Whatassessmentcan—and%cannot —do%for%teachers%of%reading% institute 7.pdf ·...

What assessment can—and cannot—do for teachers of reading

Dylan Wiliam (@dylanwiliam)

www.dylanwiliam.org

!

Don’t blame me: I’m Welsh 2

¨  A very simple code: ¤ a b c ch d dd e f ff g ng h i j l ll m n o p ph r rh s t th u w y

¨  ComplicaDons: ¤ vowels can be short or long ¤ “y” can be clear (unaccented) or obscure (accented), but ¤  the accent is (almost) always on the penulDmate syllable

¨  SimplificaDons: ¤ one leJer—>one sound ¤  if you want a diphthong, use two vowels (wy, aw, ei,…)

Overview: Science and Design

¨  We need to improve student achievement ¨  This requires improving teacher quality ¨  Improving the quality of entrants takes too long ¨  So we have to make the teachers we have beJer ¨  We can change teachers in a range of ways ¨  Some will benefit students, and some will not ¨  Those that do involve changes in teacher pracDce

¨  Changing pracDce requires new kinds of teacher learning ¨  And new models of professional development Design

Science

3

!

Raising achievement maJers 4

¨  For individuals: ¤  Increased lifeDme salary ¤  Improved health ¤ Longer life

¨  For society: ¤ Lower criminal jusDce costs ¤ Lower healthcare costs ¤  Increased economic growth:

n Net present value to the U.S. of a 25-‐point increase on PISA: $40 trillion (2.5 Dmes the NaDonal Debt)

n Net present value to the U.S. of ge^ng all students to 400 on PISA: $70 trillion

School effecDveness

Three generaDons of school effecDveness research: ¨  Raw results approaches:

¤ Different schools get different results. ¤  Conclusion: Schools make a difference.

¨  Demographic-‐based approaches: ¤ Demographic factors account for most of the variaDon. ¤  Conclusion: Schools don’t make a difference.

¨  Value-‐added approaches: ¤  School-‐level differences in value-‐added are relaDvely small. ¤  Classroom-‐level differences in value-‐added are large. ¤  Conclusion: An effecDve school is a school full of effecDve classrooms.

5

!

We need to focus on classrooms, not schools 6

¨  In the USA, variability at the classroom level is at least four Dmes that at school level. ¤ As long as you go to school, it doesn’t maJer very much which school you go to.

¤ But it maJers very much which classrooms you are in.

¨  It’s not class size. ¨  It’s not the between-‐class grouping strategy. ¨  It’s not the within-‐class grouping strategy.

!

Teacher quality and student achievement

Study Loca1on Reading Math Rockoff (2004) New Jersey 0.10 0.11

Nye, Konstantopoulos, Hedges (2004) Tennessee 0.26 0.36

Rivkin, Hanushek, and Kain (2005) Texas 0.15 0.11

Aaronson, Barrow, and Sander (2007) Chicago 0.13

Kane, Rockoff, and Staiger (2008) New York City 0.08 0.11 Jacob and Lefgren (2008) 0.12 0.26 Kane and Staiger (2008) 0.18 0.22 Koedel and BeJs (2009) San Diego 0.23 Rothstein (2010) North Carolina 0.11 0.15

Hanushek and Rivkin (2010) 0.11

CheJy et al. (2014) 0.12 0.16

Hanushek and Rivkin (2010)*

What this means 8

¨  Take a group of 50 teachers: ¤  Students taught by the most effecDve teacher in that group of 50 teachers learn in six months what those taught by the average teacher learn in a year.

¤  Students taught by the least effecDve teacher in that group of 50 teachers will take two years to achieve the same learning

(Hanushek & Rivkin, 2006) ¨  And furthermore:

¤  In the classrooms of the most effecDve teachers, students from disadvantaged backgrounds learn at the same rate as those from advantaged backgrounds

(Hamre & Pianta, 2005).

!

Improving teacher quality takes Dme

¨  A classic labor force issue with two (non-‐exclusive) soluDons: ¤ Replace exisDng teachers with beJer ones:

n Raising the bar for entry into the profession? n De-‐select (i.e., fire) ineffecDve teachers?

n  Teacher observaDons? n  Value-‐added measurements? n  Student percepDon surveys?

