For example, when considering
John’s argument, you might ask whether it is still raining, whether
it is
raining hard enough to make going outside uncomfortabie
whether you actually care about getting wet, or whether you hav
some urgent reason for going outside now that makes getting wet
ee
unimportant. Depending on the answers, you might ag
that the argument is convincing and choose to stay indoors, Oryyou
seem
COnsider
E
heading out.
the argument unpersuasive and
decide
to
risa
Because all
arguments consist of grounds, warrants, and conclusions, thinking critically requires
evaluating each of these ele-
ments. Grounds statements are claims about the
way things are. In
Tohn’s argument, the grounds are the straightforward claim
that it
is raining right now. We can evaluate this claim
by looking outside
to
check whether it really is raining, or
perhaps we might get into a
discussion about just how hard it’s raining, or about what
“raining”
actually nmeans-is it just misting or sprinkling or
lightly raining,
and is that enough precipitation to make us want to
stay indoors?
Other claims (“Poverty is caused by discrimination and other
problematic social arrangements,” say, or “Poverty is caused a culby
ture that discourages effort”) may be
supported by all sorts of more
complicated evidence-examples, statistics, definitions, and so
on-and there are various ways of approaching such
grounds statements. Do the statements seem true? Do we have
enough evidence
to evaluate the
statements? Does the evidence seem strong, or does
it have weaknesses? Are there other
things we’d like to know? And
so on.
Arguments may have elaborate grounds consisting of sev
eral statements, and there may be lots of reasons to criticize those
grounds.
Warrants are justifications; they invoke values. The warrant
that we
don’t want to get wet justifies John’s conclusion. Thinking
about warrants can be tricky. Sometimes, warrants are implicit: if
the person making the argument and the people who hear it share
the same values, it may not seem necessary to spell out the argu-
ment’s warrants. That is, John might assume that no one wants to
get wet if they can avoid it, and so not bother stating the warrant,
but simply say, “Since it is raining right now, I think we should
delay going outside until it stops.” Criticizing warrants can be
uncomfortable because it might draw attention to fundamental
THE BASICS
[9
E
SOI
C
VA
H
compelling-by, say, John asserting not just that “it’s raining,” but
that “it’s
pouring down rain”? Logic, in turn, attempts to evaluate
the strength of arguments, whether the
grounds and warrants are
sufficient to convince someone who is thinking
rationally to accept
the argument’s conclusions.
Logicians (the philosophers who
study logic) identify logical fallacies-flawed forms of
argument in
which the conclusion does not
necessarily follow. (We will have
more to say about fallacies in later
chapters.)
Every argument depends on assumptions. Often these are
unstated parts of grounds
This need not
that
or warrants
cause concern.
gravity is in play,
that
are
taken for
granted.
We
and the
routinely assume, for instance,
people we’re talking to doubtless
agree (unless our subject happens to be outer space). But there are
plenty of assumptions that can cause mischief.
When you talk to someone who holds very different
religious
beliefs or political opinions than you do, disagreement is to be
expected. A conversation about religion between someone who
confidently believes that God exists and another who simply does
is likely to end up with them talking past each other. Not only
is each making a critical assumption that the other does not accept,
not
but each may take his or her own assumption for granted, not realizing that it is just that: an assumption. It can be hard for us to even
recognize our own assumptions, and of course it is hard to think critically about them because-after all-we are already convinced they
are true.
Assumptions are necessary to argument, however. We can’t be
expected to supply the entire chain of reasoning (about gravity,
for example) that leads to every claim we make. At the same
time, we need to acknowledge that we do make assumptions,
THE BASICS
[i1]
3EverydayArguments
Lots of us enjoy discussing-even debating-different aspects of
latest
society. Anything from the
news
headline to a walk
through
ion is
our neighborhood can inspire such a conversation. An
uttered, someone else chimes in to either agree or disagree u d
we’re off.
These conversations tend to be relaxed, without many rules
about what you can or can’t contribute. The arguments that people
make-with their grounds, warrants, and conclusions-receive lit
tle close scrutiny. As a
we
result, there are no clear standards for what
might call critical thinking in these situations.
chapter examines some common elements everyday
and may even
to
arguments that are flawed. They are tempting use,
of
This
have
convincing on the surface, but they
need to be understood.
seem
limitations
that
Anecdotes
erienes:
Arguments often feature stories
about one’s own
Just the other day I saw…. “Usuallythese
exp
tales arel
provide firsthand evidence ofsomething that the speas
[14]
meansto
be
h
understood as common, or a particular
por
for
some broader claim: “I
saw two
example offered as suppeople sitting at a table in
restaurant, each oT them staring at their own
phone. We are losing the ability to talk to one another
face-to-face.”
a
In
other cases, the anecdote is not
olavs a
story that he or she heard from
firsthand. Instead, the teller
friend or on the news. But
apain, the implication is that this case is somehow
typical. Thus, an
example of someone who fraudulently claimed benefits
from a
social welfare program can be used to
argue that many of the prooram’s beneficiaries do not really need or deserve
assistance
Such anecdotes may seem quite
to the
them, but
they
a
compelling
should not be considered
people telling
especially strong
evi-
dence. The very fact that a story is distinctive or memorable
enough
to catch your attention may be a
that
this case is not at all
sign
typical. A single example (for instance, you know a
poor person who
strikes you as lazy) is a weak basis for broad generalizations (all
poor people are lazy). After all, we occupy a big world with billions
of people living all sorts of lives. A story about
something we have
witnessed can no more represent the complexity of the whole world
than any one photograph can depict everything we might see. Even
if someone can regale us with two or three or even more
examples,
we need to realize that we all travel in more or less restricted social
Circumstances. Let’s say Sally, a teacher we know, complains about
bad behavior among some of her students. Perhaps she can offer
lots of examples, perhaps she convinces us that the students in her
classroom are indeed a difficult bunch. How confident can we be
that her experiences with her class tell us much about what’s going
on
in other classrooms,
or
in other schools?
Anecdotes are almost inevitably about atypical or unusual
DEnavior-something that caught the teller’s attention and seemed
EVERYDAY ARGUMENTS
|15
to
interesting enough
fic.
we
aren’t
share with others.
After driui.
about all
likely to tell anyone
at
nessed stopping
who becomes
the
red
lights;
it is the
vingthro
taat
ught
ivers we
8h
the other
wit
o ran
driver who
the red
ran.
light
anecdote.
Suppose Carlos tells you he saw just such a red-liah
ght-running
driver, then declares, “Traffie is getting more and more da.
re dangerous
with drivers like that on the roads.” If you check the stati
s
col
lected bytraficenforcement agencies, though, you will
d that
in
fact rates oftraftic fatalities have fallen dramatically over e l
last
eweral decades. Obviously, this doesn’t mean that Carlos did
idn’t
see someone run a red light; but it might make us question hison
on
clusion that that red-light-running driver proves that
today’s road.
ways are more dangerous than they used to be.
