Article summary

ROAD TO THE CODE: EXAMINING THE NECESSITY AND SUFFICIENCY OF PROGRAM COMPONENTS

D r . Stephanie L. Schmitz University o f Northern Iowa

Sedona L oy Heartland Area Education Agency

As the ability to read proficiently is essen­ tial for success both in and out of the school setting, literacy has become an area o f partic­ ular focus in today’s classrooms. While recent assessments indicate that students are making progress in the area o f reading (e.g., National Assessment o f Educational Progress [NAEP], 2011), there continues to be many children who encounter difficulty when learning to read (Moats 1999). Some children enter kin­ dergarten with few early literacy skills, which can result in significant difficulty catching up to typically developing peers without some type o f intensive intervention (Good, Simmons, & Smith, 1998; Moats, 1999), including identification for special education services (Whitehurst & Lonigan, 2001). Such facts indicate the need for early prevention and intervention efforts in reading.

Phonological Awareness Early literacy skills are discrete skills that

provide a foundation for fluent reading (Nev- ills & Wolfe, 2009; Daly, Chafouleas, & Skin­ ner, 2005). Examples o f early literacy skills include letter naming, knowledge o f books, concepts about print, rhyming, alliteration, and identifying commons signs and symbols from the environment. Another important ear­ ly literacy skill, and one o f the earliest to de­ velop due to its connection to oral language,

is phonological awareness. Phonological awareness is described as an awareness o f both the various sound units o f speech and that they are separate from their meaning (Snow, Bums, & Griffin, 1998), as well as the ability to manipulate the sound units (Nelson, Lindstrom, Lindstrom, & Denis, 2013). Phonological awareness can be best described as a continuum o f skills (Heroman & Jones, 2004). Specifically, early phonolog­ ical awareness can first be observed through rhyming and alliteration around 2 to 3 years of age, followed by an ability to segment words by syllables and then by onsets and rimes. The most complex end o f this continuum is represented by phonemic awareness, which entails recognizing and identifying individual sounds in a word, and typically begins to de­ velop during the kindergarten year (Heroman & Jones, 2004; Ranweiler, 2004).

Phonological Awareness Instruction and Training

Phonological awareness has been found to have a significant and causal influence on the development o f early literacy skills (Hulme, Bowyer-Crane, Carroll, Duff, & Snowling, 2012), as well as a significant, causal, and bidirectional relationship with future reading success (Gillon, 2004; Snow et al., 1998). However, phonological awareness is not

341

3 4 2 / R ead in g Im p ro vem en t

always easy for children to develop, espe­ cially at the phoneme level. This difficulty is due to both the blending o f individual sounds in oral language (Gillon, 2004; Ranweiler, 2004), as well as the lack o f a true one-to-one correspondence between letters and sounds in the English language (Nevills & Wolfe, 2009; Snow et al., 1998). Such Information rein­ forces the importance o f developing a young child’s phonological awareness through ex­ plicit instruction or training. Researchers have found that providing young children with training in phonological awareness resulted in their learning to read more quickly than chil­ dren that did not receive such training (Bus & van IJzendoom, 1999; Snow et al., 1998), and they frequently maintained their early success in reading for several years (Lundberg, Frost, & Peterson, 1988; Schneider, Ennemoser, Roth, & Kuspert, 1999).

Researchers have asserted that phonolog­ ical awareness training should be provided at the phoneme level once children enter kin­ dergarten for several reasons (Gillon, 2004; Smith et al., 1998). First, o f all the skills on the phonological awareness continuum, pho­ nemic awareness has been found to be the strongest predictor o f future reading achieve­ ment (Gillon, 2004; Perez, 2008). Further, phonemic awareness is not easily developed without some type o f explicit instruction or training (Nevills & Wolfe, 2009), especially for those children experiencing difficulty with early literacy skills (National Reading Panel [NRP], 2000; National Institute o f Literacy [NIL], 2001).

While there appears to be a consensus in the literature with regards to the impor­ tance o f phonological awareness, there does not seem to be agreement on how to most effectively teach it. However, researchers have documented some features that con­ tribute to the effectiveness o f a phonologi­ cal awareness program (Good et al., 1998). Specifically, previous researchers have

found that increasing the level o f difficulty in phonological awareness tasks (i.e., tasks that progress from rhyming and alliteration to segmentation at the syllable, onset and rime, and phoneme levels) throughout the program (e.g., Good et al., 1998; Schuele & Boudreau, 2008) and integrating instruction in letter-sound correspondence (e.g., Bus & van IJzendoom, 1999; Good et al., 1998; NRP, 2000; Snow et al., 1998) have result­ ed in larger overall student gains following completion o f a phonological awareness program. Further, numerous researchers have indicated phoneme segmentation and blending to be the most important and ef­ fective phonological awareness skills due to their correlation with future reading success (NRP, 2000; Schuele & Boudreau, 2008).

Phonological Awareness Intervention and Programs

Due to the important role o f phonological awareness in the reading process, a number o f interventions and programs have been cre­ ated to develop and strengthen young chil­ dren’s phonological awareness skills (NRP, 2000; Santi, Manchetti, & Edwards, 2004). While many o f these programs are described as being research-based (NRP, 2000), few studies examining the effectiveness o f these programs were found in a review o f the liter­ ature. Several o f the phonological awareness programs have been described as including components and/or activities that have been found to be effective contributors to the ef­ fectiveness o f phonological awareness pro­ grams (Santi et al., 2004).

The program used in this study, Road to the Code (Blachman, Ball, Black, & Tangel, 2000), is described as being based upon scien­ tifically-based research and to include several o f the components that were identified in the literature as being effective contributors to the development o f phonological awareness. Road to the Code is a structured, phonological

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 343

awareness program that can be implemented with kindergarten or first-grade students who are demonstrating difficulty with early litera­ cy skills. While no studies specifically inves­ tigating the effectiveness o f the Road to the Code program were found in a review o f the literature prior to this study, the program au­ thors conducted several studies prior to devel­ opment o f the program that included similar tasks and procedures to those included in the current program. Results from these studies indicated that participating children made improvements in phonological awareness and outperformed children who did not receive such training (Ball & Blachman, 1991; Blach- man, Tangel, Ball, Black, & McGraw, 1999).

It is generally unknown how these pro­ grams, including Road to the Code, deter­ mined which and how many o f such compo­ nents to include, as studies investigating the necessity and sufficiency o f individual com­ ponents incorporated in such programs were not located in a literature review. While not examining a specific phonological awareness program, one recent study investigated both the individual effects, as well as any syner­ gistic effects, o f both code-focused (i.e. pho­ nological awareness, letter knowledge) and meaning-focused (i.e., shared and dialogic reading) interventions for 318 preschool chil­ dren (Lonigan, Purpura, Wilson, Walker, & Clancy-Menchetti, 2013). The children were randomly assigned to five groups, with each group receiving a different combination of these interventions four to five times per week in a small group format throughout the entire school year. A fifth group, receiving only the classroom curriculum, served as the control.