¤ Help exisDng teachers become even more effecDve.

9

Raise the bar for entry into teaching?

!

Teacher qualificaDons and student progress 11

Mathema1cs Reading

Primary Middle High Primary Middle High

General theory of educaDon courses

Teaching pracDce courses

Pedagogical content courses

Advanced university courses

ApDtude test scores

Harris and Sass (2007)

Mathema1cs Reading

Primary Middle High Primary Middle High

General theory of educaDon courses — Teaching pracDce courses — + Pedagogical content courses + + Advanced university courses — + ApDtude test scores —

!

Teacher licensure in California 12

¨  Three tests ¤ General knowledge

n California Basic EducaDonal Skills Test (CBEST)

¤ Subject specific tests n California Subject ExaminaDons for Teachers (CSET)

¤ Reading pedagogy (elementary only) n Reading InstrucDon Competence Assessment (RICA)

!

Reading InstrucDon Competency Assessment Which of the following informal assessments would be most appropriate to use to assess an individual student's phonemic awareness?

A.  Asking the student to idenDfy the sound at the beginning, middle, or end of a spoken word (e.g., "What sound do you hear at the end of step?").

B.  Having the student listen to a tape-‐recorded story while looking at the book and then answer several simple quesDons about the story.

C.  Asking the student to idenDfy the leJers in the alphabet that correspond to the iniDal consonant sounds of several familiar spoken words.

D.  Having the student listen to the teacher read aloud a set of words with the same beginning sound (e.g., train, trap, trouble) and then repeat the words.

RICA pracDce test, item #10 EvaluaDon Systems (2014)

!

Teacher licensure in LAUSD 14

Impact of teacher licensure scores on student achievement

Reading MathemaDcs

CBEST —

CSET —

RICA +

Buddin and Zamarro (2009)

Teacher observaDons

!

Framework for teaching (Danielson 1996)

¨  Four domains of professional pracDce 1.  Planning and preparaDon 2.  Classroom environment 3.  InstrucDon 4.  Professional responsibiliDes

¨  Links with student achievement (Sartain, et al. 2011) ¤ Domains 1 and 4: no impact on student achievement ¤ Domains 2 and 3: some impact on student achievement

16

!

A framework for teaching (Danielson, 1996)

¨  Domain 2: The classroom environment ¤  2a: CreaDng an environment of respect and rapport ¤  2b: Establishing a culture for learning ¤  2c: Managing classroom procedures ¤  2d: Managing student behavior ¤  2e: Organizing physical space

¨  Domain 3: InstrucDon ¤  3a: CommunicaDng with students ¤  3b: Using quesDoning and discussion techniques ¤  3c: Engaging students in learning ¤  3d: Using assessment in instrucDon ¤  3e: DemonstraDng flexibility and responsiveness

!

ObservaDons and teacher quality 18

-‐15

-‐10

-‐5

0

5

10

15

20

UnsaDsfactory Basic Proficient DisDnguished

Percen

tage cha

nge in ra

te of

learning

Reading MathemaDcs

Sartain, Stoelinga, Brown, Luppescu, Matsko, Miller, Durwood, Jiang, and Glazer (2011)

So, the highest rated teachers are 30% more producDve than the lowest rated

But the best teachers are 400% more producDve than the least effecDve

Imprecision in lesson observaDons 19

Achieving a reliability of 0.9 in judging teacher quality through lesson observaDon is likely to require observing a teacher teaching 6 different classes, and for each lesson to be judged by 5 independent observers.

Hill, Charalambous and Kraz (2012)

Teacher value-‐added

!

Teacher value-‐added

Value-‐added quin1le: Student fixed-‐effects model

Value-‐ad

ded qu

in1le:

Trad

i1on

al m

odel

1 2 3 4 5

1 38 22 24 16 0

2 26 28 15 20 11

3 20 20 20 24 16

4 13 24 26 13 24

5 9 5 12 28 47

Goldhaber, Goldschmidt, and Tseng (2013)

!

Teacher progress in reading

AJeberry, Loeb and Wyckoff (2013)

-‐0.3

-‐0.2

-‐0.1

0

0.1

0.2

0.3

0.4

0 1 2 3 4

Value-‐ad

ded (stude

nt SDs

)

Years of teaching experience

Quin1le of value added

5th 4th 3rd 2nd 1st

!

