In what follows I provide the scoring guide given to the GSIs. These are simply intended to be samples of adequate (3-point)
answers. Other good answers were, of course, possible.
I worried that the 17-item exam would prove to be a
little too long, and maybe it was. Still, in fact the vast majority of students
(70%) were able
to answer all 17 questions, and a couple of students answered
more than 17 items. Following the procedure indicated on the exam,
for those students we counted only the first eight items in each section.
Students attempted an average of 15.68 questions (standard deviation =
3.56). The exam yielded a mean score of
37.18 (SD = 9.91), which was a lower average, and a larger standard deviation,
than Spring 2008. And so was the average grade of a solid C (74%).
Nobody got a perfect score of 50. It had always been my intention to give everyone full
credit for Item #23, the "Quotations from Prof. Bruner".
It was just a way of giving everyone the two points remaining on the
exam, after Parts 1 and 2. As it happens, most people didn't get
two out of three right. But I bet you'll remember them now --
which is a good thing, because those three Brunerian aphorisms pretty
much sum up all of cognitive, and cognitive social, psychology.
If, at your 50th alumni reunion, all you remember are these three
aphorisms and their implications, I will have done my job.
As usual, I also conducted an item analysis to
identify bad questions. With a standard multiple-choice exam, in
which everybody must answer every item, the statistical standards for a
bad item are pretty clear: a low item pass percent, coupled with a low
item-to-total correlation, is a pretty good indication that the item in
question doesn't belong on the test. But with a short-answer (or
essay) exam, with continuous (rather than dichotomous) scoring of items,
not to mention the element of choice, the standards aren't quite so
clear, and so a little improvisation is called for. Here's how I do it.
I worried that the 17-item exam would prove to be a little too long, and maybe it was. Still, in fact the vast majority of students (70%) were able to answer all 17 questions, and a couple of students answered more than 17 items. Following the procedure indicated on the exam, for those students we counted only the first eight items in each section. Students attempted an average of 15.68 questions (standard deviation = 3.56).
The exam yielded a mean score of 37.18 (SD = 9.91), which was a lower average, and a larger standard deviation, than Spring 2008. And so was the average grade of a solid C (74%). Nobody got a perfect score of 50.
It had always been my intention to give everyone full credit for Item #23, the "Quotations from Prof. Bruner". It was just a way of giving everyone the two points remaining on the exam, after Parts 1 and 2. As it happens, most people didn't get two out of three right. But I bet you'll remember them now -- which is a good thing, because those three Brunerian aphorisms pretty much sum up all of cognitive, and cognitive social, psychology. If, at your 50th alumni reunion, all you remember are these three aphorisms and their implications, I will have done my job.
As usual, I also conducted an item analysis to identify bad questions. With a standard multiple-choice exam, in which everybody must answer every item, the statistical standards for a bad item are pretty clear: a low item pass percent, coupled with a low item-to-total correlation, is a pretty good indication that the item in question doesn't belong on the test. But with a short-answer (or essay) exam, with continuous (rather than dichotomous) scoring of items, not to mention the element of choice, the standards aren't quite so clear, and so a little improvisation is called for.
Here's how I do it.
A plausible statistical criterion for identifying bad items would be to follow the statistical "Rule of 2" -- that any value that lies more than two SDs away from the mean is an outlier, and thus suspect. That would mean that:
How do you drop an item?
Over the years, I have been trying to push the length of the exam a little, in order to achieve content validity -- that is, in an attempt to make sure that there was adequate coverage of both lectures and readings. And it's pretty clear that I overshot the mark this time. The biggest hint to this effect is the sharp drop off, beginning with Item #18 and running through Item #21, in both the percentage of students attempting an item (48%, compared to 69% for the exam as a whole), and the average score achieved by students who did attempt an item (1.60, compared to 2.29 for the exam as a whole). Taking the difference between predicted (2.29) and obtained (1.60), scores for these items, and multiplying it by 4 items yields a correction for length of 2.76 points. Not to be a jerk about it, let's make it 3 points. And again, we truncated your score if it rose above 50.
So, to calculate your final exam score:
As a result of the rescoring procedure, the mean exam score rose to 44.97 (SD = 6.05). The average grade is now an A- (90%). Both figures are comparable to those obtained the last time I taught the course.
The exam scores reported on the bSpace gradebook reflect the rescoring of the exam, as described above.
In what follows I provide information on the item analysis, as well as some additional commentary, as appropriate.
Answer eight (8) questions from each of Sections 1-2, and the single question from Section 3, for a total of 17 questions. Each of these questions is worth three (3) points. Do not answer more than eight (8) questions per section. If you answer more than eight (8) questions in any section, we will count only the first eight.
Be sure to print your name and UCB Student ID on every page of the exam.
Your responses should be very concise. In every case, less than 5 sentences will do. Write your answers in the space provided. If absolutely necessary, you may continue on the other side of the page. Write legibly in the space provided, and please use ink. Exams written in pencil will not be eligible for regrading.