Of course, if you remark that traffic fatality rates are
down,
Carlos might respond that such statistics are irrelevant: after no
all,
one
died when the driver he
saw ran
that red
light. This raises an
important point about evidence. Evidence is almost never complete or perfect. There is no way of knowing the precise rate at
which drivers
red lights; we can’t monitor every
drivers
approach to every stop light, and even if we could, we can’t go back
in
run
time to
make similar measurements, so we can’t
possibly prove
that red
light running has increased (or, for that matter, declineu
Sowe look for the best available evidence. We
assue that,
might
in contrast to fender benders, many of which may never come to
ne
attention of law
cause
enforcement, accidents serious eno
h to
fatality are almost sure to be reported, and as
*
So, it is
counts of traffic fatalities are probably reasonably
not
about the driver
unreasonable to counter Carlos’s
anecdote
who ran a
abou
red light with
evidence that traftic
a
a
accurate.
rates a r
fara
16]
EVERYDAY ARGUMENTs
declining, Presumably if reckless driving is
becoming more combe
should
aCcidents
increasing, and so should fatalities.
Tt is certainly possible to continue to
debate the value of the
fatality evidence. Carlos might suggest, for
example, that
increased recklessness may be causing a big rise in nonfatal acci-
trafic
dents. But without more evidence to
nent
has
no
support that claim, his arguteeth. The point here is that evidence is
key to a
successtul argument.
Anecdotes have another feature: they usually describe a
sequence of events-Qhappened, and then R followed, and that
led to S. It is important to appreciate that such stories or narratives
have their own limitations. Any narrative is necessarily selective; it
is
impossible to tell a tale that encompasses everything that hap-
pened. Highlighting the Q-R-S sequence inevitably ignores A
through P.
One way to think critically about a narrative is to question its
hoice of elements. Have all the relevant events been included?
Are parts of the narrative’s sequence irrelevant? That is, does it
make better sense to add elements (to tell the story as P-Q-R-S,
instead of just Q-R-S), or even to subtract some (so that we have
only Q-S)? Disagreements about why something happenedanything from how we wound up eating at this restaurant to
whether slavery caused the Civil War-often revolve around which
elements are selected to make sense of the story.
Even when we agree about the essential elements in the story,
we may interpret them differently. When Carlos tells the tale of the
Cd ight runner, he suggests that the driver was simply reckless, but
a critic might propose other possible explanations: perhaps the
ariver had an emergency or whatever. Agreeing on the relevant
EVERYDAY ARGUMENTS
|17]
does
sequence
elements in a
an interpretation.
agree on
not necessariy mean
Notice that
we
that peonl
people will
.. accept
may be inclined to
with our ideas ofwhat is relevant or true
stories that fit well
and resist others because they seem to contradict what we believe.
some
anecdotes. Stories can make things seem clas
which is why authors and journalists often begin their booke and
We all
news
earer,
use
stories with
an
example
so as to
give their topic
a
huma
uman
dimension. But anecdotes have limitations. If someone makina .
ing a
Sweeping declaration”The world is going to hell!”-is asked for
evidence”What makes you say that?-and responds with an
anecdote, about people looking at their cellphones or running a red
light, say, at first glance this simple evidence may seem sufficient
to support the conclusion. But an anecdote is always weak, mper-
fect, incomplete as evidence. We ought to try to move beyond spe-
cific examples if we want to understand social life.
Ad Hominem Arguments
An ad hominem argument is one that focuses on the person who
has said something, rather than on what has been said. Claimings
Well, that person is an environmentalist [or a conservative or
just fill in the blank], so I don’t have to listen” rejects the mes
sage because it comes from a particular messenger. This is danger
ous, because it closes off the listener from whatever ideas
tna
be
person may presenting.
Of
course, people disagree about lots of things. But it is a m
take to think that
you can simply ignore or reject out of hand w at
everthe people you disagree with might say. It is fine to
an
argument because of its weaknesses,
but not simply becausitwas
made by a kind of
person with whom
you
|18
probably disagree
EVERYDAY ARGUMENTS
tempting to fall int0 ad hominem
arguments. Most of
have complex identities that include particular
political or reli
1 G views, and we know that others
disagree with those views.
Tt
can
be
11S
People who consider themselves liberals realize that other
folks
hink of themselves as conservatives, and vice
versa. We probably
sketch rough deseriptions of what people on the other side of
the fence think, and we probably find their
arguments predictable;
we may think we already know what
they are going to say. Still, to
can
ionore argument sinmply because the person making it belongs to a
category of people who disagree with us is an error of reasoning.
The term ad hominem is Latin, meaning “to the person”; the
error involves addressing the supposed motivations or biases ofthe
person making an argument, while ignoring the argument’s intrin-
sic logic or evidence. It is a logical fallacy that was named centuries
ago, at a time when learned people wrote their analyses in Latin.3
The key to critical thinking is assessing evidence. Assessing
does not mean accepting. As we have already noted, there is nothing wrong with arguing that an anecdote is a relatively weak form
of evidence, that a description of a specific incident is a poor basis
for making broad generalizations. But that is not at all the same as
rejecting the anecdote’s relevance because the person telling the
story holds beliefs different from yours.
Intense conflicts often lead opponents to develop dismissive,
hostile names for one another-slurs based on ethnicity, religion,
or politics. These labels are hurled back and forth, and they encour-
age ad hominem
critiques: if Jane is a [derogatory label],
aont need to listen to her ideas or even to her
then
we
evidence-whether
nat s the evidence she presents in support of her own claims or in
ner critiques of our arguments. This is a seductive line of thinking
Dcause it seems to excuse us from taking our opponent seriousiy.
EVERYDAY ARGUMENTS
119]
And it
returns
opponent’s
much
us to
the t a m i l i a r
temptation:
them), as
(or just ignore
arguments
are
claiming
terribly dangerous.
in response.
ODnoe
o the
because
Ad nominem argume.
they c a u s e
us to
views, while discouraging
who share our
in critical thinking.
to engage
we
to
aboa what we
task of thinking critically
challenging
more
ourselves
ity
simpis
While this chapter is
will have occasion
later chapters dealing
tocused on
us
huddle
amc..
from using c a r e
pittalls in everyday argu
to further discuss
ad hominem
with sociological
reasoning.
argume-
sin
Mvths
Like ad hominem critiques, calling something a “myth’ is ancther
way to justify dismissing an argument
out
of hand, without consid
ering its merits. Folklorists-the people who actually studymythsuse the term to refer to origin tales about gods and goddesses and
how the word took form. Different cultures have different myths
the Greeks and Romans, the Norse people, and the Navajo: all have
their own mythologies. In everyday conversation, however, calingS
something a myth is to argue that it is false, and that only mistaken
people believe it. Presumably the reasoning is that since we con
Siaer tales teaturing Aphrodite or Thor as fictional, the key fearure
of these
myths
must
be that
they
are not true.
Social scienus
Sometimes use the term this way. For example, one can hna ists ot
ape myths-sets
of statements about rape that some people
believe but that, the
hen
analysts insist, are simply false (e.g
incite men to
rape,” “Women fantasize about being raped”)
Similarly, there are lists of marriage
myths, disaster myui
gration myths, and so on.
20
EVERYDAY ARGUMENTS
imm
As we have seen, there is nothing wrongwith reviewing the evi-
dence regarding some claim and arguing that that evidence is so
weak that the claim should be rejected. It is less clear that labeling
such claims as myths is helpful. Calling a claim a myth dismissesit,
simply by declaring it to be false: “Some people believe that X hap-
12ens, but that isn’t true; it’s just a myth.” But what does this mean?
Is the argument that it is a myth because X never happens, or that
it happens only infrequently, or what? Much like ad hominem argu-
ments, the myth label promotes dismissing an argument out of
hand without actually assessing its evidence.