Issues with value-‐added models for teachers 23

¨  Different (reasonable) models of value-‐added give very different esDmates of teacher quality (Goldhaber, Goldschmidt and Tseng, 2013)

¨  Teacher value-‐added in their first year accounts for less than 5% of the variaDon in teacher quality in their fizh year of teaching (AJeberry, Loeb and Wyckoff, 2013)

¨  Teachers benefit students for at least three years azer they stop teaching them (Rothstein, 2010)

!

Skill improves at different rates for different skills…

Rockoff (2004)

0.00

0.05

0.10

0.15

0.20

0.25

0 1 2 3 4 5 6 7 8 9 10 11

Effect size

Experience (years)

Reading Comprehension

Vocabulary

!

Reading vocabulary

Reading comprehension

Math computa1on

Math concepts

Reading vocabulary

Reading comprehension

0.27

Math computaDon

0.16 0.46

Math concepts

0.32 0.58 0.67

…and experDse is specific…

Rockoff (2004)

!

Pu^ng it all together 26

¨  “Measures of EffecDve Teaching” project ¤ WeighDngs of different sources to maximize predicDon n Teacher value-‐added raDngs (65%) n Classroom observaDons (9%) n Student percepDon surveys (25%)

¤ Reliability of teacher raDng: 0.42 ¤ ProporDon of variance explained in:

n  state test: 37% n higher-‐order test: 12%

Bill and Melinda Gates FoundaDon (2012)

Assessment: good servant, bad master

!

“They have perverted the best efforts of teachers, and narrowed and grooved their instrucDon; they have occasioned and made well nigh imperaDve the use of mechanical and rote

methods of teaching; they have occasioned cramming and the most vicious habits of study; they have caused much of the

overpressure charged upon schools, some of which is real; they have tempted both teachers and pupils to dishonesty; and last

but not least, they have permiJed a mechanical method of school supervision.”

WriJen examinaDons 28

(White, 1888 pp. 517-‐518)

!

Why all assessment should be for learning 29

“We shouldn't want a shiz to formaDve assessment because research shows how it improves learning (we don't need to be told that—it has to be true). We should want it because schools are places where learners should be learning more ozen than they are being selected, screened or tested in order to check up on their teachers. The laJer are important; the former are why schools exist.” (Silcock, 1998)

!

Ten principles for assessment

¨  Assessment ¤  is part of effecDve learning ¤  focuses on how students learn ¤  is central to classroom pracDce ¤  is a key professional skill ¤  is sensiDve and construcDve ¤  fosters moDvaDon ¤  promotes understanding of goals and criteria ¤  helps learners know how to improve ¤  develops the capacity for self-‐assessment ¤  recognises all educaDonal achievement

Assessment Reform Group (2002)

31

Mapping out the terrain

Annual

Interim

Weekly

Daily

InstrucDonal Guidance

(“formaDve”)

Describing Individuals

(“summaDve”)

InsDtuDonal Accountability (“evaluaDve”)

FuncDon

Hourly

Timescale

High-‐stakes accountability

Academic promoDon

Hinge-‐point quesDons

End-‐of-‐unit tests

Benchmarks

Exit pass

Common assessments

End-‐of-‐course exams

Before the end-‐ of-‐unit tests

Growth

32

!

Impact of background on development

Feinstein (2008)

0 10 20 30 40 50 60 70 80 90

100

20 40 60 80 100 120

Percen

1le of distribu1

on

Age in months

Q1 Low SES Q1 High-‐SES Q4 Low SES Q4 High SES

!

Meaningful differences

¨  Hour-‐long samples of family talk in 42 American families

¨  Number of words spoken to children by adults by the age of 36 months ¤  In professional families: 35 million ¤  In other working-‐class families: 20 million ¤  In families on welfare: 10 million

¨  Kinds of reinforcements: posiDve negaDve

professional 500,000 50,000 working-‐class 200,000 100,000 welfare 100,000 200,000

Hart and Risley (1995)

!

Trajectories of learning to read

Pianta, Belsky, Vandergriz, Houts, & Morrison (2008)

‘Fast’ readers

‘Normal’ readers

PK 1st 3rd 5th

!

Reading is complex…

Scarborough (2001)

!

