week 3 practice and reflection q

0.0(0)
Studied by 3 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/9

encourage image

There's no tags or description

Looks like no tags are added yet.

10 Terms

1
New cards

1. The standardized test you are using has a basal = 5 sequential correct and ceiling = 3 sequential errors. You begin administering items at item #10 which the manual states is the correct starting point for the client’s age. The client gets the first 3 items correct then misses the next item. How do you proceed (e.g., what item do you administer next or no further items)? Explain why you made this decision

because they did not hit the basal (5 sequential correct) we would go back to item 9 and then item 8. If they got both of those correct, we would have established the basal and would go to item 14 and continue to administer the assessment until we reach the ceiling (3 sequential errors)


AK:

Go back to item #9 and administer items back from 9,8,7 etc. until the basal of 5 sequential correct is reached.

2
New cards

2. The test manual lists an inter-judge reliability intraclass correlation. What does this information tell you about a test (e.g. what is the purpose of this information)? This particular test says the intraclass correlation is .40. Is this correlation favorable (e.g. would support using this test) or not

inter judge reliability is when the same assessment is conducted on the same client but with a different judge/clinician and they should get the same result. We want a correlation that is closer to 1.0/100% so this correlation is not favorable and does not support using this test


AK:

Inter-judge reliability provides an indication of how reliable or consistent different clinicians are in administering and/or scoring the test. The premise is that use of a test should be the same across clinicians. A correlation of .40 is low and suggests this is not a reliable test.

3
New cards

3. In 2-3 sentences, define the concept of consequential (purpose driven) validity and why it matters

Is the assessment we are choosing reliable, valid, and relevant for this purpose, client and situation?

-our decisions have consequences and we want to do stuff based on our client

-do not want to misuse because kids can miss services, insurance may not cover, and can be stuck with the wrong “label”


AK:

This means that the data we get from assessment and how we use that data has consequences for our clients in terms of leading to treatment recommendations, access to services, treatment targets, and other clinical decisions. We need to be sure the assessments we use, and how we use them are valid and reliable so we get meaningful data to use in making those decisions. If we make bad decisions in how we assess clients, our actions can have bad consequences for our clients.

4
New cards

What is one method that you can use to assess if there is cultural or linguistic bias in a standardized assessment for your particular client?

-does the reference sample include people from same cultural and linguistic background as MY client?

-Does the test manual provide adjustments for scoring (example for dialect)


AK:

This refers to the concept of least biased assessment. Methods we can use to assess for bias in a standardized test is to review the normative or reference sample to see if that sample represents the background of our client. We can look for evidence in the test manual about how cultural and language background was considered in test development. We can look for scoring and interpretation guides that specifically address the cultural and/or language background of our client.

 

5
New cards

The assessment you are considering for a diagnostic evaluation reports that in the reference sample there were 70 true negatives and 30 false positives. What is the specificity of the test and does this value argue for or against using this test?

specificity = TN/(TN+FP)=70/(70+30)=70/100=70% specificity

this argues against the use of this test because we want the specificity to be around 85-90%


AK: 70% - this does not support use of this test as specificity is too low.

6
New cards

6. Your client’s standard score = 83 on a test that uses a standard scale of mean = 100; SD = 15. Given just this information, which of the following best describes your client’s score:

A. The score is within normal limits

B. The score is within 1 SD of the mean

C. The score is between -1 to -2 SD from the mean

D. The score is greater than 2 SD below the mean

E. The score qualifies the child for a highly capable program

within 1 SD of mean =85-115 since they got 83 it is between -1 and -2 SD from the mean

7
New cards

7. You have the following pre-post treatment standardized data from a client from a test using the

standard scale mean = 100; SD = 15.

Pre-tx score: 75 with a 95% CI = 73 – 77

Post-tx score: 86 with a 95% CI = 82 – 90

Is this change most likely a ‘real’ change or is this change at high risk of being due to variability due to measurement error? Answer and explain why in 1 – 2 sentences

The confidence intervals show on any given day the client’s score would be in that range.

Therefore, since the confidence intervals do not overlap we can be more confident that the change is real and from therapy and not due to variability. If the confidence intervals overlapped we cannot be sure if the change is real or not


AK:

This is likely a ‘real’ change and not due to measurement error because the confidence intervals do not overlap.

8
New cards

8. What are two client characteristics that should be considered in determining if the client can be compared to the test’s reference sample?

-race

-age

-sex

-socioeconomic status


AK:

Age, sex assigned at birth (will accept sex or gender), geographic region, race, ethnicity, language background, potentially other demographic factors including SES factors.

 

9
New cards

9. A pediatric test has age bands for reference data of 1 year (e.g. 4 y.o., separate from 5 y.o., etc.). An adult test has no age bands grouping all adults together for the reference group. Would you consider this acceptable for test validity or not in each instrument? Why or why not?

Yes because in pediatrics so much change occurs especially during the critical age where speech/lang development is crucial so we need finer resolution/detail to see those changes.


Age differences are not usually relevant for adults because you would expect them to be relatively the same


AK:

This is acceptable. Pediatric tests need narrower age bands because speech and language change considerably each year during a child’s early development. Typical adults do not experience speech and language changes year to year so it is ok to group them all together.

 

10
New cards

10. Your client scores at the 45% percentile. Demonstrate your understanding of this score by describing what this score means in 1 – 2 sentences in layperson language (as if you are explaining this to a parent of the child).

Your child scored 45% better than 45% of children from the norm referenced sample (same background/demographics)


AK:

This means that in the normative or reference sample, 45% of the people in the sample scored the same as or below this client.