COMPREHENSION Understanding it Teaching it Assessing it

1
COMPREHENSIONUnderstanding itTeaching
itAssessing it

• Scott Paris
• P. David Pearson

Slides at www.scienceandliteracy.org
2
Goal

• Convince you that you can make a big difference
  for childrens comprehension of text if you
• Understand the nature of comprehension
• And the implications models have for pedagogy
• Teach it well
• Develop a sophisticated and nuanced understanding
  of strategies, both comprehension and
  metacognitive
• Worry about how you orchestrate talk about text
• Understand the role of assessment in
  comprehension
• And then
• bathe it in lots of other good curriculum and
  teaching

3
Journey Part 1

• A little history how we got to where we are
  today
• The current research base on how readers
  understand what they read
• The Mental Models Tradition, as illustrated by
  Walter Kintschs C-I model
• David Pearson will be the tour guide
• with Scott Paris picking up the pieces and
  filling in the gaps
• About 1.5 hours

4
The Journey Part 2

• Comprehension Strategies and Skills
• Why we teach them
• How we teach them
• Talk about Text
• Scott Paris as tour guide
• with David Pearson picking up the pieces and
  filling in the gaps
• About 1.5 hours

5
The Journey Part 3

• Issues in the assessment of comprehension and
  related reading tasks
• Scott Paris as tour guide
• A simulation activity
• David Pearson as tour guide
• QA
• About 1.5 hours

6
Levels of teaching

• Primary Teachers Love
• Their students
• Secondary Teachers Love
• Their subject matter
• College Teachers Love
• Themselves!

7
Building blocks for the work I describe today in
part 1

• NAS Report How Children Learn
• Walter Kintsch Comprehension A Paradigm for
  Cognition
• A number of reviews of comprehension instruction
• New IRA/New Standards Report Reading and Writing
  with Understanding Comprehension in 4th and 5th
  Grade

8
How we got to where we are todaya little history
lesson
9
A really short history of RC

• Until 1914, reading meant oral reading not
  comprehension
• Accurate, fluent, expressive (declamatory)
  reading was the test
• Comprehension enters with silent reading and
  testing
• Comprehension is the result of instruction, not
  the object of instruction
• Simple view RC LCDec
• Implicit in our instructional models until the
  early 1970s

10
1970s/1980s

• Cognitive revolution
• Schema theory (Knowledge matters!)
• Comprehension matters (connect new to known)
• Reading, like writing, is all about making
  meaning
• So what does this mean for instruction
• If it doesnt just happen, how do you teach it?
• Durkin, 1978 we test it but we dont teach it

11
1980s Attempts to achieve a research-based
approach to comprehension instruction

• Determine the skills that are associated with
  skilled reading
• In small scale experiments, teach the skills to
  kids who do not excel at them and determine
  whether learning them leads to improved
  comprehension for that passage, that skill, and
  for comprehension more generally construed.
• Build a streamlined comprehension curriculum of
  mainline skills/strategies

Comprehension Revolution, cont.
12
Research-based approach to comprehension
instruction

• By late 1980s, we, as a field, had documented the
  efficacy of a whole set of instructional routines
  and strategies
• Paris and friends
• Pressley and friends
• Pearson and friends
• Friends and more friends
• Assessment developments
• ala Australia (several sites)
• Michigan
• Illinois
• Calif CLAS,
• Maryland MSPAP
• New Standards

13
Parallel Developments in the 70s/80s

• Ascendancy of constructivist pedagogies
• Whole Language
• Literature based reading
• Process writing
• Integrated curriculum

14

• But... Then something happened!
• About 1990
• Comprehension instruction took a back seat for
  more than a decade
• Andour reading tests reverted to mc, objectives
  based (now called standards based)

15
Why did comprehension take a back seat for a
decade and a half?

• Did not really fit either of the big movements of
  the late 80s/early 90s.
• Whole Language
• The New Phonics

16
Resistance from Whole Language

• Whole language found the tradition of explicit
  instruction in comprehension strategies a little
  too skillsy, too controlling, in feel.
• Preferred to have comprehension emerge from
  genuine encounters with authentic, engaging
  texts.
• Provide good texts and good assignments and it
  will happen (maybe a mini-lesson or two along the
  way...)

