| CASL FAQ: | |
| Question |
Answer |
| Do I need to buy both the OWLS & CASL? What is the difference between the two tests? |
It depends on your reason for testing. Both the OWLS and the CASL are by the same author and are based on the same theory of language. The OWLS is a relatively quick measure of receptive (Listening Comprehension), expressive (Oral Expression), and written language (Written Expression, which is packaged separately). Listening Comprehension and Oral Expression are composed of item tasks in the four language categories of lexical/semantic, syntactic, supralinguistic, and pragmatic. However, examinees don't take enough items in each of those categories to report separate category scores in OWLS. CASL takes that next step. CASL provides multiple tests (15 total) in all these categories, so clinicians can make an in-depth evaluation of language skills.
If you are looking for a quick receptive and expressive measure of language (especially one with a written component that will give you a global language composite), choose OWLS. If you need an in-depth diagnostic measure that gives you the flexibility of 15 highly reliable, stand-alone tests, choose CASL. Many clinicians need both. OWLS helps them determine that the examinee has a receptive or expressive processing problem. CASL gives clinicians a wealth of diagnostic information to pinpoint specific areas of concern and help in planning interventions. 
|
| I understand the CASL is a battery of tests, not subtests. What is the difference? |
Each of the 15 CASL tests is a highly reliable, stand-alone test. The clinician is free to give one test or several tests and can report the score(s) with confidence.
In tests other than the
CASL, a single subtest score is generally not reported because it is not
reliable enough to stand alone.
The subtest must be given in conjunction with other subtests to form a composite score. CASL gives the clinician the flexibility to administer only the necessary tests.
|
| Can I use the CASL with bilingual children? |
The CASL was developed to assess spoken Standard American language. The standardization sample included children of Hispanic ethnicity, but they were fluent in English.
If the child is not fluent in English, then of course reporting a standard score would not be appropriate. But a descriptive analysis of a child's English skills can be done with 9 of the CASL tests.
If you are using the CASL with children who are acquiring English as a second language, I would suggest doing the following:
First administer and score selected CASL tests according to the procedures and rules provided in the manual and Test Books. (The only procedure I would alter is the start item. I would probably start with item 1.) Then use the appropriate descriptive analysis sheet to look at strengths and weaknesses in the area of semantics or syntax.
|
| Can you administer English speech and language tests to ESL children? |
You can administer any of our speech and language tests to ESL children and get some information about the child's English language skills. At least they will give you a baseline of receptive or expressive skills from which to measure progress. You just can't report a standard score. We recommend using OWLS and CASL for this purpose because of the descriptive analysis sheets, which allow you to look more closely at the structural aspects of the child's language.
|
| Do I have to give all 15 tests? |
Very seldom would anyone need to administer the entire battery of tests. Most people choose to give the Core Composite (three, four, or five tests depending on age), which takes only about an hour at most, and less with young examinees. Of course, administration also gets easier and quicker the oftener you do it.
|
| Why do the category and processing indexes require supplemental tests? |
The Index scores are not Core scores. They are additional scores that you can get when you want more information about a certain area of language. They include supplementary tests to enable comparisons within categories or processes. The Core tests for a certain age are the most representative tests for that age. They include at least one test from each of the four categories: Lexical, Syntactic, Supralinguistic (for older children), and Pragmatic. They also include at least one test from each of the two processes: Receptive and Expressive. The Core tests are best if you need to report a global language composite.
