Exam 1 Study Guide

How linear regression is used to predict performance?

Chapter 1

1) Describe major milestones in the history of testing.

­ c. 2200­2000 BC­ Chinese civil service exams

­ Carl Frederich Gauss and the normal distribution

­ Civil Service exams in Europe

­ By 1883, American Civil Service Commission established

­ Francis Galton: Sets up an anthropometric lab at the International Exposition of 1884 ­ First intelligence test: Binet­Simon Scale of 1905

­ 30 items of increasing difficulty

­ Standardized administration

­ Same instructions and format for ALL children

­ Standardization sample

­ Created norms by which performance one child can be compared with other children

­ 1916: Binet­Simon translated for US use

How scatterplots are used to describe the relationships between two variables?

­ WWI

­ Army Alpha­ required reading ability

­ Army Beta­ did not require reading ability

­ Testing “frenzy” hits between WWI and 1930s

­ Personality testing

­ Woolworth Personal Data Sheet

­ First objective personality test meant to aid in psychiatric interviews ­ Developed during WWI

­ Designed to screen out soldiers unfit for duty

­ Projective testing

­ Herman Rorscharch inkblot test (1921)

­ Started with great suspicion; first serious study in 1932

­ Symmetric colored and black/white inkblot

­ College admission tests­ SAT and ACT

­ Wechsler Intelligence Scales

How tests can be developed that produce both norm­referenced and criterion­referenced interpretations?

­ Minnesota Multiphasic Personality Inventory (MMPI)

2) Define test, measurement, and assessment.

­ test­ device or procedure in which a sample of an individual’s behavior is obtained, evaluated, and scored using standardized procedures Don't forget about the age old question of What do you call the limited nature of resources?

­ Measurement­ set of rules for assigning numbers to represent objects, traits, attributes, or behaviors We also discuss several other topics like Why must a government have checks and balances?
We also discuss several other topics like Why do birds usually hide their illness?

­ Assessment­ systematic procedure for collecting information that can be used to make inferences about the characteristics of people or objects

3) Describe and give examples of different types of tests.

­ Maximum performance tests

­ Achievement tests

­ Assess the knowledge or skills of an individual in a content domain in

which they have received instruction

­ I.e. School exams

­ Aptitude tests

­ Designed to measure the cognitive skills, abilities, and knowledge that an individual has accumulated as the result of overall life experiences

­ I.e. SAT and ACT

­ Objective vs. subjective tests

­ Objective tests­ less disagreement in how it is graded

­ I.e. True/false, multiple­choice questions

­ Subjective tests­ different graders → different scores

­ I.e. Essay questions, open­ended questions, debates

­ Speed vs. power tests

­ Speed tests­ performance only reflects differences in the speed of

performance (how much you can do in a certain amount of time)

­ Power tests­ performance reflects difficulty of the items the test taker is able to answer correctly (I.e. GRE and SAT) We also discuss several other topics like What does john stuart mill believe about psychology and philosophy?
We also discuss several other topics like What is the significance of gitlow v new york?

­ Typical response tests We also discuss several other topics like What is the standard deviation of x?

­ Objective tests

­ Use items that are not influenced by the subjective judgment of the

person scoring the test

­ Projective tests

­ Presentation of unstructured or ambiguous material that elicits an almost infinite range of responses → requires subjectivity

4) Describe and give examples of different types of score interpretations.

­ Norm­referenced scores: Interpretation is relative to other people

­ I.e. Percentile ranks

­ Criterion­referenced scores: Interpretation is absolute

­ I.e. Percent correct on an exam

5) Describe and explain the assumptions underlying psychological assessment. ­ Psychological constructs (i.e. intelligence, depression, attitudes, etc.) exist ­ Psychological constructs can be measured

­ Measurement is not perfect­­ there is usually some degree of error is inherent in all measurements

­ There are different ways to measure a construct

­ All procedures have strengths and weaknesses

­ Multiple sources of information should be part of the process (think about the college application process)

­ Performance on tests can be generalized

­ Assessment can provide information

­ Assessments can be fair

­ Assessments can benefit individuals and society

6) Describe and explain the major applications of psychological assessments. ­ diagnosis

­ Treatment planning and effectiveness

­ Instructional planning

­ Selection

­ Placement

­ Classification

­ Self­understanding

­ Evaluation

­ Licensing

­ Program evaluation

­ Scientific method

7) Explain why psychologists use tests?

­ people are not good at judging others

­ Tests provide objective information that helps us make better decisions 8) Describe the major participants in the assessment process.

­ people who develop tests

­ People who use tests

­ People who take tests

Chapter 2

1) Describe how linear regression is used to predict performance.

­ Linear Regression­ ​a relationship best represented by a straight line ­ allows you to predict values on one variable given information on another variable

2) Describe the different scales of measurement and give examples

­ Nominal Scales​­ classify people or objects to categories, classes, or sets ­ i.e. eye color, gender, type of sport, college major, music genre

­ Ordinal Scales​­ there’s a clear order, but the magnitude makes no sense ­ i.e. rank according to height, age classification

­ Interval Scales​­ rank ppl or objects like an ordinal scale, but on a scale with equal units (i.e. IQ, temperature, percentile rank)

­ no true zero

­ Ratio Scales​­ equal intervals between

­ i.e. weight in pounds, height of buildings in city, distance from one city to another, percentage score

3) Describe the measures of central tendency and their appropriate use.

