Reading Grade Levels

When discussing reading grade levels, we can get deeply into the weeds. However the purpose of this article is to help teachers deal with the parent question: “What is my child’s reading grade level?”

A recent teacher’s post in a science of reading Facebook group (name not included here) justifies the parent’s question:

…it seems completely reasonable for parents to wish to know, in practical terms, how far away their child is from [the] grade-level target.
If we have universal assessments, NAEP, state assessments, etc. that identify where students fall with regard to grade level reading, it seems appropriate to have a quality measure (in addition to the diagnostic testing) that gives a broad overview of where a student falls on that continuum? It is just one part of the assessment information within the “data folder” we keep on a student. It isn’t the driving force of instruction, but rather a way to indicate where s/he stands in relation to the overall goal.
I would comment as follows:

We have to understand what parents do not and do mean by the “reading level” question.

Few parents want to know or would understand what a normed reading assessment demonstrates. Parents don’t generally care about how well their children score relative to students in Mississippi or Connecticut.

What parents do want to know, from my experience, is can their children read and understand the “grade level” stories and textbooks used by their teacher and how well do their children read compared to others in the class?

I agree with the teacher’s post that the parent question is legitimate. However, I disagree with the teacher about needing a quality measure  “that gives a broad overview” of a student’s reading grade level” to answer the parent’s question.

Dr. Matt Burns responded to the teacher’s post regarding the validity of grade level reading assessment data:

The problem is that assessments that result in a level are so flawed as to make the data meaningless. 1. The scores have a large standard error of measure that usually equals + and – 2 levels. So a 2.8 is somewhere between 0.8 and 4.8. Two scores within that range are essentially equal. 2. Data never generalize to or from an individual. Thus, if a score equaled 2.8 and if 2.8 actually meant something, that still doesn’t mean that an individual kids at 2.8 could actually read any individual book supposedly written at 2.8. We are better off to abandon levels and describe kids in relation to skills and norms.
Dr. Burns provides a grad level course in reading assessment stats in this one helpful video: https://www.youtube.com/watch?v=kgxwnoFuDDE 

So if we shouldn’t pull out the state assessment scores indicating reading grade levels, how should we answer the parent question: “What is my child’s reading grade level?”

Remember that what the parent really wants to know includes these two sub-questions: 1. Can my child read and understand the “grade level” stories and textbooks used in class? 2. How well does my child read compared to others in the class?

1. Can my child read and understand the “grade level” stories and textbooks used in class?

In addition to a beginning of the year a diagnostic fluency assessment (I’ll share one at the end of the article), teach parents at Back-to-School Night how to create simple fluencies from “grade level” stories and textbooks with an oral or written re-tell or summary. Parents can be trained how to select an appropriate passage from grade level text to be used in class, listen to their child read and tabulate total word count-errors and the correct percentage of words read in a one-minute timing. No need to train parents to analyze prosody, attention to punctuation, types of miscues, etc.

Parents of younger children can ask, “Tell me about what you read.” Parents of older children can say, “Write a paragraph about what you read.” Creating, administering, and scoring a fluency with re-tell or summary is not rocket science, and the results perfectly answer the Can my child read and understand the “grade level” stories and textbooks used in class? question.

These two tasks help parents answer their own question and can help them complete their own progress monitoring.

As an aside, I also advise training parents how to use simple comprehension cues to discuss reading at home. Teachers might wish to check out my article, Reading Comprehension Cues.

2. How well does my child read compared to others in the class?
To address this sub-question I share two data measurements with parents: Hasbrouck and Tindal Fluency Norms

and the baseline fluency numbers of the other students in class (without names, of course, to protect privacy).

Some teachers have strong reservations about sharing assessment data of other children, even if the identities of the children are unknown. However, the public has access to grade level and school scores of state testing that is far more detailed then fluency numerical scores.

If unsure about sharing this data, don’t do so or ask your principal.

My thoughts on how to answer the parent question: “What is my child’s reading grade level?”

 

*****THE FREE DIAGNOSTIC READING FLUENCY ASSESSMENT*****

The “Pets” diagnostic expository fluency passage is leveled in a unique pyramid design: the first paragraph is at the first grade (Fleish-Kincaid) reading level; the second paragraph is at the second grade level; the third paragraph is at the third grade level; the fourth paragraph is at the fourth grade level; the fifth paragraph is at the fifth grade level; the sixth paragraph is at the sixth grade level; and the seventh paragraph is at the seventh grade level. Thus, the reader begins practice at an easier level to build confidence and then moves to more difficult academic language. As the student reads the fluency passage, the teacher will be able to note the reading levels at which the student has a high degree of accuracy and automaticity. Automaticity refers to the ability of the reader to read effortlessly without stumbling or sounding-out words. The 383 word passage permits the teacher to assess two-minute reading fluencies (a much better measurement than a one-minute timing).