The CASL enables you to be very flexible depending on the types of scores you need to report. Each CASL test is a reliable, independent test, the score of which can be reported with high reliability. If you're concerned about an examinee's grammar, you can give just the Syntactic tests to get the Syntactic Category Index score. If it's more important to report Receptive and Expressive scores, just give those tests to get the Processing Index scores. On the CASL record forms, color indicates which tests are receptive (brown) and which are expressive (blue or green).
|
| How can I report a receptive or expressive score if the child I'm testing is out of the age range for the Processing Index score? |
Because each CASL test is an independent, highly reliable test, you can report a single receptive or expressive score with confidence. However, you may also choose to average the examinee's receptive or expressive scores and report that score. On the CASL record forms, color indicates which tests are expressive (green or blue, depending on the form) and which are receptive (brown).
|
| Do responses to CASL tests have to be grammatically correct? |
The answer to this question depends on the test. On tests that require one-word responses, the response must be appropriate to the item. For example, in Sentence Completion, the word used to complete the sentence must be the proper part of speech to make that sentence grammatically correct. In Antonyms, the antonym that is elicited must be the same part of speech as the prompt given. Although the scoring of these tests does not specify grammatical correctness as such, it is understood by looking at the correct responses listed.
The expressive Syntactic tests all have a grammatical component. Because Syntax Construction is an open-ended test measuring the examinee's ability to generate sentences in correct syntactic form, the responses must be grammatically correct. However, the examinee does not always have to answer in a complete sentence. Sometimes the item focus is simply a prepositional phrase or a past tense verb. In those cases, just a grammatically correct prepositional phrase or predicate is sufficient for a response. The scoring rule for each Syntax Construction item clearly indicates what is expected. Grammatical Morphemes requires a completion of the analogy with the grammatically correct structure. Grammaticality Judgment requires the examinee to judge the grammatical correctness of a given sentence and, if it is incorrect, then change it to be correct. All responses must be grammatically correct.
The Supralinguistic and Pragmatic tests are open-ended, and responses do not have to be grammatically correct.
|
| In Syntax Construction, how do I handle responses that are semantically meaningful and grammatically correct but do not include the target grammatical form? |
In trying out the test, we tried to omit items that prompted a wide variety of responses and include only items that the majority of examinees answered with the expected structure. As with every open-ended test, however, it is impossible to make all examinees respond the way you want them to respond. If you get a response that you feel is correct but does not fit the scoring rule, count it as correct and notify AGS. It's helpful to have that information for reprint. If you question whether a response is correct, call us.
|
| What part does memory play in Paragraph Comprehension? |
The author believes that what makes understanding difficult is not the length of the sentence but the complexity of the content and syntax of the message. The comprehension of sentences in meaningful contextual material depends on examinees deriving the meaning of what is said or read, not on their memory of specific words. Once they get the gist of a paragraph, they will have little difficulty answering questions about it. The length should not influence their ability to comprehend, but the complexity of the sentences will. Even if a sentence is long, if listeners can decode it and get the gist of it, it will be understood. If individuals cannot understand or remember the paragraph because of the number of sentences in it, they will have the same kind of difficulty in any listening situation where connected speech is used.
For a discussion of the content, format, and development issues concerning Paragraph Comprehension, see Chapter 4 of the CASL manual (pages 49-50 for this issue specifically). The author also addresses the topic of memory on pages 28-29 in the manual.
|
| Why aren't scores from CASL and, for example, CELF-3 comparable? |
Score differences between tests can occur for numerous reasons: (1) The two tests are not measuring the same construct; (2) The construct is not being measured in the same way; (3) The standardization samples on which the tests are based are not the same.
In the case of CASL and CELF-3, all of the above apply. Chapter 4 in the CASL manual tells you specifically what each CASL test is measuring and how to interpret results. A big difference between CASL and CELF-3 is the standardization samples. The CELF-3 technical manual states that "no examinee in the standardization sample was receiving language therapy or had a diagnosed or identified language disorder." This means that the CELF-3 norms are based on children not identified as special education students. CASL, like all AGS tests, is standardized on subjects with a full range of ability. Special education students are included in approximately the same percentages that they occur in the population.