Reading skills: what are they really?

A manifold, contained in an intui:on which I call mine, is represented, by means of the synthesis of the understanding, as belonging to the necessary unity of self-‐consciousness; and this is effected by means of the category. What is the main idea of this passage?

1. Without a manifold, one cannot call an intuiDon ‘mine.’ 2. IntuiDon must precede understanding. 3. IntuiDon must occur through a category. 4. Self-‐consciousness is necessary to understanding

Hirsch (2006)

!

What is reading, really (part 2) 38

"Thus, as the final day dawned and a near capacity crowd lusDly cheered every run Australia mustered, much depended on PonDng and the new wizard of Oz, Mike Hussey, the two overnight batsmen. But this duo perished either side of lunch—the laJer a liJle unfortunate to be adjudged leg-‐before—and with Andrew Symonds, too, being shown the dreaded finger off an inside edge, the inevitable beckoned, bar the pyrotechnics of Michael Clarke and the ninth wicket.

Clarke clinically cut and drove to 10 fours in a 134-‐ball 81, before he stepped out to Kumble to present an easy stumping to Mahendra Singh Dhoni.” (The Guardian, January 19, 2008) Flesch-‐Kincaid Reading Ease: 42.5 Grade level: 14

Quality in assessment

!

What is an assessment? 40

¨  An assessment is a process for making inferences ¨  Key quesDon: “Once you know the assessment outcome, what do you know?”

¨  “One validates, not a test, but an interpreta:on of data arising from a specified procedure” (Cronbach, 1971; emphasis in original)

¨  Consequences ¤ No such thing as a valid (or indeed invalid) assessment ¤ No such thing as a biased assessment

!

Threats to validity 41

¨  Construct-‐irrelevant variance ¤ SystemaDc: good performance on the assessment requires abiliDes not related to the construct of interest

¤ Random: good performance is related to chance factors, such as luck (effecDvely poor reliability)

¨  Construct under-‐representaDon ¤ Good performance on the assessment can be achieved without demonstraDng all aspects of the construct of interest

Understanding reliability

!

Understanding test scores

¨  Consider a test with the following properDes ¤ average score: 50 ¤ standard deviaDon of scores: 15

¨  This means that on average: ¤ 68% of students score between 35 and 65 (±1 sd) ¤ 96% of students score between 20 and 80 (±2 sd)

¨  But the score on any one occasion may not be a good guide to the score on another occasion

¨  No measurement (either educaDonal or other) is perfectly reliable

43

!

Reliability of test scores 44

¨  Basic assumpDon of classical test theory: X = T + E

¤  In other words, the score a student gets on any occasion is their “true score” plus or minus some error

¤ This does not mean assuming that ability is fixed ¤ The “true score” is just the long-‐run average score over numerous tesDng occasions with similar tests, so n The average error is zero n For a reliable test, the errors are bunched closely n For an unreliable test, the errors are spread out n The more spread out the errors, the less reliable the test

!

Defining reliability 45

¨  To find out how spread out the errors are, we can’t use the average (because it’s zero!), so

¨  We square the errors (to get rid of the minus signs) and then average (this is called the variance)

¨  The reliability of a test is then defined as:

¨  When the variance of the errors: ¤  is zero, the test is perfectly reliable (reliability = 1) ¤ equals that of the observed scores, the test score is just noise (reliability = 0)

!

Reliability: 0.75 46

0 10 20 30 40 50 60 70 80 90

100

0 10 20 30 40 50 60 70 80 90 100

Observed score

True score

!


0 10 20 30 40 50 60 70 80 90

100

0 10 20 30 40 50 60 70 80 90 100

Observed score

True score

!

The standard error of measurement 50

¨  The “standard error of measurement” (SEM) is just the standard deviaDon of the errors, so, on any given tesDng occasion ¤ 68% of students score within 1 SEM of their true score ¤ 96% of students score within 2 SEM of their true score

¨  Re-‐arranging the defining equaDon for reliability:

!

Reliability, standard errors, and progress 51

Grade Reliability SEM as a percentage of annual progress 1 0.89 26% 2 0.85 56% 3 0.82 76% 4 0.83 39% 5 0.83 55% 6 0.89 46%

Average 0.85 49%

In other words, the standard error of measurement of this reading test is equal to six months’ progress by a typical student

!