A total of 66% of the class attempted the item; mean score = 2.18. Any three of the following will do: (a) Social and nonsocial cognition may be mediated by different brain modules or systems. (b) Social cognition blurs the subject-object distinction, because the self is both the knower and the object of knowledge. (c) In social cognition the object of cognition is a sentient being, with intelligence, consciousness, and purposes. (d) In social cognition, the object of cognition may be aware his/her status, and seek to manipulate (manage) the perceiver's impression through strategic self-presentation. [Introduction]
63% attempting, mean score = 2.57. The Thomas Theorem states that "if men define situations as real, they are real in their consequences" (paraphrase OK). The focus on the "definition of the situation" implies that the subjective, mental representation of the situation, rather than the "objective" situation (as it might be described by a third party), is decisive for behavior. If you want to understand the person's behavior, you have to understand the situation from the actor's point of view. [Cognitive Perspective]
86%, 2.81. In the self-fulfilling prophecy, a person acts in such a way as to make an originally false conception of a situation come true. In behavioral confirmation, the target's behavior is viewed by an objective, unbiased observer as confirming the actor's expectancies. In perceptual confirmation, the target's behavior is ambiguous, but is interpreted by the actor as confirming his/her expectancies. [Cognitive Perspective]
95%, 2.93. Central traits play a critical role in shaping the perceiver's impression of the target because they carry more implications for unobserved features of the person. Thus, a change in standing on a central trait implies a change in many other traits as well. Central traits appear to be closely related to major "superordinate" factors of personality, such as intellectual and social "good-bad", or the "Big Five" personality traits. [Social Perception]
79%, 2.89. Of course, people can tell us what their emotional states are. Beyond that, there are various nonverbal cues to a person's emotional state, particularly facial expressions of emotion. These facial expressions, in turn, are produced a relatively small number of facial muscles that move the lips, eyebrows, etc. [Social Perception]
55%, 2.14. The fan effect demonstrates the "paradox of interference", meaning that the more we know about a topic (such as a person), the longer it takes to retrieve any single piece of information about that topic. This provides evidence for an associative-network model of declarative social memory, in which persons are represented by nodes connected by associative links to other nodes representing their traits, attitudes, behaviors, and other facts. When the person node is activated, activation spreads throughout the network of associated links, but the process of information-retrieval searches the network serially, one link at a time. [Social Memory]
79%, 2,09. Trait and behavioral knowledge are represented independently in memory. That is, specific behavioral episodes (like "adopting a child") are not organized by the generic traits (like "kindness") that they exemplify. Any one of the following will do: (a) We know this from studies of the organization of recall: when recalling a target's behaviors, subjects do not cluster trait-related items together (at least, not very much). Also, amnesic patients can describe what they're like in general, in terms of their traits, even though they cannot remember specific episodes of behavior or experience. But mostly we know this from studies of priming. Presenting a trait term (like "kind") does not facilitate the retrieval of trait-related behavioral episodes (like "adopted a child"). [Social Memory]
65%, 2.60. Prior analyses of ingroup-outgroup relations had assumed that groups formed based on physical differences, such as gender or race, ethnic differences (such as national origin), or competition for limited economic resources -- as in Sherif's "Robbers Cave" experiment. But Tajfel formed groups based on entirely arbitrary distinctions, such as preferences for paintings, or even a coin-toss, and still found that subjects preferred their own ingroup to the outgroup. This suggests that social categories can have purely cognitive origins, as well as physical, ethnocentric, or economic ones. [Social Categorizataion]
74%, 2.37. Stereotyped traits need not be present in all, or even most, group members, and they may even be less frequently present than nonstereotyped traits. However, stereotyped traits are believed (this is important) to be relatively more likely to be observed in members of the stereotyped group, compared to the population as a whole (or perhaps to the perceiver's ingroup). [Social Categorization]
57%, 2.43. We think that Paul is loathed by Ted because of something about Paul, probably that he is loathsome. We don't attribute Ted's loathing to Ted, e.g., that he is the kind of person who loathes people, because loathing is a "mental state" verb that invokes the "stimulus-experiencer" stimulus, which drives attributes toward the stimulus, rather than the experiencer. [Social Judgment and Inference]
40%, 2.86. A reason is an explanation for intentional action, while a cause is an explanation for an unintended behavior. Reasons refer to beliefs, desires, or values, or some combination of these; there is a rational connection between the reason and the action, and it is assumed that the actor is aware of the reasons for his action. Causes apply to both social and nonsocial events, the connection between cause and effect is essentially mechanical, and the actor is not necessarily aware of the cause. [Social Judgment and inference]
90%, 2.94. The consistency-seeker is motivated to reduce any dissonance between attitudes, and between attitudes and behavior. The naive scientist engages in a rational analysis of events in the social world. The cognitive miser is motivated to reduce information-processing demand by relying on cognitive strategies that simplify complex problems. The motivated tactician chooses among available cognitive strategies based on goals. The activated actor relies on automatic, unconscious processes rather than conscious, deliberate ones. [Fiske & Taylor, Chapter 1]