You might be interested in reading an article in Language, Speech, and Hearing Services in Schools (Volume 27, Number 1, January 1996) called "Creating Language Impairments in Typically Achieving Children: The Pitfalls of 'Normal' Normative Sampling" by Teresa Ukrainetz McFadden. Many school districts are now insisting that testing be done with assessments based on representative norms that include special education students.
|
| Can CASL be administered to vision-impaired children? |
Because CASL is basically a test of spoken language, much of it can be used with blind or visually impaired examinees. Only three of the 15 tests are completely dependent upon pictures. Two of these are receptive tests: Basic Concepts and Paragraph Comprehension. The third, Syntax Construction, is an expressive test that requires very open-ended responses, so the pictures help to focus the examinee's answers. Sentence Completion, Grammatical Morphemes, and Pragmatic Judgment use a few pictures for the youngest examinees, but pictures are certainly not necessary and a few extra minutes spent explaining the examples might be all that is needed to help a visually impaired examinee to get comfortable with the task. Nine tests (Antonyms, Synonyms, Idiomatic Language, Sentence Comprehension, Grammaticality Judgment, Nonliteral Language, Meaning from Context, Inference, and Ambiguous Sentences) require no pictures.
Remember that each test is a stand-alone test, with its own standard scores, percentiles, NCEs, stanines, and test-age equivalents. Consequently, you can get much information about an examinee's language even without giving the picture-dependent tests.
|
| In the Paragraph Comprehension test, an examinee can get a basal and a ceiling on the first paragraph set if he or she answers either of the questions incorrectly. Does this give a true score for that ability? |
If an examinee reaches a basal and a ceiling in the first paragraph set, please go on to administer the second paragraph set. Chances are that the examinee won't get too far in that set, but at least you will have more to base the score on. Use the ceiling rule with sets B through H.
|
| Does the CASL have descriptive analysis worksheets similar to the OWLS? |
Yes, descriptive analysis worksheets are available for the Lexical/Semantic tests of Comprehension of Basic Concepts, Antonyms, Synonyms, and Sentence Completion. Worksheets are also available for the Syntactic tests of Syntax Construction, Grammatical Morphemes, Sentence Comprehension of Syntax, and Grammaticality Judgment. These worksheets break down each item by task. You will have a handy means of assessing which skills appear to be problematic and which appear to be strengths. These worksheets are available to be downloaded.
|
| To what extent is a test confidential? What information can parents be given? |
The child's parent has the right to "inspect and review" the completed record
form, in the presence of the qualified person who administered the test or another
qualified professional. The parent does NOT have the legal right to make a copy
of the form or to hand copy the information from the form. The only part of the
record form that can be photocopied is the page that contains the score summary
information (often the first page).We grant that permission so professionals
working with the child can share that information with each other.
We have obtained copyright protection for all test materials. Details of
test items should not be released to persons who are unqualified to administer
or interpret them. According to IDEA guidelines, the child is best served by
having an expert interpret his/her test performance. Understanding the significance
of a child's performance on a test requires theoretical grounding in child
development and measurement and expertise in interpreting technical data, including
norms and validity studies. Most parents do not have this understanding or
expertise, which is why the federal guidelines were written so that only in
the presence of a qualified professional should the parent be shown the completed
record form.
|
| In our district and state we are still required to derive a total language score and then show a discrepancy between receptive and expressive language abilities. Has anyone come up with a way of documenting this using the CASL?
|
In an attempt to make the CASL record forms more user-friendly, we have decided to use color to indicate which CASL tests are receptive and which are expressive. On the front page of each record form, in the Raw Score to Standard Score Conversion box that lists the tests, the expressive tests will be printed in brown and the receptive tests will be in green or blue (depending on form).
If examiners wish to get a receptive or expressive score but are testing a child that doesn't fit the age level for the receptive and expressive processing indexes, they can average the appropriate scores. Also, some examiners may be interested in looking at only a single category (for example, Syntactic) and would want to give one receptive and one expressive test from that category.
