Relationship between Standardized and Classroom based Assessment Principles

Relationship between Standardized and Classroom -based Assessment Principles Manuel T. Barrera, Ph. D. 1/7/2022 1

Introduction � Two methods will be discussed ◦ Principles of Standardized Assessment ◦ Curriculum-based measures of student progress Manuel T. Barrera, Ph. D. 1/7/2022 2

Assessment Principle 1 � Every test is an estimate of student performance based on ◦ ◦ Knowledge/skill taught Knowledge/skill acquired Knowledge/skill tested Additional psycho-social variables � Test = data + error Manuel T. Barrera, Ph. D. 1/7/2022 3

Assessment Principle 2 � Tests are used to measure student competence not student worth. ◦ A student receives an “A” ◦ A student is NOT an “A student” Manuel T. Barrera, Ph. D. 1/7/2022 4

Assessment Principle 3 � Tests measure different dimensions of competence ◦ Some tests measure student competence against (hypothetically) all possible other students �All standardized, norm-referenced tests ◦ Some tests measure students against a standard �Criterion-referenced ◦ Some tests measure students against what was taught �Curriculum-based assessments Manuel T. Barrera, Ph. D. 1/7/2022 5

Assessment Principle 4 � Any test can be standardized ◦ Standardization is function of turning observed results into a measure of normality Manuel T. Barrera, Ph. D. 1/7/2022 6

Assessment Principle 5 � Assessment a grade. results are not the same thing as ◦ A grade is a subjective appraisal of the teacher according to stated/unstated criteria ◦ Test results are objective appraisals of material tested Manuel T. Barrera, Ph. D. 1/7/2022 7

Assessment Principle 6 � All tests are subjective ◦ Teacher-made tests are influenced by what the teacher considers appropriate material ◦ Standardized Tests are influenced by what test makers consider appropriate material � Objectivity is a goal based on ◦ Experience ◦ Incorporation of multiple sources Manuel T. Barrera, Ph. D. 1/7/2022 8

Assessment Principle 7 � Some tests are primarily used to inform � Some tests are primarily used for decisions ◦ Formative ◦ Summative Manuel T. Barrera, Ph. D. 1/7/2022 9

Assessment Principle 8 � Assessors make decisions not tests. ◦ Tests help make informed decisions ◦ Teachers make decisions Manuel T. Barrera, Ph. D. 1/7/2022 10

Assessment Principle 9 � First, you decide WHY you are testing, then you decide what the test results mean. ◦ Test Validity is based on the congruence between what is being measured and what is being tested. Manuel T. Barrera, Ph. D. 1/7/2022 11

Assessment Principle 10 �A test is only as valid as it is consistent ◦ Reliability is based on whether tests measure what they do consistently irrespective of time, raters, or items (internal consistency). Manuel T. Barrera, Ph. D. 1/7/2022 12

What are standardized tests good for? � Designed to be consistent and valid � Represent a broad comparison measure (national, regional, or local norms) � A better measure of competence across extant curricula or competencies Manuel T. Barrera, Ph. D. 1/7/2022 13

What are criterion-referenced measures good for? � Designed to examine student competence in relation to a specific standard of competence � Represent a specific comparison measure of an existing curriculum Manuel T. Barrera, Ph. D. 1/7/2022 14

What are curriculum-based measures good for? � Identifying learner skills and knowledge in context of instructional ecology ◦ ◦ Classroom environment Classroom expectations Instructional environment Instructional expectations � Designed to examine student competence of actual material to which the student has been exposed. � When standardized, they measure student comparison to existing classmates Manuel T. Barrera, Ph. D. 1/7/2022 15

Manuel T. Barrera, Ph. D. 1/7/2022 16

Curriculum-based Assessment Model Organize Data Display Data Interpret Data Summarize Data Prepare • What will you assess? • How will you assess? Collect Data Manuel T. Barrera, Ph. D. 1/7/2022 17

Vocabulary � Percentiles � Standard Scores Manuel T. Barrera, Ph. D. 1/7/2022 18

Examples of Norm-referenced Tests � MAT � SAT � CAT (see page 196 of your text) Manuel T. Barrera, Ph. D. 1/7/2022 19