Results indicated that all four groups receiving the interventions demonstrated progress with their early literacy skills during the year, especially when compared with the control group (Lonigan et al., 2013). The in­ tervention effects were found to be specific, meaning that growth was observed only in

the skill(s) that the intervention(s) targeted; however, some generalization o f content was observed as the outcome measures assessed only broad skills rather than being specifi­ cally aligned with the intervention content. Interestingly, the authors found no synergistic effects when combining the interventions, indicating that no larger effects were found on the targeted skills as a result o f adding an additional intervention(s). Lonigan et al. (2013) assert that results from this study re­ inforce previous research that has suggested early literacy skills are domain specific and interventions that target several domains are necessary in order to impact multiple early lit­ eracy skills (e.g., Bowyer-Crane et al., 2008).

Purpose The purpose o f this study was to both repli­

cate results from a dissertation study (Schmitz, 2011) that evaluated the effectiveness o f the Road to the Code program (Blachman et al, 2000) and to extend it by determining the ne­ cessity and sufficiency o f the program com­ ponents. Three hypotheses were generated for this current study based upon both the results from the dissertation study, as well as from information gathered through a recent litera­ ture review. Results o f the dissertation study indicated that most students made gains on measured skills throughout the intervention phase that carried into the maintenance phase, but did not provide any information regarding the effectiveness o f the individual compo­ nents included in each lesson o f the program. Therefore, it was hypothesized that students receiving instruction in either a singular com­ ponent of, or through the entire Road to the Code program, would improve their phono­ logical awareness skills. A second hypothesis, based upon recent research examining the ef­ fects o f intervention components on students’ phonological awareness skills (i.e., Lonigan et al., 2013; Bowyer-Crane et al., 2008), was that students receiving instruction in one of

344 / Reading Improvement

the program components would show the largest gains on the skill(s) targeted by that particular component, and these gains would be greater than those of the students receiving instruction on the other individual, program components. Third and finally, it was hypoth­ esized that those children participating in the full Road to the Code program would demon­ strate comparable or greater growth with all measured phonological awareness skills when compared with the other groups due to their consistent practice with all components targeted by this program.

Method

Participants and Setting Twelve students from four kindergarten

classrooms in a local, Midwestern, and richly diverse elementary school were initially iden­ tified by their classroom teachers as having difficulty with phonological awareness skills. One student moved following the first few lessons o f the Road to the Code program. O f the 11 remaining students, Randy, Debbie, and Lenny were Hispanic, Keith, Rachel, and Tanya were African-American, and Katie, Linda, Jack, Lizzy, and Jordan were Cauca­ sian. All students were five- and six-years of age, with a mean age o f 5 years, 8 months. In addition, six o f the students were female and five were male. Pseudonyms are used for all students to ensure confidentiality.

All o f the referred students were concur­ rently receiving additional reading assistance outside o f the classroom. Specifically, Randy, Debbie, and Lenny were receiving literacy support as part o f the English Language Learner (ELL) services, with the remaining eight students all receiving the same sup­ ports and services through Title I reading. The school declined to remove these services during this study as the students were consid­ ered to be some o f the most at-risk students, with the exception o f those receiving special education services.

Screening Measures Students were screened using two mea­

sures o f phonological awareness. Each student was administered both the Test o f Phonological Awareness Skills (TOPAS; Newcomer & Barenbaum, 2003) and the Test o f Phonological Awareness, Second Edition Plus (TOPA-2+; Torgesen & Bryant, 2004) to assess current proficiency in various pho­ nological awareness skills. The TOPAS is de­ signed for children, ages 5 years, 0 months to 10 years, 11 months. The TOPAS consists of four subtests, with 5-year-old students com­ pleting only the first two subtests, and stu­ dents six years and older completing all four subtests. The first subtest measures rhyming through the student’s ability to produce a word that rhymes with a designated word that is orally presented by the examiner. The sec­ ond subtest requires the student to identify the missing sound in a word, measuring phoneme segmentation. At six years and older, students also complete a subtest measuring phoneme deletion (i.e., ‘rat’ without the /r/ sounds like ‘at’), as well as a subtest measuring sound se­ quencing by representing the presented series o f sounds using manipulatives.

The TOPAS is reported to produce both reliable and valid scores. Specifically, coeffi­ cient alphas ranged from .91 at 5 years to .97 at 6 to 10 years, levels that are considered to be indicative o f high reliability (Newcomer & Barenbaum, 2003). When correlated with reading composites from various measures, the mean coefficient was .71.

The second screening measure used was the kindergarten version o f the TOPA-2+ (Torgeson & Bryant, 2004). The kindergarten version contains three subtests, with two sub­ tests measuring the ability to identify either the same or different initial sounds in words, and one subtest measuring the ability to iden­ tify the letter that matches an orally presented sound. The TOPA-2+ has also been found to produce reliable and valid scores (Torgesen &

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 345

Bryant, 2004). A measure o f internal consis­ tency indicated that the kindergarten version o f the test produced an average reliability of .91 on the two phonological awareness sub­ tests and .89 on the letter sequencing subtest. With regards to test-retest reliability, the pho­ nological awareness and letter identification subtests on the kindergarten version had a reliability o f .87 and .85, respectively. Stud­ ies o f concurrent validity indicated a strong correlation with other measures o f phonolog­ ical awareness.

Authors o f both the TOPAS (Newcom­ er & Barenbaum, 2003) and the TOPA-2+ (Torgesen & Bryant, 2004) state that the pri­ mary use o f each instrument is to identify chil­ dren with deficits in phonological awareness, and that a standard score falling below the 25th percentile indicates a need for additional assistance. As each instrument measures a different combination o f phonological aware­ ness skills, not only was the student’s overall performance on both measures considered when determining eligibility for this study, but so was his/her performance on individual subtests and scales within each instrument. Specifically, while a student might achieve an overall score above the 25th percentile on one o f the instruments, analyzing his/her performance on individual subtests and scales may indicate a deficit in one or more specific phonological awareness skills. Further, pre­ test performance on the two DIBELS probes discussed in the next section, was also taken into consideration.

Dependent Variables The Pre-Reading Inventory o f Phono­

logical Awareness (PIPA; Dodd, Crosbie, McIntosh, Teitzel, & Ozanne, 2000) and the Phoneme Segmentation Fluency (PSF) and Nonsense Word Fluency (NWF) subtests from the Dynamic Indicators o f Basic Early Liter­ acy Skills, Next (DIBELS Next; Kaminski & Good, 2010) were used as pre- and post-test

measures of group performance. The PSF and NWF were also used to monitor student prog­ ress on a weekly basis throughout the study.