Get the The Pets Fluency Assessment FREE Resource:

Grammar/Mechanics ,



CORE and LETRS Phonics Screeners

Don't Assess for Assessment Sake

Don’t Assess What You Can’t Teach

In my following criticisms of the CORE and LETRS phonics screeners, I’m reminded that it’s unfair to criticize a duck for being a duck.

Of the plethora of literacy screeners, especially ones which focus on decoding (phonics) and encoding (spelling), the CORE 2nd edition and LETRS Phonics and Word Reading Survey 3rd edition are certainly the most popular. The former has been around for ages and helped pioneer the combined pseudoword and real word approach to phonics assessments. The latter has achieved prominence due to the widespread LETRS training and the omnipresent Dr. Louisa Moats.

These ducks do what ducks do. They are reliable and valid measurements of phonics acquisition. Using these assessments to place students in appropriate tiered instruction will do that job nicely, and they can be used to serve for program evaluation purposes. However, you can’t make a duck walk on all fours. And to beg the question, if you really need a quadruped, why are you using a duck?

Yes, we need to use data for program placement and evaluation; however, other animals can do these jobs and better provide teachers what they need. KEY POINT: Simply put, we need to replace screening assessments with specific and comprehensive diagnostic assessments. Specific and comprehension diagnostic assessments can accomplish all three functions: placement, program accountability (I prefer progress monitoring), and teachable data.

I would caution teachers that using the CORE and LETRS phonics screeners for instructional decision-making is ill-advised. Unfortunately, many teachers use data gleaned from these surveys to place students within small groups on in software modules which introduce or remediate phonics and spelling deficits. Both screeners have included language in their administration directions suggesting this application of their data.

The CORE 1st edition states, “These assessments are best used to plan instruction for students in the primary grades and to develop instructional groups.” Fortunately, the authors of the 2nd edition directions delete this language and suggest that other assessments should be added for instructional decision-making. The LETRS directions state, “The survey is a tool for identifying which correspondences and patterns the student has learned, and which ones the student needs to be taught.” The LETRS verbiage clearly overstates. Although much more comprehensive than the CORE screener, LETRS still uses random sampling and does not identify all correspondences and patterns learned and all those which need to be taught.

One example (the long /u/) should prove my point that the CORE and LETRS phonics screeners should not be used as diagnostics to introduce or remediate phonics and spelling. The CORE 1st edition provides only 1 long /u/ test item, lute. The CORE 2nd edition does not include any long /u/ test items. LETRS does provide 3 long /u/ words: tune, pruse, and commune; however, I’d hazard to guess that any LETRS-trained teacher could tell me why more data should be required to place students in a small group for long /u/ instruction. Most teachers would want to see not only the “u_e,” but also the “u,” “_ew,” and “_ue” sound-spellings on a decoding and spelling diagnostic to make informed instructional decisions.

My point is that we have plenty of diagnostic phonics and spelling assessments that will test those 4 long /u/ sound-spellings (including my own free-to-use diagnostics which I will provide at the end of the article).

Two more relevant points as to why we need to ditch screeners, such as the CORE and LETRS assessments. They waste precious instructional time, and teachers tend to “throw the baby out with the bath water.”

These screeners are time-consuming. In a class of, say 20 students, each assessment takes about 10-15 minutes per student to administer and a few more minutes to score and record–that’s at least 4 hours of administration time. And the CORE 1st edition recommends re-testing every 4-6 weeks. Yikes! And don’t forget the re-testing! Instructional time is reductive. Losing 4 hours of instruction is certainly not useful for any of those 20 struggling readers.

Because the data derived from these assessments is not specific and comprehensive, teachers can’t use the data to inform instruction. Giving screeners undermines the value of teachable diagnostic assessments. Assessment becomes a “have to fill in the blanks” exercise done to meet the expectations of site administrators or reading specialists/literacy coaches or flashpoints for “academic freedom, I don’t have to do them” confrontations.

Replacing screeners, such as the CORE and LETRS assessments, with diagnostic assessments that provide the data teachers can and want to use to help their students makes sense.

Now, teachers jaded by using unhelpful and time-consuming phonics and spelling assessments may be interested in trying these free phonics and spelling assessments. Each is specific and comprehensive and will help teachers make informed instructional decisions for their students. I find them particularly helpful for older reading intervention students.

Oh, how about administration? Whole class, not individual. Audio files. Self-correcting Google forms (or print if you wish). Recording matrices. Vowel Sounds Phonics Assessment (10:42), Consonant Sounds Phonics Assessment (12:07) , Diagnostic Spelling Assessment (17:32-22:38 both American and Canadian English). Oh, would you mind if I included 5 phonemic awareness assessments, a heart words assessment, a grammar assessment, a mechanics assessment, and academic vocabulary assessments? Will do. Plus, think about the ease of using the audio files and Google forms for new students! Scroll down to get the download link sent immediately to your email.

*****

Intervention Program Science of Reading

The Science of Reading Intervention Program

The Science of Reading Intervention Program: Word Recognition includes explicit, scripted, sounds to print instruction and practice with the 5 Daily Google Slide Activities every grades 4-adult reading intervention student needs: 1. Phonemic Awareness and Morphology 2. Blending, Segmenting, and Spelling 3. Sounds and Spellings (including handwriting) 4. Heart Words Practice 5. Sam and Friends Phonics Books (decodables). Plus, digital and printable sound wall cards, speech articulation songs, sounds to print games, and morphology walls. Print versions are available for all activities. First Half of the Year Program (55 minutes-per-day, 18 weeks)

The Science of Reading Intervention Program: Language Comprehension resources are designed for students who have completed the word recognition program or have demonstrated basic mastery of the alphabetic code and can read with some degree of fluency. The program features the 5 Weekly Language Comprehension Activities: 1. Background Knowledge Mentor Texts 2. Academic Language, Greek and Latin Morphology, Figures of Speech, Connotations, Multiple Meaning Words 3. Syntax in Reading 4. Reading Comprehension Strategies 5. Literacy Knowledge (Narrative and Expository). Second Half of the Year Program (30 minutes-per-day, 18 weeks)

The Science of Reading Intervention Program: Assessment-based Instruction provides diagnostically-based “second chance” instructional resources. The program includes 13 comprehensive assessments and matching instructional resources to fill in the yet-to-be-mastered gaps in phonemic awareness, alphabetic awareness, phonics, fluency (with YouTube modeled readings), Heart Words and Phonics Games, spelling patterns, grammar, usage, and mechanics, syllabication and morphology, executive function shills. Second Half of the Year Program (25 minutes-per-day, 18 weeks)

The Science of Reading Intervention Program BUNDLE  includes all 3 program components for the comprehensive, state-of-the-art (and science) grades 4-adult full-year program. Scripted, easy-to-teach, no prep, no need for time-consuming (albeit valuable) LETRS training or O-G certification… Learn as you teach and get results NOW for your students. Print to speech with plenty of speech to print instructional components.

Click the SCIENCE OF READING INTERVENTION PROGRAM RESOURCES for detailed program description, sample lessons, and video overviews. Click the links to get these ready-to-use resources, developed by a teacher (Mark Pennington, MA reading specialist) for teachers and their students.

Get the Diagnostic ELA and Reading Assessments FREE Resource:

*****

Grammar/Mechanics , , , , ,



Vocabulary Instructional Sequence

Thought I’d share my grades 4–8 comprehensive vocabulary instructional scope and sequence. Includes multiple meanings, Greek and Latin morphology, word relationships, figures of speech, connotations, and academic language: Grades 4-8 Vocabulary Instructional Scope and Sequence

All too often we tend to teach the same content over and over again, or we skip teaching the content because we assume the previous grade-level teacher “must have covered it.” We generally work well in grade-level teams, but rarely have the chance to plan with other teachers at the same school, the previous or next school, or among departments and special programs. This document remedies these realities with systematic and coordinated vocabulary content. Note that although it tells teachers “what to teach”; it does not tell teachers “how to teach it.”

The planning document provides specific vocabulary words to teach each instructional vocabulary component (except figures of speech) for twice-per-week explicit vocabulary instruction. Words have been chosen according to frequency studies when possible. The instructional components mirror the old Common Core Anchor Standards for Language L. 4, 5, and 6. Although many have faulted areas of the Common Core State Standards and states have tended to adopt their own standards, I’ve never heard any complaints about the vocabulary standards. After all, who would have issues with contributing author, Isabel Beck and colleagues? https://achieve.lausd.net/cms/lib/CA01000043/Centricity/domain/219/ccss%20docs/ELA%20Appendix%20A.pdf

The vocabulary instructional scope and sequence corresponds to my grades 4-8 Comprehensive Vocabulary programs.

Grades 4-8 Comprehensive Vocabulary

Comprehensive Vocabulary

Here’s how your students will master each slice of the vocabulary pie with one or more of these grade-level programs:

Multiple Meanings

Students practice grade-level homonyms (same spelling and sound) in context clue sentences which show the different meanings and function (part of speech) for each word.

Greek and Latin Morphology

Three criteria were applied to choose the grade-level prefixes, bases, and suffixes:

1. Frequency research 2. Utility for grade-level Tier 2 words 3. Word Part Memory Pairs

Each odd-numbered vocabulary worksheet pairs a Greek or Latin prefix-base or base-suffix combination to enhance memorization and to demonstrate utility of the Greek and Latin word parts.

Students review spelling patterns (orthography), pronunciations, and meanings in Google slides (or print) by creating word matrices with drag and drop prefixes, bases, and suffixes and spelling the Greek and Latin word parts, using word sums.

