Delivery and Evaluation of Synchronous Online Reading Tutoring to Students At-Risk of Reading Failure Eleazar Vasquez III, Ph.D. of Central Florida David E. Forbush, Ph.D. Cache County School District Lee L. Mason, M.Ed. April R. Lockwood Linsey deed Utah State University

Abstract In a technological society, demands for higher literacy are ever increasing, creating grievous consequences for those who fail to meet these demands (National Research Council, 1998). Both legislative mandates, such as NCLB, and developments in the provision of reading instruction, such as Rtl, have increased the demand for high quality tutoring services. However, the U.S. Department of (USDOE, 2005) and state officials (Richard, 2005) report that access to intensive supplemental tutoring may be limited in rural areas due to limited' availabilit)' of qualified providers. One way to overcome this obstacle is to supplement classroom instruction with live systematic, comprehen- sive, and explicit online reading tutoring. The purpose of this manuscript is to describe and discuss a preliminary investigation of the effects of live online reading tutoring on 4''' grade students in a high minorit)', high povert)' school in Philadelphia, PA. Although this study was not conducted in a rural location, the implications are great for rural schools where qualified tutors may be nonexistent or inaccessible. Using a distance teleconferencing system, trained and supervised undergraduate students located at Utah State University delivered tutoring. The researchers employed a multiple-baseline across participants design to determine, the effects of 1:1 online reading tutoring on reading fluency scores for 3 students. In addition, they compared data on tutors' face-to-face and online tutoring behaviors. Data collected from parents, , tutees, and tutors evaluating the effectiveness, efficiency, and feasibilit)- of online tutoring for remediating students' skill deficiencies are presented and discussed.

The ability to read accurately and fluently is a future, more complex and technical, environments highly valued skill in any literate society. Upon entering (e.g., working with computers, ATMs, access). school, many children learn to read without great In spite of decades of attention by educators and difficulty; however, each year, a portion of children lawmakers, many students still struggle with reading, experience significant problems learning to read (Catts, For example, the College Board recently released SAT Fey, Tomblin, & Zhang, 2002). Authors of the 2005 test scores for die class of 2007, which reaffirmed the National Academy of Science's report on Preventing need for more support for our students. The class of Reading Difficulties in Toung Children stated, "current 2007 SAT test-takers was the largest and most diverse difficulties in reading largely originate from rising to date. However, there was a continued decline in demands for literacy, not from declining absolute reading and math scores for the second consecutive literacy levels. In a technological society, the demands year (Spellings, 2007). In 2007 the National for higher literacy are ever increasing, creating more Assessment of Educational Progress (NAEP) released grievous consequences for those who fall short" (Snow, the Nation's Report Card indicating mixed results in Burns, & Griffen, 1998, p. 1). Clearly, reading is reading performance. Despite recent attention to essential to success in modern American society, reading instruction, nationally, fourth graders' average Reading is required to execute critical tasks (e.g., score in 2007 was only one point higher than 2005 reading instructions, reading magazines and (218 to 219 on a 500 point scale; NAEP, 2007). While newspapers, gaining access to literature, completing tax recent aggregated NAEP scores show minimal positive forms and paying bills, and accessing the internet) growth, disaggregated data continue to show within school, home, and community environments, substantial perforrnance differences across genders. Lack of reading skills produces immediate functional ethnic/racial groups, and social classes (SES levels), deficits in response to environmental demands (e.g.. For example, in 2007, NAEP scores for Caucasian and driving, shopping, reading warning signs) but also Asian students in the 4* grade averaged 26.7 points produces long-term limitations for individuals in higher than for American Indian, Hispanic, and Black

Author's Note Address all correspondence to Eleazar Vasquez III (eleazar.vasquez@ucfedu).