The PIPA was chosen as a pre- and post­ test measure for this study due to both the number o f phonological awareness skills included in this instrument, as well as their association with the skills included in the Road to the Code phonological awareness program. The PIPA was developed for chil­ dren, ages 4 years, 0 months to 6 years, 11 months, and designed to identify kindergarten and first-grade students who have phonologi­ cal awareness deficits (Dodd et al., 2000). The PIPA contains six subtests, with each measur­ ing a separate phonological awareness skill, including rhyming, alliteration, syllable and phoneme segmentation, and letter-sound cor­ respondence. Scores in the 0 to 29th percentile indicate Emerging or Below Basic skill in phonological awareness, scores in the 3 04 to the 69th percentile indicate Basic skill, and scores above the 70th percentile suggest Pro­ ficient skill. According to the PIPA authors, students whose skill in phonological aware­ ness is considered Emerging or Below Basic require some type o f intensive intervention to assist with the acquisition o f those skills

The PIPA has been found to produce both reliable and valid scores. Test-retest reliability coefficients ranged from .67 on Syllable Seg­ mentation to .97 on Letter-Sound Knowledge. For concurrent validity, correlations between the kindergarten age group on the Early Read­ ing Diagnostic Assessment (ERDA) and the PIPA subtest total scores ranged from moder­ ate to high.

The DIBELS Next PSF and NWF probes are brief, one-minute probes that are often used to progress monitor student perfor­ mance on early literacy skills. Cut points are provided for all DIBELS subtests, which give information about a student’s level of performance when compared to same-aged peers. As this study was completed during

346 / Reading Improvement

the spring, the end-of-kindergarten cut points were used. A random sampling with replace­ ment procedure, in which three probes were randomly selected from the total pool for each administration, was used for both the PSF and NWF probes, for both pre- and post-testing and for progress monitoring. This procedure accounts for any practice effects as the num­ ber o f needed probes exceeded the number o f available probes.

The NWF probe is an indicator o f the stu­ dent’s knowledge o f the alphabetic principle, and measures phoneme segmentation and blending, as well as pseudoword decoding. On this probe, a student is visually presented with two to three letter nonsense words that are made up o f common patterns o f either consonant-vowel or consonant-vowel-con­ sonant words, and is asked to pronounce the sounds in the word. The ultimate goal for this measure is for the student to blend each sound, or phoneme, to orally produce the word. According to NWF cut points provided for the end o f the kindergarten year, if a score under 15 is obtained, a student is considered to be well below benchmark and to likely require intensive support. A score between 15 and 27 indicates a student is below bench­ mark and will likely require strategic support, and a student is considered to be at or above benchmark and to likely need only core sup­ port when receiving a score above 28.

The PSF probe assesses phoneme segmen­ tation, as students are asked to orally segment a word containing either three or four sounds that was verbally presented by the examiner into its individual sounds, or phonemes. As with the NWF probe, cut points are provided for the PSF measure at the end o f the kinder­ garten year. If a score under 24 is obtained, a student is considered to be well below bench­ mark and to likely require intensive support. A score between 25 and 39 indicates a student is below benchmark and will likely require strategic support, and a student is considered

to be at or above benchmark and to likely need only core support when receiving a score above 40.

For the PSF probe, alternate form reliabil­ ity for kindergarten was .70 when using three probes, according to early research conducted on the DIBELS Next (Dynamic Measurement Group, 2011). Alternative form reliability for the NWF subtest was found to be sufficient for making screening decisions, at .88 for kindergarten when using three probes. Fur­ ther, criterion-related validity was found to be moderate for the PSF subtest and moderate to strong for NWF.

Reading Intervention The Road to the Code phonological aware­

ness program consists o f 44 lessons, with each lesson containing three sections (Blachman et al., 2000). While this program was designed to be implemented over an 11-week period in 15 to 20 minute, small group sessions, the manual does allow for flexible pacing o f the program based upon student understanding o f and performance with included content.

The first section o f the Road to the Code program consists o f a Say-It-and-Move-It (SIMI) activity in which each student is asked to repeat a target sound(s) and to then move a small tile while repeating the sound(s) (Blachman et al., 2000). Letter tiles are intro­ duced to this activity about midway through the program. The second section o f each pro­ gram lesson consists o f an activity reinforcing letter-sound correspondence. Eight letters are explicitly taught as part of the Road to the Code program; specifically, short a, m, t, short i, s, r, b, and f were taught. The third and final section in each lesson of the Road to the Code program includes a general, phonological awareness activity in which one or more pho­ nological skills are emphasized (i.e., rhyming, alliteration, phoneme segmentation, phoneme blending, and letter-sound correspondence).

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 347

Experimental Design and Data Analysis A dismantling design, also known as a

component analysis, was used in this study to examine and compare the efficacy o f Road to the Code program components. Dismantling an intervention into its individual components and then evaluating the effectiveness o f each, or a combination o f each, compared to that of the full program gives information on what component or components were responsible for the behavior change (Nock, Janis, & Wed- ig, 2007).

To analyze the data, T-tests were per­ formed to determine whether there were any differences between groups at the beginning o f the study, and a one-way Analysis o f Vari­ ance (ANOVA) was used to examine any dif­ ferences between baseline and post-test for all three dependent variables. A Tukey HSD post hoc analysis was conducted to look at specific comparisons o f interest. Effect sizes were also calculated. Due to the small number o f stu­ dents participating in this study and the pos­ sible impact of this small sample size on the statistical results, the ongoing performance o f both the individual and groups o f students was analyzed through a visual inspection of the data obtained on the PSF and NWF probes administered throughout the study; specifical­ ly, the trend and variability o f the data was examined. Finally, cut points provided by the DIBELS were also examined.

Procedures Two examiners, one graduate student and

one faculty member in a local school psy­ chology program, conducted all phases o f this study. Both examiners had background in administering assessment instruments to young children and reviewed each instrument prior to this study. Further, as one examiner had previously implemented the Road to the Code program, instruction on the activities included within the program was provided to the second examiner.

Screening. After receiving informed con­ sent for, and assent from, all students, each student was screened to ensure that they were demonstrating difficulty in the area of phonological awareness. While there was some variability in screening scores, 11 stu­ dents were initially found to meet eligibility criteria for participation in this study as then- scores indicated difficulty in at least one of the measured phonological awareness skills on one or across both screening measures. As Jack’s overall scores fell into the Average range across both screening measures, which contradicted teacher reports, his performance on both the PSF and NWF probes that would be used to monitor his progress throughout the study was examined. Jack was considered to be well below end-of-the-year kindergarten cut points on the PSF measure and below cut points on the NWF measure; therefore, he was included in the study. Table 1 includes both the individual screening scores and de­ scriptive categories for each o f the 11 students participating in the study.