Students add each of the 54 Greek and Latin word parts to personal Greek and Latin Morphology Walls, lesson by lesson, and drag and drop to create new academic language words with the word parts that have mastered to orthographic memory.

Language Resources

Students look up the Greek and Latin whole word in a dictionary (print or online) to compare and contrast their educated guesses to the denotative definition of the word. Students divide the vocabulary word into syl/la/bles, mark its primary áccent, list its part of speech, and write its primary definition.

Additionally, students write synonyms, antonyms, or inflected forms of the word, using either the dictionary or thesaurus (print or online). This activity helps students develop a more precise understanding of the word.

Figures of Speech

Students learn a variety of figures of speech (non-literal expression used by a certain group of people). The Standards assign specific types of figures of speech to each grade level. Students must interpret sentences which use the figures of speech on the biweekly unit tests.

Word Relationships

Students use context clue strategies to figure out the different meanings of homonyms in our Multiple Meaning Words section. In the Word Relationships section, students must apply context clues strategies to show the different meanings of word pairs. After students have mastered decoding, the program’s S.A.L.E. Context Clues Strategies will help students problem-solve the meanings of unknown words in their reading.

Students learn two new grade-level vocabulary words which have similar denotative meanings, but different connotative meanings. From the provided definitions, students write these new words on a semantic spectrum to fit in with two similar words, which most of your students will already know. For example, the two new words, abundant and scarce would fit in with the already known words, plentiful and rare in this semantic order: abundant–plentiful–scarce–rare.

Academic Language

Tier 2 words (academic vocabulary) should be a key focus of vocabulary instruction. The Academic Language section of the vocabulary lessons provides two grade-level words from the research-based Academic Word List. Students use the Frayer model four square (definition, synonym, antonym, and example-characteristic-picture) method to learn these words. The Common Core authors and reading specialists (like me) refer to this process as learning vocabulary with depth of instruction.

Vocabulary Study Guides

Vocabulary study guides are provided for each of the weekly paired lessons for whole-class review, vocabulary games, and individual practice. Print back-to-back and have students fold to study.

Vocabulary Tests

Biweekly vocabulary tests assess both memorization and application. The first section of each test is simple matching. The second section of each test requires students to apply the vocabulary in the writing context. Answers follow.

Syllable Blending, Syllable Worksheets, and Derivatives Worksheets

Whole class syllable blending “openers” will help your students learn the rules of structural analysis, including proper pronunciation, syllable division, accent placement, and derivatives. Each “opener” includes a Syllable Worksheet and a Derivatives Worksheet for individual practice. Answers follow.

Context Clues Strategies

Students learn the FP’S BAG SALE approach to learning the meanings of unknown words through surrounding context clues. Context clue worksheets will help students master the SALE Context Clue Strategies.

Vocabulary Acquisition and Use Resources

Greek and Latin word parts lists, vocabulary review games, vocabulary steps, and semantic spectrums provide additional vocabulary instructional resources.

Students who complete each of the Comprehensive Vocabulary Grades 4–8 grade-level programs will have practiced and learned much of the Academic Word Corpus and all of the skills of vocabulary acquisition. These students will have gained a comprehensive understanding of academic language and will be well-equipped to apply the skills of context clues strategies and structural analysis to read well and write with precision.

Grammar/Mechanics , ,



Structured Word Inquiry

If you’ve been reading up on morphology and its importance in teaching our students to read and write well, you’ve been hearing these buzz phrases: structured word inquiry, word matrices, and word sums.

Let me assist with some of the terminology, point to some relevant research, and get you a fantastic free resource to use tomorrow in your classes.

Dr. Pete Bowers coined the instructional procedure, Structured Word Inquiry (SWI) to describe specific techniques to help students understand how morphology (the study of morphemes–the smallest units of meaning) is the one linguistic component that connects orthographic (spelling) patterns, pronunciation, and meanings of words.

Dr. Bowers and his colleague’s research, Morphological Instruction and Literacy (Kirby & Bowers 2017), points to the importance of reframing our approach to vocabulary, reading, spelling, and writing instruction.

Those of us who are into brain research to a certain degree recognize the three features of how we develop sight word automaticity: the orthographical, phonological, and semantic interplay.

Instruction which helps students see how words are formed and helps them practice doing so produces big dividends.

Knowledge of word structure and how words are formed is linked to both greater vocabulary development and stronger reading comprehension (Prince, 2009; Wolter & Green, 2013).

Having a firm foundation in morphology has been shown to predict how well students perform on reading and spelling tasks (Bowers, Kirby, & Deacon 2010).

Teachers see how morphological study helps their students understand challenging academic vocabulary. If a student understands a morpheme, such as pre, that student is empowered with a problem-solving skill to gain clues to meaning for a multitude of words ( 1360 according to Word Find) words with that prefix.

