Sources of Reading Comprehension Difficulties among Adolescent Spanish-speaking Language Minority Learners

Nonie K. Lesaux

Jeannette Mancilla-Martinez

Armida Lizárraga

Julie RussThis project effort was supported by Grant Number 5P01HD039530-09 from the Eunice

Kennedy Shriver National Institute of Child Health and Human Development. The content is

solely the responsibility of the authors and does not necessarily represent the official views of

the Eunice Kennedy Shriver National Institute of Child Health and Human Development or the

National Institutes of Health.

2

English Language Learners (ELLs)

U.S-born ELLs from Spanish-speaking homes represent the largest and

fastest growing segment of the ELL school-aged population

(Fry & Gonzales, 2008; Planty et al., 2009)

Account for a disproportionate percentage of struggling readers

(August & Shanahan, 2006)

Yet, little is known about their language and literacy development,

especially for learners who are past the primary grades

longitudinal study needed (L1, L2)

few models of L2 comprehension

e.g., Droop & Verhoeven, 2003; Hoover & Gough, 1990; Proctor et al., 2005, 2006

Study Goals

1) To improve our understanding of developmental trajectories of vocabulary & reading of Spanish-speaking ELLs

2) To examine the influence of social , cultural , and linguistic factors on developmental trajectories of reading & vocabulary

3) To gain insight into the source of difficulties of ELLs who are struggling readers

3

Research Design

4

Wave 1 Wave 2 Wave 3 Wave 4 Wave 5

Fall of Preschool

(2001)

Spring of

Preschool

(2002)

Kindergarten

(2003)

1st Grade

(2004)

2nd Grade

(2005)

Age 4.5 Age 5 Age 6 Age 7 Age 8

Wave 6 Wave 7 Wave 8 Wave 9

5th Grade

(2008)

6th Grade

(2009)

7th Grade

(2010)

8th Grade

(2011)

Age 11 Age 12 Age 13 Age 14

Early Childhood Phase

Follow-Up Phase

Participants

387 children recruited from Head Start and public

preschool programs at age 4.5

Spanish-speaking, predominantly immigrant parents (89%)

75% living in or near poverty

Predominantly U.S.-born children of immigrants (89%)

173 children recruited for follow-up at age 11 (5th grade)

Enrolled in 75 public schools (English instruction)

20 had been retained a year

188 children in follow-up sample at age 13 (7th grade)

Enrolled in 73 public schools (English instruction)

5

School Characteristics

6

95 % Public schools81% Title I65% Low-income80% Minority students51% Needs improvement/warning/failing state tests

58% Latino

Measures

Home Language Use survey (parent report)

Productive Vocabulary (WLPB-R, English & Spanish)

Letter-Word Identification (WLPB-R, English & Spanish)

Reading Comprehension

Passage Comprehension (WLPB-R, English)

TORC-3 Syntactic Similarities

Gates MacGinitie Reading Comprehension

Three Sets of Findings

1. Developmental Trajectories of Vocabulary & Word Reading

2. The Influence of Early Home Language Use on Later Outcomes

3. Modeling Reading Comprehension Full sample using SEM

Mixed-Methods study with a subsample

8

1. Normative Growth

Patterns of development of students’ Spanish and English word

reading and oral language skills from age 4.5 to 11

WLPB-R Productive Vocabulary, English & Spanish

WLPB-R Word Reading, English & Spanish

Rates of growth as compared to national norms in each language

Individual Growth Modeling (IGM) using the multilevel model

for change (Singer & Willett, 2003), with age in months used to

index time

9

320

340

360

380

400

420

440

460

480

500

520L

ette

r-W

ord

ID

WL

PB

-R W

Sco

res

Age 4.5 Age 5 Age 6 Age 7 Age 8 Age 11

(fall preK) (spring preK) (kinder) (1st grade) (2nd grade) (5th grade)

