Midterm Notes

10 Pages
163 Views
Unlock Document

Department
Psychology
Course
PSY371H5
Professor
Christine Burton
Semester
Winter

Description
PSY371: Midterm Notes Friday Feb. 15 : 10am-12pm Section A: 5 Short Answer Questions (you must answer all of them)  Lecture material  Background issues related to the more controversial theories covered in the class discussions Section B: 2 Essay Questions (you must choose ONE)  Questions that require you to weigh the evidence surrounding an unresolved topic and argue your opinion  Draw on lecture material, class discussions, and the assigned readings (readings are only intended to give you evidence to support your argument)  Know the main point of each of the papers and some evidence the authors use to support their argument  Have an argument ready for both sides of the issue  Know the main author of each of the readings *Executive function will not be on the test Lecture 1: Historical Approaches to Intelligence Historical Approaches to Intelligence  Francis Galton was concerned with differences between individual’s intelligence o Large influence on testing and hereditary view of intelligence o Individual differences in measuring and classifying people according to their ability o Big influence on the idea that intelligence is inherited  Alfred Binet was concerned with identifying group milestones o Influenced information processing view of intelligence and cognitive development o Not interested in individual differences but comparing people to group milestones (are they where they should be?) o Thought that intelligence was something that developed with age – was NOT stable; it could change Galton (1822-1911)  Privileged, intelligent, and successful  Major life interests included measurement and ranking of ability  Highly influenced by cousin Darwin’s Origin of Species  Went to med school on account of his father’s wishes  Figured that fingerprints were all different  His Anthropometric Laboratory measured keenness of sensation and individual differences in intelligence  Invented regression and correlation to describe the relationship between items that tended to be related  In Hereditary Genius he proposed that eminence was passed down from generation to generation  Introduced “nature vs. nurture” debate, twin studies of intelligence and eugenics movement  If intelligence is inherited, then why bother educating the less eminent and their resources should be revoked – therefore really controversial issue Binet (1857-1911)  Trained as a lawyer, Binet was interested in science and human development  Believed intelligence develops with age, is highly malleable by experience, and is not unitary  With his graduate student, Theodore Simon, developed a cognitive abilities test for French schoolchildren  Introduced idea of standardizing mental tests on average groups of people  In order to design a school readiness test, he started looking at skills you need in school  Not interested in comparing students – just find the benchmark; what is the minimum that a child needs to succeed  He was designing a test but it was not designed to rank Testing-based intelligence theories  Using factor analysis, Spearman argued for the single intelligence construct, “g” o Extracted this one factor that he called general intelligence o If you are high in g, you are smart and you will be able to succeed  Thurstone’s Theory of Primary Mental Abilities proposed 7 intelligence components o Also used factor analysis and developed 7 intelligence factors  R. Cattell and Horn’s g(f)-g(c) model views intelligence as 2 major components o Argued that there are 2 factors that explain intelligence o Fluid intelligence and crystallized intelligence o Fluid intelligence is your ability to solve problems that are abstract and everyday problems; everyday problems – things you don’t already know o Crystallized intelligence are things you already know; language, arithmetic, etc. Information processing-based intelligence theories  Some believe processing speed serves as a good index of intelligence (e.g. Eysenck, Jenson, Vernon)  Sternberg’s Triarchic Theory assumes intelligence is a result of the interaction of mental components within ourselves and the external world o Analytic (which is essentially g) o Practical (street smarts) o Creative (thinking outside the box)  Gardner’s Multiple Intelligences theory proposes 8 independent types of intelligence Early Intelligence tests  Earliest intelligence tests were not predictive  Two major developments toward modern intelligence testing o Binet & Simon’s cognitive abilities test  Mental age; not useful after a certain point (i.e. in adults)  Expensive and slow o Army Alpha and Beta  Group test; used multiple choice which meant it could be scored easily and quickly, it could be used on adults (age wasn’t the comparison) Structure of Intelligence Tests  Most IQ batteries consist of scales made up of subtests  A good test should have reliability and validity Reliability  Any test score consists of the true score + error o Varobs=vartruar err o Error can be random or systematic o If it is random error, it affects people randomly and does not affect the average score, it changes variability around the average o If it is systematic error, it influences the entire sample equally which is a problem because it shifts the average score o Reliability tells us how much of the observed score reflects the true score (low error variance means a reliable score) Validity  A valid test measures what it is supposed to measure o Content validity  Face validity  Reflects the knowledge you are trying to get o Criterion-related validity  Concurrent validity  Predictive validity o Construct validity  Convergent validity  Discriminant validity o Incremental validity o Item validity  Particular question or item – is that question valid? Is it testing what it is supposed to test? Developments in test theory  We can compute an item difficulty index (p) and an item discrimination index (D) for each test question o Item difficulty index – whether or not the question does a good job between discriminating between high and low poeple  According to the Item response theory (IRT) the probability of a correct response is a function of a latent trait of an individual and a set of item parameters o The probability of getting a question right depends on the individual’s intelligence and parameters related to the question itself – those parameters are item difficulty and item discrimination  The slope of the curve in the graph represents the item discrimination index – how well it can discriminate between people  If IRT assumptions are met, adaptive testing can be used o Allows shorter, reliable tests  Adaptive testing could allow continuous item validation and AI item generation What do intelligence tests measure? Factor analysis  Determines how many dimensions are required to account for the variability between scores  All tests measure variability due to specific abilities and some degree of communality between subtest o Factor loadings represent the communality  Most of the score on the subtest can be explained by the factor  One way to assess validity is to look at factor loadings o Subtest  scale  composite  A scale is an observable variable, but a factor is a hypothetical unobserved variable  What do the factors represent? How can scale scores be interpreted?  “Strong” models extract common variance first then look for residual variance o Many IQ tests use this approach  Spearman described the positive manifold of the subtests of cognitive ability  The positive manifold leads to a single first factor in factor analysis – Spearman called this g (general intelligence)  IQ test must measure something real o They are stable, reliable, and have predictive validity  IQ  school performance (r^2~0.25)  IQ  years of education (r^~0.30)  IQ  SES (r^2~0.11) BUT…  Years of education  SES  Years of education  IQ  School performance  years of education Lecture 2: One vs. Many Intelligences G Most IQ tests claim to measure g, which is heritable and has some biological correlates A positive manifold exists; so do specific abilities Shared environments become less influential on IQ with age People with widely different IQs will not learn to the same degree IQ is the best predictor of many social outcomes and predicts mortality to some degree IQ does not guarantee a successful life IQ is not completely fixed; environments influence learning and IQ Individuals have different patterns of skills and abilities The existence of g is not, by itself, a value judgment A happy, successful life requires more than intelligence Multiple Intelligences There are multiple, independent modules for storing and processing different types of information that have value within a culture Intelligence criteria:  Potential for brain isolation  Savants and prodigies 
More Less

Related notes for PSY371H5

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit