prevention: the promise and the power · 2020-07-10 · keynote presentation dmg superinstitute las...

© 2018, Dynamic Measurement Group 1

Keynote Presentation DMG SuperInstitute Las Vegas, NV, July, 2018

1

Testing Environment

Prevention:The Promise and the Power

Keynote AddressDIBELS Next Super Institute

Las Vegas, NevadaJuly 10, 2018

Ruth KaminskiDynamic Measurement Group

2

Back to the Basics

“Back to the Basics” From Urban Dictionary:

If you talk about getting back to basics, you are suggesting that people have become too concerned with complicated details or new theories, and that they should concentrate on simple, important ideas or activities.



3

Back to Basics

4

The Chaos

I am not sure that I can do DIBELS now because:

now we are doing RTI/MTSSnow we have the Common CoreNow we have to focus on teaching children to comprehend more complex textwe have to identify children who have dyslexia



5

Back to Basics: DIBELS

PREVENTION

6

1 2 3 4 5 6

Grade

Grade 1

Cohort

Grade 2

Cohort

Grade 3

Cohort

Grade 4

Cohort

Grade 5

Cohort

Wor

ds P

er M

inut

e

Middle10%

Low 10%

Good, R. H., Simmons, D. C., & Smith, S. B. (1998). Effective academic interventions in the United States: Evaluating and enhancing the acquisition of early reading

skills. School Psychology Review, 27, 740-753. [Joint publication with Educational and Child Psychology.]

Why DIBELS?



7

0

20

40

60

80

100

120

140

0.0 1.0 2.0

Need for DIBELS®: Prevent Reading Difficulties

7

Wor

ds P

er M

inut

e

Grade1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9

Grade

Months

Good, R. H., Simmons, D. C., & Smith, S. B. (1998). Effective academic interventions in the United States: Evaluating and enhancing the acquisition of early reading skills. School Psychology Review, 27, 740-753. [Joint publication with Educational and Child Psychology.]

8

The Issue: Differences in Precursors to Reading Trajectories Begin Even Earlier (Hart & Risley, 1995)

Hart & Risley, 1995



9

Effects of Preschool?

Marked variation in findings of lasting effects of preschool:

Some evidence of long-term benefitsSizeable short-term benefits in children’s academic skills

Initial gains tend to dissipate as children progress through elementary school

Campbell et al., 2012; Jenkins et al., 2011; Philips, Gormley, & Anderson, 2016; Schweinhart, 2005; Yoshikawa, Weiland, & Brooks-Gunn, 2016

10

Role of Kindergarten

Complementary

Compensatory

Discordant

Enhance

Compensate

Dissipate

Bailey, Duncan, Odgers, & Yu, 2017; Phillips et al., 2017



11

Preschool Early Literacy Indicators (PELI)

12

PELI to Kindergarten DIBELS Data

604 children with PELI and DIBELS Next Data at BOY, MOY, and EOY in PreK and K

50 preschool classrooms in 26 schools from 8 different states105 kindergarten classrooms in 42 schools



13

Effectiveness of Preschool and Kindergarten in Promoting and Maintaining Early Literacy Skills

Poster Presented at the National Research Conference in Early Childhood

Arlington, VA

Ruth KaminskiRoland H. Good

Mary AbbottRachael LatimerAmy Warnock

14

Research Questions

1. What is the linkage between children’s early literacy skills at the end of preschool and their performance on early literacy skills at the beginning of kindergarten?

2. Are there programmatic differences in growth of early literacy skills during preschool?

3. Are there programmatic differences in the degree to which kindergarten programs (i.e., kindergarten context) maintain and increase early literacy skills acquired in preschool?



15

PCS PCS PCS DCS DCS DCS.74 .80 .61 .74 .81(.71*)

PA PA PA FSF FSF.57 .64 .57 .51

PSF PSF.54

.52 .60

Phonological Awareness Linkages

Alphabet Knowledge Linkages

AK AK AK LNF LNF LNF.70 .83 .65 .75 .75

Composite Score Linkages

1. Linkage of Skills Preschool to Kindergarten

---------------PELI--------------- -----------DIBELS Next---------BOY MOY EOY BOY MOY EOY

16

Differences in Growth

Individual Slope of Progress

Classroom Mean Slope of Progress(Based on 35 PK45 classrooms with at least 10 students)



17

Kindergarten HLM ModelLevel 1, individual student progress model

Level 2, between student model of progress

RLexileti= π0i+ π1i*TOYti+ π2i*TOY2ti + eti

π0i= β00+ β01*PK45INTi + β02*PK45SLPi + r0i

π1i= β10+ β11*PATH345Ci+ r1i

π2i= β20+ β21*PATH345Ci + r2i

Where:PK45INTi = Preschool intercept (initial skills) for student i

from the preschool model of growthPK45SLPi = Preschool slope (rate of progress) for student i

from the preschool model of growthPATH345Ci = Proportion of kindergarten context students

making at least typical progress on DIBELS Next.