National Monolingual Norms

English

Spanish

English ES: 0.5 0.1 -0.2 -0.2 0.1 0.04

Spanish ES: 0.8 0.7 0.9 1.3 1.4 1.1

Absolute Rate of Growth_National: 135 W-score Points

Absolute Rate of Growth_English: 145 W-score Points

Absolute Rate of Growth_Spanish: 124 W-score Points

Mancilla-Martinez & Lesaux, in press

320

340

360

380

400

420

440

460

480

500

520P

roduct

ive

Voca

bula

ry W

LP

B-R

W S

core

s

Age 4.5 Age 5 Age 6 Age 7 Age 8 Age 11

(fall preK) (spring preK) (kinder) (1st grade) (2nd grade) (5th grade)

National Monolingual Norms

English Spanish

English ES: 1.8 1.8 1.6 1.3 0.9 1.0

Spanish ES: 2.3 2.6 2.9 3.0 2.6 3.4

Absolute Rate of Growth_National: 45 W-score Points

Absolute Rate of Growth_English: 60 W-score Points

Absolute Rate of Growth_Spanish: 34 W-score Points

Mancilla-Martinez & Lesaux, in press

2. Early Home Language Use & Later Vocabulary Development

Research Questions To what extent does Spanish-speaking LM learners’ Spanish and

English vocabulary development from age 4.5 to 12 vary by early patterns of home language use?

Do rates of vocabulary growth in Spanish and English vary by early patterns of home language use (i.e., Mostly Spanish, Equal Amounts, and Mostly English) and how do these rates compare to national norms in each language?

Home Language Classification

Because the correlation between overall language exposure and overall language use was significant and strong (r = .7, p<.001), we combined it into a single language variable Mostly Spanish = 39% (n = 70)

Equal Amounts = 46% (n = 83)

Mostly English = 15% (n = 27)

Note that this is a time-invariant classification refers to children’s home language use patterns at age 4.5

children who were identified as Mostly English were the reference group

Key Findings

Absolute vocabulary differences Children from homes where English was used much more than

Spanish had higher levels of English vocabulary + lower levels of Spanish vocabulary, initially (age 4.5) and through age 12 (and vice versa)

Persistent, widening gaps in Spanish vocabulary development

A more complicated story related to English patterns of vocabulary growth Children from homes where English was used much more than

Spanish at age 4.5 had higher levels of English vocabulary

But… the two other language groups experienced significantly higher rates of English vocabulary growth and lower rates of Spanish deceleration

Mancilla-Martinez & Lesaux, in press

330

350

370

390

410

430

450

470

490

510

530

Engli

sh P

roduct

ive

Voca

bula

ry W

LP

B-R

W S

core

s

Age 4.5 Age 8 Age 12

(fall of preschool) (spring of 2nd grade) (spring of 6th grade)

National Monolingual Norms

Mostly English Group

Equal Amounts Group

Mostly Spanish Group

English Vocabulary

330

350

370

390

410

430

450

470

490

510

530

Age 4.5 Age 8 Age 12

(fall of preschool) (spring of 2nd grade) (spring of 6th grade)

Span

ish P

roduct

ive

Vo

cabula

ry W

LP

B-R

W S

core

s

National Monolingual Norms

Mostly English Group

Equal Amounts Group

Mostly Spanish Group

Spanish Vocabulary

3. Modeling Reading Comprehension

Spanish and English vocabulary and word reading initial status at age 4.5 and

growth rates from age 4.5 to 11 as predictors of English reading

comprehension at age 11

WLPB-R Productive Vocabulary, English & Spanish

WLPB-R Word Reading, English & Spanish

English Reading Comprehension

WLPB-R Passage Comprehension

TORC-3 Syntactic Similarities

Gates MacGinitie Reading Comprehension

Longitudinal Structural Equation Models (SEMs) of latent growth

curves using Mplus (Muthén & Muthén, 2006)

