The Scientific Model
- A study by Penelope Lockwood and Zivz Kunda was conducted about the effects of
information about “superstars” on other individuals it is a social comparison
- Lockwood and Kunda hypothesized that being exposed to information about an
extremely successful person would make some participants feel negatively about
- This prediction is consistent to the study done by Thornton and Moore,
which showed that participants who sat beside a poster-board displaying
photos of attractive people rated themselves more negatively than did
participants in control settings.
- But Lockwood and Kunda also predicted that some other participants would
actually feel better about themselves after exposure to a superstar.
- Lockwood and Kunda suggest that the critical factor is whether or not people think
that they themselves have the potential to be very successful.
- They predicted that students who were still in their first year would be inspired by
the stellar student because they still had lots of time to achieve similar success.
Fourth year students, however, would feel worse because they had little to no
chance of achieving similar success.
- The results confirm these predictions.
- Social psychology is a science, it is not, however, simple to study social behavior
- Social psychologists are interested in understanding spontaneous social behavior,
which is difficult to study in tightly controlled settings.
- Social psychologists had to be very creative in using the scientific method to study
how individuals are influenced by other people.
Theories and Hypotheses
- Two key elements of the scientific method are theories and hypotheses.
- Theories are like explanations, and hypotheses are predictions.
- More specifically, a theory is a scientist’s explanation of why an event or outcome
occurs; it identifies the underlying causes of something the scientist has observed.
- In psychology theories typically focus on psychological processes to explain events.
- Whereas theories provide a framework for understanding why something occurs,
hypotheses are specific predictions about what should occur if a theory is valid.
- Hypotheses are derived from a theory and, therefore, provide a means for testing
- Most theories build on prior scientific work, including previous theories that have
ben shown to be inaccurate or limited.
- Theories often involve applying a concept or principle from one field to another
related but distinct field.
- theories also frequently rely on scientists’ intuition analysis of problems, including
their personal experiences. - Some theories are the result of collaborations between scientists who have
- In developing a theory, scientists aim for simplicity, coherence and testability,
because these features make it more likely that the theory will generate new ideas
and new discoveries.
- Melvin Lerner proposed Just World Theory in which he suggested that humans
need to believe that the world is a fair and just place.
- Lerner proposed that we are all motivated to believe that people usually receive
what they deserve: hard work and honest bring rewards, whereas laziness and
dishonesty do not pay.
- Lerner argued that if we did not believe that that the world was largely fair, then
we would fear that our own efforts and investments might not pay off.
- That is, if we believed that the world is unjust and that hard-working people do not
necessarily succeed, then we would fear that our own hard work would be for
- Hypotheses that can be derived from this theory are:
When asked, most people will say that the world is generally fair. After all, if
just world theory is valid, then people should report that the world is just.
Lerner also derived a more direct hypothesis that we want to protect out
beliefs in a just world. He made the prediction that people would blame or
derogate victims who are suffering from negative circumstances, especially
when their suffering is expected to continue.
o This prediction was based on the logic that suffering victims threaten
the belief that the world is fair, unless the victims are either
responsible for their suffering or are bad people who in some sense
deserve their suffering. Therefore, if we are truly motivated to defend
our belief that the world is fair, we might try to convince ourselves
that people who are suffering brought it on themselves or are bad
people who deserve their fate.
- Lerner’s hypothesis seems consistent with some people’s devaluations of these
Translating Theoretical ideas into Testable Questions: Operational Definitions
- Theories and hypotheses in social psychology are typically expressed in conceptual
terms: they refer to abstract ideas or concepts that cannot be observed directly.
- Therefore, to test theories or hypotheses, researchers must somehow translate
abstract ideas into concrete, objective measures.
- An operational definition of a concept is a specific, observable response that will
be used to measure the concept.
- For instance, people’s attitudes towards religion could be operationally
defined by asking respondents to indicate how unfavorable of favorable they
are toward religion on a scale of 0 to 10.
- The concept is attitudes towards religion; the operation definition is scores
from 0 to 10 on a response scale. - Two types of measures are most common in social psychology: self-report
measures and behavioral measures.
