Formal Assessment Handout

and

RICA Domains Study Grid (A Sample)

Types: norm-referenced and criterion referenced

Norm referenced tests (standardized)

•There are statistics or norms for comparing the performance of a student to a large sample of similar students (the norm sample).

•The norm sample represents the general population

•Each student in the norm sample takes the test and norms are established for how well, for example a fourth grader does on the test

•Procedures for test administration have to strictly adhered to in order to compare students in your classroom to the norm sample accurately.

Criterion referenced tests

•A student's performance is compared to a specific standard or criterion rather than to the norm sample

•This test determines whether a student has mastered certain competencies or skills (Can the student recognize ing endings or find the main idea in a paragraph)

•These tests measure mastery rather than grade level, or they describe rather than compare

Bias in testing

•The content of the test should represent the experiences and values of all groups taking the test (a test should not contain vocabulary, pictures or stories that are unfamiliar to certain populations taking the test)

Scores on norm referenced tests

•Scores on a norm referenced test indicate how an individual student has performed compared to students of the same grade or age level who are in the norm sample

Raw scores represent the number correct on the test

•Raw scores are then converted into derived scores which can be used for interpreting the student's performance

•Derived scores can be reported as standard scores, reading grade scores, percentiles, normal curve equivalents, and stanines.

•Many scores are based on a normal curve distribution of scores (the hump or where most of the scores fall)

•The highest point on the curve is the mean or average point

Standard scores

•Standard scores refer to scores in which the mean and the standard deviation (which is a measure of variations) have been assigned preset values. For example, in the WISC III, the mean is set at 100 and the standard deviation at 15.

Reading grade score

•This score indicates how well a student reads in terms of grade level. For example, a score of 4.5 (the fifth month of a fourth grader) indicates that the student correctly answered the same number of questions on this particular test as the average pupil in the fifth month of the fourth grade.

•Reading scores do not indicate an absolute performance; they indicate how the student performed in relation to the students in the morn sample population

Percentiles

•Percentiles can be understood as a rank within 100, expressed in numbers from 1 to 99.

•A perecntile ranl is the percentage of students that scored lower than the student being tested. For example, a percentile score of 57 indicates that this student scored higher than 57 percent of the comparison group and lower than 42 percent.

•The 50th percentile indicates the median (or middle) score

•The highest percentile is 99; the lowest is 1

•Equal distances in percentiles do not indicate equal difference in raw score (the correct number on the test) points.

•Since many scores group towards the middle, the difference between the 50th and the 60th may be only a few raw score points. However, the difference between the 99th and the 80th represents a great many raw score points.

Normal curve equivalent scores (NCE)

•NCE scores are similar to percentiles in tat they range from 1 to 99 with a mean of 50.

•They differ from percentiles because they have been transformed into equal units of reading achievement. For example, the difference between the 50th and the 60th NCE and the 80th and the 90th NCE is the same in raw point scores (the number correct on the test)

Stanines

•The stanine score ranks pupils from 1 to 9.

•The lowest score is 1, the median is 5 and the highest is 9

•Stanine scores represent a normal distribution. Thus, in an average class, most students will receive stanine scores of 4, 5, or 6 and a few will receive stanines of 1 or 9.

•The word stanine is a contraction of standard nine and is based on the fact that the score runs from 1 to 9.

Standardization, Validity, Reliability

•To standardize a test, it is given to a large number of students (the norm sample)

•Based on data derived from the norm sample, inferences are made about other students who take the test.

•Look at the norm sample and consider whether the norm was large enough to establish performance norms. Does the norm sample include students who are being tested?

•Some school districts develop their own local norms.

Validity refers to whether a test measures what it is supposed to measure

•There are two types of validity: content validity and criterion validity

Content validity involves inspecting the test to see whether the items are valid for testing purposes. For example, a valid reading comprehension test would probably contain passages with questions. A comprehension test that asked students to match words would have questionable validity

Criterion validity refers to how the test compares with some other aspects of achievement such as grade point average. The comparison is done with a statistical correlation (high positive: +1.0 to low: -1.0)

•For acceptab;le criterion validity, the correlation should be positive and high

•Reliability refers to the stability of the test.

•If a test is reliable, the person will receive the same score on repeated testings.

•To be useful for making decisions about an individual, a test must have high reliability

•If test scores are reported for groups for general administration purposes or screening, a reliability score of .80 is acceptable.

•For individual placement such as Title I, the reliability of the test should be .90

•There are two forma of reliability: test-retest and internal

•In test-retest, student take the test two times. Then the scores are correlated to determine whether individual students perform about the same in the first and the second administration

•In internal reliability items within a test are compared with each other. In one form of internal reliability, split half reliability, a group of students' scores on one half of the test items is correlated with performance on the other half of the items on the test.

 

Sample RICA Grid

 

Domain How to Assess How to Teach It (Strategies) Accommodations
I. (1)Ongoing assessement:

Multiple measures

•Small group

•Individual students

•Class progress

Communicate results to parents, teachers, administrators, guardians

(2) Planning, organizing and managing reading instruction

Informal assessment:

Miscue analysis

Informal reading inventory

Running records

Writing sample

Retelling

Comprehension questions

Independent reading level 99-100%

Instructional reading level 98-95%

Frustration reading level 90 and below

Formal assessment

SAT 9

Percentiles

Stanines

Validity

Reliability

Grade equivalent score

Short term and long term planning: •differentiated instruction

•Timely interventions

•flexible grouping

•Individualized instruction

Use assessment results to select appropriate materials at students' independent reading level to support independent reading.

•leveled books (decodable text)

•predictable text

•variety of genres

Domain How to Assess How to Teach It (Strategies) Accommodations
II. Developing Phonological and Other Linguistic Processes Related to Reading

(3) phonemic awareness

(4) concepts about print

(5) Systematic, explicit phonics and other word identification strategies

Yopp-Singer Test of Phonemic Awareness:

•assess auditory awareness

•discrimination of sounds

•segment words

•blend the sounds

Test understanding about book knowledge, letter knowledge, words, left to right directionality while reading, understanding of language conventions (punctuation, capitalization) - use a self-made survey or Jerry Johns IRI or Marie Clay Observation Survey

Use decoding tests such as the Sheffelbine Test of phonics, Jerry Johns IRI that has a test of word identification, San Diego Quick Test of word identification, timed fluency tests (for rate and accuracy), sight word checks (Dolch Word List, Frye Sight Word List)

Use a word wall

Have students make words using syllables, onsets, rimes, phonemes

Teach about onsets and rimes

Teach students to rhyme, blend, substitute sounds and letters, delete sounds

Teach about word boundaries (syllables)

Teach letter recognition - upper case and lower case. Use multisensory techniques such as felt letters, tracing letters and words in the sand, pudding, etc.

Teach about their names and the letters with which their names begin.

Select appropriate materials for second language learners and special education students

Select materials that support the learning of phonemic awareness such as song, rhymes, stories that repeatedly use the same phonemes

Provide lots of opportunities for special needs students to have exposure to plastic letters, magnetic letters, sand paper letters, song with the alphabet, materials that focus on individual letters

Return to Credential Students