17

Key Findings

Students’ English reading comprehension performance

is at a 2nd grade level

Spanish vocabulary and word reading skills do not

contribute unique variance to the model once English

skills are accounted for

Word reading achievement is a stronger predictor of

reading comprehension, compared to vocabulary

18

19

WLPB-R

Passage

Comp

TORC-3

Syntactic

Similarities

Gates

MacGinitie

Reading

Comp

English

Reading

Comprehensione

e

e

.79

.62

.69

20

WLPB-R

Passage

Comp

TORC-3

Syntactic

Similarities

Gates

MacGinitie

Reading

Comp

English

Reading

Comprehension

ev1

Eng Vocab

Initial Status

ev2 ev3 ev4 ev5 ev6

Eng Vocab

Initial Rate of

Change

Eng Vocab

Curvature

ew1 ew2 ew3 ew4 ew5 ew6

Eng Word

Rd Initial

Status

Eng Word

Rd Initial Rate

of Change

Eng Word Rd

Curvature

e e e e e e

e eeeee

e

e

e

1.0 1.0 1.0 1.01.0 1.0

0.00.5 1.5

2.5 3.5 6.50.0 0.252.25 6.25 12.25 42.25

1.0

1.0 1.0 1.01.0 1.0

0.0 0.5 1.5

2.5

3.56.5

0.00.25 2.25 6.25

12.25 42.25

.79

.62

.69

21

WLPB-R

Passage

Comp

TORC-3

Syntactic

Similarities

Gates

MacGinitie

Reading

Comp

English

Reading

Comprehension

ev1

Eng Vocab

Initial Status

ev2 ev3 ev4 ev5 ev6

Eng Vocab

Initial Rate of

Change

Eng Vocab

Curvature

ew1 ew2 ew3 ew4 ew5 ew6

Eng Word

Rd Initial

Status

Eng Word

Rd Initial Rate

of Change

Eng Word Rd

Curvature

e e e e e e

e eeeee

e

e

e

.79

.62

.69

-.82

-.51

-.48

-.33

.66 .37

.33.42

.68

1.0 1.0 1.0 1.01.0 1.0

0.00.5 1.5

2.5 3.5 6.50.0 0.252.25 6.25 12.25 42.25

1.0

1.0 1.0 1.01.0 1.0

0.0 0.5 1.5

2.5

3.56.5

0.00.25 2.25 6.25

12.25 42.25

R2 = .88

χ2(94,N=173)=462.81,p=0.00, RMSEA=.15, CFI=.76

.97

21

1 SD below ½ SD below average ½ SD above 1 SD above

Late

nt

Re

ad

ing

Co

mp

reh

en

sio

n S

co

re

Average Fitted Score1.6

4 S

D

5th Grade National Norms

58

Digging Deeper into Comprehension Scores

Investigated the processes underlying the global

reading scores of 41 students scoring below the 35th

percentile (51% male; M age = 12 years, 11 months;

1 Expository passage from GMRT alternate form and

corresponding multiple choice questions

Think-aloud protocol (or “semi-structured clinical

interview”) following the completion of the multiple

choice questions

(Russ & Lesaux, under review)23

Results

They are “active” comprehenders Report strategy use and have clearly reasoned processes for finding

information.

Primary sources of difficulty are in domains of

language skills and knowledge relevant to the passage Tended to construct inappropriate and/or inaccurate representations of

the text.

Students appeared to over-rely on strategies Tended to construct inappropriate and/or inaccurate representations of

the text

(Russ & Lesaux, under review)

Overall Project: Implications to Date

Need for increased and sustained attention to

promoting this population’s language development Word reading-word knowledge gap in students’ developmental

trajectories

Early Spanish home language use does not appear to

bear on overall English outcomes, and (of course)

contributes to Spanish maintenance

Theoretical models of comprehension should account

for students’ skill profiles and abilities

Instructional attention to the role of vocabulary and

background knowledge in effective strategy use25

2626

Acknowledgments

Christina Kelley Emma Billard Rachel Slama Michelle Hastings Laura Salinas Sylvia de La Torre Spencer, Montgomery County Public Schools Jason Sachs, Boston Public Schools Lynn Catarius Petetit, Lawrence Public Schools Susan McGilvray-Rivet, Framingham Public Schools

Presentation Link: http://www.cal.org/vias/subproject1/index.html