This is a tactic that can be used by anyone who wants to chal-
lenge particular ideas. Try Googling global warming myths or inequality myths-or virtually any social issue + myth. All of these
folks are using the term myth to say, in effect, that some misguided
people may believe K, but X is just wrong, wrong, wrong.
Notice, too, that people with competing views often declare the
other side’s assertions to be myths. Thus, a Hufington Post piece
entitled “10 Abortion Myths That Need To Be Busted” begins:
“1. MYTH: Abortion is dangerous”; while “1o Pro-Abortion Myths
That Need to Be Completely Debunked,” an article posted on
LifeNews.com, leads off with “1. MYTH: Abortion is safe.”* Or take
competing lists about guns: the second myth discussed in the
Federalist’s”7 Gun Control Myths That Just Won’t Die” is “Nobody’s
Demanding Gun Confiscation”; yet “1o Pro-Gun Myths, Shot
Down,” from Mother Jones, features as Myth #1: “They’re coming
for your
guns.”5 Such examples of contradictory myth-spotting sug-
gest that simply branding claims as false-or as myths-may be
overly simplistic.
We can suspect that it would help to define some of these
terms.
What precisely do these folks mean by “safe,” “dangerous,”
EVERYDAY ARGUMENTs
21
“contiscation,”
seem
or
“coming for “?
These claims to
isn’t
something
absolutism:
coma kind of
to argue for
false. Clarifying definition.
must be absolutely
true, then it
if
pletely
may
identify mnths
resolve s o m e
dangerous?
confusion.
ofthis
approach might
One
con
Take abortion-is it safe or ie
be to acknowledge that abortion
medical
procedure
carriee
procedure, and that every
wrong. We can, however, suspect that
risk that something mightgo
doctors-like the vast
of abortions performed by
is
a
the
medical
vast
majority
not lead to
serious medical
majority of, say, appendectomies-do
that some very small number of aborcomplications, and still agree
the issue is not whether
tions may result in problems.” Perhaps
abortion is perfectly safe (in the sense that no woman who undersuffers harm), but whether it is relatively safe in
goes abortion ever
the sense that other well-established medical procedures that
rarely lead to harm are considered safe. This definition might lead
us to argue that abortion is about as safe as other common medical
treatments. On the other hand, a different definition-say, that any
evidence of harm having occurred justifies considering abortion
risky-might lead to acknowledgment that it and lotsofother medi
cal procedures involve some danger. Understanding either claim
requires that we examine both the definitions being used and the
evidence; we can’t simply impose the word myth and consider the
matter settled.
But
examining the evidence
is
precisely what calling some
thing a myth discourages. Giving reasons why a particular belief
mayor may not hold up to scrutiny is a form ofcritical thinking, but
simplyresponding toa claim with “That’s a myth” is, in effect, an
argument that there is no need for reasoning, that the matter is set
tled. Critical thinking demands that we review the evidence. Tns
will not necessarily end debate-reasonable people may dstill
22
EVERYDAY ARGUMENTS
disagreeabout how to interpret the evidence; but at least it offers a
solid basis for discussion.
more
Folk Wisdom
and
Metaphors
addition to studying myths, some folklorists study aphorismshace
little sayings
that are invoked to
support
everyday
argu-
ents, Aphorisms are often contradictory. Imagine a conversation
here
Bob says he’s having difficulty making a decision regarding
work. Maria urges him on by remarking, “He who hesitates islost.
Raut then Vince adds, “Look before you leap.” These two time-worn
bits of advice advocate opposite courses of action, and probably
won’t be ofmuch use to Bob. In other words, folk wisdom tends to
be awfully flexible: it is usually possible to drag out some aphorism
to support whatever argument one wants to make.
A related form of talk is the invocation of metaphors. The
course that Bob says he’s considering taking may sound on the sur-
face reasonable, but Vince might comment, “Sounds like a slippery
slope to me,” or remarks that it could be just the tip of the iceberg,
meaning, respectively, that making a small concession now will
inevitably lead to further concessions, or that whatever is visible
may be only a small part ofthe whole. Metaphors can make conversations more colorful, at least until they become so overly familiar
that people dismiss them as clichés. But their real purpose is to con
aense a larger argument into a single, familiar bit of folk wisdom.
he problem with metaphors is that they can discourage think-
ng critically about the claim being made. We all know that only
the
P OT the iceberg, about 10 percent ofthe whole, is above the water
une and visible. When the metaphor is used to describe, say, some
l a l issue, we are being asked to imagine a hidden, vastly larger
EVERYDAY ARGUMENTS
|23|
problem
that
be
that may
can’t see
really 90
ddressed c
to be
be aaddressed
eventually. Of
Of ecourse,
are some
ases uwhere we
there probably
would
have to
cases
true;
issues.
the underlying
the
(as in
But what p r o p o r t i o n
case
percent
s0 percent?
Or is it only
to imagine
that the
L
simply
.
dden? 1s it
real iceberg’s
iceberg’s hidden
of a real
ass)?
less? Withn.
a
hout presentin
Or significantly
iceberg metaphor
than it may in
issue is much, larger
fact
the way ofevidence,
anything in
us
in
the
be
Aphorisms and metaphors are verbal shortcuts; they packay
into just
strings ofreasoning
a
This is
is val.
valuable,
few, familiar words.This
thinking would he ie..
we
Imag1ne
even necessary.
how slu8gIsh our
because
act on them. Yet
they simpitycomplexity, metaphors.car
could not use metaphorical reasoning to recognize similarities and
also easily misdirect us. We need to think critically about where
ere
they are leading us and
whether that’s where
we want
to go.
Facts
Our commonsense understanding of fact is that it refers to some
thing that is simply true. The declaration “That’s just a fact!”is
often intended as a kind of argumentative trump card-a state
ment that cannot be
disputed. At the same time, we know thatpeo
ple sometimes get into arguments over just what the facts are. HOw
is this possible?
A better way to think about facts is to realize that facts depend
on social agreement. Imagine a gathering of people who belong to
religion, who all agree
partucular
is the word
a
particular book is holy,
of God. Within that gathering of believers, pe
agreethat it is a “fact” that that book reveals God’s will.
that it
nay
sup
pose other
people with different beliefs join the gatherug perhaps
theydon’t believe in God, or perhaps they believe thata
ditferent
24
EVERYDAY ARGUMENTS
book reveals God’s will.
Suddenly there
is going to be
ment among those present about what is factual.
disagree-
This example demonstrates that facts are social; they depend
on people agreeing about the evidence-and those agreements can
change. Today, small children learn that the earth is one of eight
planets that revolve around the sun in our solar system; this is
taught as a fact. When I was in school, though, I was taught that
there were nine planets. And a thousand years ago, people were
confident that the sun revolved around the earth-this was consid-
ered a fact. Similarly, in seventeenth-century Massachusetts, peo
ple considered the existence of witches to be a fact; today we
dismiss that beliefas ridiculous. We explain these changes in what
is considered factual in terms of improvements in people’s understanding of the evidence; this allows us to dismiss earlier factual
claims as erroneous.
What is deemed factual can also vary from group to group.
Whether it is considered a fact that a particular book is the actual
word of God depends on whom you ask. A group of believers may
affirm it as a fact, but a collection of people with more diverse reli-
gious beliefs will not necessarily agree.