17
Resistance from pockets in the Big Phonics

• Does not really fit the zeitgeist of the new
  phonics renaissance either
• Some question the efficacy of skill and strategy
  instruction
• If you want to build oral language, fine.
• If you want to engage kids in comprehension
  activities, fine,
• But comprehension strategies probably dont
  matter that much.

18
We (well at least some of us) seem to be ready
for a comprehension renaissance

• WHY?
• Realization that no matter how important the code
  is, it is not the point of reading

19
More on why a renaissance

• The 4th grade slump
• The 7th grade cliff
• Enabling skills wont get you to where you need
  to be
• Phonics and phonemic awareness may take you
  through the first couple of levels of tests, but
• They wont take you to
• The big ideas and rich content of later
  curriculum
• Even assessments
• Mounting evidence that comprehension instruction
  really does matter.

20
The evidence

• National Reading Panel implicates comprehension
  and vocabulary along with phonics, phonemic
  awareness and phluency (no fluency!).
• More Recent Summaries of Comprehension Instruction

21
From our own work

• Taylor and Pearson
• In low income schools, the amount of high level
  talk about text, challenging assignments,
  student-centered instruction, and high levels of
  student engagement predicts growth in student
  achievement on a variety of measures.
• Teaching for cognitive engagement
• Conversely, phonics instruction predicted growth
  only in Grade 1.

22
Recent Meta-analysis on discussion by Wilkinson,
Murphy, Soter

• Review studies on discussion
• Three types of emphasis
• Efferent (unpacking the facts of the text)
• Aesthetic-gt expressive (say what you
  think--affective response)
• Critical-analytic
• Debate ideas
• Interrogate the text, the author, the issue
• Bottom line Pretty much get what you pay for
  (more later)

23
My current stance

• Present a vision of comprehension reflecting 30
  years of cognitive and instructional research
• Mine the text
• Mine ones store of knowledge
• Build a model of meaning that fits the current
  data available
• Move from comprehension to the acquisition of new
  knowledge
• Start all over again
• Knowledge begets comprehension begets knowledge
  begets comprehension

24
So whats new and different in this approach?

• Beyond Schema Theory and Reader Response
• Based on the evolving Construction-Integration
  model of Walter Kintsch, a prominent cognitive
  psychologist
• Focuses on how readers
• Build a text base for a text they read
• And filter it through their knowledge base to
  build
• A mental model (situation model) that balances
  the facts of the text base in relation to the
  facts of their knowledge base

25
Reader
Text
Reading Comprehension
Context
Most models of reading have tried to explain how
reader factors, text factors and context factors
interact when readers make meaning.
26
Bottom up and New Criticism Text-centric
Reader
Text
Reading
Reading Comprehension
Context
The bottom up cognitive models of the 60s were
very text centric, as was the new criticism
model of literature from the 40s and 50s (I.A.
Richards)
27
Pedagogy for Bottom up and New Criticism
Text-centric

• Since the meaning is in the text, we need to go
  dig it out
• Leads to Questions that
• Interrogate the facts of the text
• Get to the right interpretation
• Writerly readings or textual readings

28
Reader
Schema and Reader Response Reader-centric
Text
Reading Comprehension
Context
The schema based cognitive models of the 70s and
the reader response models (Rosenblatt) of the
80s focused more on reader factors--knowledge or
interpretation mattered most
29
Pedagogy for Reader-centric

• Since the meaning is largely in the reader, we
  need to go dig it out
• Spend a lot of time on
• Building background knowledge
• Inferences needed to build a coherent model of
  meaning
• Readers impressions, expressions, unbridled
  response
• Readerly readings

