PSY320 Midterm Flashcards

Belief Based Model (Syllogistic Model)

In understanding our attitudes, we must consider these structural linkages (a belief network)

A belief network and how they link and the consequence of those links.

Attitudes = Σ (belief x evaluation)

What is the link between beliefs and evaluations

Beliefs are a full set of what is known about the object. Each of these beliefs has an evaluative component.

Belief premise

The “cognitive” component of attitude

Evaluative premise

A person’s affective or emotional reaction to the belief.

Example of belief premise + evaluative premise = conclusion

Belief : defense spending creates jobs

Evaluative premise : Creating jobs is good

Conclusion : Defense spending is good

Horizontal structure

Using another example to come to a conclusion

When an attitude is linked to more than one underlying set of beliefs.

Ex. Attitude 1 ) Defense spending is good and creating jobs is good. Attitude 2 ) Defense spending prevents war and preventing war is good —> therefore, defense spending is good.

Vertical structure

Can elaborate on your reasoning and come to conclusions

Signifies a minor belief is derived from or dependent on a primary belief

Example : Preventing war can save lives, saving lives is good, therefore preventing war is good. —> Defense spending can prevent war. Preventing war is good. Therefore, defense spending is good.

Degrees of beliefs

How many lives were saved?

A. Defense spending will produce a first strike

B. A first strike is bad

C. Defense spending is bad.

Combinations of probabilities

p(C)= p(A)p(B)+p(K)

p(K) meaning

Other resources that the evaluation / belief could be bad.

Unidimensional models

I assume you have these attitudes, I’m not going to worry about the fact that you might like things for different reasons.

You like the thing simply because you like the thing. Positive and negative elements are stored as opposite ends of a single dimension.

Until 1960’s

Affective or emotional orientation to an object along a single dimension of favorability or unfavorability.

Affective responses to an object

Semantic differential

Maybe we just need to know that final evaluation at the end and we don’t need to know what causes that feeling.

Two-dimensional models

Focuses on attitudes on two separate dimensions of positive and negative elements

Affective versus Cognitive

We have cognitions but they need to be translated into affects that guide behavior.

Example : British Columbia bridge experiment

Schachter & Singer Model

Stimulus —> perception / interpretation —>(+) specific pattern of arousal —> Particular emotion experienced

Instead of having cognitions and then finding the emotions, we have emotions and we need to understand the cognitive meaning.

Emotions are derived from a combination of arousal and cognitions used to explain arousal.

Russell’s Circumplex Model

Arousal and Valence

Maybe your attitude simply is the emotional responses you have in the end.

X-axis is a continuum of pleasant and unpleasant feelings, while the y-axis is activation level.

Does this capture all the emotions? No

Evaluative space Model

Part of the positivity and negativity model.

Attitudes might not b best defined as either good or bad evaluations, sometimes it could be ambivalence or indifference.

Considers all the inputs into creating and evaluations in a grid like way, which captures positive, negative, ambivalent and indifferent scale of attitudes.

Evaluative space model formula

Attitude = .4 (P+C)^0.5 - 0.6(N)^.5

P and N are degrees of positive activations and negative activation in any given point

Negativity Bias

Given a positive and a negative piece of information will produce a slightly negative attitude. In other words, negative things are more salient.

Why could this be function? Survival. Better to be safe than sorry

It’s the reason why it’s .6(N) in the formula.

The reason for .5 powers?

We want to curve the scales.

Our brain processes numbers on a logarithmic scale which means that low numbers are more salient to us, so small differences in high numbers will yield less response than differences in lower numbers, since it’s before the logarithmic leveling off point.

Positivity offset

People tend to evaluate novel attitude objects as slightly positive. Why? because we want to be slightly exploratory if not knowing things. We’re curious and neutral things become positive if you’re curious.

This is why there’s a +C in the formula.

Automatic and controlled bidimensional models

Has to do with dual processing thinking.

Believes we have two separate attitudes, a fast one to tell you positive or negative. Without reasoning you might have a different view of something.

Dual processing theory

The idea that we have two cognitive systems, a fast, automatic, parallel, and effortless one, and a slow, serial, controlled and effortful one.

Tripartite Model

Three-dimensional model

Takes an affective, behavioral, and cognitive component to all come together to form attitudes about the things we interact with

Affective component

Emotions, moods, and feelings towards the attitude/object.

I feel sad when I walk past the cat shelter.

Behavior component

Observations of how one behaves towards the object.