Examples of Criterion-Referenced Tests � Profiles of Learning � Tests developed by book publishers for adopted texts Manuel T. Barrera, Ph. D. 1/7/2022 20

Examples of Curriculum-based Assessments � Curriculum-based measures ◦ Reading of classroom-based texts ◦ Timed writing samples ◦ Timed math computation Manuel T. Barrera, Ph. D. 1/7/2022 21

Percentiles � Derived score indicating the percentage of scores that fall below a given score. � Distribution is based on the median of scores � %ile = %below score + (0. 5)(% getting a score) Manuel T. Barrera, Ph. D. 1/7/2022 22

Calculating a percentile � order all scores highest to lowest � place equal scores one above the other � take a targeted score and calculate percent all those geting the score � multiply target score percentage by 0. 5 � calculate percentage of all scores below the target score � add 0. 5*%getting the score with % below the score. Manuel T. Barrera, Ph. D. 1/7/2022 23

Derived Scores: Measures of Relative Position � z-scores � T-scores � IQ scores Manuel T. Barrera, Ph. D. 1/7/2022 24

z-scores � Defined Z= as a mean of 0 and a SD of 1 X-X SD Manuel T. Barrera, Ph. D. 1/7/2022 25

T-Scores � Derived 10 score with a mean of 50 and SD of T = 50 + 10(z) Manuel T. Barrera, Ph. D. 1/7/2022 26

IQ scores � Derived 15 score with a mean of 100 and SD of ◦ in some cases SD = 16 � IQ = 100 + 15(z) More Broadly: SS= ss + (sss) (z) Manuel T. Barrera, Ph. D. 1/7/2022 27

Percentiles � Derived score indicated the percentage of scores that fall below a given score. � Distribution is based on the median of scores � %ile = %below score + (0. 5)(% getting a score) Manuel T. Barrera, Ph. D. 1/7/2022 28

Calculating a percentile � order all scores highest to lowest � place equal scores one above the other � take a targeted score and calculate percent all those geting the score � multiply target score percentage by 0. 5 � calculate percentage of all scores below the target score � add 0. 5*%getting the score with % below the score. Manuel T. Barrera, Ph. D. 1/7/2022 29

Student retelling � Students read and paraphrase as they go along � Students read entire passage and paraphrase at the end � Procedures for scoring include ◦ Percent total words retold per words in the passage ◦ Percent content words retold per content words in the passage Manuel T. Barrera, Ph. D. 1/7/2022 30

Timed Oral Reading � Fluency of reading is highly associated with reading comprehension ◦ Measure words read correctly per total words read ◦ Correct words per minute � See Salvia and Hughes (1990) for detailed procedures Manuel T. Barrera, Ph. D. 1/7/2022 31

Curriculum-Based Measures of Written Expression (Shinn, 1989) � Story starters ◦ Read one-sentence story starter ◦ Student thinks about what to write for one minute ◦ Student continuously writes a story for three minutes � Scoring options ◦ Total words written (correct or incorrect spelling, no numbers not spelled, title, names) ◦ Words spelled correctly ◦ Total letters written (spelled correct or incorrect) ◦ Correct word sequences (information units): connected words = one sequence, numbers next to words are not. Manuel T. Barrera, Ph. D. 1/7/2022 32

Sample CBM (Shinn, 1989) � Correct math digits CBM ◦ Single skill-Computation �Provide worksheet (s) of specific computation problems on one of four basic operations +, -, *, / at grade level � 2 -minute probes for answering all problems possible �Hard problems: mark an X on them and go on ◦ Scoring: count the number of correct digits in the work sample �Place value is important �Point values assigned for more complicated problems (e. g. , long division) Manuel T. Barrera, Ph. D. 1/7/2022 33

Variability (Dispersion) � Measures of Central Tendency ◦ Mean, Median, & Mode ◦ Variance and Standard Deviation � Descriptive Statistics Manuel T. Barrera, Ph. D. 1/7/2022 34

Relationship of Derived Scores z -2 -1 0 1 30 40 50 60 70 70 85 IQ 100 115 130 T 2 Percentiles 1 5 10 20 30 40 50 60 70 80 90 95 Manuel T. Barrera, Ph. D. 99 1/7/2022 35