READING REMEDIATION IN THE MIDDLE SCHOOL

1
READING REMEDIATION IN THE MIDDLE SCHOOL
2

Dr. Geri Marshall Mohler Assistant Professor
Reading/Literacy

• California State University Bakersfield
• gmohler_at_csub.edu

3
What is the purpose of the research?
4
This study explored the effects of a program of
direct instruction in

• phonemic awareness and spelling,
• multisensory phonics,
• and fluency
• on the basic reading skills of 25 low-ability
  seventh grade students in a Midwest middle
  school.

5
What were the research questions asked?
6
Research Questions

• 1. What changes in overall scores occur for
• total reading (TRS),
• word recognition (WRC),
• phonemic awareness (CLS),
• spelling (WSC) and
• reading fluency (ORF)
• when 7th grade low readers receive
  instruction in phonemic awareness, multisensory
  phonics, and fluency?

7
Research Questions

• 2. Can the analysis of these scores be used to
  understand how subgroups (gender, special needs
  and initial reading level) and individuals differ
  in their response to that treatment?

8
What was the theory claimed?
9
Theory

• 1. PA, spelling, and fluency are developmental
  and receptive to remediation in older students
  and are useful measures to demonstrate reading
  growth.
• 2. Individual scores will vary widely from
  group averages and are the most valuable data in
  the analysis.

10
Who were the researched?
11
Who?

• 25 seventh grade students reading 4 or more
  years below grade level.
• 12 boys 13 girls
• 18 RSP (resource) 7 ELL
• 4 1st grade level
• 7 2nd grade level
• 14 3rd grade level

12
Who provided the intervention?
13

• In 2 Classrooms 43 mins/day
• 2 Reading Teachers
• 2 RSP (resource) Teachers
• 1 ELL Teacher
• 1 ELL Interpreter
• 1 Student Teacher
• 1 Parent Volunteer (part-time)
• 1 Student Body Guard

14
What instruments were used and how?
15
Instruments

• Total Reading (TRS)
• Stanford Diagnostic Reading Test
• Word Recognition (WRC)
• Slosson Oral Reading Test
• Words Spelled Correctly(WSC)
• Correct Letter Sequence (CLS)
• CBM Spelling w/ Morrison McCall Words
• Oral Reading Fluency (ORF)
• CBM Fluency

16
What was the research design?
17
Research Design

• 9 months of instruction.
• One-half of group in 1st and 2nd grade reading
• level.
• One-half of group in low 3rd grade reading
  level.
• Students were on 2 different middle level teams.
  
• Team teaching with the teacher participants
  included whole group, small group, and individual
  instruction.

18
What was the intervention?
19
The Intervention

• Phonemic Awareness
• Spalding Phonics Program w/ phonograms and
  spelling
• Reading of independent and instructional level
  texts appropriate for 7th grade readers
• Repeated Reading

20
What was reported?
21

• All assessments were analyzed for pre- and
  post-test differences.
• WSC, CLS, and ORF scores were recorded multiple
  times and analyzed using regression analyses for
  rate of growth for individuals and subgroups in
  order to determine effect size of the
  intervention.

22
What were the overall results of the
intervention?
23
Table 1. Pre- Post test Comparisons Using
Repeated Measures T-Tests for TRS, WRC, WSC, CLS,
and ORF Reading Measures Measurement n
Pre-Mean Post-Mean Std. Dev.
t-value Sig. TRS 17
17.76 24.97 7.8832 3.769 .002 WRC
25 75.08 104.56
21.8250 6.754 .000 WSC 25
15.00 19.78 5.265 4.356
.000 CLS 25 61.50
69.33 14.153 2.711 .012 ORF
25 73.64 85.18 11.287
4.212 .001 ____________________________
____________________________________ p , lt .05
24
What is effect size?
25
www.uccs.edu/lbecker/psy590/escalc3.htmmeans
and standard deviations

• Effect size (ES) is a name given to a family of
  indices that measure the magnitude of a treatment
  effect. Unlike significance tests, these indices
  are independent of sample size. At least 5
  measures should be recorded before performing
  regression analyses. Then the df and SD can be
  used to establish effect size at the website
  above
• 2.0 is two standard deviations from 0 and shows
  significant
• According to Cohen (1988), Effect Size,
  calculated with the t-value and the df of a
  measurement analysis, can be classified as
• .8 strong, .5 moderate, .2 small

26

• Out of the 75 scores using regression analysis
  and significance testing, only 11 individuals had
  a significant effect.
• I should therefore conclude that the intervention
  was not successful even though the average growth
  was significant.