Following screening, the students were randomly assigned to four groups o f three. Group 1 (SIMI) completed only the first section o f each lesson o f Road to the Code, the Say It and Move It activity, and was com­ prised of Tanya, Rachel, and Jordan. Group 2 (LS) was made up of Keith and Randy after the third student moved, and completed only the second section o f each lesson that included activities focused on developing letter-sound correspondence for eight letters. Group 3 (PA) completed only the third section o f the program, which focused on improving overall phonological awareness skills, and was comprised o f Lenny, Debbie, and Lizzy. Linda, Katie, and Jack made up Group 4 (FP) and completed all three sections o f each les­ son, or the entire program.

Pretest. Once eligibility was determined for all students, pretesting was completed using the PIPA, as well as the PSF and NWF

3 4 8 / R ead in g Im p ro vem en t

measures from the DIBELS Next assessment. Student performance on the DIBELS mea­ sures served as a baseline from which to com­ pare weekly progress throughout the study, and overall progress during the intervention.

When looking at group differences on the three pre-test measures, results from T-tests indicated that both Group 2 (LS) (t = -7.949, p = .004) and Group 3 (PA) (t = -3.677, p - .021) performed significantly better than Group 1 (SIMI) on the PSF probe. Group 4 (FP) performed significantly better than Group 1 (SIMI) on both the Syllable Seg­ mentation subtest on the PIPA (t = -2.907, p = .044) and the NWF probe from the DIBELS (t = -2.988, p = .040).

Intervention. For this study, the Road to the Code program was implemented for all four groups o f students, four to five times per week over a nine-week period. The program does recommend implementation four times per week over an 11-week period, while allow­ ing for some flexibility in pacing. However, due to the amount o f the time available before

the end o f the school year, this standard pacing was not possible. For those groups completing only one section, or component, o f each les­ son, each session lasted approximately five to 15 minutes. For the group completing all three sections, or components, o f each lesson, each session lasted approximately 10 to 25 minutes. Each student also completed three PSF and three NWF probes on a weekly basis, either be­ fore or after completion o f the program lesson. Each student’s and each group’s performance on both probes was scored and graphed to look at their progress.

A reward program was implemented throughout the duration o f this study. As part o f this program, students in each group could intermittently earn marbles during each les­ son if they remained in their seat and followed directions. Marbles were not ever removed from the jar. Students selected a reward (e.g., bubbles, pencils, erasers) once they reached a predetermined criterion. After selecting their reward, the marbles were emptied and the cycle began again.

Table 1. Student screening scores

Student Group

TOPAS TOPAS TOPA-2+ TOPA-2+ TOPA-2+ TOPA-2+

Percentile Rank

Descriptive Rating

PA PA LS LS

Percentile Rank

Descriptive Rating

Percentile Rank

Descriptive Rating

Tanya 1 9 Below Average 1 Very Poor 14 Below Average

Rachel 1 12 Below Average 5 Poor <1 Very Poor

Jordan 1 27 Average 6 Poor 24 Average

Keith 2 92 Superior 25 Average 16 Below Average

Randy 2 8 Poor 17 Below Average 24 Average

Lenny 3 7 Poor 8 Poor 4 Poor

Debbie 3 8 Poor 6 Poor 75 Average

Lizzy 3 8 Poor 51 Average 75 Average

Linda 4 8 Poor 51 Average 75 Average

Katie 4 58 Average 2 Poor 39 Average

Jack 4 45 Average 37 Average 30 Average

Note. PA = Phonological Awareness subtests; LS = Letter-Sound Knowledge subtests.

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 3 4 9

Treatment Integrity and Interrater Reliability Following completion o f the study, each

examiner reviewed a random sample o f the other’s audiotaped lessons using a checklist containing each step o f the Road to the Code lesson to examine treatment integrity. Thir­ ty-five percent o f the total number o f lessons across all groups were reviewed, with an equal number randomly selected from each group. A review o f the checklists corre­ sponding to the randomly selected lessons indicated the level o f treatment integrity was 91.4%.

Interrater reliability was examined by reviewing the other’s screening and pre- and post-test measures, as well as randomly selected DIBELS probes used for progress monitoring. Interrater reliability was cal­ culated by dividing the smaller number by the larger number and then multiplying by 100 to obtain the final percentage. A total o f 44% o f PSF and NWF probes were randomly selected and reviewed, with an equal number for each group. Interrater agreement for the PSF probes was 89%, and interrater agree­ ment for the NWF probes was 91%.

Results

Descriptive Statistics When comparing mean performance on

all six subtests o f the PIPA, as well as on the PSF and NWF probes, results indicated most individual students and groups to have made gains on most measures. Further, most signif­ icant, group differences found at pre-testing disappeared at post-testing, with the excep­ tion being Group 4 ’s (FP) performance on the Syllable Segmentation o f the PIPA when compared to Group 1 ’s (SIMI). This will be discussed in more detail later in the paper. Table 2 provides group means at both pre-test and post-test for all six PIPA subtests.

A one-way ANOVA was computed, which compared post-test group scores on all six subtests o f the PIPA, as well as on the PSF and the NWF probes. Results from the one­ way ANOVA showed significance on both the Syllable Segmentation subtest on the PIPA (F = 10.106, p = .006) and on the PSF subtest (F = 7.641, p = .013). A Tukey HSD post-hoc comparison was calculated to determine the specific nature o f the significant differences between the group scores. Results indicated that Group 4 (FP) demonstrated signifi­ cantly higher performance on the Syllable

Table 2. Group PIPA Pretest and Posttest Means (total number o f items)

Group RA Pre (12)

RA Post (12)

SySeg Pre (12)

SySeg Post (12)

AA Pre (12)

AA Post (12)

SI Pre (12)

SI Post (12)

SoSeg Pre (12)

SoSeg Post (12)

LS Pre (30)

LS Post (30)

1 (SIMI) 2.67 3.33 2.33 2 2.33 5 5.67 10 .67 1.67 13.3 15.67

2 (LS) 4.5 5.5 7.5 8 3 6.5 5 10.5 1 2.5 9 16.5 3 (PA) 3.33 5 5 3.67 5.33 6 6.33 8.67 2.67 2 13 17 4 (FP) 7.67 6.33 6.67 9.67 5.33 8.67 10 11 .67 2 17 21.67

Note. RA = Rhyme Awareness subtest; SySeg = Syllable Segmentation subtest; AA = Alliteration Aware­ ness subtest; S I = Sound Isolation subtest; SoSeg = Sound Segmentation subtest; L S = Letter-Sound subtest. Each subtest contains 12 items (12).

350 / Reading Improvement

Table 3. Individual Student and Group Means and Cut Points fo r Correct Sounds (CS) at Pre­ test and Post-test on the Phoneme Segmentation Fluency (PSF) Measure

Student Group PSF

Pre-test Score

PSF cut point (pre-test)

PSF Post-test

Score

PSF cut point (Post-test)

PSF Range

Tanya 1 10 Well Below Benchmark 29 Below Benchmark 10-35

Rachel 1 6 Well Below Benchmark 39 Below Benchmark 0-39

Jordan 1 7 Well Below Benchmark 17 Well Below Benchmark 7-17

Total 1 7.67 Well Below Benchmark 28.33 Below Benchmark 0-39

Keith 2 20 Well Below Benchmark 39 Below Benchmark 20-43

Randy 2 20 Well Below Benchmark 34 Below Benchmark 17-34

Total 2 20 Well Below Benchmark 36.5 Below Benchmark 17-43

Lenny 3 13 Well Below Benchmark 6 Well Below Benchmark 6-20

Debbie 3 20 Well Below Benchmark 33 Below Benchmark 17-33

Lizzy 3 16 Well Below Benchmark 21 Well Below Benchmark 16-28

Total 3 1 6 3 Well Below Benchmark 20 Well Below Benchmark 6-33

Linda 4 6 Well Below Benchmark 51 At or Above Benchmark 6-51

Katie 4 14 Well Below Benchmark 52 A t or Above Benchmark 14-52

Jack 4 21 Well Below Benchmark 51 A t or Above Benchmark 21-51

Total 4 13.67 Well Below Benchmark 513 3 At or Above Benchmark 6-51

Note. Group 1 (SIMI); Group 2 (LS); Group 3 (PA); Group 4 (FP).

Segmentation subtest on the PIPA when com­ pared to both Group 1 (SIMI) (M = 2.33, SD = 2.517) and Group 3 (PA) (M = 9.67, SD = 1.16), as well as on the PSF measure when compared with Group 3 (PA) ( M - 51.33, SD = .577).

Effect sizes were calculated using Cohen’s d for significant results on post-test measures, specifically on the Syllable Segmentation subtest o f the PIPA and on the PSF measure. The effect size for the significant difference between Groups 1 (SIMI) and 4 (FP) on the Syllable Segmentation subtest o f the PIPA (d = 3.07) is considered o f very large magnitude (Keppel & Wickens, 2004), indicating the performance o f these two groups on this mea­ sure differs by over three standard deviations. Similarly, the effect size for the significant difference between Groups 3 (PA) and 4 (FP) on the Syllable Segmentation post-test (d =

1.57) is also considered o f very large mag­ nitude, and indicates that the performance o f these two groups on this measure differs by approximately one and one-half standard deviations. Finally, the effect size for the significant difference between Groups 3 (PA) and 4 (FP) on the PSF measure ( d = .34) is considered o f small to medium magnitude and indicates the performance o f these two groups on the PSF measure differs by approximately .34 standard deviations.

Progress Monitoring Data A visual inspection o f the data from the

PSF and NWF probes, as well as through a calculation o f means, indicated student means to have increased for all four groups and for 10 o f the 11 students on the PSF measure and for all four groups, as well as for all four groups and for nine o f the 11

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 351

Figure 1. Progress Monitoring Group Means Figure 2. Progress Monitoring Group Means fo r the PSF Measure fo r the PSF Measure

students on the NWF measure. Figure 1 il­ lustrates the weekly mean growth of all four groups on the PSF measure, and Figure 2 il­ lustrates the weekly mean growth of all four groups on the NWF measure. Table 3 pro­ vides individual student and group means on the PSF probe at both pre- and post-test, along with end-of-the-year kindergarten

cut points, and Table 4 provides individual student and group means on the NWF probe at both pre- and post-test, along with end- of-the-year kindergarten cut points.

PSF Probe. When looking at each group’s performance on the PSF probe, both a visu­ al inspection of the data and a calculation of group means indicate Group 4 (FP) to

Table 4. Individual Student and Group Means and Cut Points fo r Correct Letter Sequences (CLS) at Pre-test and Post-test on the Nonsense Word Fluency (NWF) Measure

Student Group NWF Pre-test

Mean

NWF cut point de­ scriptor

(Pre-test)

NWF Post-test

Mean

NWF cut point descriptor (Post-test)

NWF Range

Tanya 1 2 Well Below Benchmark 15 Below Benchmark 2-17

Rachel 1 8 Well Below Benchmark 10 Well Below Benchmark 8-14 Jordan 1 1 Well Below Benchmark 27 Below Benchmark 1-27 Total 1 3.67 Well Below Benchmark 1 733 Benchmark 1-27 Keith 2 7 Well Below Benchmark 21 Below Benchmark 7-23 Randy 2 12 Well Below Benchmark 20 Below Benchmark 12-22 Total 2 9.5 Well Below Benchmark 20.5 Below Benchmark 7-23 Lenny 3 2 Well Below Benchmark 2 Well Below Benchmark 2-4 Debbie 3 20 Below Benchmark 31 A t or Above Benchmark 20-32 Lizzy 3 22 Below Benchmark 34 A t or Above Benchmark 18-34 Total 3 14.67 Well Below Benchmark 2 2 3 Benchmark 2-34 Linda 4 16 Below Benchmark 26 Below Benchmark 16-26 Katie 4 9 Well Below Benchmark 21 Below Benchmark 9-24 Jack 4 19 Below Benchmark 12 Well Below Benchmark 12-28 Total 4 14.67 Well Below Benchmark 19.67 Benchmark 9-28

Note. Group 1 (SIMI); Group 2 (LS); Group 3 (PA); Group 4 (FP)

352 / Reading Improvement

have made the most progress throughout the course of the Road to the Code program when compared to the other three groups that each completed only one component of each pro­ gram lesson. Specifically, the overall mean for Group 4 (FP) at pre-test was 13.67 Correct Sounds (CS), with group means increasing on a weekly basis through post-test, at which the overall mean was 51.33 CS, a growth of 37.66 CS. The average weekly increase for Group 4 was 5.38 CS.

While the other three groups also made progress from the beginning to the end of the intervention, the progress and change for each group was smaller than for Group 4 (FP). Group 1 (SIMI) is the group expected to show the most similar results to Group 4 (FP) on this subtest as they also completed the SIMI activity for each lesson, which is most comparable to the PSF probe. The overall mean for Group 1 (SIMI) at the be­ ginning of the intervention was 5.67 CS and was 28.33 CS at the end of the intervention, a growth of 20.66 CS. While Group 1 (SIMI) demonstrated gains on the PSF probe, they were not as large as those of students in Group 4 (FP), as evidenced by Group l ’s (SIMI) average weekly increase of 3.44 CS.

The overall mean on the PSF probe for Group 2 (LS) at the beginning of the inter­ vention was 20 CS, and the overall mean at the end of the intervention was 36.5 CS, a growth of 16.5 CS. While both students in this group also showed progress with pho­ neme segmentation as seen through both individual scores and group mean, it was not as large as either Group 4 (FP) or Group 1 (SIMI), as evidenced by Group 2’s (LS) av­ erage weekly increase of 2.36 CS.