30
Critical literacy models Context-centric
Reader
Text
Reading
Reading Comprehension
Context
The sociocultural models of the 90s focused on
the central role of context (purpose, situation,
discourse community)
31
Pedagogy for Critical literacy models

• Since the meaning is largely in the context, we
  need to go dig it out
• Questions that get at the social, political and
  economic underbelly of the text (no neutral or
  autonomous texts)
• Whose interests are served by this text?
• Whos not there?
• What is the author trying to get us to believe?
• What features of the text contribute to the
  interpretation that money is evil?

32
Those from Australia will see another way to name
these movements
33
Bottom up and New Criticism Text-centric
Reader
Text
Reading
Reading Comprehension
Reader as Decoder
Context
The bottom up cognitive models of the 60s were
very text centric, as was the new criticism
model of literature from the 40s and 50s (I.A.
Richards)
34
Reader
Schema and Reader Response Reader-centric
Text
Reading Comprehension
Reader as Meaning Maker
Context
The schema based cognitive models of the 70s and
the reader response models (Rosenblatt) of the
80s focused more on reader factors--knowledge or
interpretation mattered most
35
Critical literacy models Context-centric
Reader
Text
Reading
Reading Comprehension
Reader as Text User and Text Critic
Context
The sociocultural models of the 90s focused on
the central role of context (purpose, situation,
discourse community)
36
CI Balance Reader and Text little c for
context
Reader
Text
Reading Comprehension
Context
In Kintschs model, Reader and Text factors are
balanced, and context plays a background
role--in purpose and motivation.
37
Pedagogical implications for CI

• Since the meaning is in this reader text
  interface, we need to go dig it out
• Query the accuracy of the text base to build up
  the microsructure and the macrostructure.
• What is going on in this part here where it says
• What does it mean when it says
• I was confused by this part
• Ascertain the situation model.
• So what is going on here?
• What do you know that we didnt know before?

38
New and different

• Most important A new model of the comprehension
  process
• Text (what the author left on the page)
• Text base (the version a reader creates on a
  veridical reading)
• Knowledge (what the reader brings from prior
  experience)
• Model of meaning for a text
• Dubbed the Situation Model (mental model)
• A model that accounts for all the facts and
  resources available in the current situation

39
Whats inside the Knowledge box?

• World knowledge (everyday stuff, including social
  and cultural norms)
• Topical knowledge (dogs and canines)
• Disciplinary knowledge (how history works)
• Linguistic knowledge
• Phonology
• Lexical and morphological
• Syntax
• Genre
• Pragmatics (how language works in the world)
  Discourse, register, academic language, intention
• Orthography (how print relates to speech

40
Kintchian Model
Context
Text
3 Knowledge Base
1 Text Base
2 Mental Model
Experience
Out in the world
Inside the head
41
How does a reader build a text base?
Excerpt from Chapter 8 of Hatchet
42

• Some of the quills were driven in deeper than
  others and they tore when they came out. He
  breathed deeply twice, let half of the breath
  out, and went back to work. Jerk, pause, jerk
  and three more times before he lay back in the
  darkness, done. The pain filled his leg now, and
  with it came new waves of self-pity. Sitting
  alone in the dark, his leg aching, some
  mosquitoes finding him again, he started crying.
  It was all too much, just too much, and he
  couldnt take it. Not the way it was.

43

• The pain filled his leg now, and with it came new
  waves of self-pity. Sitting alone in the dark,
  his leg aching, some mosquitoes finding him
  again, he started crying. It was all too much,
  just too much, and he couldnt take it. Not the
  way it was.

44

• I cant take it this way, alone with no fire and
  in the dark, and next time it might be something
  worse, maybe a bear, and it wouldnt be just
  quills in the leg, it would be worse. I cant do
  this, he thought, again and again. I cant. Brian
  pulled himself up until he was sitting upright
  back in the corner of the cave. He put his head
  down on his arms across his knees, with stiffness
  taking his left leg, and cried until he was cried
  out.

