Woodcock Johnson-IV Tests of Cognitive Abilities and Tests of Achievement New York Association of School Psychologists Conference November 6, 2014 John M. Garruto, D.Ed., NCSP Educational Consultant-Houghton Mifflin Harcourt/Riverside Goals of presentation • Participants will have an understanding of the differences between the WJ-III and WJ-IV. • Participants will learn the new clusters that are available in the WJ-IV. • Participants will be familiar with the new subtests that are in the WJ-IV. • Participants will be able to understand the difference between standard scores and the Relative Proficiency Index (RPI). • Participants will understand the procedures of the intravariation and comparison discrepancy procedures. Assumptions of the Presentation • Participants will already be familiar with the WJ-III COG/ACH. • Participants will already be familiar with CHC theory as it was conceptualized for the WJ-III COG. Special Thanks… Changes from the WJ-III CHC theory • • • • • • • WJ-III Gc-Comprehension-Knowledge Gv-Visual-Spatial Thinking Gf-Fluid Reasoning Gsm-Short-Term Memory Gs-Processing Speed Ga-Auditory Processing Glr-Long-Term Retrieval • • • • • • • WJ-IV Gc-Comprehension-Knowledge Gv-Visual Processing Gf-Fluid Reasoning Gwm-Short-Term Working Memory Gs-Cognitive Processing Speed Ga-Auditory Processing Glr-Long-Term Retrieval Changes from the WJ-III: Global Clusters WJ-III • GIA-General Intellectual Ability • BIA-Brief Intellectual Ability WJ-IV • GIA-General Intellectual Ability • BIA-Brief Intellectual Ability • Gf-Gc Composite Changes from the WJ-III Narrow Ability and Clinical Clusters • • • • • • • WJ-III (Not including the DS or ACH) Verbal Ability Thinking Ability Cognitive Efficiency Broad Attention Executive Processes Cognitive Fluency Phonemic Awareness • • • • • • WJ-IV Quantitative Reasoning Auditory Memory Span Number Facility Perceptual Speed Vocabulary Cognitive Efficiency Changes from the WJ-III: Subtests moved • Sound Blending-moved to the Oral Language • Rapid Picture Naming-moved to the Oral Language • Retrieval Fluency-Moved to the Oral Language Changes from the WJ-III: Voted Off the Island • • • • Incomplete Words Decision Speed Visual-Auditory Learning Delayed Planning Fun time-The Top 5 signs you’ve given the WJ COG too much 1. You name your children Bob and Jeff…and one of them is a girl. 2. You know how to sound just like the audio voice for sound blending. 3. When your significant other asks you what you want for breakfast, you reply, “Waffles…they have to be both big and round.” 4. When you ask your child to do their homework, you say, “Tell me if you have finished before I say stop.” 5. When your significant other asks you what you want for your birthday, you think of as many items as you can in one minute. Changes from the WJ-III: New and Different Subtests • Oral Vocabulary (was Verbal Comprehension)-only has synonyms and antonyms • Number Series (was the second half of Quantitative Concepts on the WJ-III ACH) • Verbal Attention • Letter-Pattern Matching • Phonological Processing Changes from the WJ-III: New and Different Subtests • • • • • • Story Recall (moved from the WJ-III ACH) Visualization (combined spatial relations and block rotation from the DS) Numbers Reversed-now has a strict basal and ceiling rule-no more groups Number-Pattern Matching-was called Visual Matching Nonword Repetition Object-Number Sequencing-was called Auditory Working Memory Selective Testing Table Psychometrics…briefly • RELIABILITY-Aim for the WJ-IV was to have cluster reliabilities at .90 or higher and subtest reliabilities at .80 or higher. Reliabilities almost always met this standard with a couple of exceptions. The median reliability for picture recognition is .74. The median reliability for visual processing is .86. Otherwise, the goals were achieved (17 clusters and 17 subtests). Psychometrics…briefly • Validity-Established through intercorrelations with other tests: • • • • • • • GIA-FSIQ=.86 Gf-Gc Composite-FSIQ=.83 Gc-VCI-.79 Gf-PRI-.70 Gv-PRI-.55 Gwm-WMI-.72 Gs-PSI-.55 Other tests (WPPSI-III, WAIS-IV, K-ABC II, S-B5 AND DAS-2 in the TM) WJ-IV COG Subtests Oral Vocabulary • This is like Verbal Comprehension from the WJ-III, though it only has synonyms and antonyms. The pictures and analogies have been removed. • This subtest loads on the GIA, BIA, Gf-Gc cognitive composites, as well as the CHC of Gc, and the narrow ability of Vocabulary (along with Picture Vocabulary from the OL.) • The median reliability is .89 for ages 5-19 and .92 for adults. Number Series• This was once the second half of the Quantitative Concepts subtest in the WJ-III ACH. • This subtest is part of the GIA, BIA, Gf-Gc Composite, Fluid Reasoning (Gf) and Quantitative Reasoning (RQ) clusters. It measures the narrow abilities of quantitative reasoning and inductive reasoning. • This subtest measures one’s ability to determine what number logically fits in a missing series after determining the overall rule. It has a median reliability of .91 for children and .90 for adults • Subtest has a one minute guideline. However, if the examinee is actively working on the problem, allow them to continue working. Verbal Attention • This subtest loads on the Gwm cluster and also on the cognitive efficiency extended cluster. • This is a working memory task that also requires attention to determine what the demand will be. According to the manual, it is a measure of working memory and attentional control. The median reliability is .86 for children and .83 for adults. • Example: “Dog, 5, 2, Cat…what was the second number?” Letter-Pattern Matching • This subtest loads on the GIA cluster, the Cognitive Processing Speed (Gs) factor, the narrow Perceptual Speed (P) factor, and also the Cognitive Efficiency Cluster. • Letter-Pattern Matching is a Cognitive Processing Speed task (Gs) that requires examinees to determine which letter or letter groups are the same. Circle the identical letters or letter groups: bl va dl bl na • This subtest is primarily a perceptual speed task with test-retest reliabilities of .88-.91. Phonological Processing • This is actually a subtest with three parts. It loads on the GIA and Auditory Processing (Ga) indexes. • Parts A and C measure Ga with the narrow ability of phonetic coding. Part B is also a measure of the speed of lexical access (which is a Glr skill). • This subtest has a median reliability of .83 for children and .9 for adults. Story Recall • This subtest was moved from the WJ-III Achievement. It measures the GIA and the broad ability of Long-term Retrieval (Glr). It is a measure of the narrow ability of Meaningful Memory (although this is not a score you get.) • The examinee listens to a story and then gives the story back. Bolded words must be present to get credit. • Median reliabilities are .93 for children and .91 for adults. Visualization • This is a combination of the subtest Spatial Relations from the WJ-III Cog and Block Rotation from the WJ-III DS. It measures the GIA and the Visual Processing (Gv) factor. • Spatial Relations is the same…only this time…examinees are not allowed partial credit; they must get all components correct. Block Rotation is also the same-the examinee must look at a three dimensional figure and identify the same figure that is rotated (two options are always correct.) • This subtest has a median reliability of .83 for children and .87 for adults. General Information • Just like in the WJ-III, it is a measure of background knowledge. The examinee identifies where certain objects may be found and what may done with certain objects. • It is a measure of the Gf-Gc Cognitive Composite and Gc-Comprehension Knowledge. • It has a median reliability of .84 for ages 5-19 and .91 for adults. Concept Formation • Exactly like on the WJ-III…the examinee must figure out the underlying rule that governs the problem-it is a measure of inductive reasoning and also measures cognitive flexibility. • The subtest loads on the Gf-Gc cluster as well as the Gf-Fluid Reasoning Cluster. • It has a median reliability of .93 for ages 5-19 and .95 for adults. Numbers Reversed • Like the WJ-III although this time, rather than test in groups, there’s a straight basal and ceiling. The examinee provides a series of numbers in reverse, which is a measure of working memory. • The subtest loads on the Gwm-Short Term Working Memory Cluster as well as the Number Facility and Cognitive Efficiency Clusters. • The subtest has a median reliability of .84 for ages 5-19 and .91 for adults. Number-Pattern Matching • Renamed from “Visual Matching” on the WJ-III. The task is exactly the same-circle the two numbers in each row as quickly as possible. • The test loads on the Number Facility, Perceptual Speed, and Cognitive Efficiency-Extended Cluster. • The subtest has a test-retest reliability of .85 for ages 7-11, .84 for 14-17 and .88 for adults. Nonword Repetition • This loads on the Auditory Processing (Ga) factor. It is a mixed measure of Ga and Gwm (specifically what is known in Baddeley’s model as the phonological loop.) • Examinees listen to a nonsense word and must say what the nonsense word is. The words become increasingly complex (e.g. “Bambledyegoneegone”) • This subtest has a median reliability of .90 for all ages. Visual Auditory Learning • Test is exactly the same as the WJ-III, only there is no delayed recall task. This controlled learning task teaches the examinee the name