1Ó Rural Special education Quarteriy 2011 30(3) students. Fourth grade students eligible for free and Research also has supported the use of Corrective reduced lunch score, on average, scored 27 points Reading through studies evaluating the curriculum's lower on reading assessments than students who did effect on students. For example, Schacter (1999) not qualify for free and reduced lunch. As a whole, reported in Reading Programs that Work: A Review of these outcomes are alarming in light of data suggesting Programs for Kindergarten to ^^ Grade that Direct that students who are not proficient readers by fourth Instruction was among one of the most effective grade are not likely to become proficient adult readers reading programs for children. In fact, Adams and (Campbell, Hombo, & Mazzeo, 2000). Engelmann (1996) reported in their meta-analysis that Many reading researchers (e.g., Blackman 1996, the mean effect size across 43 studies was .69. 1997; Felton, 1993; Fletcher & Lyon, 1998; Similarly, Building on the Best, Learning from What Torgesen, 1997) have argued that the vast majority of Works (AFT, 1999) reported a 3-year study in which a reading problems can be prevented through early group of rural third grade students were instructed identification and high quality research-based using Corrective Reading and compared to students instruction. Lyon (2001) estimated that, through not receiving DI. They reported an effect size of .61 in appropriate, explicit, comprehensive, and intensive favor of the students who had received Corrective early instruction, we can decrease the number of Reading when compared to the non-DI group (AFT, children experiencing reading failure by approximately 1999). two-thirds. In addition, through an extensive review of The No Child Left Behind Act (NCLB, 2001) and thousands of research studies, members of the National the Individuals with Disabilities Education Reading Panel, (NRP, 2000), identified five critical Improvement Act (IDEIA, 2004) require teachers to elements of reading (i.e., phonemic awareness, use research-based practices and instructional phonics, fluency, vocabulary, reading comprehension) arrangements in their classrooms (Odom et al., 2005). that, if attended to instructionally, may result in skill As students progress from grade to grade, curriculum improvements for many children. demands vary along with classroom settings and One reading program identified as effective is expectations (Haisley, Tell, & Andrews, 1981). Corrective Reading (AÍnetican Federation of Teachers This magnifies the need for educators to incorporate [AFT], 1999). The design of this Direct Instruction teaching practices and instructional arrangements to (DI) program incorporates five main elements: (a) reduce the skill disparity between students. analysis of the content including identification of Tutoring is one instructional arrangement that has generalizable strategies to help students learn more been used extensively to increase student learning in efficiently, (b) instruction designed to minimize elementary settings (e.g., Delquadri, Greenwood, vagueness for students, (c) structured dialogue Stretton, & Hall, 1983; Greenwood, Delquadri, & between the teacher and student, (d) skills taught in a Hall, 1989; Greenwood et al., 1987; Johnson & sequential manner, and (e) organized instructional Johnson, 1984; Maheady & Harper, 1987; Nelson, objectives to allow for systematic skill development Johnson, & Marchand-Martella, 1996). Through the throughout the programs (Watkins & Slocum, 2004). use of tutors, content may be individualized to allow Within this framework. Corrective Reading aligns with students extensive opportunities to respond and the recommendations of the NRP and Put Reading receive feedback (Greenwood, Carta, 8c Kamps, 1990). First reports. The Corrective Reading program includes In fact. Greenwood et al. (1987) suggested that an explicit phonics based approach; sequential opportunities for individual responding are higher introduction to concepts and skills; systematic review during tutoring than during traditional classroom of previously taught material; and fluency, vocabulary, instruction. Moreover, tutoring might be used to and comprehension building (Stein & Kinder, 2004). increase students' engaged time during instructional Specifically, the Corrective Reading-Decoding series is a activities when managing students is often difficult, systematic, phonics program containing four levels (A, such as independent practice. Providing more engaged Bl, B2, and C) which provides hierarchical skill time allows struggling students increased opportunities development in reading for students in grades 3 to cover the material within the curriculum which through 12 (Engelmann, Camine, & Johnson, 1999; otherwise may be hampered due to classroom Engelmann et al., 1999; Engelmann, Meyer, Johnson, constraints and extended breaks from school. & Camine, 1999). Researchers have documented the When implementing a tutoring program, it is effectiveness of the Corrective Reading-Decoding series important that tutors are trained to engage in specific with remedial readers (Campbell, 1988; Gregory, teaching behaviors. Marchand-Martella, Blakely, and Hackney, & Gregory, 1982), non-categorical poor Schaefer (2004) provided seven guidelines of what readers (Holdsworth, 1984-85; Kasendorf & tutors should be trained to do during tutoting sessions: McQuaid, 1987), and special education students (a) begin tutoring when the session begins, (b) use (Thomson, 1992; Thorne, 1978). clear instruction, (c) use explicit teaching methods, (d)

