-
preevaluation process (environmental, linguistic and cultural variables)
- prior to formal assessments
- comprehensive teacher input on student's classroom performance and comparison to peers
- interview teacher
- teacher questionnaire
- interview teacher with questionnaire
- conduct at least one classroom observation (preferably of core curriculum activity)
- assess student's lang in l1 and l2
- remember to consider subtractive bilingualism, time and quality of instruction in each lang and time second lang has been spoken
-
assessments
- standarized tests
- -most commonly used for articulation or lang disorders
- -allow comparison of individuals performance to the group, answering the question "how does this person compare to the average
- -indicates whether individual is at, above or below the average
- -may not always be used for CLD
- -distribution normally uses bell-shaped curve
- -mean determines the peak demonstrating average performance
- - mean also demonstrates the peak which is the middle of the distribution
- -mode which is the most frequently occurring score
- -current eligibility standard is
- -standard deviation = 1.5 sd below the mean or 7th percentile
-
assessments
what are the terms for the bell-curve
mean-the sum of list of numbers, divided by the total number
-
assessments
criterion referenced tests
- used for neurogenic disorders, fluency disorders and voice disorders
- comparison to an expected level of performmance
-
assessments
authentic assessment approach
alternative assessment/nontraditional assessment
-
assessments
portfolio assessments
- identifies what a person can and cannot do
- most effective to observe in a real-life situations
-
validity
a test truly measures what it claims to measure (face validity)
-
content validity
completeness of the test-are items relevant to measuring what test says it measures and do the items measure the full range of skills
-
construct validity
degree to which the test is able to measure a predetermined behavior or attribute (theoretical construct). considered key to test development stanford-binet
-
concurrent validity
degree to which a new test correlates with an established test of known validity. EVT correalating to the PPVT
-
predictive validity
test's ability to possibly predict performance. GRE predicting future academic performance (some people may not do well on these types of tests)
-
how can standardized tests be misused
- (assessment needs to match the population you are testing)
- used with the wrong population
- age
- ses
- may not identify all areas of disability
- may not fully describe the problem (may only touch the surface)
- how many items on the test for each area (one question doesn't identify a disability)
- outdated version
- revised versions usually have different or expanded populations and norms
-
what are some considerations for using standardized tests with cld:
- does the child feel comfortable in the testing situation with the tester
- will they guess if they don't know the answer
- will they try to do their best
- understand what they are being asked to do
- standardized tests may ask children to do something that is contrary to their customs therefore them must match their normed group
-
bias on tests
- 4 types:
- Cultrual linguistic
- value
- format
- examiner
-
bias on tests
cultural-linguistic bias
- content bias
- activities that don't match with the child's experimental base
-
bias on tests
value bias
test items assume a value system different than the child's system
-
bias on tests
format bias
- situational bias
- use of testing procedures, materials or both that is less familiar to some children (not willing to talk to strangers)
-
bias on tests
examiner bias
- how they administer or interpret tests
- how to modify standardized tests
- collecting lang smaples
- -take over several sessions in many settings
- -have them talk to several people including family
- -take into consideration lang differences
- -analyze both form and content in both langs
- -use structured lang sample techniques to ensure they talk
-
what are some alternatives to standardized assessment problems
- look for assessment strategies that are not biased
- extend time limits
- reword instruction
- provide practice of types of items
- evaluate lang of test (ask for things not in the child's native lang)
- use norm referenced tests that allow description of child's skills
- -CLD children do better on verbal tests so add to battery
-
larry p 1986
suit that claimed that iq test was biased against aa children so in state of ca aa children can not be tested on iq. invalid test.
-
assessment alternatives
- assessment of narrative skills to determine if student can retell story with details
- sequence
- organization of info
- relevant/irrelevant comments
- portfolio assessments
-
use of interpreters
- attempt to use those who are from same geographic area/speak the same dialect
- train interpreter
- demonstrate how to use test, how to document student responses
- -ensure they translate the info to discuss it with you when they have questions
- -watch them administer the test
- -take to them about their scoring
|
|