for a symbol and asks the examinee to recall the name (yes, Bob and Jeff are still there!) • This subtest loads on the Glr (Long Term Retrieval) and measures the narrow ability of associative memory. • This subtest has a median reliability of .96 in the 5-19 age and .98 in the adult age. Picture Recognition • Just like in the WJ-III…only this time-the examinee must get all the items correct-no partial credit. Also, straight basals and ceilings are done rather than group rules. • No more wall phones…telephone Z is gone. However, windows R&S are the two windows you saw before. • PR loads on the Visual Processing (Gv) factor and measures the narrow ability of visual memory. • PR has a median reliability of .71 for ages 5-19 and .73 for adults. Analysis-Synthesis • Exactly like the WJ-III…individuals look at a key and determine what colors go in the missing squares. • This subtest loads on the Fluid Reasoning (Gf) Extended and Quantitative Reasoning factors. • A-S has a median reliability of .92 for ages 5-19 and .94 for adults. Object-Number Sequencing • Was called Auditory Working Memory on the WJ-III. The subtest is exactly the same, although there is no partial credit and you do not test by groups. • This subtest loads on the factor of Short-Term Working Memory-Extended (Gwm). • This subtest has a median reliability of .89 for ages 5-19 and .87 for adults. Pair Cancellation • Exactly like the WJ-III…examinee circles each instance of the ball followed by the dog. • It loads on the Processing Speed (Gs) cluster and also provides information about attentional control. • PC has a test-retest reliability of .89 for ages 7-11, .89 for 14-17 and .95 for adults. Memory for Words • Exactly like the WJ-III, only you no longer test by item blocks. Examinee listens to multiple words spoken and must repeat the words in the same sequence. Many of the words cannot be visualized. Rhyming words get credit. • This subtest loads on the narrow ability of auditory memory span combined with sentence repetition from the oral language battery. • The median reliability is .82 for ages 5-19 and .82 for adults. A note about basals and ceilings • Sometimes people misunderstand the basal and ceiling rule. You must test by complete pages. • The complete page rule is that if you’re testing and you meet the ceiling, you keep testing until you have reached the bottom of a page if there is visual information on the examinee’s side. If the examinee gets any of the items correct after the “perceived” ceiling, not only are those responses credited…but you keep testing. You only stop when you meet the ceiling at the end of a page. GIA score • To get an overall GIA (g) score-you must give the first 7 subtests. This gives one measure of each of the CHC abilities. BIA score • To get a BIA score-you give the first 3 subtests. It’s one measure of Gc, one of Gf, and one of Gwm. The purpose is primarily for screening. Gf-Gc • This is a four-factor composite that is new to the WJ-IV. It is two measures of Gc (Oral Vocabulary and General Information) and two measures of Gf (Number Series and Concept Formation). • This was developed so that measures of Gwm and Gs do not attenuate the overall composite. Subtests you must administer for composites: • Comprehension-Knowledge-Gc: Oral Vocabulary and General Information. • Fluid Reasoning-Gf: Number Series and Concept Formation (Analysis Synthesis gives you Gf-Ext.) • Short-Term Working Memory-Gsm: Verbal Attention and Numbers Reversed (Object-Number Sequencing gives you Gsm-Ext.) • Cognitive Processing Speed-Gs: Letter-Pattern Matching and Pair Cancellation. • Long-Term Retrieval-Glr: Story Recall and Visual Auditory Learning • Visual Processing-Gv: Visualization and Picture Recognition Subtests you must administer for composites: • • • • • • Quantitative Reasoning: Number Series and Analysis-Synthesis Auditory Memory Span: Memory for Words and Memory for Sentences (OL) Number Facility: Numbers Reversed and Number Pattern Matching Perceptual Speed: Letter-Pattern Matching and Number Pattern Matching Vocabulary: Oral Vocabulary and Picture Vocabulary (OL) Cognitive Efficiency: Letter-Pattern Matching and Numbers Reversed (Administering Verbal Attention and Number-Pattern Matching gives you Cognitive Efficiency-Ext.) Hierarchy of WJ-IV scores (Mather & Wendling, 2014). • Level 1-Qualitative • This is about examining an examinee’s test behavior. The examinee who has word finding difficulties, vs. the one who gives antonyms for synonyms, vs. the one who says “dunno” for items may all get the same raw score but the interpretation may be different. • Use the cover of the test protocol to identify the types of behaviors noted. • The achievement test actually