Ruroi Specioi Cducolion Quorteriv 2011 30(3) 17 provide positive feedback for correct responses, (e) tutoring can be made available during school or provide corrective feedback for incorrect responses, (f) outside of school hours; (c) online tutoring can be provide several opportunities to respond to increase accessed from home, school, or other locations, student engaged time, and (g) ignore undesired tutee potentially increasing access, and (d) online tutoring in behavior. Interestingly, these teaching behaviors align the home may increase parent involvement by with those supported by the NRP, Put Reading Eirst, providing frequent updates and debriefing of each and the DI reading and math curricula. Thus, the tutoring session. desired tutoring behaviors align well with the Branzberg (2004) identified seven for-profit curriculum used in this study. companies using synchronous and/or asynchronous Barker, Gersten, and Keating (2000) assessed the online technology to deliver instruction. He found that effects of tutoring on reading accuracy, fluency, and there is little evidence about student outcomes or the comprehension of elementary students across 2 extent to which services provide an effective research- academic years. Tutoring sessions were 30 min. twice based curriculum. Reportedly, each of these businesses per week fbr 6 months during each academic year. The provided some homework assistance (e.g., daily comparison group received typical classroom reading independent practice of math facts or reading instruction. At the end of the 1" grade, the effect size assistance), and two programs provided tutoring for word identification was .42, oral reading fluency services (e.g., use of company created curriculum). was .53, and passage comprehension was .43 when None of the companies offered synchronous audio/ comparing the students who received tutoring to video delivery of curriculum with individualized students who received the regular classroom reading instruction (Branzburg, 2004). None of the companies instruction. At the end of the 2nd grade, the effect size reported the use of research-based curriculum or had for word identification was .44, oral reading fluency any empirical evidence to support their claims of being was .53, and passage comprehension was .32. Their highly effective (Branzburg, 2004). In fact, all seven results indicate that tutoring had an important companies relied on client testimonials as the basis for educational influence on the tutees' reading skills. making their claims. In addition, schools or parents Along with the use of an evidence-based reading also incurred a substantial cost to purchase these program, these results support the use of tutoring as an services—cost range from $35 to $140 per hr. effective method for struggling students. (Branzburg, 2004). Thus, it appears that there is a While several years of research supports the use of need for additional research and development of tutoring as an effective practice to teach reading, it may systems to provide research-based online tutoring that be difficult for some students to receive these services is accessible and affordable. due to a lack of resources in certain rural areas. Eor Researchers have found that there is little example, the USDOE reported that only 226,000 out difference between face-to-face and online instruction of 2 million students (8.9%) nationwide who qualified (Bernard et al., 2004; Meyer, 2002). Bernard ct al. for supplemental tutoring services actually received reported in their meta-analysis of those services (USDOE, 2005). It is difficult to see that face-to-face communication benefited students in this outcome as anything but deplorable considering synchronous distance education. They recommended the potential for supplemental services through online that more research should be conducted in making tutoring. There are numerous advantages of using online learning more accessible to learners located in online delivery for tutoring. Eor example, the use of an rural areas and those with disabilities. In addition, they online format allows the tutor to be flexible in recommended that a comparison be made between instruction with regard to materials, delivery to students provided instruction face-to-facc and students geographic areas, and the time during which tutoring provided instruction at a distance. is delivered (Beth-Marom, Saporta, & Caspi, 2005). In Einally, the NRP (2000) argued that the effects of addition, tutors do not have to exert resources (e.g., computer technology must be studied in partnership time and money) to commute to rural areas (Collins & with the content in which the technology is delivering. Galyon Keramidas, 2006). Using online distance Computer technology "is not a method in itself tutoring may be one way to meet students' needs in (NRP, p. 17). Thus, it is imperative that any locations in which high quality tutoring is not examination of technology should be examined in available. conjunction with the delivery of content through the Synchronous online tutoring refers to 1:1 tutoring technology system. Direct Instruction, a carefully taking place in real time in which the tutor is physically sequenced method of instruction with over 25 years of distant from the student. Synchronous online tutoring research that support its efficacy for use with struggling may offer important advantages, including the readers, may be an effective method to use in an online following: (a) access to skilled tutors is extended environment because one element of Corrective beyond students' geographic boundaries; (b) online Reading is clear, precise instructional delivery.