Finally, Group 3’s (PA) overall mean at the beginning of the intervention was 16.33 CS and was 20 CS at the end of the inter­ vention, a growth of 3.7 CS. Group 3 (PA) demonstrated the least amount of progress on the PSF measure, as indicated by their

average weekly increase of .6 CS. It should be noted that the students in this group demonstrated variability with their week­ ly progress throughout the duration of the intervention.

NWF Probe. When looking at each group’s performance on the NWF probe, both a visual inspection of the data and a calculation of group means indicated Group 1 (SIMI) to have made the most progress on the NWF probe throughout the course of the Road to the Code program. Specifically, the overall mean for Group 1 (SIMI) at the be­ ginning of the intervention was 3.67 Correct Letter Sequences (CLS). The overall mean at post-test was 17.33 CLS, a growth of 13.66 CLS. The average weekly increase for Group 1 (SIMI) was 2.28 CLS.

While the other three groups also made progress from the beginning to the end of the intervention, the progress and change for each group was smaller than for Group 1 (SIMI). Specifically, the overall mean for Group 2 (LS) at the beginning of the inter­ vention was 9.5 CLS, and was 20.5 CLS at the end of the intervention, a growth of 11 CLS.

The overall mean for Group 3 (PA) at the beginning of the intervention was 14.67 CLS and was 22.3 CLS, a growth of 7.63 CLS at the end of the intervention. This group also demonstrated progress on this measure, but it was not as large as either Group 1 (SIMI) or Group 2 (LS), as seen by Group 3’s (PA) average weekly increase of 1.27 CLS.

Finally, Group 4 ’s (FP) overall mean at the beginning of the intervention was 14.67 CLS and was 19.67 CLS at the end of the in­ tervention, a growth of 5 CLS. Group 4 (FP) demonstrated the least amount of progress on the NWF measure, as indicated by their average weekly increase of .71 CLS.

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 353

Discussion

Overall Development o f Phonological Aware­ ness skills

The first hypothesis focused on the develop­ ment and improvement o f the students’ phono­ logical awareness skills, skills that researchers have consistently found to be important with regards to a student’s future success in reading (Gillon, 2004; Good et al., 1998). Researchers have also shown that instruction and/or train­ ing in phonological awareness can increase a child’s skills in phonological awareness (Gil­ lon, 2004; Schuele & Boudreau, 2008; Snow et al., 1998), especially when instruction in the more complex skill of phonemic awareness begins in kindergarten (NIL, 2001).

Overall, results from this study support such research, as most students made growth with measured phonological awareness skills throughout the duration o f this study when comparing pre- and post-test results on both the PIPA and the PSF and NWF probes on the DIBELS, as well as when examining both group and individual student progress moni­ toring data, especially in the area o f segmen­ tation. Significant improvements in syllable and phoneme segmentation were found across all groups. While these findings were not sur­ prising for Groups 1 (SIMI), 3 (PA), and 4 (FP) as two o f the three lesson components in the Road to the Code program include activi­ ties focusing on segmentation, the remaining component, Letter-Sound Knowledge, does not (i.e., Group 2 (LS)). Data from this cur­ rent study support this hypothesis.

Specific Skill Gains While important components that should

be included in a phonological awareness training program have been identified (Good et al., 1998), little research focusing on the necessity and sufficiency o f such components was found during a recent literature search. Results from one recent study investigating both the individual and possible synergistic

effects o f both code-focused and meaning-fo­ cused early literacy interventions on the emergent literacy skills o f preschool children showed growth in the specific skills targeted by the intervention (Lonigan et al., 2013). The authors stated that such results further support previous research (e.g., Bowyer-Crane et al., 2008) that early literacy skills are domain specific and interventions must include mul­ tiple domains to have an impact on multiple early literacy skills.

Results from this study show mixed results in this regard. First, the component for Group 1 (SIMI) focused on phoneme segmentation, and they demonstrated the largest mean gain on the measure (i.e., PSF) most closely relat­ ed to the program component they completed when compared with the other two groups receiving only one component o f each Road to the Code lesson. Other data supporting Group l ’s (SIMI) growth on the PSF probe was the magnitude o f change from pre- to post-test when compared to the other groups. Interestingly, Group 2 (LS) scored and Group 4 (FP) continued to score significantly higher than Group 1 (SIMI) on the Syllable Segmen­ tation subtest o f the PIPA. These results sug­ gest that while Group 1 (SIMI) experienced success with phoneme segmentation, which is a focus o f the first program component, it did not necessarily generalize to segmentation at a broader level (i.e., syllable). This finding could be seen as supporting results from pre­ vious studies (i.e., Bowyer-Crane et al., 2008; Lonigan et al., 2013), which stated that early literacy skills are domain specific and inter­ ventions must incorporate multiple domains to influence multiple early literacy skills.

The fact that Group 1 (SIMI) also showed the largest gains on the NWF probe could in­ dicate an ability to generalize their learning o f phoneme segmentation to the NWF probe when combined with their current letter-sound knowledge. However, it should be noted that NWF scores for both Group 3 (PA) and Group

354 / Reading Improvement

4 (FP) were affected by one or more students in each group beginning to blend letter-sound combinations to attempt to form whole, non­ sense words. While this is the ultimate goal o f this measure, it did affect their CLS scores as they progressed more slowly through each word. Overall, data support this hypothesis for Group 1.

The component administered to Group 2 (LS) included activities reinforcing the letter-sound correspondence o f eight letters. Group 2 (LS) not only demonstrated then- largest mean P1PA gain on the Letter Sound subtest, but also the largest mean gain on this subtest when compared to the other three groups. Another interesting note is that while only eight letters are specifically taught as part o f the Road to the Code program, both students in Group 2 (LS) also recognized letters not explicitly taught in the program at post-testing, with one student moving from zero recognized letters at pre-test to 15 recognized letters at post-test. Group 2 demonstrated the second largest gain on the NWF subtest when compared with the oth­ er groups, possibly indicating an ability to generalize their knowledge o f letter-sound combinations to those being seen on the NWF measure. Overall, the data support this hypothesis for Group 2 (LS).

With regards to Group 3 (PA), the students were completing activities that targeted dif­ ferent phonological awareness skills, such as rhyming, alliteration, syllable segmentation, phoneme segmentation, phoneme blending, and letter-sound correspondence throughout the program. Their overall and individual performances on the P1PA, PSF and NWF measures were also inconsistent, with one student’s scores actually decreasing on both measures over the last several weeks of the program. It is difficult to determine the reason for the inconsistency o f this group’s scores across measures. One student did be­ gin blending on the NWF measure, which

did appear to affect her overall CLS score as the blending slowed down her progress on this probe. Further, perhaps practicing multiple skills, rather than one on a consis­ tent basis, affected the lack o f consistency in any one skill across post-test measures. Therefore, the hypothesis is not supported for Group 3 (PA).