has qualitative norms. You check off how the student attacked the various problems (e.g. read words quickly and fluently) and the manual has actual norms on the percentage of respondents in the standardization sample who achieved a similar rating. Hierarchy of WJ-IV scores (Mather & Wendling, 2014). • Level 2-Level of Development: • Use of age and grade equivalent. • As we know, age and grade equivalents are misunderstood metrics. The layperson will think that a grade equivalent of 2.4 means the child is reading on a grade 2, 4 month reading level. Actually it means they have the same number of items correct as the student who was at grade 2.4 in the norm sample who performed at the 50th percentile. Hierarchy of WJ-IV scores (Mather & Wendling, 2014). • Level 2-W scores • W scores are not easily understood by many. This has to do with Rasch scaling and item response theory. • W scores are calibrated arbitrarily. In this case, a W score of 500 was calibrated for the average 10-0 or 5.0 student (at the 50th percentile). The W score is the basis for both the RPI and the instructional zone of proficiency (which are level 3 scores). Hierarchy of WJ-IV scores (Mather & Wendling, 2014). • Level 3-Proficiency: • W scores and W difference scores. A W score reflects a probability that the individual will get 50% of the items correct and 50% of items incorrect. So a student who earns a W score of 500 gets a grade equivalent of 5.0. The average individual at that grade level with a proficiency at W of 500 would likely get half of the items correct and half of the items incorrect. • 50% of items correct would not suggest proficiency. The authors move proficiency to 90% proficient. Therefore, based on the difficulty of the items, it is figured out which W score a student would need to get 90% of the items correct. For example, for an average 5.0 student, perhaps it’s a W score of 450 (I’m just making this up but let’s use this number). • Now, let’s assume you’re evaluating a student in grade 5.0. You want to determine how proficient your examinee would be against the average 5.0 would be with a W difficulty for getting 90% correct. This means the Reference W would be 450 (not 500 as that is only 50% proficiency.) Hierarchy of WJ-IV scores (Mather & Wendling, 2014). • Level 3-Proficiency • So let’s assume your examinee was administered Number Series. Your examinee gets a W score of 420. That’s 30 points lower than the W score of 450. So at a W of 450perhaps your examinee will only be at 65% proficiency (again I just made up the score). • That’s where the RPI comes into place. The RPI score for your student is 65/90. This can be viewed like a Snellen chart. It means that what the average 5th grade student would get with 90% accuracy your student would get with 65% accuracy. Hierarchy of WJ-IV scores (Mather & Wendling, 2014). • Level 4-Relative Standing in a Group • Standard Scores/percentile ranks. Basically these are scores most school psychologists are used to. They identify how many more people their score is stronger than in the standardized population. So a score at the 25th percentile means your examinee performed better than 25% of the standardization sample. RPIs and standard scores • The RPI and standard score do not always correlate. A way to conceptualize this was created by Joseph Claeys (Account Executive for HMH-Midwest). • Assume you have 10 horses in a race. If your horse is in 3rd-you’re at the 30th percentile. 70% of the horses are faster than you. You’re still in the average range. • However, the horses do not run at the same speeds. The horse at the 50th percentile may run 10 mph. The horse at the 40th percentile may run 5 mph. Your horse may run 4 mph. So yes-horses in the 30th percentile still run better than 30% of the horses-but 30% of the horses run significantly slower than the horse at the halfway point. • This is why you may have many students who don’t qualify (use of level 4 scores) but who still struggle (use of level 3 scores). Intravariation procedure • The intravariation procedure is an analysis of the examinee’s strengths and weaknesses. • Basically, to get an intavariation analysis, you must complete the first seven subtests. • The predicted score of the subtest will be based on an average of the other 6 (it’s not a straight average, it is adjusted based on the psychometrics of the test). • So the oral language predicted score comes from the adjusted mean of number series, verbal attention, letter-pattern matching, phonological processing, story recall, and visualization. Intravariation Procedure • For other subtests, the predicted score is still based on the first 7, but the subtest under observation is replaced by the most