18 Rural Special education Quarteriy 201 30(3) Given NCLB mandates along with recent nadonal distance-tutoring lab was equipped with high-speed test results, there is a clear need for broad access to Internet capabilities, and each computer tutoring high quality supplemental reading instruction station contained a web camera and voice over IP including individual tutoring. The lack of instrucdonal phone with headset. The Philadelphia Elementary resources in rural and inner city locadons paired with School was equipped with high-speed Internet time-intensive demands of teaching reading skills capabilides, and each tutoring stadon contained a web suggest that there is a strong need to examine camera, phone with headset, landline phone alternadve pracdces of providing supplemental reading connection, desktop microphone, and electronic services. Synchronous online tutoring is one promising wridng board. method for increasing access to these services. The After distance-tutoring pardcipants were identified, purpose of this study was to evaluate the effect of the online delivery system was set up in the elementary online reading tutoring for at-risk 4* grade students' school site and participants were administered three reading skills as measured by the DIBELS oral reading assessments (i.e., Woodcock-Johnson III., DIBEES, and fluency probes. In addidon, we also wanted to idendfy Placement Test). Previous research by the authors on the extent to which tutors' direct instruction assessing students over an online system with the observadon checklist scores coUected from online and Woodcock-Johnson III (WJ-III) indicated that face-to-face instructional sessions compare across assessments can be delivered over an online system muldple dimensions (i.e., cue, pause, signal, response with as valid outcomes as traditional face-to-face error correcdon, instrucdonal pace, praise rate). Last, delivery (see Forbush et al., 2007). The 3 pardcipants' we wanted to find out the students', tutors', teachers', scores are reported in Table I. All 3 students' scores and school administrators' perceptions of the were below the 20* percendle on the three subtests effecdveness and desirability of online tutoring. delivered from the WJ-III. Method Table 1. Participants and Setting Participants' Scores: Woodcock-Johnson-III Tests of This study included 3 fourth grade African Academic Achievement American students and 4 undergraduate college tutors as pardcipants. The researchers used three criteria to Letter Word Picture select student participants. First, their classroom Student Word ID Attack Vocabulary Mean teacher identified students needing reading intervendon. Second, confirming the need for reading intervention, students were required to score below 1 18 10 33 20.3 the 20* percendle on the state reading achievement 2 7 10 1 6 test; finally, students responded to a Corrective 3 26 33 40 33 Reading placement test. The researchers invited students who placed in the same Corrective Reading Note: Scoresrepresent percentile scores curriculum level and the other selecdon criteria to be pardcipants in this study. School personnel identified the elementary school students selected as pardcipants Apparatus for this study as scoring below the 20'^ percendle on Adobe Connect Internet Protocol Video software state reading achievement tests, and they placed was used to allow tutors and students to see one similarly in the Corrective Reading curriculum. The another and to work with instructional materials. placement test ensured that students were grouped in a Adobe Connect's capabilities included real-time relatively homogeneous decoding ability. The communication (i.e., audio, video, use of virtual researchers recruited tutors from Utah State whiteboard for presentation of curriculum) and University's (USU) preservice teacher program and document sharing within the virtual tutoring room. taught them to deliver 1:1 online assessments and The wridng tablet allowed users to write on documents online tutoring. Tutors were female undergraduates digitally. This enabled tutees to complete written work with 2 years experience delivering Direct Instruction. and allowed tutors to monitor and give feedback to Tutors received ongoing supervision in the delivery of students on their workbook acdvides in real dme. (See key assessments and teaching behaviors. Figure 1 for a Screenshot of Adobe Connect as used in The researchers conducted this study in two this invesdgadon.) setrings: (a) USU's Center for Academic Intervenrions tutoring lab and (b) a computer lab at an inner city Measures elementary school in Philadelphia, PA. The USU The primary dependent variable was students' oral

Rural Special education Quorteriy 2011 30(3) 19 Figure 1.

Screen shot of synchronous online tutoring delivered over the Adobe Connect system.

A Winter Day on the Fam »-it.