45
Building a Text Base

• Some of the quills were driven in (into what?
  His leg) deeper than others (other what? Quills)
  and they (the quills that were driven in deeper)
  tore when they (the deeper-in quills) came out
  (of his leg). He (Brian) breathed deeply twice,
  let half the breath out, and went back to work
  (work on what? Dont know yet. Suspense. Expect
  to find out in next sentence). Jerk, pause, jerk
  (the work is jerking quills out) and three more
  times (jerking quills out) he (Brian) lay back in
  the darkness, done (all the quills jerked out).

46

• The pain filled his (Brians) leg now, and with
  it (the pain) came new waves (what were the old
  waves?) of self-pity. (Brian) Sitting alone in
  the dark, his (Brians) leg aching, some
  mosquitoes finding him (Brian) again, he (Brian)
  started crying. It (the whole situation Brian was
  in) was all too much, just too much, and he
  (Brian) couldnt take it (the situation). Not the
  way it (the situation) was. (What way was the
  situation? Dont know yet. Suspense. Expect to
  find out in the next paragraph.)

47

• I (Brian) cant take it (the situation) this way
  (what way? Still dont know. Suspense), alone
  with no fire and in the dark (now we know this
  way means alone with no fire and in the dark),
  and next time it (the next situation) might be
  something worse (than this situation), maybe a
  bear, and it (the problem that will define the
  situation) wouldnt be just quills in the leg, it
  (the problem) would be worse (than quills in the
  leg). I (Brian) cant do this (deal with the
  problem situation), he (Brian) thought, again and
  again. I (Brian) cant do this (deal with the
  problem situation). Brian pulled himself (Brian)
  up until he (Brian) was sitting upright back in
  the corner of the cave. He (Brian) put his
  (Brians) head down on his (Brians) arms across
  his (Brians) knees, with stiffness taking his
  (Brians) left leg, and cried until he (Brian)
  was cried out.

48
Some key moves in building a text base

• Processing words and attaching meaning to them
• Using syntax to solidify key relations among
  ideas
• Microstructure
• Macrostructure
• Resolving reference--things that stand for other
  things (mainly pronouns and nouns)
• Using logical connectives (before, after,
  because, so, then, when, while, but) to figure
  out the relations among ideas
• Inferring omitted connectives (e.g., figuring out
  that A is the cause of B) based on PK about the
  world
• Posing questions for short term resolution
• Identifying ambiguities for later resolution
  (wait and see)

49
So how about building a situation model?

• The knowledge-comprehension relationship
• We use our knowledge to build a situation model
  for a text
• The information in the situation model is now
  available to become part of our long term memory
  and store of knowledge
• To assist in processing the next bit.

50
Situation Model for Hatchet Passage
51
The blurb from the jacket of Hatchet gives a
preview of the book

• Thirteen-year old Brian Robeson is on his way to
  visit his father when the single engine plane in
  which he is flying crashes. Suddenly, Brian finds
  himself alone in the Canadian wilderness with
  nothing but his clothing, a tattered windbreaker
  and the hatchet his mother has given him as a
  present and the dreadful secret that has been
  tearing him apart since his parents divorce. But
  now Brian has no time for anger, self-pity or
  despair it will take all his know-how and
  determination, and more courage than he knew he
  possessed, to survive.

52
What a reader knows by Chapter 8

• Brian is stranded in the Canadian wilderness
  with a hatchet and his wits as his only tools for
  survival. He already has overcome several
  obstacles, including surviving the plane crash,
  building a small shelter and finding food.
• In chapter eight, Brian awakens in the night to
  realize that there is an animal in his shelter.
  He throws his hatchet at the animal but misses.
  The hatchet makes sparks when it hits the wall of
  the cave. Brian then feels a pain in his leg. He
  sees the creature scuttle out of his shelter.
  Brian figures out that the animal was a porcupine
  because there are quills in his leg.