How to Use Structured Word Inquiry

Structured Word Inquiry Word Matrix

Structured Word Inquiry

  1. Use a base word in the middle of a matrix–either a free base (stands on its own as a base word) or a bound base (needs an affix to complete a word).

    Grades 4-8 Comprehensive Vocabulary

    Comprehensive Vocabulary

  2. Add prefixes and suffixes, which will add to the base to form multi-syllabic words.
  3. Students print (or with my free program below–drag and drop in Google slides) the affixes to form the words. Stickies work well, too.
  4. Students print (or type in my free program) the word sums. For example, “sub” + “tract” + “ion” → “subtraction.”

About my free program… I developed structured inquiry lessons for one of my Comprehensive Vocabulary grade-level programs (grade 4) and included directions for how to easily construct your own interactive Google slide matrices and word sums, just like mine! By the way, my Comprehensive Vocabulary programs each include print copy word matrices as shown in the graphic. All five grade-level Comprehensive Vocabulary programs are included in The Science of Reading Intervention Program: Language Comprehension.

https://docs.google.com/presentation/d/1gT2kNkA0x3Ijum1ClChn-nFL3z7j05jO1d_YYvzH5gM/copy?fbclid=IwAR0MVUXgVBxwJPjITaBvhOhz5GrSAm9JhGIhvdxvBeJrhms2bLcOfwA9l0g 

Grammar/Mechanics



Diagnostic Academic Language Assessments

Three questions about Diagnostic Academic Language Assessments: 1. What if you were able to administer comprehensive diagnostic language assessments to determine exactly which Tier 2 academic words your students do and do not know? 2. What if these assessments were ordered by research-based frequency to help you pinpoint instructional levels?  There are, and these assessments are absolutely free to administer in self-correcting Google forms.

Rationale: In addition to explicit instruction in Greek and Latin Morphology, figures of speech, word relationships, multiple meanings, and connotations, teachers know the importance of teaching academic language, thanks to the research of Beck, Isabel L. McKeown, M., & Kucan, L. (2002).

Yes, but which words? High frequency and utility seem like sensible selection criteria. Accordingly, I put together Dr. Averil Coxhead’s research-based Academic Word List in self-correcting Google forms as diagnostic assessments to inform instruction. I divided into grades 4, 5, 6, 7, and 8 assessments, each with 56 words and matching definitions. Happy to share: https://blog.penningtonpublishing.com/wp-content/uploads/2023/01/Academic-Language-Diagnostic-Assessments-1.pdf

And question 3… What if there were comprehensive grade-level vocabulary programs (4, 5, 6, 7, and 8), which provided the corresponding instructional resources for each academic language word on the assessments and Greek and Latin Morphology (with Morphology Walls in interactive Google slides), figures of speech, word relationships, multiple meanings, and connotations? There are… check out the extensive previews for Comprehensive Vocabulary by Pennington Publishing.

Grades 4-8 Comprehensive Vocabulary

Comprehensive Vocabulary

If you would like to read more about how Dr. Coxhead developed her Academic Word List and download a free copy of the list, visit https://blog.penningtonpublishing.com/reading/academic-word-list/

Grammar/Mechanics , , ,



Free Morphology Walls

You might like my free Greek and Latin Morphology Walls in Google slides, featuring 100 high frequency/utility prefixes, bases, and suffixes from my reading intervention program. Here’s the

word part list and links to 18 slides–fully editable if you wish to add your own. https://blog.penningtonpublishing.com/wp-content/uploads/2023/01/Morphology-List-1.pdf?fbclid=IwAR2lJ3byOfmXbGMuP9frlruKIx0ZNP_Lw7gxP9eZnh8ICE-NPlO5gcKG2Nk

If  you want the definitions and practice with these morphemes before reviewing and extending learning with these walls, get an entire year-long program of 56 lessons in Greek and Latin Morphology.

The program features high frequency word parts, cleverly combined to assist in orthographic mapping. The result is 14 key anchor words, which will help students place the morphemes in their long term memories.

The entire Greek and Morphology program is one component of The Science of Reading Intervention Program: Word Recognition.

No prep. Simply teach the morphemes and share the Greek and Latin Morphology Wall slides (or print them if you like) as review. You’ll be amazed at the morphological connections that your students make as they drag and drop prefixes, bases, and suffixes to create new words. Enjoy the lessons!

Word Recognition

56 Lessons with Greek and Latin Morphology

Greek and Latin Morphology

 

 

 

Grammar/Mechanics , , ,



Create Decodable Text with Artificial Intelligence

Decodable text can be created with this tool.

Create Decodable Text

Reading specialists, language coaches, SLPs, EL/EB, SPED, and reading teachers will appreciate a new free tool to create decodable text with artificial intelligence (AI). The recently released open artificial intelligence site, ChatGPT (link will follow), is free and easy to use. It follows your directions to compose text with your own parameters. No more brainstorming sentences featuring r-controlled vowels; no more searching in vain for decodable narrative and expository text with the sound-spellings you have introduced; Wow! This is a game-changer.