I enjoy playing with cats

Cognitive component

Beliefs and thoughts about the attitude object

I believe cats deserve to be in loving homes instead of shelters.

An attitude is a summary evaluation

Thoughts, feelings, and behaviors don’t have to coincide, the overall of these combined is what makes the evaluation.

Evaluative consistency

Associations between each of the components and within component elements.

A change in one component (theoretically) “should” result in a change in the other two.

Non-dimensional Models

Attitude objects are nodes in an associative network

Attitudes are associations between an object and an evaluation

More frequently the link is activated, the stronger the attitude.

Operational definition

Specifies explicitly how to measure a variable so that one can get assigned a ranking for the contract (high, low, medium, etc)

Object Appraisal Function

The ability of attitudes to summarize the positive and negative attributes of objects in our social world.

Attitudes help us classify things in our environment, thus organizing and simplifying our lives.

Daniel Katz

Expanded on the notion that attitudes serve functions. He classified attitudes into 4 groups of function: Utilitarian, Knowledge, Ego-defense, Value-expressive, Social-adjustment.

Utilitarian function

Provides us with general approach/avoidance tendencies.

People adopt attitudes that are rewarding and that help them avoid punishment.

Any attitude that is adopted in a person’s own self-interest serves a utilitarian function.

Knowledge function

Help people organized and interpret new information.

People need to maintain an organized, meaningful, stable view of the world. Important values and general principles provide a framework for our knowledge : equality, belief in a Just world, etc.

Attitudes can help them achieve this goal by making things fit together and make sense.

System Justification Theory

People are motivated to believe that the social system they’re in is a good one because it’s terrifying to not know what the alternative is.

Ego-defensive function

This function involves psychoanalytic principles : People use defense mechanisms to protect themselves from psychological harm.

Mechanisms include : Denial, repression, projection, rationalization.

Downward Comparison Theory

By derogating a less fortunate other, we can increase our own subjective well being.

This is likely to happen after we have ourselves suffered misfortune or frustration.

ex. After losing money on Wall street, people have more negative attitudes towards the homeless.

Value-expressive functions

Used to express central values or beliefs. Serves to express one’s central values and self-concept. Central values, establish our identity, and gain social approval.

Example : attitudes towards a controversial political issue.

Social-adjustment function

Attitudes affect our social relationships (help or hurt). Attitudes serve to help people connect to others and “fit in” to a reference group. Core attitudes gain social approval and change to reflect current social trends.

Ex. attitudes towards fashion.

Proposed sixth function : Existential

Give our lives meaning and provide meaning to our activities.

We want to feel like there’s a meaning and purpose behind our actions and existence.

Terror Management Theory

Conscious awareness of mortality which creates the potential for overwhelming terror.

Cultural worldview exist as an anxiety buffer to manage terror or mortality by providing real and symbolic immortality to those who live up to standards.

Two buffering components

Validity of worldview

Meeting/exceeding the standard

When mortality is made salient, people :

Respond positively to those who bolster and respond negatively to those who threaten. This leads us to become much more polarized when faced with uncertainty and simply sticking to our group and worldview and belittling other worldviews.

Experiment on terror management

Had people think about their mortality and then answer the appropriate bond for a prostitute and the appropriate bond for the person turning in the prostitute. When thinking about morality, punishment for prostitute and reward for the one who turned them in went up.

What charismatic leaders do to draw in people

Allay their fears, Self Confident, Emphasizes the positive qualities of the worldview.

Self monitoring scale (Snyder, 1974)

To see if they’re sensitive to certain functions of attitudes. Some people are social chameleons while others are like accept me or leave it.

Thus, we ask questions like “I guess I put on a show to impress or entertain others” and “At parties and social gatherings, I do not attempt to do or say things that others will like”

People who are high in self monitoring will most likely behave inconsistently because they behave according to the situation. —> chameleon

People who are low in self monitoring draw on feelings and attitudes when behaving. Their behavior is normally much more consistent. —> accept me or leave it.

Symbolic politics

Certain “political attitudes” mask other related attitudes.

Attitudes towards affirmative action may reflect attitudes towards race and ethnicity.

This allows some attitudes to be expressed in a more socially desirable way.

Nomological network defintion

Derived from Greek and means “lawful” thus it means “lawful network”

This would include the theoretical framework for what you are trying to measure. An empirical framework for how you are going to measure it and specification of the linkage among and between these two frameworks.

Nomological net

The set of constructs to construct relationships derived from theory and state at the abstract theoretical level.

