- the most generic term

- gathering data on students and using it to make decisions. 

Measurment 

- assigning meaningful numbers (or sometimes names) based on the degree to which a student possess a characterictic. 

 - Ex: scores on exams 

- mechanical porcess

- counting up the number of things a student gets right 

assigning a value to a number

- EX: 93 is an A 

Formal 

vs

 Informal Assessment 

- any assessment technique where the entire class gets assessed in a standardized way, using the same instrument. Ex: exams, quizes, graded homework.

- individualized btwn students. teacher walking around the room. Common in younger grades, used too little in older grades 

- anytime assessment device is used while a teaching unit is in progress. 

- Purpose is ti give teacher feedback about how a unit is in progress

- Ex: teaching prerequisite skill.

- should not be used for grading . 

- assessment done at the end of a unit or chapter. 

- a summary 

- Used to calculate grades 

- collges use 

- tells teacher where students need to be 

- reading diagnostics 

- used to diagnose a problem 

Maximum 

vs

 Typical Performance Assessment 

- what we use in this class *** 

- how well a student can do 

- not measuring how well, but what is typical 

- how do you interpret a raw score ? 

- 4 ways to interprert test scores 

- 2 problematic; 2 used frequently 

Ability Referenced Interpretatons 

- how does the actual score compare to what the student is capable of doing ? 

- PROBLEM: how do we know what a student should be able to do?

- influenced by more than ability 

- so many factors can affect, it is hard to judge

- not the best method,. 

- pre test to post test --> observe the growth between the two 

- post - pre score- improvment score 

- PROBLEMS:

1. every test contains some sort of measurment error. 

2. different score is much more unreliable than they seen

3. learning curve: some students plateau earlier than others. Never know for sure where each student is at at a given point. 

- you interpret students score by comparing it to the scores of others who took the same test

- or compare students to others in the clasroom 

- competitive systems: competing with other students

- students tend to aviod challenging course work 

- tells how they do compared to others, but know how much theye learned. 

- Unit tests 

- score is compared to some pre set criteria 

- the grading scale 

- domain sampling model: does the test cover the full domain? each item on test should measure each objective given. 

- growing pop.; not competitive 

- if done right, very easy to see how much is learned 

- chapter tests 

Developing objectives 

Learning vs performance 

- instruction = learning = performance 

- assumption is the better you perform, the more you learn 

Standards 

Goals 

Objectives 

- set by nations professional organizations. developing appropritae standards that are adopted by states that are adopted by schools. 

- set by teachers. a general statement of anticipated outcome 

- set by teachers. much more specific. based on task analysis 

- business to militaty to education 

- writing objectives 

- 3 domains 

- cognigitve- what students should know 

- affective 

- psychomotor- physical movement learning 

Knowledge- remembering 

Comprehension- understanding 

 Application- apply to real life

Analysis- taking something apart

Systhesis- putting back toegeher

Evaluation- look at idea and recognize flaws.  

- 1962

- approach to writing objectives 

- program instruction textbook 

1. Action Verb- students demonstrate using action. List, calculate, solve, read. Aviod: know, understand, remember. 

2. Criterion- how well do the students have to perform in order to say theyve met the objective 

3. Conditions of Assessment- under what circumstances must the student perform under. When, time limit, group, single, memory. 

- looked at learning theoires and translated them into instruction 

- own approach to objectives 

1. building blocks for lesson plans 

2. used in building tests 

- way to sample objectives for exams 

- used Blooms taxonomy 

- characteristic of measurment device 

- reliability = consistency 

- is the test consistent ? 

- observed score = true scores + measurment error: anything that causes observed score to be different than the true score 

1. Subject effects: anything about a perosn that influences the way they do on a test. Ex: ill, medication, sleep, events. 

2. Test Effects: bad instructions, ambiguous questions, non objective scoring, time limits 

3. Environmental Effects: open window, temperature, lighting, crowding, desk size, time limits. 

- test a student once, later test the student again to see if there scores are the same. 

- issues- if you take the tests now and then again in 2 weeks, you could remember the test. 

- if you wait to long, you could change. 

- not good for classroom tests 

- two differend exams that are measuring the same objectives 

- common on standardized tests 

- only make one test and give to students one time. 

- are all the items measuring the same skill set ? 

- college exams variety of skill sets --> low internal consistency 

2 types 

Split Half Reliability- look at how students do on odd or even numbers on exams. 

KR 20- compare item variances vs total score variance 

- standardized tests we should see reliability about about .9 or higher 

- if skill is designed to measure intell/personality (test retest), reliability should be at .8 or higher 

- classroom tests reliability should be btwn .6 or .8 

- hard to deal with subject effects 

- can control test affects 

- can improve the way we score tests 

- most important 

- does the test measure what we think it should be measuring? 

- using a test to make 2 different decisions 

- most important for classroom tests 

- domain sampling model- how well does test measure what was taught 

3 types 

1. Instructional Validity- how good of a match is there btwn what you taught and what you tested 

2. Curricular Validity- how well does your test match the officail curriculum? - content standards 

3. Face Validity- do the items seem appropriate? 

- think of test as a short cut to measure a skill that you could measure in another way that would take longer. 

- question is how good is the shortened version ? 

1. Concurrent Validity- measuring the two skills at the same time 

2. Predictive Validity- give test now to try to predict how youll perform in the future (SATS, Job placement) 

- Construct --> theory 

- is the test measuring what the theory says it should 

- IQ test 

- EX: correlating IQ scores with problem solving scores 

- validity in most important 

- relaiabilty effects validity 

- if test not reliable --> measurment error --> cannot measure what supposed to

- reliability is a necessary, but not sufficient condition of validity 

- reliabilty sets the upper limit for validity 

- can have high reliable, but not valid 

- validity can never be higher than reliability