Turning away ftom tfie window's tigb, Sam puHed tte coven up tightly around his shoulders. He hated to think about WÊÊÊÊÊÊÊI^ leaving the wannth of his bed to do chores. Sighing in Mrt Ktnatt resignation, he forced his feet onto the cold floor and sccambled for his clothes. Sam had recently moved with his &nily to a ftrm on the prairie. He missed all his fHends from town. Each weekday, they had gathered in the one

reading fluency, measured with the Oral Reading Table 2. Fluency (ORF) subtest of the Dynamic Indicators of Basic Early Literacy Skills (DIBELS; http:// Treatment Integrity dibels,uoregon,edu/) reading battery. Good, (Consistent Delivery ofDIBELS-ORF Probes) Simmons, and Kame'enui (2001) provide a complete description of the DIBELS-OPLF assessment, an Tutor Face-to Face Online individually administered test of accuracy and fluency 1 100% 100% with connected text. Each student performs a series of 2 100% 89.25% 1-min, readings on three different passages. All words 3 100% 100% accurately read within 1 min, were counted as correct. Errors included words omitted, substitutions, and M 100% 96.4% hesitations of more than 3 s. Words self-corrected within 3 s were scored as accurate. Each day's score was Note: Scores represent percent items given correctly based the median correct words per min, from the three on the DIBELS integrity checklist passages.

Reliability undergraduate tutors delivered ORF probes Before data collection began, test administrators, consistendy and reliably. primary observers, and interobserver agreement (IOA) Thirty five percent of treatment integrity observers received training. The researchers observations included a second observer to establish electronically archived all distance tutoring and testing IOA, The treatment integrity protocol consisted of sessions for purposes of data collection and IOA specific behaviors identified as critical behaviors from observations. The researchers observed 30% of all DIBELS authors. Given each critical behavior was archived distance-tutoring sessions in real time and observed by independent observers, a check mark was scored them to assess treatment integrity. Table 2 made next to the appropriate definition. Overall, IOA shows a comparison of the consistency in the accuracy for delivering ORF probes online was 91,1% with a of delivery of face-to-face and online delivery of range of 76% to 100% Face-to-face IOA for the DIBELS oral reading fluency probes. Across both DIBELS was 97,1%, ranging from 82% to 100%, conditions (i.e., face-to-face and online). Importantly, the difference in scores may have been