Senator Daniel Patrick Moynihan (who was a social scientist
before he entered politics) reportedly said, “Everyone is entitled to
his own opinion, but not to his own facts.” This reveals our commonsense understanding that two contradictory statements can-
not both be factual. This is why the expression “alternative facts”
quickly became a target of ridicule. Critical thinking requires that,
when we are confronted with two antithetical claims, we weigh the
evidence. But there are other, less critically satisfying responses,
such
announcing that because you know that what your group
believes is true, anyone who says something different is wrong.
as
EVERYDAY ARGUMENTS
25]
Weighing
the
on
evidence
what the
agreement
evidence
or
the way
will not necessarily
toiimmediate
questi
People may
People may question
tacts are.
that
hat
lead
interpreted.
evidence is interpreted.
evidence
ediate
another’s
People who
they believe,eu
old
faceeof
cling
beliefs often
evidence
that strikes others as compelling. The historical:recordis
strong
filled with
to what
cases
where
people
believed
th
the end
prophecies that
hat
those
of
predictions har
So far,
of
was nigh.
the world
continued to hold
true b e l i e v e r s
and
most
yet
Wrong,
all
proved
tosh.
convictions.” Nor is the tendency to cling to discredited theoriae limited to religious
accept findings
believers. Scientists
that seemed to
have been
discredit
known to be slou
their positions.
We like to think that the facts are the facts, that thev are true, a
sort
be disputed. But what is considered
of last word that cannot
factual always reflects some social consensus: at some particular
time, there is agreement among
some
specific people that some-
thing is true. Critical thinking is a tool that can help us sort through
evidence for and against claims that something is factual. We may
conclude that that evidence supports the consensus, that we can
agree that a claim that something is
a
fact is well founded; but we
also need to understand that claiming that something is a fact is
not, in and of itself, enough to end debate.
Everyday Reasoning
Critical thinking is
something we all do, every day. We argue W
one another about
such everyvday matters as our tastes in mo
food, sports, and politics. Disagreeing with others,standing upfor
Our own
ideas, or being persuaded by someone else’s
ents
arg
can be fun; or we can agree to disagree, even tease those we disa-
Bree with about
their preferences. Most of these
discussio
126
EVERYDAY ARGUMENTS
re
casual and not very consequential, so we don’t worry too much
about the quality of the reasoning. But sometimes disagreements
grow heated, and we become frustrated when others don’t accept
our reasoning. As this chapter has tried to point out, mundane rea-
soning can be flawed, and it can help if we are able to examine it
critically.
We can be pretty good about thinking critically in the context
of everyday arguments, at least when we care enough to disagree.
Listen to two people debating the relative merits oftheir favorite
quarterbacks or their favorite television shows, and you can find
them offering up evidence to support their own positions and criti-
cizing the evidence for the other side. But in other cases, when we
already agree with one another or when we just don’t care very
much, we may not bother thinking critically about the evidence.
We just nod along with an anecdote or ignore ad hominem attacks.
That said, when flawed arguments spill over into serious
attempts to understand the world, critical thinking becomes very
important. The efforts of social scientists trying to improve our
understanding ofsocial life, for example, merit criticalevaluation.
This is the subject of the remaining chapters.
Critical Thinking Takeaways
Anecdotes are a weak form of evidence.
Ad hominem arguments and dismissing claims as “myths”
are ways of avoiding critical thinking.
Aphorisms and metaphors may contain assumptions that
need inspection.
Facts depend on social agreement.
EVERYDAY ARGUMENTs
[27]
The Logic of SocialScience
4
The goal of science is to better understand the world. Scientific
claims are evaluated by a particular set of standards: we make
observations of the world to acquire evidence for a claim; anv
claims inconsistent with that evidence are rejected.
The social sciences seek to apply these scientific standards to
understanding human behavior. This means that for sociologists
and other social scientists, critical thinking centers on eval-
uating evidence and the resulting explanations about how people
behave.
Patterns
Social science begins with trying to recognize patterns in social
These
patterns vary. Some
are
easy to spot:
men
life.
cannot bear chilD9o
dren, only women can do so. Others are harder to recognize.
students who sit near the front of a classroom earn better grades
than those who sit in the back? We might suspect this could be
true;
front
imagine that not all students sitting near the
near
receive particularly high grades, just as some students sitting
De
the back do really well. Still, we might predict that there wil
but we
can
also
a
28]
tendency for those who sit near the front to
that is, that there will be a pattern.
If we want to go beyond
exist, however,
ourselves
that
we
ll need to
receive higher grades-
speculating that this pattern might
gather evidence, both to confirm to
prediction is correct and to convince others. We
could, for example, keep track of where students sit in a
particular
class, then check their grades. But even if our hypothesis is confirmed and we find that students who sat near the front tended to
receive higher grades, other people might challenge our finding
our
arguing, for instance, that evidence from that one class hardly
proves that the same pattern will be found in all classes. Such chal-
lenges are a form of critical thinking, and most research will face
such critiques. As we will see in later chapters, deciding how to
gather the best evidence can be complicated.
Causality
Identifying a pattern is not enough. People are likely to ask why that
particular pattern exists: they will want an explanation for the pattern. Explanations involve an argument that a certain cause pro-
has to meet
duces a certain effect. Basically, every causal argument
four criteria.’The namesgiven to these criteria vary, but it is impor
tant to understand what each involves.
Precedence
The first and
OCcur
simplest
criterion is
before the effect. In
our
precedence: the
example,
the grades
The classroom occurs first, and
SO we can say
cause
where students sit in
they receive come
that where
that it is at least plausible
THE
LOGIC
OF
SOCIAL
has to
SCIENCE
29|
later.
students
sit
at
may infuence-be
least
part
of the c a u s e – f o r s
the grades they
receive.
Notice that it
would make
no sense
to argue that t
ades stuthey
but they
them
em ta
to sit
dents received afterthe
where
This probably seems obvious.
met.
the class
class ended
caused
did when
even
errar
this
make
researchers occasionally
or instan
tinguished
Howard S. Becker,
a
very fine sociologist, argued thas
Congrress
passed the Marihuana Tax Actof1937 (the originalfederallaw
hibiting marijuana)
after the Federal Bureau of Narcoties
waged a
public relations campaign that led major magazines to publisha
cles about the drug’s dangers. These articles then aroused n.t
ublic
opinion, which led to pressure on Congress to pass the bill. pre-
senting evidence to support his argument, Becker noted thath
the
Reader’s Guide to Periodical Literature (in those days the leadina
index for articles in popular magaz1nes) showed that coverage of
marijuana peaked in the index volume covering July 1937-June
1939: there were seventeen articles about marijuana indexed in
that volume of the Reader’s Guide, whereas no other volume of the
index between 1925 and 1951 listed morethan 4 articles. Presumably
those magazine articles must have helped inspire the public con-
that led to the bill’s passage in July
1937
While at first Becker’s argument might
cern
seem
persuasive,
Donald T. Dickson looked more carefully at the dates when tne
magazine articles were published and noted that “no articles
appeared in the five months preceding the House committee
nca
ngs on the act in late April and early May, one appearea m July
1937, and the rest
appeared after the bill was signed into la
on
August2,1937” In other words, what Becker identifiedas acause
magazine articles that
supposedly aroused the pudile
130
THE LOGIC OF
sOCIAL SCIEN
Congress take
action-actualy occurred after the
supposed effect
h e act’s passage). In this case, the
criterion of
precedence is
violated.
In
many cases, precedence cannot be
acclear-cut
tice,
as
the
precedence
infuence Y, but
complicated
established by something
publication dates of magazine articles. In
praccan be
complicated by feedback: that is, X
may
then Y
proceeds
to
influence
X.