The receptive tests on Record Form 1 are:
- Basic Concepts
-
Paragraph Comprehension
The expressive tests on Record Form 1 are:
- Antonyms
-
Sentence Completion
-
Syntax Construction
-
Pragmatic Judgment
The receptive tests on Record Form 2 are:
- Synonyms
-
Paragraph Comprehension
-
Sentence Comprehension
The expressive tests on Record Form 2 are:
- Antonyms
-
Sentence Completion
-
Idiomatic Language
-
Syntax Construction
-
Grammatical Morphemes
-
Grammaticality Judgment
-
Nonliteral Language
-
Meaning From Context
-
Inference
-
Ambiguous Sentences
-
Pragmatic Judgment
These tests are designated as receptive or expressive solely on the basis of whether the examinee has to vocalize a response. Chapter 3 in the manual presents an excellent explanation of why very few of the tests are purely receptive (comprehension) or purely expressive. Many are a combination. It may be helpful to review this chapter and Chapter 4, which describes the tests in detail and provides information about interpreting results.
The clinician may also benefit from using the CASL Descriptive Analysis Worksheets.
|
| Does the CASL ASSIST software allow the Score Summary to be copied or exported to a word processing program, such as Microsoft Word?
|
Generally, our software doesn't allow exportation of test data. While many of
our users would like to copy and paste report data into other documents, the
APA guidelines regarding the integrity and privacy of computer-prepared test
data are rather rigid.
In order to address our users' requests while remaining within APA guidelines, we have added an export feature to the CASL software. This allows you to export the narrative report and open it in a word processing program. You may lose some formatting, but all of the information is still present, and you can have flexibility with the document. There is also an "Export Research File" feature which allows you to export a list of standard scores for research purposes.
In summary, the software doesn't allow you to copy and paste the Score Summary, but we have tried to address that need by offering the narrative report export and research file export features. I hope this helps you with your CASL administration and use.
I also heard that you may want to only print out the score summary, and no other reports. To do that, simply go to the Options menu and choose Report Options. Deselect all report options by clicking on the check marks. This leaves you with the report default, which is the Score Summary. It will be the only report printed out until you return to Report Options and change your selections.
|
| Please explain why the CASL Core Composite standard score can sometimes be lower or higher than any of the individual Core test standard scores.
|
The Core Composite standard score may be lower or higher than any of the Core test standard scores when these scores have a collective pattern of being relatively low or high. The reason for this is based on the theory and structure of the CASL. Each CASL test was carefully designed to measure a different aspect of oral language. (See Chapters 3 and 4.) Each test stands alone as an independent, highly reliable measure. When an examinee scores low or toward the lower end of the distribution on all or most of his or her Core tests, the Core Composite score will put him or her at a lower point in the distribution of all examinees who took that set of Core tests. The opposite is true if the examinee scores high or relatively high on all Core tests.
To understand this better, consider a group of examinees 7 to 10 years of age who take separate conormed tests of math and reading. Some students are good at both and will score high on both. Some are better at math and will have higher math scores. Another group will be better at reading and have higher reading scores. A fourth group will have problems with both and score lower on both measures. Those in this fourth group who score lower on both will form the lower end of the distribution of all individuals if a composite score of reading and math is derived.
The same is happening when an examinee is given a set of conormed CASL Core tests, each of which measures a distinct aspect of oral language. The Core Composite standard scores are derived from a distribution of the average of the Core test standard scores. (See Chapter 7.) If the person scores toward the lower end on all Core tests, the average of these test scores will put this examinee at the lower end of the distribution of averages. The resulting standard score derived from this distribution could be lower than any of the individual Core test scores. Consider the following example.
The examinee is aged 12 years 6 months. Appropriately, to get the Core Composite, this examinee was given the following CASL tests and received the standard scores indicated:
| Antonyms | 74 |
| Grammatical Morphemes | 80 |
| Sentence Comprehension | 87 |
| Nonliteral Language | 76 |
| Pragmatic Judgment | 73 |
The sum of these five scores is 390. According to Table 3 in the Norms Book, the Core Composite is 72. Remember that the Core Composite is derived from the distribution of the averages of the tests comprising the composite for a specific age band. So, even though the average of 390 is 78, this average equates to a standard score of 72 when considering the distribution of averages for all standardization examinees aged 11-0 to 12-11.
|