ImplicaDons: TesDng ever semester 52

¨  If we test students every six months: ¤ 16% of students who have made normal progress will appear to have made no progress at all, or appear to have gone backwards.

¤ 16% of the students who have made no progress will appear to be progressing faster than normal.

!

ImplicaDons: Bimonthly tesDng 53

¨  If we test students every two months: ¤ 37% of students who have made normal progress will appear to have made no progress at all, or appear to have gone backwards.

¤ 37% of the students who have made no progress will appear to be progressing faster than average.

!

If you must measure progress… 54

¨  As rules of thumb: ¤ For individual students, progress measures are meaningful only if the progress is more than twice the standard error of measurement of the test being used to measure progress.

¤ For a class of 25 students, progress measures are meaningful if the progress is more than half the standard error of measurement of the test being used to measure progress.

Curriculum-‐based measurement

!

Curriculum-‐based measurement 56

¨  Used for a variety of purposes including screening, benchmarking, and progress monitoring

¨  Avoids the problems of measuring change scores because it focuses on mulDple assessments of status.

¨  Depends on a clear view of what will be learned by the end of the instrucDonal sequence.

¨  However, it is not a panacea

!

Review of studies of CBM-‐R 57

“Such studies suggest that even under the best condiDons (i.e., high-‐quality probe sets and Dghtly controlled condiDons), (a) a minimum of 5 or 6 weeks of data with mulDple data points collected per week are needed to inform rouDne instrucDonal decisions and (b) a minimum of 12 weeks of data with mulDple data points collected per week are needed to make special educaDon eligibility decisions ” (p. 12)

Ardoin, Christ, Morena, Cormier, and Klingbeil (2013)

!

58

“… at this point, there are no studies to suggest that an individual student's progress can be accurately determined using CBM-‐R progress monitoring data” (p. 14) “Furthermore, trainers and publishers of CBM-‐R materials should neither suggest to school teachers and other educators that CBM-‐R progress monitoring data can be used as a primary outcome measure to evaluate individual student growth over short periods of Dme nor train them in current CBM-‐R decision rules.” (pp. 14-‐15)

Ardoin, Christ, Morena, Cormier, and Klingbeil (2013)

DiagnosDc tesDng

!

The limits of diagnosDc tesDng 60

¨  120-‐item mulDple choice test for teacher licensure ¤ Four major subject areas

n  language arts/reading n mathemaDcs n  social studies n  science

¤ 30 items per subject area ¤ Sub-‐score reliabiliDes range from 0.71 to 0.83

!

How reliable are 10-‐item subtest scores? 61

¨  Items for each subject area ranked in order of difficulty (i.e., 1 to 30)

¨  Three parallel 10-‐item forms created in each subject area: ¤ Form A: items 1, 4, 7, … 28 ¤ Form B: items 2, 5, 8, … 29 ¤ Form C: items 3, 6, 9, … 30

¨  Sub-‐score reliabiliDes in the range 0.40 to 0.60 ¨  On form A, 271 examinees scored 7 in mathemaDcs and 3 in science

!

Scores of 271 students on form B

Science subscore 1 2 3 4 5 6 7 8 9 10

Math subscore

1 0 0 0 1 1 1 0 0 0 0

2 0 0 0 1 3 1 2 0 0 0

3 1 0 0 1 2 4 3 1 1 1

4 0 0 2 7 7 6 4 0 1 0

5 0 1 1 1 10 14 8 5 1 1

6 2 0 1 5 10 11 15 8 1 1

7 0 1 4 4 4 11 10 7 4 0

8 0 1 1 5 12 13 7 5 4 0

9 0 0 1 1 6 3 7 4 3 0

10 0 0 0 1 1 2 1 1 0 0

Sinharay, Gautam and Halberman (2010)

110 out of 271 (41%) examinees got a beJer form B score in science than mathemaDcs

!

63

¨  A student scoring 7 on mathemaDcs and 3 on science would probably want to improve the laJer

¨  But 110 of the 271 examinees got a beJer score in science than mathemaDcs on Form B

¨  CorrelaDon of science subscores on Forms A and B is 0.48

¨  CorrelaDon of science subscore on Form A with total score on Form B is 0.63

¨  In other words, the total score on a test is a beJer guide to the score on a sub-‐test than another score on the same sub-‐test

So what’s to be done?