Performance o f Group 4, Full Participation Group

Previous researchers have identified several tasks that contribute to the overall effectiveness o f a phonological awareness program (Good et al., 1998), with a majority of them being in­ cluded in the Road to the Code program. Other researchers have indicated early literacy skills to be domain specific and that several domains must be targeted to impact multiple early liter­ acy skills (Bowyer-Crane et al., 2008; Lonigan et al., 2013). Such findings suggest that Group 4 (FP) would make comparable or higher gains on target skills across all measures when compared to the other three groups as they are receiving consistent instruction and practice on multiple phonological skills throughout the duration o f this program. Taken together, the data support this hypothesis.

Results from this study indicated that most students who received the full Road to the Code program (FP) typically made more progress than students who only received one component. This was particularly seen in the area o f segmentation, both at the syllable level when compared with Groups 3 (PA) and Group 1 (SIMI), and at the phoneme level when compared with Group 3 (PA). Other information that supports this hypothesis may not be seen through scores alone. For exam­ ple, students in Group 4 (FP) typically scored higher on most PIPA subtests at pre-test, giv­ ing them less room to grow with these skills by post-test. Further, while Group 4 ’s (FP) gains on the NWF probe did not reach the lev­ el o f either Group 1 (SIMI) or Group 2 (LS),

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 355

it is important to note that two o f the students began blending phonemes to attempt to pro­ duce the whole nonsense word, which is the ultimate goal o f this subtest. While students are given credit for this blending through a Words Read Correctly (WRC) score, it does not account for incorrect attempts. Further, at­ tempting to blend the phonemes slowed down the students’ progress on the measure, which affected the CLS score.

Limitations While there are several interesting and

encouraging findings as a result o f this study, there are also several limitations. A first limita­ tion is in regards to the timeline for this study; specifically, the number o f lessons per week and per day. Due to a number o f recruitment and scheduling issues, implementation of the intervention did not begin until the end of March, leaving approximately two months to complete the intervention. Further, due to schedule constraints, there were several days on which the intervention could not be implement­ ed. Therefore, with the end o f the school year approaching, the number o f lessons per day and per week increased for all groups. All groups participated in approximately the same amount of groups per day and per week, but increasing the number o f lessons per day and week at the end o f the intervention was not optimal.

A second limitation is the small number of students who participated in this study, due primarily to the number of people available to implement the intervention. While this number was sufficient to form four groups for this study, it did prevent the possibility o f forming groups in which other combinations (i.e., Components 1 and 2, 1 and 3, 2 and 3) o f the intervention could be implemented. Further, the smaller number o f students who participated in the study could have reduced its general izability and impacted statistical procedures and results, which is one reason why progress monitoring data was collected and analyzed.

A final limitation o f this study is the fact that all students were receiving outside ser­ vices throughout the duration o f this study. It is typical that those students considered to be most at-risk will be receiving some type o f additional assistance and that the school will want those services to continue as other assistance is added. However, the presence o f additional interventions and/or strategies makes it difficult to determine the individual effectiveness o f each particular service.

Future Research There are several issues and further ques­

tions that can be addressed through additional research investigating the Road the Code pro­ gram. First, beginning any study using this par­ ticular program earlier in the school year is nec­ essary in order to avoid the need to administer a larger number o f lessons per day and per week than what is suggested in the manual. Further having additional time would also allow for more flexible pacing based upon student needs.

A second recommendation for future research would be to include more students when evaluating the Road to the Code pro­ gram. Not only would this increase the ability to generalize results to other kindergarten stu­ dents and in other settings, but it would also allow for the evaluation o f all individual les­ son components, as well as a combination o f components. Further, a larger sample size will further enhance confidence in results obtained through statistical procedures.

A third recommendation for future research would be to consider if and/or the type o f out­ side assistance that students are receiving while participating in the intervention. While it may not be possible to work with those students to be considered the “most at-risk” without the pres­ ence of other assistance, attempting to include students who were all receiving the same type and amount of assistance may help in deter­ mining both the impact o f the Road to the Code program as well as the other type of service(s).

3 5 6 / R ead in g Im p ro vem en t

Implications fo r Practice There are several implications for prac­

tice as a result o f this study. As phonological awareness has been found to be causally related to future reading success (Gillon et al., 2004; Snow et al., 1998), it is a skill that educators may want to strongly consider targeting within both their classroom, group, and/or individual interventions. Results from both this study and a previous dissertation study (Schmitz, 2011) indicate the Road to the Code program may have a positive im­ pact on the development and improvement o f a young child’s phonological awareness skills. Further, this program is considered to be a research-based program (Blachman et al., 2000), includes instruction in phonemic awareness, which has been found to be the phonological awareness skill most closely tied to future reading achievement (Bus & van Ijzendoom, 1999; Good et al., 1998), and has a user-friendly format.

Another implication for practice is in regards to the current shift o f educational systems to the Response to Intervention (Rtl) model. As part o f this model, students who experience continued difficulty in one or more target areas are given additional assistance through an evidence-based intervention that matches their current area and level o f need (Daly, Martens, Barnett, Witt, & Olson, 2007). Further, their progress as part of this interven­ tion is monitored through ongoing data col­ lection to determine whether an intervention should be continued, modified, or terminated, and what, if any, additional supports should be provided. As the Road to the Code is reported to be based upon scientifically-based research, it can be considered an evidence-based inter­ vention. Further, when incorporating consis­ tent progress monitoring, decisions could be made throughout the intervention to determine the student’s progress and if more flexible pacing, as outlined in the manual, would be a consideration for that particular student.

Conclusion While most students in this current study,

regardless o f group assignment, made prog­ ress on measured phonological awareness skills, it does appear that those students who participated in the full program demonstrated the largest gains overall, especially with the skill o f segmentation.

Results also indicated that students receiv­ ing instruction in one o f the first two compo­ nents o f the program targeting either phoneme segmentation or letter-sound correspondence made the largest gains on those specific skills. The third component o f the program targets multiple phonological awareness skills, and results indicated that students in this group typically made some growth in multiple skills.

Overall, while more research is recom­ mended to address both the limitations and the suggestions to further explore possible performance differences across individual and combination o f components, these re­ sults are encouraging. They suggest that the Road to the Code program can have positive effects on young children’s development and progress with phonological awareness skills, especially when all program components are administered. Further, the results from this study appear to support research indicating that early literacy skills are domain specific; therefore, interventions that target multiple skills should be implemented if the intent is to promote consistent gains across multiple phonological awareness skills.

Road to the Code: Examining the Necessity and Sufficiency of Program Components / 3 5 7

References Ball, E.W., & Blachman, B.A. (1991). Does phoneme

awareness training in kindergarten make a differ­ ence in early word recognition and developmental spelling? Reading Research Quarterly; 26, 49-66. doi:10.1598/RRQ.26.1.3

Blachman, B.A., Ball, E.W., Black, R., & Tangel, D.M. (2000). R oad to the code: A phonological awareness program f o r young children. Baltimore, MD: Paul H. Brookes Publishing, Co.