Now of course the AI has its potential drawbacks. For example, students can enter their teacher’s writing prompt and ChatGPT will churn out a well-crafted, albeit a bit formulaic, essay with MLA citations. And each ChatGPT-generated writing text is unique; the AI learns from its experience and multiple submissions produce better and more detailed information. Secondary teachers are beginning to brainstorm ChatGPT detection ideas. At least I hope this free platform sounds the death knell sounding for all those essay-writing services stalking our students.

However, beyond the real concerns of plagiarism and questions about how to adapt to this new tool, I’m sure you will find plenty of useful applications to teaching reading. I’ll share a few helpful suggestions to create those in-demand reading sources that reading teachers are always asking about–decodable texts.

To direct ChatGPT (link will follow, I promise!), simply type (or copy and paste) what you want it to produce in its textbook. So easy to use, but the more specific parameters you assign, the better the results will be. It takes a bit of trial and error to get precisely what you want for your students.

One important caveat… True decodable text depends upon what has been explicitly taught and provides practice on those targeted sound-spellings. The AI produced examples I’ll share are not completely decodable texts by any means. BTW, I’ve written my own set of fully decodable Sam and Friends Phonics Books to align with my own reading intervention program and it’s tough work if one’s goal is fidelity to lessons explicitly and previously taught. However, I will say that my work would have been significantly easier had I been able to use this amazing AI tool. See what you think!

EXAMPLES

I typed…

Write decodable text focusing on words with the long a sound.

ChatGPT produced…

  1. “The snake lay in the grass and basked in the sun’s rays.”
  2. “Kate made a cake for Jake’s birthday and it was a huge hit.”
  3. “The rain came down in sheets as the thunder crashed.”
  4. “The waves of the ocean were calm and the boats swayed gently.”
  5. “The game was tied and the players battled for the ball.”

Not bad at all! It actually churned out quite a few more with other long /a/ sound-spellings.

Next, I decided to challenge the AI a bit more. I like how Marni Ginsberg uses these types of memory hook sentences in her Reading Simplified program (hers are better). Great practice orthographic mapping, connecting the sound-spellings to words to meaning to syntax. I typed…

Write a sentence including these long o words: go, boat, though, toe, show, poke.

ChatGPT (the link’s coming) came up with this long /o/ memory hook sentence…

I was going to take the boat out on the lake, though I knew I might poke my toe on the way there because I always seem to trip and show off my clumsiness.

With a bit of revision, that might do rather nicely.

Write a story for second graders, using these words: rain, made, laid, rays, play, game, waves, wait, weigh, eight, Jay, Kate, great.

ChatGPT produced this cute short story… Notice that my parameters directed the level of controlled text (second grade) and a set of randomly chosen long /a/ nouns and verbs. I add the bolded long /a/ words and threw in a few revisions in (parentheses).

Once upon a time, in a small town called Sunnyville (Maytown), Jay and Kate were best friends. They loved to play outside and explore the world around them.

One day, it started to rain outside, so Jay and Kate decided to stay inside and play a game. They looked for a fun game to play, and finally decided on a card game called “Go Fish.” (“Pay Day.”)

As they played, they could hear the rain hitting the roof and the waves crashing against the shore. They loved the sound of the rain and the waves, and they enjoyed the cozy feeling of being inside while it was raining outside.

After a while, they started to get a little hungry. Kate said, “I wish we had something to eat. I’m starving!” Jay thought for a moment and then said, “I know! Let’s make some cookies!” (bake a cake!)

So they gathered all the ingredients they needed (So they opened the cake mix) and started to mix and stir. They carefully measured out the flour (cake mix, eggs,) and sugar, and then added some chocolate chips for extra yumminess.

As the cookies (cake) baked in the oven, the rain outside began to ease up. The sun started to peek out from behind the clouds, and the rays of sunshine streamed through the windows.

Finally, the cookies were (cake was) ready! Jay and Kate took them (it) out of the oven and let them (it) cool for a few minutes. Then they laid them out on a plate and waited patiently for them to cool down completely.

When the cookies were (cake was) finally cool enough to eat, Jay and Kate each took a bite. (slice.) They were delicious! (It was so tasty!) They ate as many cookies as they could, (the whole cake,) and then weighed themselves on the scale to see how much they had gained.

“I think we gained about eight pounds each!” Kate laughed.

Jay and Kate had a great time that day, even though it had rained outside. They learned that sometimes the best days are the ones that are a little bit unexpected, and that it’s always fun to have a good friend by your side.