The evidence base for formaDve assessment

¨  Fuchs & Fuchs (1986) ¨  Natriello (1987) ¨  Crooks (1988) ¨  Bangert-‐Drowns, et al. (1991) ¨  Dempster (1991, 1992) ¨  Elshout-‐Mohr (1994) ¨  Kluger & DeNisi (1996) ¨  Black & Wiliam (1998)

¨  Nyquist (2003) ¨  Brookhart (2004) ¨  Allal & Lopez (2005) ¨  Köller (2005) ¨  Brookhart (2007) ¨  Wiliam (2007) ¨  Ha^e & Timperley (2007) ¨  Shute (2008)

65

!

Which of these are formaDve?

A.  A district science supervisor uses test results to plan professional development workshops for teachers

B.  Teachers doing item-‐by-‐item analysis of 5th grade math tests to review their 5th grade curriculum

C.  A school tests students every 10 weeks to predict which students are “on course” to pass the state test in March

D.  “Three-‐fourths of the way through a unit” test E.  Students who fail a test on Friday have to come back on

Saturday F.  Exit pass quesDon: “What is the difference between mass

and weight?” G.  “Sketch the graph of y equals one over one plus

x squared on your mini-‐white boards.”

66

!

The formaDve assessment hijack

¨  Long-‐cycle: ¤  Span: across units, terms ¤  Length: four weeks to one year ¤  Impact: Student monitoring; curriculum alignment

¨  Medium-‐cycle: ¤  Span: within and between teaching units ¤  Length: one to four weeks ¤  Impact: Improved, student-‐involved assessment; teacher cogniDon about learning

¨  Short-‐cycle: ¤  Span: within and between lessons ¤  Length:

n  day-‐by-‐day: 24 to 48 hours n  minute-‐by-‐minute: five seconds to two hours

¤  Impact: classroom pracDce; student engagement

67

!

Main approaches to formaDve assessment 68

¨  Professional Learning CommuniDes “…an inclusive group of people, moDvated by a shared learning vision, who support and work with each other, finding ways, inside and outside their immediate community, to enquire on their pracDce and together learn new and beJer approaches that will enhance all pupils’ learning.” (Stoll et al., 2006)

¨  Two main approaches ¤ Focus on outcomes for students ¤ Focus on increased teacher capacity

Complementary processes

•  Quality control •  Common assessments •  Improvement through beJer

team work and systems •  Focus on individual outcomes

for students •  Regular meeDngs focused on

data •  16 points on PISA (in two to

three years)

•  Quality assurance •  Highly structured meeDngs •  Improvement through

increased teacher capacity •  Focus on teachers’ individual

accountability for change •  Regular meeDngs focused on

teacher change •  30 points on PISA (in two to

three years)

69

Instruc1onal data teams Teacher skill development

Unpacking formaDve assessment

Where the learner is going Where the learner is How to get there

Teacher

Peer

Learner

Clarifying, sharing and

understanding learning intenDons

Engineering effecDve discussions, tasks, and acDviDes that elicit evidence of learning

Providing feedback that moves learners

forward

AcDvaDng students as learning resources for one another

AcDvaDng students as owners of their own learning

70

!

FormaDve assessment and other prioriDes

¨  FormaDve assessment is an integral part of many current policy prioriDes: ¤ Framework for teaching (Danielson) ¤ DifferenDated instrucDon (Tomlinson) ¤ Response to (instrucDon and) intervenDon

71

!

Framework for teaching (Danielson 1996)

¨  Four domains of professional pracDce 1.  Planning and preparaDon 2.  Classroom environment 3.  InstrucDon 4.  Professional responsibiliDes

¨  Links with student achievement (Sartain, et al. 2011) ¤ Domains 1 and 4: no impact on student achievement ¤ Domains 2 and 3: some impact on student achievement

72

!