closely related subtest . So normally, the oral vocabulary predicted score is based on: number series, verbal attention, letterpattern matching, phonological processing, story recall, and visualization. • However, if you want to know the predicted score for concept formation, it replaces number series (the other Gf test). So the predicted score is based on oral vocabulary, verbal attention, letter-pattern matching, phonological processing, story recall, and visualization. It still uses the top 7 GIA tests as predictors. Intravariation Procedure • One unique characteristic to the intravariation procedure for the WJ-IV is it does not include the subtest being measured. Most cognitive batteries include the subtest in the predicted score, which may not be best because the subtest you’re looking at may bring the score either up or down. The WJ-IV intravariation procedure ensures that predicted score is not attenuated by the actual subtest score. Comparison Procedures • Comparison procedures are basically the discrepancy procedures you can use. The various types of procedures will allow you to determine whether you wish to use a straight cognitive/achievement discrepancy, or you could look for a pattern of strengths and weaknesses. The various procedures include: • GIA vs. Achievement • Gf-Gc vs. Achievement • Oral Language vs. Achievement • Scholastic Aptitude vs. Achievement • Academic Knowledge vs. Achievement Caveat to the multitude of comparison procedures • Avoid fishing expeditions. You get a lot of discrepancies-basically the program will provide for you whatever you want-you get to choose. You’re the clinician. • Remember, a Type I error means you rejected the null hypothesis when you shouldn’t have (you say there’s a problem when there really isn’t.) A Type II error means you failed to reject the null hypothesis when you should have rejected (you say there’s no problem when there actually is.) • Remember also that the more comparisons you make, the more restrictive you should make the discrepancy severity. The program defaults at +/- 1.5 SD. Consider adjusting if you’re making multiple comparisons. Comparison Procedures-GIA • This procedure uses the GIA to predict the achievement scores in each area. A discrepancy means that the achievement score is significantly higher or lower than predicted using the GIA. Comparison Procedures-Gf-Gc • This procedure uses the Gf-Gc composite to predict the achievement scores in each area. A discrepancy means that the achievement score is significantly higher or lower than predicted using the Gf-Gc. Consider using the Gf-Gc composite if you believe that the actual disability is a short-term memory or processing speed deficit and you’re concerned that the GIA could be attenuated by the verbal attention or letter-pattern matching subtests. Comparison Procedure-Oral Language • This uses the Oral Language Composite as the basis for comparing OL to Achievement. You may have an individual with appropriate oral language skills but achievement deficits. Comparison Procedures-Scholastic Aptitude vs. Achievement • The Scholastic Aptitude (SAPT) was called Predicted Achievement in the WJ-III. It uses the cognitive tests that are most related to the achievement tests to predict the achievement test. • The utility of this cluster is when you’re looking for consistency-discrepancy. If certain cognitive skills are bringing down achievement and achievement is actually brought down-you have consistency. • Discussion question: If you’re using the scholastic aptitude comparison procedure and you’re looking for a pattern of strengths and weaknesses, would you be looking for a discrepancy or the absence of a discrepancy to detect a problem? Regarding Pattern of Strengths and Weaknesses • The WJ-IV presents with utility in determining PSW. At the intravariation level, you will see which areas are significantly different from each other. • There are many who view a true PSW as cognitive weaknesses that relate to academic weaknesses and cognitive strengths that do not relate to academic weaknesses. • The GIA and Gf-Gc composite will identify skills not attenuated by Gwm or Gs to help predict achievement. The SAPT will actually identify subtests that predict achievement to be low. To that end, this framework can be utilized with the WJ. Questions on the WJ-COG? Quick note about WJ-Oral Language • Clusters: • Subtests: • Picture Vocabulary • • • • Understanding Directions Oral Comprehension Sound Blending Segmentation Retrieval Fluency Rapid Picture Naming Sound Awareness Sentence Repetition Spanish Subtests • • • • • • Oral Language Broad Oral Language Oral Expression Listening Comprehension Phonetic Coding Speed of Lexical Access Focus