What is the relationship between two or more constructs according to a theory at an abstract and theoretical level. Becomes the starting point for operational definitions.

Observed score

True score + systematic error + random error

True score

Represents your attitude

Systematic error

A way to factor in the possibility that we don’t measure what we think or intend to measure.

Random error

Sometimes people make error, like not knowing the scale, misreading the item, missaying the actual score, etc.

Reliability

true variance/ total variance

How much of that total variability is related to the true variance with what you’re actually interested in.

Test-retest reliability

Seeing if a person gets the same results/response when examined at two different points in time on the same measure.

Problems with test-retest reliability

These’s other variables at pay which make the answers not indicative of your attitudes. Even if you have these attitudes, it’s possible the score could change when retaken because of missaying or putting down the wrong response.

Inter-rater reliability

Two or more independent judges score the test.

Want more than one person to examine something to make sure the results are the same and not skewed or subjective.

Parallel-forms reliability

Comparing two different tests that were created using the same content.

If they’re actually measuring the same thing, the results should be similar in both tests.

Internal consistency

Similar results across items of the same test

Stats programs where you don’t have to take two tests you can simulate a version of this

What you do is give everyone 50 questions and you do the same statistically by re-sorting them over and over and over and taking the average correlation of the subsets.

Can we get rid of random error?

Cannot get rid of it but can reduce it.

Asking more questions can help the random error go away, since the more items you have, the less likelihood for error to occur.

Validity

Does the measure actually measure the true attitudes or is it untrue?

Ex. a scale can be saying the same number (reliability) but if it’s 10 grams off, it’s not valid.

Face validity

If a measure appears to measure what it is supposed to measure.

Ex. if you’re asking someone whether or not they like a store, asking them what their favorite color is have low face validity since it’s not measuring attitudes towards the store they like.

Convergent Validity

The degree to which scores on a test correlate with (or are related to) scores on other tests that were designed to assess the same construct.

Ex. How much you like products at a store and how often you visit a store both measure enjoyment of that store and will be more valid if there’s a similar response to both.

Discriminant validity

The degree which scores on a test do not correlate with scores on another test that are not designed to assess the same construct.

Ex. If you’re trying to measure attitudes towards a political view with a opposite view also present, you could ask questions which measure attitudes towards view 1 and then measure attitudes towards view 2, and if they rate view 1 favorable and view 2 unfavorably then one can conclude what view is common fairly validly.

Unstructured (open ended) self-report measures

Most difficult but important and shows how nomological nets require revision

Often psych researchers jump to scales, but this means the questions and scales generated might guide participants into a certain answer instead of testing your theories validity beforehand and whether people naturally respond that way, instead of guided by questions.

Structured self-report measures (scales)

A continuum from one extreme to another extreme view on an attitude. Attitudes fall somewhere on this continuum.

Measures have numerous questions on one attitude object

Bogardus’s Social Distance Scale

The scale of stereotypical bathroom characters standing in certain distances from each other and asked to circle which one group or person fits into

Mirroring a behavior measure by subtly measuring how close or distance you want to be with someone.

Example of Bogardus’s Social Distance Scale

A respondent’s score is the closest distance at which he or she finds the relationship with members of the minority group acceptable :

Would exclude from my country, Would accept as visitor only to my country, Would accept to citizenship to my country, Would accept for employment in my occupation in my country, Would accept to my street as neighbors, Would accept to my club with personal friends, Would accept close kinship by marriage.

Thurstone’s Method of Equal-Appearing Intervals

An attitude scale in which many possible scale items are rated by a panel of judges.

These judges sort the statements into groups (Ex. only a person with this kind of statement would have a negative view of immigration, etc.) and these groups range from very positive to neutral to very negative.

The groups have to be equally as negative, positive, etc attitude, from the other in the bin and must be an equal distance of attitude from the previous and next bin.

Statements used in the final scale are a sample of statements from each group that has the highest level of agreement among the judges. Respondents are asked to state if they agree with each of the statements.

Attitude scores consist of the average value of the items agreed with.

Likert’s Method Of Summated Rating

An approach where respondents indicated the extent to which they agree or disagree with a statement.

Items usually have five response options : Strongly agree, moderately agree, neither agree nor disagree, moderately disagree, strongly disagree.

A person’s score is the sum of the values of the options they picked across all items (some items are reverse coded).

To ensure scale homogeneity, only items that correlate well with the total score are kept on the scale. This process of refinement is normally done during the developmental stage of the scale.