The framework in detail

¨  Domain 2: The classroom environment ¤  2a: CreaDng an environment of respect and rapport ¤  2b: Establishing a culture for learning ¤  2c: Managing classroom procedures ¤  2d: Managing student behavior ¤  2e: Organizing physical space

¨  Domain 3: InstrucDon ¤  3a: CommunicaDng with students ¤  3b: Using quesDoning and discussion techniques ¤  3c: Engaging students in learning ¤  3d: Using assessment in instrucDon ¤  3e: DemonstraDng flexibility and responsiveness

73

FormaDve assessment and domain 3

¨  CommunicaDng with students

¨  Using quesDoning and discussion techniques

¨  Engaging students in learning

¨  Using assessment in instrucDon

¨  DemonstraDng flexibil-‐ ity and responsiveness

¨  Sharing learning intenDons with students

¨  EliciDng evidence

¨  Feedback

¨  Students as learning resources

¨  Students as owners of their learning

Framework for teaching Classroom forma1ve assessment

74

!

DifferenDated instrucDon: not a new idea

¨  DifferenDaDon in acDon (Stradling & Saunders, 1993) ¨  Differences in

¤ educaDonal goals ¤ curriculum structure ¤ course content ¤  learning tasks ¤  teaching approach ¤ pace of learning ¤ assessment ¤  review

75

!

Most definiDons of DI are vague

“While the concept of ‘differenDated instrucDon’ can be defined in many ways, as good a definiDon as any is ensuring that what a student learns, how he/she learns it, and how the student demonstrates what he/she has learned is a match for that student's readiness level, interests, and preferred mode of learning.” (Tomlinson, 2004 p. 188) “To differenDate instrucDon is to recognize students' varying background knowledge, readiness, language, preferences in learning and interests; and to react responsively. DifferenDated instrucDon is a process to teaching and learning for students of differing abiliDes in the same class.” (Hall, Strangman, & Meyer, 2011)

76

DifferenDated instrucDon and formaDve assessment

77

Aspects of differen1ated instruc1on (Hall, Strangman & Meyer, 2008) FA?

Conten

t Several elements and materials are used

Align tasks and objecDves to learning goals ✓

InstrucDon is concept-‐focused and principle-‐driven

Process Flexible grouping is consistently used

Classroom management benefits students and teachers

Prod

ucts IniDal and on-‐going assessment of student readiness and growth ✓

Students are acDve and responsible explorers ✓

Vary expectaDons and requirements for student responses ✓

Misc

celane

ous Clarify key concepts and generalizaDons

Use assessment as a teaching tool ✓

Emphasize criDcal and creaDve thinking as a goal in lesson design

Engaging all learners is essenDal ✓

Balance between teacher-‐assigned and student-‐selected tasks

!

Response to (instrucDon and) intervenDon

“Response to intervenDon integrates assessment and intervenDon within a mulD-‐level prevenDon system to maximize student achievement and reduce behavior problems. With RTI, schools idenDfy students at risk for poor learning outcomes, monitor student progress, provide evidence-‐based intervenDons and adjust the intensity and nature of those intervenDons depending on a student’s responsiveness, and idenDfy students with learning disabiliDes.” (NaDonal Center on Response to IntervenDon, 2010)

¨  Two “creaDon myths” for RT(I)I ¤  A protocol for prevenDng academic failure (progress monitoring, early—research-‐based—intervenDon)

¤  An alternaDve to IQ tesDng in the idenDficaDon of learning disabiliDes

78

!

Response to (instrucDon and) intervenDon

¨  Key points ¤ Tier 1 must be high-‐quality, evidence-‐based instrucDon

¤ Student progress must be monitored ¤ Failure to progress triggers addiDonal support

¨  FormaDve assessment ¤ Makes Der 1 instrucDon as effecDve as it can be ¤ Allows assessment of progress (for Der 2 assessment)

79

And one big idea

Where the learner is going Where the learner is How to get there

Teacher

Peer

Learner

80

Using evidence of achievement to adapt

what happens in classrooms to meet

learner needs

!

An educaDonal posiDoning system

¨  A good teacher: ¤ Establishes where the students are in their learning ¤  IdenDfies the learning desDnaDon ¤ Carefully plans a route ¤ Begins the learning journey ¤ Makes regular checks on progress on the way ¤ Makes adjustments to the course as condiDons dictate

81

Whatassessmentcan—and%cannot —do%for%teachers%of%reading% institute 7.pdf ·...

Documents

Transcript of Whatassessmentcan—and%cannot —do%for%teachers%of%reading% institute 7.pdf ·...