on Tests of Achievement Psychometrics…briefly • Reliability-The same goals were set (.90 for clusters and .80 for subtests). Median reliabilities were above .80 for all subtests. All cluster scores were above .90. • Validity (K-TEA II): • Broad Reading-Reading Composite-.92 • Broad Math-Math-.91 • Broad Writing-Written Language-.84 Clusters-Test of Achievement-Reading • • • • Reading (Was Brief Reading) (Letter-Word ID and Passage Comp)Broad Reading (Same as above but also Sentence Reading Fluency) Basic Reading Skills (LWID and Word Attack) Reading Comprehension (Passage Comp and Reading Recall (Reading Vocab creates Reading Comp ext.) • Reading Fluency (Sentence Reading Fluency and Oral Reading) • Reading Rate (Sentence Reading Fluency and Word Reading Fluency) Clusters-Test of Achievement-Math • • • • Mathematics (Was Brief Math) (Calculation and Applied Problems) Broad Mathematics (Same as above but also Math Facts Fluency) Math Calculation Skills (Calculation and Math Facts Fluency) Math Problem Solving (Applied Problems and Number Matrices) Clusters-Test of Achievement-Writing • Written Language (was Brief Writing) (Spelling and Writing Samples) • Broad Written Language (Same as above but includes Sentence Writing Fluency) • Basic Writing Skills (Spelling and Editing) • Written Expression (Writing Samples and Sentence Writing Fluency) Think of the past clusters…did anyone see anything interesting? • How about IDEA areas at the cluster level?! Cross-Domain Clusters • Academic Skills • Think of your basic skills. These include LWID, Calculation, and Spelling. • Academic Fluency • Think of how efficiently you perform yours skills. These include sentence reading fluency, sentence writing fluency, and math facts fluency. • Academic Applications • Think of the final product. These include applied problems, writing samples, and passage comprehension. Changes from the WJ-III: Subtests moved • • • • Story Recall-Moved to the COG Understanding Directions-Moved to the OL Picture Vocabulary-Moved to the OL Sound Awareness-Moved to the OL (except substitution, which was moved to Phonological Awareness in the COG) Changes from the WJ-III: Voted Off the Island • Story Recall-Delayed • Punctuation and Capitalization • Quantitative Concepts (second half…number series…moved to the COG) Selective Testing Table WJ-ACH Subtests Letter-Word Identification • Similar to the WJ-III. Examinee begins with letters and sounds and then moves into word identification. • Loads on the Reading, Broad Reading, Basic Reading Skills, and Academic Skills clusters. • Has a median reliability of .92 for ages 5-19 and .94 for adults. Applied Problems • Similar to the WJ-III (though I do not think they have pay phones anymore). Examinee listens to word problems. • Loads on the clusters of Mathematics, Broad Mathematics, Math Problem Solving, and Academic Applications. • Has a median reliability of .91 for ages 5-19 and .92 for adults. Spelling • Like the WJ-III. Examinee begins with basic line formation, then letters, and finally words. • Loads on the Written Language, Broad Written Language, Basic Writing Skills, and Academic Skills Cluster. • Has a median reliability of .91 for ages 5-19 and .93 for adults. Passage Comprehension • Much like the WJ-III. Examinee begins with basic rebuses, moves to pictures to tell about words, and finally engages in cloze reading sentences. • Loads on the Reading, Broad Reading, Reading Comprehension, and Academic Applications clusters. • Has a median reliability of .89 for ages 5-19 and .91 for adults. Calculation • Much like the WJ-III. Examinee engages in various computation problems. • Loads on the Mathematics, Broad Mathematics, Math Calculation Skills, and Academic Skills clusters. • Has a median reliability of .93 for ages 5-19 and also for adults. Writing Samples • Much like the WJ-III. Examinee writes sentences to examiner-directed prompts. • Must use the manual to score. Follow manual closely for scoring, particularly .5 responses. • Loads on the Written Language, Broad Written Language, Written Expression, and Academic Applications clusters. • Has a median reliability of .90 for ages 5-19 and .89 for adults. Word Attack • Much like the WJ-III. Begin with basic skills and then move to decoding nonsense words. Moved to the standard battery. • Loads on the Basic Reading Skills and Phoneme-Grapheme Knowledge clusters. • Has a median reliability of .90 for ages 5-19 and .93 for adults. Oral Reading • New to the WJ-IV. It’s a measure of reading fluency but not as much about how quickly one reads but how fluently, which includes prosody. • Examinee reads sentences out loud and examiner scores for accuracy. Optional scoring includes