- To measure many social psychological concepts, the easiest strategy is to ask
- So long as a concept is something that people are able and willing to report,
measuring it via self-report questions make sense.
- Of course, even when people are able and willing to report a concept, researchers
must be careful to express self-report questions clearly.
- It can be difficult to answer a question if the meaning is not well specified.
- The same question can often be asked in different ways, and subtle changes in
phrasing and wording can significantly affect responses. As an illustration of this,
Elizabeth Loftus and John Palmer asked people to watch a short film about a car
accident. Everybody saw the same film, but different questions were asked to
different groups of viewers using different verbs, one asked “how fast were the cars
going when they hit” and when asked “how fast were the cars going when they
smashed.” On average, the people who had the word hit answer 13kM, while the
smashed group answered 18kM.
- These different answers occurred because of the word choices the
researchers chose to use.
- Sometimes the assumption that people are able and willing to report a concept is
- People may believe that they can answer these questions correctly, but their
responses will probably be guesses.
- Perhaps even more common than being unaware of a concept, participants in
social psychological research may not want to report some things honestly.
- For instance respndents may be motivated to create a positive impression
of themselves. Responding in this manner is known as socially desirable
responding giving answers that portray the respondent in a favorable
- In order to control contamination of their measures, social psychologists can avoid
phrasing terms in ways that make responses more socially desirable than others.
- A second strategy is to obtain a measure of participants’ general tendencies to
respond in a desirable manner on all measures, which can then be used to control
desirable responding on a specific measure.
- Example: Delroy Paulhus developed a scale called the Balance Inventory of
Desirable Responding, which provides scores for respondents’ tendencies to
present themselves in a favorable light.
- Because people may be unable or unwilling to report some things accurately,
researchers sometimes measure concepts by observing individuals’ behaviors. - One advantage of behavioral measures is that they are often unobtrusive
measures, which means that participants do not realize that the measure is being
- If participants are unaware that a measure is being taken, then presumably they
will not try to alter their actions to create a favorable impression.
- A disadvantage of behavioral measures is they that can be difficult or time-
consuming to obtain.
- A second disadvantage is that behavioral measures are very difficult or impossible
for some concepts.
- Self-report measure are much more common is social psychology.
- No matter how a concept is operationally defined, the goal is to measure the
- The problem of the accuracy of psychological measures is so challenging that an
entire sub discipline within psychology – psychometrics – is devoted to
understanding and refining methods of psychological measurements.
- Psychometrics tells us to focus on two properties of measures that
represent accuracy: reliability and validity.
- A good measure should be stable and steady. It should always give us the same
- Reliability refers to the consistency or stability of scores on a measure. A reliable
measure is one that produces consistent scores, free from random or unexplained
- We can think of consistency in two ways:
- Over time: a reliable measure is one that produces stable scores for the
same object on repeated use,
- Across judge: reliability occurs when an object receives similar scores from
- A good measure does more than yield a reliable score. It also yields a score that
truly reflects the concept it is intended to measure.
- Validity: the extent to which scores on the measures really represent the
underlying concept they are intended to represent.
- The most common way to demonstrate validity is by showing that scores on a
measure correspond reasonably well to scores on other measures that should be
- These other measures presumably assess at least some aspects of the same
concept; so showing that they overlap with the focal measure supports the validity
of that measure. Correlation Research
- In correlation research, investigators measure two or more concepts and see
whether the concepts are associated with one another- that is whether measures of
the concepts go together, or co-relate.
- The various kinds of correlation studies differ primarily in how researchers obtain
the data: by asking questions, using historical information, or by watching behavior.
- When two measures are correlated, it means that scores on the measures are
systematically related: as scores on one measure change, scores on the other
measure also change in a consistent fashion.
- If the two measures change in the same direction, it is a positive correlation
- If two measures change in the opposite direction, it is an opposite
Example of correlation:
- Robert Altemeyer and other social psychologists have found that the amount of
contact people have had with members of an ethnic group is negatively correlated
with their prejudice toward that group: people who report more contact with
members of an ethnic group also report less prejudice toward that group.