Osgood’s Semantic Differential

An attitude scale which includes various sub-scales that measure the connotative meaning of the attitude object.

THese subscales are bipolar and respondents are asked to check along the line between the two bipolar opposites (ex. good ←→ bad, cruel ←→kind)

A respondent’s score consists of the average of the ratings.

Problem is that it doesn’t tell researchers what you’re thinking about in your response and thus lacks nuance.

Three elements Osgood’s Semantic Differential Approach is based on

Evaluation (good/bad)

Potency (strong/weak)

Activity (active/ passive)

One item scales

Although there are good psychometric reasons for using several items to measure an attitude, often a single item does the trick. Such an item can consist of a question that asks how positively or negatively one feels about the attitude objects.

Ex. For measuring self-esteem, one of the best measures is by just asking “do you have high self-esteem”. 

Can take the form of “thermometer scales” where one is asked to indicate on a scale from 0 to 100 how warmly one feels towards the attiude object.

Problem with positive scale versus negative to positive scale in direct measures

How happy were you when you were a child? 0-10 scale versus a -5 to 5 scale.

When you use these different scales, the first one is 31% of people were unhappy children, but when asked the second way, 13% were unhappy children.

The negative scale requires more negative experiences to evaluate it as a negative number, versus a positive number someone might rank slightly lower on the scale simply because they didn’t get a pony.

Problem with specific wording and responses on direct measures

Do you support the government giving money to the poor versus giving money to welfare.

The poor = more positive, but welfare has a different connotation so it shows negative attitudes towards that word, even though both mean the same thing.

Social desirability bias

Survey respondents provide answers according to society's expectations, rather than their own beliefs or experiences.

For sensitive issues it’s typically an underestimate since people don’t want to share something that is personal or controversial.

In order to measure this, one way is to get people to answer 50 questions on risky behaviors, and then you have another question, which is you put the really horrible thing nobody is going to tell you after that and see the number difference.

Context sensitivity and problems with direct measures

Preceding questions prime a response. It’s been studied how you can trick attitudes by asking certain questions.

Word scales like “I’ve thought of it before” versus a numerical scale like “I’ve done it 10 times” can change the response in further questions.

Question working implies differences.

Solution of ambigious items

Items with different variabilities can help with this.

Example of an ambiguous item : “Men and Women should have equal rights”

Could trigger social desirability bias. Might be too general and not expressing underlying attitudes.

Example of solution : All men should be removed from the country, you can never trust a man, the world would be a better place without men, I feel negatively towards some men, etc.

Extremity bias

Some people are just more willing to use the end points of a scale

Solution would be half od the questions in one direction and the other coded for the other direction.

Should cancel out if the actually have extreme beliefs rather than simply putting extreme answers.

Left right bias

People like one side of the paper more than the other.

Could be because where they’re sitting or orientation of paper.

Solution : Can be handled statistically

Bogus pipeline

Detective in a County Police department were trying to get a confession and they pretended to use an old Xerox machine as a lie detector but putting a he’s lying card into the machine, putting a metal colander over the suspects head, and wiring the colander to the Xerox machine.

When the suspect gave an answer the detectives didn’t believe, they pushed the copy button and the machine spewed out paper saying, “He’s lying”. Faced with such advanced level police tactics, the person finally confessed.

Solution : Lying (self-deception/social desirability)

Bogus pipeline

Speeded responses— decreasing response time increases truthful responses because it takes less time to tell the truth. Problem with this is that it could lead them to missaying something or other random bias.

Computer administration — less trying to be fake good for an experimenter and feel more anonymous doing it . However anonymity leads people to not take the experiment as seriously and might give bizarre responses which might or might not represent their true attitude.

Research with expert wine tests

Got professional wine tasters, took all the wines, mixed it all up with the bottle labels and they all put price tags on it.

These super famous wine tasters could not discriminate between the amazing 400 dollar wines and the cheaper wines. Gradations was completely gone. What determined their winetasters appreciation of the wine was the theory about what was in the wine. fMRI looked at the reward circuitry, they weren’t lying, they literally were enjoying the wine more than they thought was the better one.

Sometimes what you think you’re experiencing will drive the response you’re having.

Clustered attitudes

People think attitudes come from experience with an attitude object which leads to your attitude, however, it’s not one object and doesn’t need to be direct experience to infer attitudes.

If you have one attitude you’re more likely have other attitudes which cluster around the single attitude that are similar and mutually reinforcing so you have a clear pattern of beliefs and associations with the least mental conflict.