Where:RLexileti = Research Lexile for student i at time tTOYti = time of year (1 = beginning, 2 = middle, 3 = end)

18

Figure 3. HLM Model 1: Suppose a child enters preschool with average skills, makes average growth in preschool, and makes average growth in kindergarten

Res

earc

h Le

xile

Beginning of Year Middle of Year End of Year

-1,400

-1,200

-1,000

-800

-600

-400

-200

0

Above BenchmarkBenchmarkCut Point for Risk

DIBELS Next benchmark goals

Kindergarten Progress



19

Figure 3. HLM Model 1: Suppose a child enters preschool with average skills, makes average growth in preschool, and makes average growth in kindergarten

Res

earc

h Le

xile


-1,400

-1,200

-1,000

-800

-600

-400

-200

0

+ 1 SDMean context- 1 SD



Kindergarten classroom context

Average Kindergarten Progress

20

Figure 4. HLM Model 2: Suppose a child enters preschool with low skills, makes high growth in preschool, and makes low growth in kindergarten.

-1,400

-1,200

-1,000

-800

-600

-400

-200

0

Res

earc

h Le

xile






Dissipating Kindergarten Progress



21

Figure 5. HLM Model 3: Suppose a child enters preschool with low skills, makes low growth in preschool, and makes high growth in kindergarten.

-1,600

-1,400

-1,200

-1,000

-800

-600

-400

-200

0R

esea

rch

Lexi

le






Compensatory Kindergarten Progress

22

Figure 6. HLM Model 4: Suppose a child enters preschool with low skills, makes high growth in preschool, and makes high growth in kindergarten.

-1,400

-1,200

-1,000

-800

-600

-400

-200

0

200

Res

earc

h Le

xile






Complementary Kindergarten Progress



23

How to Sustain Preschool Effects

23

Hypotheses proposed:Focused kindergarten curricula teaching advanced contentAlignment of instructional content between preschool and kindergartenOnsite support/coaching for teachersSummer schoolFamily involvement

Entwisle, Alexander, & Olson, 2001; Galindo & Sheldon, 2012; Yoshikowa et al., 2016

24

Thinking is RequiredOne Size Does Not Fit All



25

Formative Assessment

25

John Hattie evaluated more than 800 meta-analyses of 138 influences on student achievement: • Student• Teacher• Teaching• Curricula• School• Home

26

Formative Evaluation and Progress Monitoring, d = 0.90

26

-0.2

-0.1

0.0

0.10.2

0.3 0.4 0.5 0.60.7

0.8

0.91.0

1.11.2

High

Medium

Low

Neg

ativ

e

Formative Evaluation, d = 0.90

Zone of desired effects

Teacher effects

Developmental effects

Reverse effects

The 3rd largest effect on student achievement out of 138 possible influences.



27

How to Sustain Preschool Effects

How about if we do formative assessment and progress monitoring?

28

Notion of Measures as Indicators

Teachers have to understand the skillsWhat they areHow they developHow to teach them



29

What Are DIBELS®?

D

98.6

I

B E L S

ynamic

ndicators

asic iteracyarly killsof

30

Understand Bench Mark Goals

If the student achieves a score

The odds of achieving subsequent early literacy

goals are:

The likely need for support to achieve subsequent early literacy goals is:

at or above the benchmark goal __________ ____________

below the benchmark goal and at or above the cut point for risk

__________ _______________

below the cut point for risk __________ _______________

80% – 90%

40% – 60%

10% – 20%

Core Support

Strategic Support

Intensive Support

What they areWhat they are notHow to use them



31

Benchmark Status What it Means:

Design Specifications

Likelihood of Meeting Later

Reading Goals

At or Above

Benchmark

Below Benchmark

Well Below

Benchmark

Overall likelihood of achieving subsequent early literacy/reading

goals is 80% to 90%


goals is 40% to 60%


goals is 10% to 20%

> 90%

90%

80%

70%

60%

50%

40%

30%

20%

10%

< 10%

There are Shades of Red, Yellow, and Green

80% to 90%

40% to 60%

10% to 20%

32

Just weighing the pig doesn’t make itfatter: Use the data



3333

34

Testing Environment

Thank You!

For more information:[email protected]

[email protected]

mailto:[email protected]

prevention: the promise and the power · 2020-07-10 · keynote presentation dmg superinstitute las...

Documents