- This correlation between contact and prejudice can be used to illustrate an
important limitation of correlational research: correlations do not show that
measures are causally connected.
- When two measures are correlated, that cause of this action cannot be known for
certainty. Just because one measure correlated with another measure does not
necessarily mean that the first measure causes the second, nor that the second
causes the first.
- The reason correlational researh is ambiguous about cause is that the investigator
does not control factors in the environment that might partly or completely explain
an obtained correlation. Instead, the researcher simply measures two ore more
concepts and examines their associations. Why associations did or did not occur is
- This problem with correlation data applies to all of the methods we describe in the
- Notwithstanding this limitation of correlation research, these methods do have
strengths. They are flexible and can be designed to explore many different issues,
and are easily than setting up experiments.
- Also they are obtained in a naturalistic setting rather than a laboratory research.
- The most common kind of correlation research
- Survey: a study in which the researcher asks questions to respondents.
- Many surveys are distributed different ways, including, printed questionnaires,
distributed in a laboratory, mailed to the participant, computer-based, telephone,
and finally face-to-face.
- Christopher Davis and colleagues were interested in counterfactual thinking, which
refers to thoughts about how past events could have turned out differently. - They wanted to see that whether people who frequently think about how a tragic
event that occurred to them could have been avoided would also report about
- To test this hypothesis, the researchers surveyed 93 people who had lost a loved
one in a motor vehicle accident four to seven years previously.
- Participants were visited in their homes by a researcher who asked them a number
- Results showed that people who reported more frequent counterfactual thinking
about the event also reported greater emotional distress.
- Representative Samples:
- Researchers who conduct surveys sometimes want to be able to generalize their
results to a large population.
- If this ability to generalize is important, than the researchers must ensure that the
sample of people who complete the survey is typical of the population.
- To make their findings generalizable to a large population, researchers must
recruit a representative sample to complete the survey. This is a group or
respondents that reflect the larger population accurately: it has rations of various
subgroups that are similar to those in the larger population.
- The typical way to obtain a representative sample is via random sampling, which
refers to a recruitment process in which every person in the larger population has
exactly the same probability of being in the study. It is a difficult and time-
- Archival Research refers to investigations that are based on pre-existing
information obtained by the researcher.
- Common in social psychology
- Ex. Brian Mullen’s study showing lynching on Black men.
- David Phillips identified 20 publicized suicides (famous people) that occurred
between ’66 and ’73.
- He also obtained information about car accident fatalities in the days prior to and
following the appearance of these suicide stories.
- Across the 20 stories, Phillips found a consistent and significant rise in mote
vehicle fatalities 3 days after the appearance of the story.
- Phillips speculated that the suicide by well-known individuals prompted copycat
suicides using motor vehicle by depressed, vulnerable members of the public.
- Observational studies refer to research in which the investigator watches
participants and codes measures from the observed behavior.
- The scoring of behavior can either be “live,” where the observer is actually
watching participants, or the behavior can also be videotaped.
- Participants in observable studies are sometimes aware that they are being
observed and sometimes unaware. - Also, observable studies can focus on behavior in either a naturalistic setting or a
- These features are interconnected.
- Ex. If behavior is observed in a natural setting, participants may be
unaware. If behavior is in a lab setting, participants are usually aware of
- Geraldine Downey was interest in the interaction styles of dating couples and
whether certain styles were dysfunctional. Participants were 39 university-age
couples who had been dating for an average of 3 months. The couples were asked to
discuss a topic they often argued about for 20 minutes; these discussions were
videotaped. They were then separated and had to report how angry they had felt in
- Results showed significant effects for the women’s interaction style but not the
men’s. Specifically, women who exhibited more frequent negative behaviors during
the discussion tended to have male partners who reported greater anger after the
discussion. Thus, a negative interpersonal style by the women during the discussion
was associated with greater partner anger after the discussion.
- Another observational study was conducted in a parking lot by Barry Ruback and
Daniel Juieng. They suggest that most humans are naturally territorial and wanted
to protect their space from intruders, even when it is public space.
- To test this idea they went into a shopping mall parking lot and watched 200
different drivers get into their cars and drive away. They looked for two things: how
long it took the driver