This
can
chicken-or-egg debates
lead
to
about, for instance, whether
eulture precedes and thereby causes particular social structures
to
emerge, or whether social structure precedes and similarly causes
particular cultures to develop.
Patterned Variation
This is a relatively straightforward idea: there needs to be a pattern
between the cause and effect. If I flip the light switch up and the
light goes on-or down, and it goes off-this pattern makes it reasonable to suspect that flipping the switch causes the light to tum
on or off. That is, our cause and effect need to vary in a patterned
often are not so straightforward.
way. Of course, causal patterns
correlated
We probably won’t find that course grades are perfectly
such that all of the stuwith where students sit in our classroom,
to the front, and so on.
dents with high grades are sitting closest
toward the front
likely to find that students sitting
Rather,
do
grades-they tend to
are somewhat more likely to get higher
we are
better.
Similarly,
researchers
find that smokers
are more
likely to
some
even though
than nonsmokers,
develop various diseases
real world,
nonsmokers do. In the
Smokers don’t get sick and some
tendencies-the cause
patterned variation involves
SCIENCE
OF
THE LOGIC
SOCIAL
makes it
131
more
effect
ikely that the
terns
will occur. Identifying and evaluatinae
often requires using
that some possible
cause
staustics
that
measure
shapes the effect.
such pat-
the litel:
kelihood
Rationale
for establishing causality involves our abilitu s
The third criterion
to
explain why the cause ought to shape the effect. Thus, I ght
explain that flipping the light switch up closes an electrical
uit,
which causes the current to fiow to the light bulb, in which a heated
filament produces light; or I might argue that students who sit in
the front of the classroom are more likely to pay attention and less
likely to be distracted by posts on social media than those who sit
farther back, so that the students in front learn more and thereby
perform better on tests, leading to higher grades. And I mightfur
ther be able to connect my explanations to authorities who have
written about how electric currents work or howfocusingimproves
learning. All of this is fairly straightforward. All causal arguments
need such rationales.
Nonspuriousness
NOnspuriousness: a fancy word, but an important one. An appar
ently causal relationship-one that meets the standards ot
prece
aence, patterned variation, and rationale-stil might be invalid
because that
factor.
ere
relationship is spurious,: that is, caused by sonie
t may help to begin with a silly example. SupPpose
hat,
alter observing the switch-light pattern-up, on; dow, o*
announce that
flicking the switch down is causing the light to turn
32
THE LOIC OF
SOCIAL SCIENCE
off, and vice versa. But Tonya responds,
“No, the light is controlled
invisible leprechauns who
mischievously choose to turn the
light on or off whenever you flick the switch.
The real cause of the
on
is
the
light going
leprechauns’ magical powers!”
by
es or ait
AsI say, this is a
ridiculous
out of hand-but
ectrical.cuca
which aheae
ents who sit
ention and es
solid
rationale,
it
works, theories that
have been tested in countless
experiments, so we have a great deal
of confidence in our rationale.
Moreover, we don’t have any evi-
dence that
leprechauns exist. “But,” Tonya responds, “that’s
because leprechauns’ magical powers allow them to avoid
being
detected.”
re and there
Can we
nd Imightfe
why? Well, first of all, we have a
elaborate theories of electricity and how
those who s.
n
objection, one we are likely to reject
to go
absolutely prove that leprechauns don’t cause the lights
on and off? Well, no. But there is a
very old philosophical
principle-often called Occam’s razor-that when we have two
aties who hare
explanations (in this case, the electrical-circuit explanation and the
singimproves
electrical-circuit-plus-leprechauns explanation) thatpredictequally
sal argumens
well, we should favor the simpler explanation. Thatis,ifwe canade
quately explain the lights going on and off without incorporating
leprechauns in our explanation, we should lose the leprechauns.
Occam’s razor allows us to dismiss explanations that invoke
various unobservable causes (such as leprechauns). But charges of
spuriousness-that some other cause is at work-can take serious
n e . A na p a r
ards
o prer
t
t
b e
i n a l i
Y
S O mte
hui
forms. Suppose we want to argue that smoking causes lung cancer.
Ted might object: smokers, he observes, tend to drink more alcohol
than nonsmokers; perhaps it is alcohol that causes lung cancer, or
perhaps it is the combination oftobacco plus alcohol. Ted’s critique
may seem more plausible than invoking leprechauns, and it cannot
Y p o s et h
be dismissed out of hand. We will need to look for more evidence,
perhaps by comparing lung cancer rates among various groups:
ghr
ot7
SCIENCE
THE LOGIC OF SOCIAL
[33
drinkers
who don’t smoke
who don’t drink,
Ke,
nonsmokers, smokers
o u r n e w evidence shows that
smokers who drink. And suppose
and
increase the risk of
lung cancer,
even
after
smoking does seem to
into account. “Okay,” Ted might say, “but smok.
we take drinking
ers
also drink
up a
new
more
critique that sets
coffee than nonsmokers”-a
round of tests.
When can we absolutely, positively declare that a relationship
is nonspurious? That
is,
when
can we
say
we
have
identified
the cause of some effect, and there is no other possible explana
tion? The answer may seem a little disturbing: never. It is alwavs
possible for a critic to argue that some
other factor may explain the
relationship between what we think is the cause
and what we con-
sider the effect of that cause. Now, to be sure, we may compile vasi
amounts of evidence consistent with our explanation, such as the
thousands ofstudies supporting the conclusion that smoking dam-
ages human health-so much evidence that it seems very unlikely
that tobacco isn’t dangerous and we have great confidence in
declaring that it is harmful. And yet, we can never completely
exclude the possibility that this well-documented relationship
might be spurious.
This is why critical thinking is so important. Every explanation
can be challenged. But those challenges can themselves be evalu-
ated. One cannot simply announce that all scientific knowledge
must be wrong, that the world is in fact being run by leprechauns.
In discussions among scientists, challenges must be subject to the
same sorts of evaluations as the arguments being challenged. That
is, we expect both those offering explanations and those challeng
ing them to support their claims with evidence, and all thatevi
dence must be weighed
andjudged. We musthold both explanations
and challenges to the same
high standards.
34]
THE L0GIC OF SOCIAL
SCIENCE
COns abc
e manner in
Judging Social Scientific Claims
Tudging scientific
reasoning-and
that includes the
social
sciences-revolves around assessing evidence. Claims must
be
supported by presenting evidence consistent with what
is being
claimed, and critics must be able to evaluate that
evidence.
Because evidence is central to
science, scientists are under an
to
be
honest in
obligation
reporting their evidence. They are
expected to find the best possible evidence, to explain
clearly how
they went about assembling and analyzing that
evidence, and to
their
report
findings in a
and
thorough
accurate manner. It is
con
sidered scandalous when scientists are found to have
behaved dis-
honestly, and being implicated in a single scandal can destroy an
individual’s entire scientific reputation.5
Such scandals aside, debates often arise around the
quality and
interpretation of evidence. Any one study is inevitably flawed
there is
such
thing as perfection-and critics can always raise
legitimate questions about its evidence. For instance, they can
no
argue that the manner in which the researcher gathered evidence
or
the methods chosen to
analyze that evidence may have affected
the results. A single research
report is unlikely to be taken as the
last word on anything, which is
why the news media’s tendency to
hype dramatic new research “breakthroughs” often fosters the
spread of bad information. Because every study has limitations
that
might have affected its results-and researchers-like everyone
elsemay find it difficult to think critically about what might
be
wrong with their research. Therefore, other researchers, rather
than
simply accepting the original claims, may be inspired to conduct their own studies: either to
replicate the first study to see
whether following the reported procedures a second time
THE LOGIC 0F SOCIAL SCIENCE
[35]
different proceduress
or to use slightly
produces the same findings,
have shaped the find
the original techniques may
to see whether
reveal whether the pat
of this work can help
ings. The
outcome
terned variation is spurious.
Debating evidence is the central focus of most critical thinking
about research in the social sciences. If this seems surprising, it
shouldn’t. This chapter has used arguments that may seem
straightforward. I chose the example of leprechauns causing lights
to go on and off precisely because it is ridiculous. And while it
might seem reasonable that sitting near the front of the classroom
leads to higher grades, there are no doubt many, many reasons why
students get the grades they do-how much and how well they
studied, were they healthy or sick when they took their tests, and
on and on. The claims that smoking causes disease is an argument
that is now very familiar, but it has a long history. The tobacco
industry waged a decades-long campaign to challenge researchers
claims that
smoking was dangerous; it mounted dozens of argu-
ments that the apparent link between smoking and disease was a
spurious relationship, that the true culprit might be alcohol or coffee
or…
you get the idea. Eventually, a vast research literature
consisting of thousands of studies using different research designs
established an edifice of evidence that has convinced
that smoking is indeed
All scientific
risky.6
most
people
knowledge, then, rests on a foundation of evi
foundation, the more confidence
dence. The bigger and better the
have in what we know. Given
the mountain of assembled evidence, very few people today doubt that
smoking is harmful. Still,
it is
always possible that a relationship
generally considered causal
is in fact
spurious. What we think we know
today may be challengea
if
compelling new evidence emerges
tomorrow. And most ques
we
36)
THE LOGIc OF
SOCIAL SCIENCE
tions that we care
about-what causes
disease, for example, or what
leads to better grades-are likely to have
that evaluating the evidence can
become
This is why most
undergraduate
and
complicated answers, so
a
very elaborate process.
social sciences feature required courses ingraduate programs in the
statistics and methodology. While on the
surface these
topics may seem less
interesting
substantive subjects, they
provide essential lessons for
social scientists who need to understand how to
than
more
so as to
produce
conduct research
the most
convincing evidence possible. Indeed,
understanding best practices can give all students-not just those
who
plan
become researchers-the tools needed
to assess
reports about research results. Everyone needs to understand the
pitfalls of doing research in ways that are likely to distort one’s findto
ings, because throughout our lives we will encounter claims about
what researchers have
found,
and
being
an
informed citizen
requires being able to think critically about those reports.
The Importance ofEvidence
Social science involves a search for knowledge that is supported by
the best available evidence. This evidence is never perfect; it is
always subject to critical evaluation. Science advances not through
pronouncements about what is true, but by dialogues between
those making claims and critics seeking to weigh the strength of
evidence.
Evidence is central to all the social sciences, but because the
various disciplines examine somewhat different topics and ask dis-
tinctive questions, the specific challenges for critical thinking vary
among the social sciences.. So let’s turn our attention from the
sOCial sciences in general and focus on sociology.
THE LOGIC OF soCIAL SCIENCE
37]
Critical Thinking Takeaways
Causal explanations
dence, patterned
is never
are
judged by the standards of Drer
ce
variation, rationale,
possible to establish
that
a
and nonspuriose
usness.
relationship absoluteiwis
i.
not spurous.
Most critical thinking in the social sciences involves judging
the quality ofevidence.
research design
Three Purposes of Research
• Exploration
• Get a better understanding
• Test feasibility of more extensive study
• Description
• Detail phenomenon
• Explanation
• Answer “why” questions
The Logic of Nomothetic
Explanation
•Goal: to find a few factors that can
account for many of the variations
in a given phenomenon, but not
necessarily all
Causation vs. Correlation
Three Criteria for Nomothetic
Causality
The variables must be correlated
1.
■
Correlation – An empirical relationship between two variables
such that changes in one are associated with changes in the
other, or particular attributes in one are associated with
particular attributes in the other.
The cause takes place before the effect (time order)
The variables are non-spurious
2.
3.
■
Spurious Relationship – A coincidental statistical correlation
between two variables shown to be caused by some third
variable
Spurious Relationship
Example
Necessary and Sufficient Causes
• A necessary cause represents a condition that
must be present for the effect to follow.
• A sufficient cause represents a condition that, if
present, guarantees the effect in question.
• Most satisfying outcomes in research include
both necessary and sufficient causes.
Units of Analysis
• Individuals :
information about
individuals is
aggregated and
generalized
• Groups : information
about a collective is
aggregated and
generalized
• Organizations :
information about formal
social organizations is
aggregated and
generalized
• Social Interactions :
information about what
goes on between
individuals, groups or
organizations
• Social Artifacts :
information from a
product of social beings
is aggregated and
generalized
Ecological Fallacy in 3 Min
The Time Dimension
• Cross-Sectional Study – A study based on
observations representing a single point in time, a
cross section of a population.
The Time Dimension
• Longitudinal Study – A study design involving the
collection of data at different points in time.
Conceptualization &
Operationalization
Levels of Measurement
Why does this matter?
OK to compute….
Nominal
Ordinal
Interval
Ratio
frequency distribution.
Yes
Yes
Yes
Yes
median and percentiles.
No
Yes
Yes
Yes
add or subtract.
No
No
Yes
Yes
mean, standard deviation,
standard error of the mean.
No
No
Yes
Yes
ratio, or coefficient of variation.
No
No
No
Yes
Reliability & Validity
Reliability & Validity
Conceptualization
and Measurement
Observing the social world
⦿
Everyday observations are haphazard,
inconsistent, and filled with bias
⦿
Scientific Measurement – Careful,
deliberate observations of the real world
for the purpose of describing objects
and events in terms of the attributes
composing the variable.
What are we measuring?
⦿
Conceptualization – The process through
which we specify what we mean when
we use particular terms in research.
⦿
Uses
› Deductive—translate abstract into testable
hypotheses involving specific variables
› Inductive—make sense of related
observations
What are we measuring?
⦿
Concepts are fuzzy notions of things in the
real world
›
›
⦿
Race
Self-esteem
An attribute is a characteristic or quality of
something (ex: female, old, student).
A variable is a logical set of attributes (ex:
gender, age).
⦿
›
›
The attributes composing it should be
exhaustive.
Attributes must be mutually exclusive.
What are we measuring?
⦿
Selecting Variables
›
›
›
›
Theories
Utility of variables in previous research
Constraints and opportunities in research
setting
Relevance of variables to analysis at hand
How will we measure?
⦿
Operationalization is the process to
devise operations that actually measure
intended concepts
› Goal🡪 achieve measurement validity
1.
2.
3.
4.
Define concepts
Choose variables , considering purpose of
research
Identify construct indicators
Feasibility
Levels of Measurement
⦿
The mathematical precision with which
the values of a variable can be expressed
› The nominal level of measurement, which is
qualitative, has no mathematical interpretation
› The quantitative levels of measurement—ordinal,
interval, and ratio—are progressively more precise
mathematically.
⦿
Level of measurement is not inherent in
variable, but is chosen by the researcher.
Levels of Measurement–NOIR
Reliability and Validity
Increasing Reliability
Test-Retest Method
■ Make the same measurement more than
once.