Ex. I don’t like weapons, guns are a weapon → therefore I don’t like guns. → Gun control means less guns, → and I don’t like guns, → therefore I like gun control.

Ambivalent states of attitudes and mental state

Leads to an uncomfortable and a weird state. We don’t like feeling positive and negative at the same time, so through these networks we try to resolve our mixed feelings to make things coherent.

Psychological Reactance Theory (Brehm, 1966)

Elimination of an available option creates a need to restore that lost option, called reactance.

Paradoxically, this is expected to lead the person to value the lost option more highly than before.

Ex. Mother forbids children to play in mud will increase the attractiveness of that activity in the mud.

Evaluative Conditioning (Automatic forms of Attitude Learning)

Changes in the liking of a stimulus that are due to the fact that the stimulus has been paired with other, positive or negative stimuli.

Similar to higher order conditioning. It’s not called higher order conditioning when it’s human participants and when emotions and attitudes rather than behavior are the dependent variable.

Higher order conditioning

When a conditioned stimulus becomes associated with a new unconditioned stimulus

Ex. Food paired with a bell leads to salivation → then the bell is paired with the cupboard opening, → so the cupboard opening eventually gets paired with the bell and food and thus salivation.

Grossman and TIll — Conditions of attitude

Materials : Ficticious brand names (Jorro toothpaste) and Pleasant pictures (beaches, etc)

Procedure : Brand names were presented with neutral pictures (control) or with pleasant pictures (positive). Participants rated how much they liked the brand (immediately or 3 weeks later)

Results : Immediately after, the positive group found Jorro toothpaste better simply when paired with something positive. Three weeks later the control went down a lot and was negative, while the experimental group was still slightly positive.

Can conclude that : Evaluative conditioning produced a more favorable attitude immediately after a product was paired with pleasant pictures and simple associations of a stimulus with affective stimuli have lasting effects on attitudes.

Overall conclusions drawn from evaluative conditioning

Attitudes are not only influenced by reasoning and careful processing of relevant information.

Attitudes are also influenced by many factors outside of our awareness.

Unconscious influences on attitudes may explain why advertising is effective even if most people believe that they are not influenced by it.

Mere exposure effect (Automatic forms of attitude learning)

Zajonc 1968

Participants say novel stimuli (Turkish signs, Chinese characters and unknown faces with different frequencies)

Afterwards participants rated for each stimulus how much they liked it. The higher exposure frequency the higher the favorability of attitudes for all stimuli.

Conclusions : liking increases with the frequency of presentations, the effect was consistent across three different type of stimuli.

Subliminal exposure (Automatic forms of attitude learning)

Participants might have caught onto the mere exposure study

Subsequent studies demonstrated that exposure even increases liking of novel stimuli if exposure occurs without awareness (Flashing on the screen and can’t report what it is)

Results : The effect was sometimes larger than when people can see the stimuli.

This is rare because oftentimes they believe that it will have the effect but lower, but to have it higher effect isn’t common.

Manipulating Attribution (Bronstein & D’Agostino 1994)

Photo of people were presented subliminally to participants.

Participants in one group were told that they had been shown some of the photos subliminally.

The participants in the second group were told that they had never seen the photo before. The second group ended up with a much larger mere exposure effect than the first group.

The participants in the first group did not misattribute their perceptual fluency, they were able to identify why they could perceive certain photos more easily than others (they had seen them before). On the other hand, the second group tended to misattribute their perceptual fluency to likability.

Heritability and attitudes

Studies using identical twins that were reared apart show that they have similar attitudes despite growing up in different environments.

Tesser 1993 suggest that while genes do not determine attitudes, the genetic influence on sensory structures, body chemistry, intelligence, temperament and activity level and conditionality impact our attitudes.

Attribution theory — Kelley 1967

The process of inferring the causes of events or behaviors.

Three different kind of information important to that : consensus, distinctiveness, and consistency.

Consensus information — Attribution theory

Do all or only a few people respond to the stimulus the same way as the target person?

Distinctiveness information — attribution theory

Does the target person respond the same way to other stimuli as well?

Consistency information — Attribution theory

Does the target person always respond in the same way to this stimulus?

High consensus, high distinctiveness, high consistency attribution theory

Everyone laughed at the comedian

Bob rarely laughs at comedians

Bob laughed at this comedian every time he sees her

Attribution : Bob really likes this comedian and it’s a good comedian.