Next, I went all reading specialist nerdy and entered all the long /i/ words from the Fry’s 300 list and wrote an expository prompt. Let’s see if I can get a controlled vocabulary and syntax text with plenty of decodable long /i/ words. Will ChatGPT produce some decent background knowledge text along with the word recognition practice? I typed…

Write an essay at third grade reading level about George Washington’s achievements as President, using all of these words as possible: I, time, like, write, my, find, right, why, try, kind, high, eyes, light, while, might, life, side, mile, night, white, idea, sometimes

ChatGPT typed out the following. Not as good as the story, but useful for its purposes of building vocabulary, background knowledge, and practicing those long /i/ sound-spellings (I boldfaced). Of course, it would be far better if I had added other high utility long /i/ words, but I limited it to the Fry list to put it through its paces.

Hi there! I’m here to write an essay about George Washington’s achievements as President. George Washington was the first President of the United States, and he did a lot of important things during his time in office.

One of the things that I like about George Washington is that he was very fair and honest. He always tried to do the right thing, even when it was hard. He was kind to everyone, and he always treated people with respect.

Another thing that I find interesting about George Washington is that he was very strong and brave. He was a leader who was not afraid to make difficult decisions. He was also a great military leader and helped lead the Continental Army to victory during the American Revolution.

I think that it is important to learn about George Washington because he played a huge role in the history of our country. He helped to create the United States and set the standards for future presidents to follow. He is a role model for all of us, and his achievements should be celebrated.

Sometimes, it can be hard to understand why certain things happen in life. But I think that it is important to try to learn as much as we can about the people and events that have shaped our world. By learning about George Washington and his accomplishments, we can gain a better understanding of our own history and the world around us.

So, I encourage you to learn more about George Washington and the important role that he played in our country’s history. You might be surprised by how much you find out and how much you can learn from this amazing leader.

Think you’ll agree that the ChatGPT tool will be useful for us reading teachers. Finally, Mark, give us the link so that we can play with this amazing tool! To produce an AI essay in ChatGPT, simply copy and paste a writing prompt into the textbox on this site: https://openai.com/. Cheers!

Sam and Friends Phonics Books

*****

Mark Pennington is the author of The Science of Reading Intervention Program–a full year word recognition and language comprehension program for grades 4-adult. His 54 Sam and Friends Phonics books feature teenage characters and plots with comic illustrations by the amazing David Rickert. Full decodable with two Heart Words per book, your students will love these stories. Formatted for booklet printing, tablet, and phone.

Grammar/Mechanics , , , ,



ChatGPT Essay Detection

ChatGPT Detector

How can we English teachers detect ChatGPT-produced essays? We can bury our heads in the sand like snooty ostriches and claim that our superior literary backgrounds will detect artificial essays submitted by our students in the new year, but having spent some time researching this artificial intelligence (AI), I will say that it’s time we English teachers need to admit that “Houston, we’ve got a problem.”

To produce an AI essay in ChatGPT, simply copy and paste a writing prompt into the textbox on this site: https://openai.com/. However, before you do so, I suggest reading the rest of the article to find answers to some of the questions you will have as you start to play with this tool.

I selected a commonly-used writing prompt used with The Outsiders and pasted the following in the ChatGPT textbox:

In S.E. Hinton’s The Outsiders, each of the Greasers could be considered a hero. Include your own definition of hero in an essay, and provide textual evidence to explain how Johnny, Ponyboy, and Dally matches your definition. Include MLA citations with page numbers.

The results? A coherent, well-written five paragraph essay. A bit formulaic and repetitive. But that’s not the point, is it? As a first draft, the essay is more than passably good. With revision, the essay could be AP quality. I’ll share one of the body paragraphs as an example:

Dally is a hero because of his determination. Despite his tough exterior and rebellious behavior, Dally is deeply loyal to his friends and willing to do whatever it takes to help them (Hinton 48). When Johnny and Ponyboy run away, Dally helps them by giving them money and a place to stay (Hinton 66). He also helps them come up with a plan to turn themselves in and clear Johnny’s name (Hinton 76). Dally’s determination to stand by his friends and do what is right, even when it is difficult, makes him a hero.

Works Cited Hinton, S.E. The Outsiders. Viking Press, 1967.

And, yes, I generated another essay from the same prompt. Results were equally good, if not better, because the AI learns from its experience. It predicts based upon previous writing samples and so improves. Additionally, the essay content was largely different than that of the first essay. Teachers also say that the AI will vary the vocabulary and syntax to reflect that reflective of specific grade levels.

Furthermore, the ChatGPT is quite capable of revising student first drafts. I found it alarmingly, surprisingly, encouragingly good at syntax revision, for example.

In the December 9, 2022 issue of The Atlantic, author Daniel Herman (a high school English teacher in the San Francisco Bay Area) shares this AI capability:

I passed it (a student’s first draft of an essay) along to OpenAI’s bots. “Can you fix this essay up and make it better?” Turns out, it could. It kept the student’s words intact but employed them more gracefully; it removed the clutter so the ideas were able to shine through. It was like magic.”