80%, 2.87. Cognitive psychology is concerned with the acquisition, representation, and use of knowledge by individual minds (and brains). Cognitive sociology is concerned with individuals as members of thought communities, and with the role of cognitive socialization in shaping the individual's thought processes. Cognitive sociology begins with the assumption that different historical epochs, different cultures, and different subcultures are characterized by distinct differences in both the content and the mode of thought. [Zerubavel, Chapter 1]
74%, 2.84. A person can become salient by being novel or distinctive in the immediate context; by violating prior expectations or beliefs; or by being relevant to the perceiver's goals, or otherwise the object of directed attention. [F&T, Chapter 3]
79%, 2.87. Cognitive dissonance refers to the individual's perception of inconsistency between his or her attitudes and behavior -- not to an objective inconsistency, but to the subjective experience of inconsistency.. The perception of inconsistency causes arousal and creates an unpleasant emotional state. This unpleasant state can be reduced by changing the behavior, or changing the attitude, or -- most likely -- by avoiding or altering attitude-relevant cognitions. [F&T, Chapter 9]
37%, 0.85. A bad item. I didn't talk about implicit attitudes at all, but Fiske & Taylor spent a lot of time on the concept (if you're really interested, we discuss implicit attitudes a little in my consciousness course). Traditional theory assumes that people are aware of their attitudes -- which is why they try to achieve balance among their attitudes, experience states like cognitive dissonance when they engage in counterattitudinal behavior. However, it is possible that people have attitudes of which they are unaware, and which affect their behavior outside of conscious awareness and control. On simple logical grounds, it's difficult if not impossible to change an attitude that you don't know you have. [F&T, Chapter 10]
83%, 2.63. The prototype view holds that categories are groups of objects which share a family resemblance -- they tend to have a set of characteristic features in common, even though there is no set of features. The concept, or mental representation of the category, consists of a summary list of these characteristic features, which constitute the category prototype. The exemplar view abandons the view of concepts as any kind of summary, or average, representation of category members, but rather holds that concepts are represented as a "list" of the instances, or exemplars, in the category itself. [F&T, Chapter 4]
66%, 1.54. Both of these are effects of stereotyping on the target -- the stereotyped individual, rather than the perceiver who holds the stereotype. In attributional ambiguity, the target is uncertain whether to attribute the perceiver's behavior to his or her own personal qualities, or to the stereotype. Stereotype threat is a variant on the self-fulfilling prophecy in which awareness of the possibility of being stereotyped leads the individual to behave in a stereotype-confirming manner. [F&T, Chapter 11]
37%, 1.63. I didn't talk about this at all, either, though Fiske & Taylor spent a lot of time on it. Correspondent inference theory argues that perceivers assume that targets have attitudes and dispositions that correspond to their behavior. These attributions are rarely qualified by the situational context in which the behavior occurs. By attributing behavior to actors' dispositions, rather than to situational factors, correspondent inference theory assumes that people are prone to making the fundamental attribution error of attributing behavior to internal, personal dispositions rather than to external, situational demands. [F&T, Chapter 6]
48%, 1.88. In the anchoring and adjustment heuristic, people base their judgments on "early data returns" and do not sufficiently adjust their initial estimates to take account of late-arriving information. The clearest illustration of anchoring and adjustment is the order effect in impression formation, discovered by Asch: positive traits appearing early in a trait ensemble give rise to a more favorable impression than if the same traits appeared later in the list. [F&T, Chapter 7]
39%, 1.35. When information is expressed in quantitative form that can be processed according to a constant decision rule, computers or other statistical aids always do as well, and usually do better, than human judges forming "clinical" impressions. This is because "statistical" decision-making employs empirically valid weightings of various data sources, and is more accurate and consistent in applying the decision rules. [F&T, Chapter 8]
92%, 2.49. Personal memories are stored (somehow) in the minds (and brains) of individuals, and reflect the individual's more or less idiosyncratic mental representation of some episode of experience or behavior. Collective memories are shared by members of an entire "mnemonic community", such that each member of the community has the same representation of the event in question - -regardless of whether he or she personally experienced the event. Moreover, collective memories are jointly remembered by members of the mnemonic community on particular occasions (such as September 11), in a phenomenon known as "mnemonic synchronization". [Zerubavel, Chapter 6]
84%, 0.85. Any two can be correct, one point per aphorism: (a) The perceiver must go beyond the information given in the stimulus. (b) Every act of perception is an act of categorization; (c) The purpose of perception is action. [Hey, I had to get to 50 points somehow: last time we were reduced to giving students five points for writing their names on every page of the exam. And if you haven't heard me recite the first two Brunerian aphorisms about a dozen times each in lecture, then you've been sleeping. Admittedly, I don't think I mentioned the third one at all, though it's an easy inference to make -- and now you know yet another Brunerian aphorism!]
Exams will be graded and returned (in discussion sections) as soon as possible.
Requests for regrading must be made to your GSI
no later than 1 week after exams are returned.