Established Measures
Already tested for reliability
Reliability of Research Workers
Inter-observer reliability
QUALITATIVE FIELD
RESEARCH
Topics Appropriate for Field Research
◻
◻
◻
Topics that defy simple quantification
Attitudes and behaviors best understood in their
natural setting
Social processes over time
Elements of Social Life Appropriate to
Field Research
🞑
🞑
🞑
🞑
🞑
🞑
🞑
🞑
🞑
Practices–talking
Episodes–short life events like divorce, crime, illness
Encounters–dating
Roles and Social Types–occupations, social positions
Social and Personal Relationships–family relationships
Groups and Cliques–athletic teams, Greek life
Organizations–workplaces, schools, hospitals
Settlements and Habitats–ethnic enclaves, suburban
neighborhoods
Subcultures and Lifestyles–polyamory, the “elite”,
soccer fans
Special Considerations in Qualitative
Field Research
◻
Roles of the Observer
🞑
Participant, Researcher, Observer
■
◻
Reactivity – The problem that the subjects of social
research may react to the fact of being studied, thus
altering their behavior from what it would have been
normally.
Relations to Subjects
🞑
🞑
🞑
Objectivity
Insider/outsider
Reflexivity
Approaches to Qualitative Research
◻
Naturalism – An approach to field research based
on the assumption that an objective social reality
exists and can be observed and reported
accurately.
🞑
Ethnography – A report on social life that focuses on
detailed and accurate description rather than
explanation.
■
Netnography
Approaches to Qualitative Research
◻
◻
Case Studies – The in-depth examination of a
single instance of some social phenomenon.
Extended Case Method – A technique in which
case study observations are used to discover flaws
in and to improve existing social theories.
Approaches to Qualitative Research
◻
Qualitative Interview – Contrasted with survey
interviewing, the qualitative interview is based on
a set of topics to be discussed in depth rather than
based on the use of standardized questions.
🞑
“Miner” or “Traveler”
Approaches to Qualitative Research
◻
Focus Group – A group of unrelated subjects
interviewed together, prompting a discussion.
🞑
🞑
Advantages: real-life data, flexible, high degree of face
validity, fast, inexpensive
Disadvantages: not representative, little interviewer
control, difficult analysis, interviewer/moderator
skills, difficult logistically
Approaches to Qualitative Research
◻
Field Research—conducted within the context of
everyday activities
🞑
🞑
🞑
Sometimes called participant observation
Uses various forms of qualitative research
Good for understanding mechanism and seeing the
social world as the research subject sees it
Conducting Qualitative Field Research
◻
Preparing for the Field
🞑
🞑
🞑
Be familiar with relevant research
Discuss your plans with others
Identify and meet informants (when appropriate)
■
■
🞑
First impressions are important
Establish rapport (an open and trusting relationship)
Ethical considerations?
Conducting Qualitative Field Research
◻
Recording Observations
🞑
🞑
🞑
🞑
🞑
Take detailed notes, but balance with observations
Rewrite notes with observations soon after being “in
the field” with filled in details
Record empirical observations and interpretations
Record everything
Anticipate observations
Strengths and Weaknesses of
Qualitative Field Research
◻
Strengths of Qualitative Field Research
🞑
🞑
🞑
◻
Effective for studying subtle nuances in attitudes and
behaviors and social processes over time
Flexibility
Inexpensive
Weaknesses of Qualitative Field Research
🞑
No appropriate statistical analyses
Sampling
This is a great lesson for
flashcards and/or visual notes (i.e.
draw things out!)
Key Concepts
■
Element
■
■
Population
■
■
Collection of elements from which the sampling frame is actually
selected (e.g. Twitter accounts in a particular community)
Sampling Frame
■
■
Specified collection of study elements (e.g. all Twitter accounts)
Study Population
■
■
The unit of which a population is composed and which is selected in a
sample. (e.g. a Twitter account)
List of elements (e.g. list of Twitter accounts in a particular
community)
Sample
■
Who actually gets surveyed, interviewed, etc. (e.g. ILUVSTARWARS,
WookieLife, and LeiaRulesTheGalaxy accounts)
Why sample?
■
Research rests on observation
■
■
■
■
Must decide what to and what not to observe
Can’t observe every element usually
Sampling is the process of selecting
observations
Two Types
■
■
Non-probability
Probability
Non-Probability Sampling
■
May introduce bias because those
selected are not typical or necessarily
representative of the larger population
■
■
Cannot generalize from data to population
Cannot use statistics to support conclusions
Non-Probability Sampling
Techniques
• Easily accessible
Convenience/
• Questionable
Availability
generalizability
Snowball
• Good for difficult to
locate/identify pop.
• Bias
Non-Probability Sampling
Purposive
Quota
• Characteristic driven
• Representativeness
• Measured Proportions
• Addresses
representativeness…sort of
Non-Probability Sampling
Techniques
Theory and Logic of
Probability Sampling
Populations are heterogeneous.
Samples should approximate populations.
■
Reduce bias
■
■
Sample should be typical of population its drawn
from
Bias results from conscious and unconscious
processes
■
■
■
Convenience
Don’t “look” like someone who can help me
Self-selection biases
Theory and Logic of
Probability Sampling
■
Representativeness
■
■
Probability of Selection
■
■
Aggregate characteristics of interest should match
Equal non-zero chance of selection into sample
Key Advantages
■
■
More likely to match population
Level of inaccuracy can be measured
Perfect Probability Sample
Less-Than-Perfect Probability
Sample
In Other Words…
Probability Sampling
Techniques
■
Simple Random Sampling
■
■
■
Method of sampling that relies on a
random, or chance, selection of elements
Every element of the sampling frame has a
known probability of being selected
Process
■
■
■
Assign numbers to sampling frame elements
Generate set of random numbers
Select sample from sampling frame using
generated numbers
Probability Sampling
Techniques
■
Systematic Random Sampling
■
Method of sampling in which sample elements are
selected from a list or from sequential files
■
■
■
■
Sampling Interval, or K
■
■
■
Randomize list to avoid periodicity
Random start
Every Kth element is selected
Frame=1000, N=200, K=1000/200 or 5
Every 5th unit is selected
Sampling ratio
■
1/5
Periodicity
●
If the sampling interval is 8
for a study in this
neighborhood, every element
of the sample will be a house
on the northwest
corner—and thus the sample
will be biased.
●
That’s not good! Corner
houses, for instance,
regularly differ in how much
they pay in taxes because
they are often larger lots, so
this may introduce bias.
Probability Sampling
Techniques–Simplest
Probability Sampling
Techniques
■
Stratified Random Sampling
■
Method of sampling in which sample elements are
selected separately from strata identified in
advance by the researcher
■
■
■
■
■
Modification of previous two designs
Stratify into homogenous groups first
Reduces sampling error to zero on the stratified
variable (e.g., race) because subset is
homogeneous
Proportionate Stratified Sampling
Disproportionate Stratified Sampling
Probability Sampling
Techniques
■
(Multistage) Cluster Sampling
■
■
Clusters—naturally occurring, mixed aggregates of
elements in a population
■
■
■
■
■
Sampling in which elements are selected in two or more stages,
with the first stage being the random selection of naturally
occurring clusters and the last stage being the random selection of
elements within clusters
Schools🡪students
Counties🡪general population
Businesses🡪employees
Blocks🡪city residents
Repetition of listing and sampling
■
E.g., all church members, all elementary students
Probability Sampling
Techniques–More Complex
Remember…
■
■
One of the determinants of sample
quality is sample size.