27
75 Individual Scores 24 Subgroup Scores on
CLS, WSC, and ORF

• Gender
• Male Female
• Special Needs
• RSP DLP ELL
• Initial Reading Level
• 1st 2nd 3rd

28
What was the effect size for the subgroups?
29

• Table 2
• Mean Effect Size for CLS, WSC, and ORF for
  the Subgroups Gender, Special Need (SN), and
  Initial Reading Level (IRL)
• CLS WSC ORF
  
  Gender
• Male .51 3.27 .22
• Female 2.87 4.02 .83
• SN
• RSP .66 1.65 .31
• DLP .83 .48 2.31
• ELL 2.87 5.37 4.13
• .8 strong effect, .5 moderate effect, .2
  small effect.

30

• Mean Effect Size for CLS, WSC, and ORF for
  the Subgroups Gender, Special Need (SN), and
  Initial Reading Level (IRL)
• IRL Group
• IRL1 1.92 2.43 2.26
• IRL2 .51 1.48 .38
• IRL3 .72 2.58
  1.07
• .8 strong effect, .5 moderate effect, .2
  small effect.

31
What was the effect size for individuals?
32

• Gender SN IRL CLS WSC ORF
  
  
  
• 1 M DLP 2 .54 5.40 .78
• 2 F ELL 1 5.89 1.80 1.43
• 3 F ELL 2 1.00 1.88 .62
• 4 M RSP 2 2.34 .35 4.14
• 5 M RSP 3 1.01 .77 1.62
• 6 F ELL 1 2.63 5.23 .47
• 7 M RSP 2 .52 .17 .57
• 8 F ELL 3 .54 1.35 .52
• 9 F RSP 2 .53 1.01 .64
• 10 F RSP 3 .10 .35 1.72

33

• Gender SN IRL CLS WSC
  ORF
• 11 M RSP 1 .82 2.11 2.20
• 12 M RSP 1 .34 .21 .00
• 13 F DLP 2 .23 .66 .47
• 14 M RSP 2 1.31 .39 2.16
• 15 F RSP 3 1.74 1.49 .03
• 16 F RSP 3 .87 .47 .78
• 17 M ELL 3 .78 1.17 .89
• 18 M RSP 3 1.47
  .98 .13
• 19 F ELL 3 1.01 1.08 2.58
• 20 F RSP 3 .79 .08 .36
• 21 M ELL 3 .97 .69 1.57
• 22 M ELL 3 .93 2.79 4.16
  
  
  
• 23 F RSP 3 .20 1.32 1.24
• 24 F RSP 3 .29 .44 1.75
• 25 M RSP 3 .68 .80 .03

34
However--

• Using regression analyses on the three
  measures
• Out of 75 possible scores for individuals, only
  11 were significant but 39 were considered strong
  effects
• Out of 24 measures in the 8 subgroups, only 10
  were significant but 16 were considered strong
  effects
• Only 3 individuals did not have at least one
  variable with a strong effect

35
Value of Effect Size

• A closer look at both subgroup effect sizes
  and individual effect sizes makes it possible to
  determine who responded, who did not respond, and
  to compare individual responses to subgroup
  effect sizes to determine how any one individual
  compared to like individuals receiving the same
  intervention treatment.

36
What was the range of scores for the individuals?

37
What was concluded?
38
Conclusions

• Back to basics instruction appears to be
  effective even when students are in middle
  school.
• Girls, overall, did better than boys.
• ELL, overall, did better than RSP.
• 1st grade reading level students had greater
  gains.
• Standardized and informal measures are both
  important.
• Significant effects are not likely when children
  are RSP or ELL, so effect size is a better
  indicator of whether the treatment was effective
  for subgroups and individuals.

39
READING REMEDIATION IN THE MIDDLE SCHOOL
Dr. Geri Marshall MohlerAssistant Professor
Reading/Literacy California State University
Bakersfieldgmohler_at_csub.edu