­ Mean​­ simple, arithmetic average

­ essential when calculating many useful stats

­ Median​­ point that divides the distribution in half

­ often preferred with skewed distribution

­ Mode​­ most frequent score

­ useful with nominal level data

4) Describe the measures of variability and their appropriate use.

­ Range = highest score ­ lower score

­ standard deviation­ average distance that scores vary from mean ­ variance­ special meaning as a theoretical concept

5) Correctly interpret descriptive statistics (e.g., skew, kurtosis). ­ skewness

­ ­ ­ kurtosis

­ leptokurtic

­ ­ platykurtic

­ ­ mesokurtic

­ 6) Explain the meaning of correlation coefficients and how they are used. ­ Correlation coefficients­ ​quantitative measures of the relationship between two variables

­ designated by r

­ range from ­1.0 to +1.0

­ sign of the coefficient

­ indicates the direction of the relationship

­ magnitude or absolute size

­ indicates strength of the relationship

7) Explain how scatterplots are used to describe the relationships between two variables. ­

8) Describe major types of correlation coefficients.

­ Pearson­Product Moment Correlation Coefficient

­ when variables are on an interval or ratio scale

­ Spearman Rank Correlation Coefficient

­ when variables are on an ordinal scale

­ Point­Biserial Correlation Coefficient

­ one variable dichotomous; one on an interval or ratio scale

­ i.e. comparing gender with test scores

9) Distinguish between correlation and causation.

Correlation DOES NOT MEAN causation

Chapter 3

1) Describe “raw scores” and explain their limitations.

­ raw scores​­ simply the number of items scored in a specific manner ­ often of limited use

­ to give raw scores more meaning, we need to transform them (i.e. scaled; derived; standard) scores

2) Define norm­referenced and criterion­referenced score interpretations and explain their major characteristics.

­ norm­referenced​­ examinee’s performance is compared to a specific level of performance

­ most of tests psychologists use are norm­referenced

­ criterion­referenced​interpretations are absolute

­ compared to an absolute standard

­ often used in educational settings

3) List and explain the important criteria for evaluating standardization data. ­ representative

­ appropriate for application

­ current (i.e. Flynn Effect)

­ adequate size (bigger the sample, the better the applicability)

4) Describe the normal curve and explain its importance in interpreting test scores. 5) Describe the major types of standard scores.

­ z­scores: mean of 0 and SD of 1

­ t­scores: mean of 50 and SD of 10

­ IQs: mean of 100 and SD of 15

­ CEEB scores (SAT/GRE): mean of 500 and SD of 100

6) Convert standard scores (e.g., z­scores, T­scores, etc.,) from one format to another. ­ formula:

x−xSS1 

­ new standard score = x SS2 + SDSS2 x ( ) SDSS1 

­ X = original standard score

­ x SS1 = mean of original standard score format

­ SDSS1 = standard deviation of original standard score format

­ x SS2 = mean new standard score format

­ SDSS2 = mean of new standard score format

7) Define normalized standard scores and describe the major types of normalized standard scores. ­ Stanine Scores: mean of 5 and SD of 2

­ Wechsler scaled scores: mean of 10 and SD of 3

­ Normal Curve equivalents (NCE): mean of 50 and SD of 21.06

8) Define percentile rank and explain its interpretation.

­ percentile rank­ reflects the percentage of people scoring below a given point 9) Define grade (and age) equivalents and explain their limitations.

­ grade equivalents­ based on percentile ranks

­ many limitations

­ subject to misinterpretation

­ age equivalents­ same limitations as grade equivalents

10) Describe some common applications of criterion­referenced score interpretations. ­ percentage correct

­ i.e. score of 85% on classroom test

­ mastery testing

­ i.e. pass/fail on driver’s license exam

­ standards­based interpretations

­ i.e. assigned an “A” to reflect superior work

11) Explain how tests can be developed that produce both norm­referenced and criterion­referenced interpretations.

­ “For example, it would be possible to interpret a student’s test performance as ‘by correctly answering 75% of the multiplication problems (criterion), the student scored better than 60% of the students in the class (norm).’”

12) Describe Item Response Theory and the properties of IRT or Rasch­type scores. ­ IRT scores are similar to traditional raw scores

­ they can be transformed to norm­ or criterion­referenced scores

­ but­ they are interval level scores, and they have stable standard deviations across age groups

­ go by different names:

­ Rasch or IRT­scores

­ Change Sensitive Scores (CSS)

­ W Scores (WJ­III)

13) Give an example of a qualitative score description and explain its purpose. ­ Qualitative Descriptions of test scores

­ helps communicate test results in an accurate and consistent manner ­ i.e. Stanford­Binet Intelligence Scale