Blachman, B.A., Tangel, D.M., Ball, E.W., Black, R., & McGraw, C.K. (1999). Developing phonological awareness and word recognition skills: A two-year intervention with low-income, inner-city children. Reading and Writing: A n Interdisciplinary Journal, 11, 239-273. doi:10.1023/A: 1008050403932

Bowyer-Crane, C., Snowling, M.J., Duff, F.J., Fieldsend, E., Carroll, J.M., Miles, J., Hulme, C. (2008). Improving early language and literacy skills: Differential effects o f an oral language versus a phonology with reading intervention. The Journal o f Child Psychology a nd Psychiatry, 49, 422-432. doi:10.1111/j.l469-7610.2007.01840.x

Bus, A.G., & van Uzendoom, M.H. (1999). Phonological awareness and early reading: A meta-analy­ sis o f experimental training studies. Journal o f Educational Psychology, 3, 403-414. doi: 10.1037/0022-0663.91.3.403

Daly III, E.J., Chafouleas, S., & Skinner, C.H. (2005). Interventions f o r reading problems: Designing and evaluating effective strategies. New York, NY: Guilford Press.

Daly, E.J., Martens, B.K., Barnett, D., Witt, J.C., & Olson, S.C. (2007). Varying intervention delivery in response to intervention: Confronting and resolv­ ing challenges with measurement, instruction, and intensity. School Psychology Review, 36, 562-581. Retrieved from http://nasponline.org/publications/ spr/index.aspx?vol=41 &issue=4

Dodd, B., Crosbie, S, McIntosh, B., Teitzel, T., & Ozanne, A. (2000). Pre-reading inventory o f phonological awareness. San Antonio, TX: Pearson.

Dynamic Measurement Group, Inc. (2011). DIBELS Next technical manual. Retrieved from http://www. oregonrti.org/wp-content/uploads/2013/09/DIBELS- Next-Technical Manual.pdf.

Gillon, G.T. (2004). Phonological awareness: From research to practice. New York, NY: The Guilford Press.

Good, R.H., Simmons, D.C., & Smith, S.B. (1998). Effective academic interventions in the United States: Evaluating and enhancing the acquisition o f early reading skills. School Psychology Review, 27,

45-56. Retrieved from http://nasponline.org/publica- tions/spr/index.aspx?vol=41 &issue=4

Heroman, C., & Jones, C. (2000). Literacy: The creative curriculum approach. Washington, D. C.: Teaching Strategies, Inc.

Hulme, C., Bowyer-Crane, C., Carroll, J.M., Duff, F.J., & Snowling, M.J. (2012). The causal role o f phoneme awareness and letter-sound knowledge in learning to read: Combining intervention studies with medi­ ation analyses. Psychological Science, 23, 572-577. doi: 10.1177/0956797611435921

Keppel, G., & Wickens, T.D. (2004). Design and analy­ sis: A researcher’s handbook (4th ed.). Upper Saddle River, NJ: Pearson.

Lonigan, C.J., Purpura, D.J., Wilson, S.B., Walker, P.M., & Clancy-Menchetti, J. (2013). Evaluating the components o f an emergent literacy intervention for preschool children at risk for reading difficulties. Journal o f Experimental Child Psychology, 114, 111- 130. doi: 10.1016/j.jecp.2012.08.010

Lundberg, I., Frost, J., & Petersen, O. (1988). Effects o f an extensive program for Stimulating phonological awareness in preschool children. Reading Research Quarterly, 23, 263-284. Doi: 10.1598/RRQ.23.3.1

Moats, L.C. (1999). Teaching reading is rocket science: What expert teachers o f reading should know and be able to do. [paper] Washington, DC: American Federation o f Teachers.

National Assessment o f Educational Progress. (2011). The nation’s report card: Reading assessment. Washington, DC: National Center for Education Statistics.

National Institute for Literacy. (2001). P ut reading first: The research building blocks f o r teaching children to read. Jessup, MD: ED Publishers.

National Reading Panel. (2000). Teaching children to read: A n evidence-based assessment o f the scientific research literature on reading and its implications f o r reading instruction. Washington, DC: U.S. Government Printing Office.

Nelson, J.M., Lindstrom, J.H., Lindstrom, W., & Denis, D. (2012). The structure o f phonological processing and its relationship to basic reading. Exceptionality, 20, 179-196. doi: 10.1080/09362835.2012.694612

Nevills, P., & Wolfe, P. (2009). Building the reading brain: PreK-3. (2nd Ed.) Thousand Oaks, CA: Corwin Press.

Newcomer, P., & Barenbaum, E. (2003). Test o f phono­ logical awareness skills. Austin, TX: PRO-ED.

Nock, M.K., Janis, I.B., & Wedig, M.M. (2008). Research designs. In A.M. Nezu & C.M. Nezu (Ed.), Evidence-based outcome research: A practical guide to conducting randomized controlled trials fo r p sy ­ chosocial interventions, (pp. 201-218). New York, NY: Oxford University Press.

358 / Reading Improvement

Perez, I.R. (2008). Phonemic awareness: A step by step approach fo r success in early reading. Lanham, MD: Rowman & Littlefield Education.

Ranweiler, L.W. (2004). Preschool readers and writers: Early literacy strategies f o r teachers. Ypsilanti, MI: High Scope Press.

Santi, K.L., Menchetti, B.M., & Edwards, B.J. (2004). A comparison o f eight kindergarten phonemic awareness programs based on empirically validat­ ed instructional principles. Remedial and Special Education, 25, 189-196. doi:10.1177/07419325040 250030601

Schmitz, S. L. (2011). The development o f phonolog­ ical awareness in young children: Examining the effectiveness o f a phonological awareness program. (Unpublished doctoral dissertation). University o f Nebraska, Lincoln.

Schneider, W., Ennemoser, M., Roth, E., & Kuspert, P. (1999). Kindergarten prevention o f dyslexia: Does training in phonological awareness work for every­ body? Journal o f Learning Disabilities, 3 2 ,429-436. Doi: 10.1177/002221949903200508

Schuele, C.M. & Boudreau, D. (2008). Phonological awareness intervention: Beyond the basics. Language, Speech, and Hearing in Schools, 3 9 ,3-20. doi: 10.1044/0161 -1461 (2008/002)

Snow, C.E., Bums, M.S., & Griffin, P. (Eds.). (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press.

Torgesen, J. K., & Bryant, B. R. (2004). Test o f phonolog­ ical awareness-Second Edition: PLUS. Austin, TX: PRO-ED.

Whitehurst, G.J., & Lonigan, G.J. (2001). Emergent lit­ eracy: Development from prereaders to readers. In S.B. Neuman, & D.K. Dickinson (Eds.), Handbook o f early literacy research (pp. 11 -29). New York, NY: The Guilford Press.

Copyright of Reading Improvement is the property of Project Innovation, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.