53
Some prior knowledge that a 5th grader might bring

• What sparks look like
• How it feels to be scared by an animal
• How big porcupines are
• To survive you have to have food, water and
  shelter
• To survive you have to be strong

54
An actual retelling of key parts of chapter 8
from Sam, a 5th grade reader

• The same text for which we just examined the text
  base

55
(No Transcript)
56
(No Transcript)
57
Why is this model of iteratively constructing and
integrating so important?

• The mental (situation) model is central to
  knowledge construction
• Building a mental model transforms new ideas and
  information into a form that can be added to
  memory, where they endure as knowledge that can
  be retrieved in the future. Unless readers build
  a mental model, the information they derive from
  the text is not likely to connect to their stored
  knowledge. The new information will be forgotten
  or lost.
• Key role of knowledge
• Knowledge involved in even the most literal of
  processing
• Knowledge begets comprehension begets knowledge
• Knowledge is available immediately dynamic
  store

58
How can we help students build solid text bases
and rich and accurate situation models?

• Do a good job of teaching subject matter in
  social studies, science, mathematics, and
  literature
• Dont let reading remain our curricular bully!

59
How can we help students build rich and accurate
mental models?

• Assist students in selecting appropriate
  knowledge frameworks to guide their construction
  process
• Do everything possible to build as many
  connections as possible with other texts,
  experiences, knowledge domains
• Do lots of what does this remind you of?
• What is this like? How is it different from what
  its like?

60
How can we help students build rich and accurate
mental models?

• A different model of guided reading
• Stop every once in a while and give the kids a
  chance to construct/revise their current mental
  model
• Research study
• interview protocol proved to be very
  instructive

61
Begin with very general probes before getting
specific

• So whats going on in this part?
• What do we know now that we didnt know before?
• Whats new?
• What was the author trying to get us to
  understand here?
• Well!say something!

62
Invite and support clarifications of tricky parts

• Anyone want to share something that was tricky or
  confusing?
• How about this part herewhere it says?
• I got confused by What do you think about this
  part? What was the author trying to get us to
  think.

63
Follow up general probes and invitations for
clarification with specific probes.

• So which of these things happened first? Why is
  that important?
• In this paragraph, they use a lot of pronouns.
  Lets check out our understanding of who or what
  they refer to..
• Typical discussion questions are OK too--just to
  make sure are the tricky parts get clarified.
• View questions as a scaffold for understanding
  the big picture not as a quiz.

64
The general model for guided reading

• A set for stock-taking
• More specific probes to scaffold the construction
  of the text base and situation model
• Results in a pretty good summary of the
  selection--story, article, etc.

65
Developing Text Bases and Mental Models

• Ensure that students have a full tool box (set
  of strategies) to haul out when things dont just
  happen automaticallyfor
• Connecting the known to the new
• Connecting texts and parts of texts
• Working toward coherence among potentially
  unconnected ideas
• Recognizing and resolving ambiguities.

66
The Vulnerabilities

• Clumsiness with motivation
• A nod to interest and an assumption that readers
  are motivated
• Gloss over critical reading
• Assumes a liberal humanist critical thinking
  perspective, not a post-modern critical
  theoretical stance

67
One more time
Context
Text
3 Knowledge Base
1 Text Base
2 Situation Model
Experience
Out in the world
Inside the head
68
Key References

• Duke, N. Pearson, P.D. (2002). Effective
  practices for developing reading comprehension.
  In A. Farstrup J. Samuels (Eds.), What research
  has to say about reading instruction (3rd ed.)
  (pp. 205-242). Newark DE International Reading
  Association.
• Chapter 6 in Hampton, S., Resnick. L. Reading
  and Writing with Understanding. New IRA
  Publication