Possible Solutions to Chat GPT-Produced Essays

1. In the December 29, 2022 Washington Post article, “Handwritten Essays Defeat CHATGPT,” a link to an opinion piece, “Here’s how teachers can foil CHATGPT,” author, Markham Heid, suggests that in-class handwritten essays can solve the plagiarism problem of artificially-produced digital writing. Moreover, the author posits that handwriting produces valuable cognitive benefits. Heid states,

For one thing, neuroscience research has revealed that, to the human brain… stimulates greater activity in a broader group of brain regions when compared with typing. However, the author admits that “much of the research comparing the differing neurological effects of handwriting and typing has focused on children or younger students.”

2.  Fight fire with fire. Artificial intelligence can also be used to predict with remarkable accuracy the likelihood of “predicted probabilities” that an essay has been created via AI. I used the https://huggingface.co/openai-detector and found upwards of 99% probability that an essay had been generated by ChatGPT with numerous submissions. Of course this company does require payment after numerous submissions; however, it shows what may be possible to detect cheating that is currently beyond the scope of TurnItIn or Google. Additionally, the software builds on its expertise by experience and lists the number of “tokens” (I presume similar submissions) to provide teachers with some sense of reliability.

3. TEACHER SOLUTIONS?

I posed this question on several FB groups and received some brilliant responses. I will summarize for the sake of brevity and keep names private:

  • Give an on demand essay to serve as a baseline. Easy to compare with AI-generated essays and provides good backup for teachers who may be subject to student and/or parent pushback.
  • Focus on smaller segments of writing craft. Perhaps one paragraph in-class. Know your students’ writing and you won’t get fooled. For example, students who consistently make capitalization, comma, or subject-verb errors are easy to spot. However, teachers report that telling the program to include errors in spelling, grammar, etc. typical of specified grade levels and the AI integrates these into its product.
  • Use screen monitoring tools, such as GoGuardian provides, or Google draft histories. I hear Draftback is an easy-to-use app to monitor writing history in Google docs. However, these won’t help with students opening separate windows and manually writing or typing from the AI-generated sources.

Making Lemonade out of Lemons

Perhaps rather than looking at AI and ChatGPT solely as opportunities for devious students to plagiarize or avoid writing essays and trying to figure out ways to outsmart this new threat, we might do better if we embrace the tool and design our instruction accordingly. A few ideas…

  • The AI is quite good at lower level cognitive tasks, such as data collection. To use Jane Schaffer’s writing terminology, the ChatGPT excels at gathering concrete details and citations. Less time thumbing through a novel for quotations. Less time Googling “global warming warning signs” on dozens of sites (and ads). More time analyzing the data and composing original commentary on the concrete details. More time revising word choice and syntax to create essays will original style and voice. In other words, more time for teachers to teach the important stuff about writing.
  • As a reading specialist, I’m excited to see AI used to produce relevant background knowledge. Think of it as a filter (and teach students how to do so) to refine a directed task for ChatGPT. Via trial and error I got more and more out of the AI with more specific parameters in my directions. These refinements will lead to the AI learning how to produce a quality knowledge base. Having access to this background information is key to advanced reading comprehension.
  • Another idea is to use ChatGPT to provide drafts for revision. I find this solution enticing, as it’s the revision step of the writing process that improves students’ writing sophistication and provides us the opportunity to share our writing teacher set of tools that we can share with our developing writers.

***** ADVERT*****

The easy-to-use e-Comments app.

Better Writing Feedback

One such set of revision tools that will teach students the art and science of writing revision for both student-produced and ChatGPT-produced essays is my e-Comments Chrome ExtensionThis easy-to-use app allows you to insert hundreds of canned comments from each of four different comment levels (including AP) into Google docs and slides with just one click from the pop-up e-Comments menu. Each instructional comment identifies, explains, and shows your writers how to revise a specific writing issue in stories, essays, and reports. These comprehensive comments don’t simply flag errors, they help your writers learn and improve. Plus, you can add your own comments! Works in all LMS and online Word 365. Add links, audio and video comments, and record the screen. Check out the free 10-day trial on the Chrome Web Store.

“An incomparable time saver for annotating student essays. Custom comments are easily added and modified, and each generation of this extension has improved significantly upon the last. Truly excellent.” K Mason Schecter

“This is an amazing extension. It not only allows you to grade quickly using ACTIONABLE feedback, but it also allows you to leave video and audio feedback all in one place! AWESOME!” Dr. Desiree Alexander

“Open an assignment in Google Classroom and it’s ready to go! Furthermore, create custom sets aside from the existing (overwhelmingly thorough and comprehensive) sets. While the pre-loaded sets deal with writing and conventions, the customizable sets could easily be utilized by teachers from other content areas who wish to expedite the grading of short/long answers and essays with comprehensive feedback. This extension isn’t just for English teachers – it’s for teachers.” Adam Huttenga

Grammar/Mechanics , , , , ,