Samples will be more representative
of the population if they are relatively
large and selected through probability
sampling methods.
SURVEY RESEARCH
Topics Appropriate for Survey
Research
◻
◻
Descriptive, exploratory, and explanatory
Surveys are most appropriate when
researchers
🞑
Want to collect large samples
🞑
Want to collect original data
🞑
Are interested in measuring attitudes and/or orientations
◻ Any decision about the best survey design
for any particular study must take into
account the unique features and goals of the
study.
Designing a Survey
What types of
questions are
appropriate for
questionnaires?
What are the dos
and don’ts for
creating “good”
questions?
What are the parts
of a
questionnaire?
Questionnaire Design
Process
Good and Bad Survey
Design
Great
examples
that will help
you
complete
this week’s
assignment!
Survey Administration Evolution
Interview Surveys
◻
◻
Interview – A data-collection encounter
in which one person (interviewer) asks
questions of another (respondent).
This is different than a qualitative
interview!
🞑
🞑
Interviewer must administer every
questionnaire exactly the same to avoid
introducing bias.
Interviewer does not add words or ask followup questions.
Interview Surveys–Part 2
◻
◻
◻
◻
◻
Interviewers solicit higher response
rates (80-85%) than mail surveys.
Interviews minimize “don’t know” and
“no answer.”
Interviewers serve as a guard against
confusion.
Interviewers can observe respondents
while completing the questionnaire.
Costly and time consuming relative to
other methods of administering surveys
Comparison of the Different
Survey Methods
◻
Mailed Surveys
Cheaper and faster than face-to-face interviews
🞑 Low cost to mail
🞑 Requires small staff
🞑 Greater willingness to answer controversial items
🞑
◻
Interview Surveys
Fewer incomplete questionnaires
🞑 More effective for complicated questionnaires
🞑 Face-to-face is more intimate
🞑
◻
Telephone Surveys
Cheaper and more time efficient
🞑 Increasingly difficult when caller ID became widespread
🞑
◻
Online/Email Surveys
Lots of available software and websites for survey construction
🞑 Arguably the norm now
🞑
Strengths and Weaknesses of
Survey Research
◻
Strengths
🞑
🞑
🞑
🞑
◻
Useful in describing large populations
Make large samples possible
Surveys are flexible
Standardized questions
Weaknesses
🞑
🞑
🞑
🞑
🞑
Round pegs in square holes
Seldom deal with context of responses
Inflexible
Artificial
Weak on validity (but strong on reliability)
Common Problems and Alternatives
◻ Check out this article, specifically the table, to see examples
of common problems and better alternatives
Firefox
File
Edit
View
History
Bookmarks
Tools
Window
Help
A
00
*
Wed Nov 3 8:33 PM
k Kindle Cloud Reader
Х
+
E
个
C
IS THAT TRUE?
from games of chance involving flipping coins,
rolling dice, or dealing cards. These are easily
understood: if you flip a fairly balanced coin, there
is a 50 percent chance it will come up heads; flip
two coins, and there is a 25 percent chance both
will be heads. This is because there is a 50 percent
chance that the first coin will land heads, and on
50 percent of those occasions, the second coin will
also show heads (.5 .5 = .25). It is nearly as simple
to determine that rolling one fairly balanced, six-
sided die will show one dot one-sixth (16.67
percent) of the time; if you roll two dice, the chance
that both will come up ones is 2.78 percent (.1667
x.1667 = .0278), or once every thirty-six rolls.
These are neatly bounded examples; the whole
point of dice is to produce a random outcome on
each roll, yet we know that this will produce clear
patterns if we roll them enough times. Thus, on
average, for every thirty-six rolls of two dice, we
should get a total of two once, while we can expect
to get some combination of seven (1 + 6,2 + 5,3 +
4) six times in all.
While it is possible to apply probabilistic
thinking to the patterns in people’s lives, we know
that social life is not neatly bounded. Insurance
provides a relatively clear example. Insurance
companies employ actuaries, people who calculate
the odds of bad things happening-such as auto
accidents, fires, or deaths—and then set rates
based on those odds. Most drivers won’t have a
>
=
www.
dom
the why
Page 129
900 SOBOSS BORSOZAOKE
https://read.amazon.com/KindleReaderApp#
353
NOV
3
ty
A
.
280
Firefox
File
Edit
View
History
Bookmarks
Tools
Window
Help
A
00
*
Wed Nov 3 8:33 PM
k Kindle Cloud Reader
Х
+
E
个
C
IS THAT TRUE?
costly accident in the coming year, but some will,
and the insurance company is willing to take your
bet: you pay a premium and they promise to pay
off if you have a costly accident. Actuaries know
that some drivers, such as experienced drivers who
aren’t too old, or people who haven’t received a lot
of tickets, are less likely to get in accidents, and
so the insurance company can charge those low-
risk drivers lower premiums. There are a couple
hundred million drivers in the United States,
so actuaries have plenty of data with which to
work. They can’t know precisely which drivers will
have an accident this year, but they wouldn’t be
surprised to learn that Xavier was not involved in
an accident this year, or that Wanda was. The point
is, they have a pretty good idea about the general
pattern—what the total number of accidents will
be—and that allows them to calculate appropriate
premiums. This is little different than a casino
knowing the odds of different outcomes in dice
games and setting a payoff structure that insures
that in the long run it will make a profit-except
that the odds calculated by actuaries are not as
precise as those for games of chance.
In effect, when sociologists conduct research and
identify patterns (for instance, that this category
of people is more likely to do X than some other
category of people), they are producing very crude
data of the sort that provides the basis for casinos’
and actuaries’ calculations. Notice that actuaries
>
wal
Page 130
9000 BOROSZOR
https://read.amazon.com/KindleReaderApp#
353
NOV
3
ty
A
.
280
Firefox
File
Edit
View
History
Bookmarks
Tools
Window
Help
A
00
*
Wed Nov 3 8:33 PM
k Kindle Cloud Reader
Х
+
E
个
C
IS THAT TRUE?
can draw upon various sources for their data-
such as police reports of traffic accidents and
insurance claim records from previous years—to
predict how many accidents will occur in the
coming year. Sociologists usually work with much
less data-often just what they have themselves
collected—so any estimates they make are likely to
be much rougher than the actuaries’ forecasts.
But just as actuaries use their data to predict
the number of traffic accidents, which then
serves a basis for calculating reasonable
premiums, sociologists use their findings to make
generalizations about patterns in social life. To
be sure, they cannot confidently predict precisely
how Sarah will act, but they can describe a pattern
based on the behavior of people in the category
that Sarah belongs to.
This is why sociologists become impatient with
people who use anecdotes to discount their
findings. Imagine a sociologist whose research
shows that older people tend to have conservative
political beliefs. In response, Paul says, “That’s not
true—my grandma and grandpa are very liberal.”
”
This might be a telling critique if our sociologist
had said that all older people are conservative.
In that case, finding even one contradictory
example would be enough to challenge the claim.
But in identifying a tendency for older people
to be conservative, the sociologist acknowledges
that there will be some liberals among older
>
as
but you per a mund the prope
m.
wing the one in die
www
krimine These che
wake. They
war…
Page 131
900 OSLOBO
https://read.amazon.com/KindleReaderApp#
353
NOV
3
ty
A
.
280