mispronunciations, hesitations, transpositions, etc. • Loads on the Reading Fluency cluster. • Has a median reliability of .97 for ages 5-19 and .95 for adults. A note about the fluency subtests • Sentence reading fluency, math facts fluency, and sentence writing fluency are at the end of stimulus book 1. • Although the subtests have been grouped together, the authors recommend that one does not test one fluency test after another, but rather intersperses them throughout the assessment. Sentence Reading Fluency • Was once called “Reading Fluency”. Like the WJ-III but no more pink milk! Examinees read sentences and quickly decide whether the answer is true or false. • Loads on the Broad Reading, Reading Fluency, Reading Rate, and Academic Fluency Clusters. • Has a test-retest reliability of .92 for ages 7-11, .91 for 14-17 and .93 for adults. Math Facts Fluency • Was called “Math Fluency” on the WJ-III. Examinees solve computation problems as quickly as possible with changing operations. • Loads on the Broad Mathematics, Math Calculation Skills, and Academic Fluency clusters. • Has test-retest reliabilities of .95 for ages 7-11, .97 for 14-17 and .95 for adults. Sentence Writing Fluency • Was called writing fluency in the WJ-III. Examinees take three words and combine them into sentences. Subtest is now 5 minutes rather than 7. • Loads on the Broad Written Language, Written Expression, and Academic Fluency Clusters • Median test-Retest reliabilities are .73 for ages 7-11, .76 for 14-17 and .88 for adults. Reading Recall • New to the WJ-IV. Like story recall, but instead of hearing the story, the examinee reads the story and, when done, must retell the story. • Loads on the Reading Comprehension cluster and is also a measure of meaningful memory. • Median reliability of .97 for ages 5-19 and .86 for adults. Number Matrices • Moved from the WJ-III Diagnostic Supplement. Like number series, only this time the examinee must solve the number grids both horizontally and vertically. • Loads on the Math Problem Solving Cluster. • Has a median reliability of .91 for ages 5-19 and .93 for adults. Editing • Like the WJ-III except this time no words can be read to the examinee. Examinee is shown sentences with an error in them and the examinee corrects the error. • Loads on the Basic Writing Skills cluster. • Has a median reliability of .90 for age 5-19 and .92 for adults. Word Reading Fluency • New subtest. Almost like decision speed from the WJ-III COG except instead of pictures, the examinee circles words that go together as quickly as possible. • Loads on the Reading Rate Cluster. • Has test-retest reliabilities of .92 for ages 7-11, .91 for 14-17 and .93 for adults. Spelling of Sounds • Like the WJ-III only this time…the examinee cannot get partial credit. The examinee hears a nonsense word (like blick) and must write the word the way it sounds. It’s the only subtest that uses the audio recording. • This subtest loads on the Phoneme-Grapheme Relationships Cluster. • Has a median reliability of .88 for ages 5-19 and .92 for adults. Reading Vocabulary • Similar to the WJ-III but is now an optional subtest for reading comprehension. Examinee provides synonyms and antonyms of words (no more analogies). • Loads on the Reading Comprehension-Extended Cluster. • Has a median reliability of .85 for ages 5-19 and .92 for adults. Science, Social Studies and Humanities • Like the WJ-III, only they’re separate subtests. Together, they create the Academic Knowledge Cluster (which can be used for a Comparison Procedure.) • Median reliabilities range from a low of .76 in science for ages 5-19 to a high of .94 for humanities for adults. Intravariation Procedures • Like the WJ-IV COG, variation procedures are based on the first 6 subtests (LWID, App Prob, Spelling, Pass Comp, Calculation, and Writing Samples). • Again, each subtest is pulled out and compared to the weighted average of the other 5. • Additionally-you get other important composites such as Academic Skills (your low level skills-LWID, Calc, Spelling), fluency (rate of skill-Sent. Read Fluency, Math Facts Flu, and Sent Writing Fluency) and applications (final product-passage comprehension, applied problems, writing samples). This will help you see where there are strengths and weaknesses for your student and determine what, if anything really pops out. Comparison Procedures • Like the WJ-IV COG-there’s an extra comparison procedure and it’s the Academic Knowledge/Achievement cluster. • You can get this comparison by giving only the WJ-IV ACH. • It’s a way to get at those students who have strong a strong content knowledge base (Sci, SS, and Humanities) but struggle in reading, writing, or math. QUESTIONS???? THANK YOU!!!!!!