20 Rural Special education Quarteriy 201 30(3) due to technological problems. During this pilot study, lesson readings. Fifth, tutors conducted the researchers had instances of "dropped packets," comprehension checks at the end of each reading defined as losing data from the audio stream. lesson. The approximate duration for each tutoring The researchers also assessed IOA on 35% of session was 50 min. DIBELS-ORF probes and included a second observer During online tutoring via Adobe Connect, the to establish reliability between the tutor and an tutors used Corrective Reading Direct Instruction independent observer. Overall, IOA for ORF probes materials in PDF/Flash Paper format, video via a web was 94.3% widi a range of 81% to 100%. camera, and audio though a voice over Internet protocol service to deliver instruction. Tutors collected Experimental Desi^rn and Conditions data from tutees' responses to oral questions, reading The researchers used a multiple baseline across connected text, producing isolated sounds, sound subjects design to determine the extent to which blends, responses to comprehension questions, online tutoring increased oral reading fluency for 3 completed written practice exercises, reading text fourth grade participants. During baseline condition, online, and responses to fluency building activities. assigned tutors administered DIBELS oral reading Face-to-face tutoring sessions were identical to online fluency progress monitoring probes to each participant tutoring sessions except they were delivered in person. two times weekly. Throughout the study, participants Researchers compared data collected from face-to-face continued to receive reading instruction from their tutoring sessions to the data collected in the online general education teachers. For each participant, tutors tutoring sessions to determine the extent to which collected a minimum of three data points in baseline tutors' behaviors matched across conditions. condition prior to implementing online tutoring. The tutors were responsible for delivering the DIBELS Results probes and collecting the words read correctly and The results (presented in Figure 2) display the incorrectly per min. Prior to implementing online effects of online reading instruction on the oral reading tutoring for each participant, the researchers visually fluency for 3 participants. Overall, supplemental inspected the data trend of the oral reading fluency reading instruction in an online format led to a marked probes to determine if data patterns were stable and increase in the participants' oral reading fluency. For predictable, establishing an effective baseline and Subject 1, the number of correct words read per min. context for observing the effect of online tutoring on (CWRPM) during baseline ranged from 19 to 36, with students' oral reading fluency. Tutors initiated online a mean of 26.25. In contrast, the mean number of tutoring with one participant at a time, with other CWRPM during the intervention phase was 41.50 participants' oral reading probes serving as controls to with a range of 20 to 59. identify the presence of extraneous variables acting In addition, the percent of non overlapping data upon students' oral reading fluency. After the points (PND) was 66% for Subject 1. During baseline intervention was applied with the first student, and condition, the number of CWRPM made by Subject 2 baseline data for the other participants remained stable, ranged from 31 to 65 with a mean of 50.75. During and the pattern of data for the student receiving online the intervention phase, the mean number of CWRPM tutoring was stable and predictable; then, the was 58.45 with a PND of 36%. For Subject 3, the intervention, or one-on-one tutoring, was introduced number of CWRPM during baseline condition ranged with the second participant and so on. from 34 to 65 with a mean of 56.82. The mean number of correct words read per min. during the Intervention intervention phase was 64.57 with a range of 63 to 75. Each online tutoring session consisted of five steps. Tables 3, 4, and 5 display the extent to which First, tutors and tutees discussed the average words tutors' direct instruction observation checklist scores read daily at home to establish rapport. Second, to collected from online and face-to-face instructional measure changes in reading skills, tutors assessed sessions compared across multiple dimensions (i.e., tutees' oral reading fluency using the DIBELS progress cue, pause, signal, response error correction, monitoring assessment. Third, tutors provided reading instructional pace, praise rate). instruction to tutees at each student's predetermined Table 3 displays the tutor presentation reading level. For this study, using the Corrective characteristics. Across both online and face-to-face Reading placement test, all students placed in level Bl instructional settings, all 3 tutors had a 100% match on of the Corrective Reading program. Fourth, tutors the discrete behaviors' cue, pause, and signal. worked with tutees to complete other Corrective Interobserver agreement for these three behaviors was Reading program related activities including 100% across all 3 tutors. workbook, spelling checks, and read to write Table 4 displays the teacher error correction comprehension checks relating to connected text from procedures across both online and face-to-face settings.

Rural Specioi education Quorteriv 2011 30(3) 21 Figure 2.

Improvement Trends for Oral Reading Fluency related to online tutoting for three 4"^ grade students at-risk of reading failure. Online Ibtoring

Table 3.

Tutor Presentation Characteristics—Direct InstructionObservation Checklist Cue Pause Signal Mean Tutor Online FtF Online FtF Online FtF Online FtF 1 100 100 100 100 100 100 100 100 2 100 100 100 100 100 100 100 100 3 100 100 100 100 100 100 100 100 4 100 100 100 100 100 100 100 100

M 100 100 100 100 100 100 100 100

Note:. FtF=Face-to-face; Scores represent percent correct; IOA=100%

22 Ruroi Specioi Cducotion Ouorteriv 2011 30(3) Table 4.

Teacher CorrectionSequence—Direct InstructionObservation Form Model Test Mean Tutor Online EtE Online EtE Online EtE 1 N/0 87.5 N/0 87.5 N/O 87.5 2 87.5 96.6 87.5 96.6 87.5 97.7 3 80 66 80 66 80 66 4 100 100 97 100 98.5 100

M 89.2 87.5 88.2 87.5 88.67 87.8

Note; EtE=Eace-to-face; Scores represent percent eorrect; IOA=86% Table 5.

Teacher Praise—Direct Instruction ObservationForm

Specific Praise General Praise Mean Tutor Online EtE Online EtE Online EtE

1 2 2 3.5 3.5 2.75 2.75 2 4 3.5 2.5 3 3.25 3.25 3 .3 1 2 4.6 1.15 2.8 4 2.3 1 5.6 2.6 3.95 1.8

M 2.15 1.87 3.4 3.43 2.78 2.65

Note. EtE= Eace-to-face; Scores represent responses per minute; IOA=96%

Tutor 1 had no opportunities during online 3.5 with a mean of 1.87. Online, general praise ranged observations to deliver a model or test since the tutee from 2 to 5.6 with a mean of 3.40. Eace-to-face general did not make any errors. However, in face-to-face praise averaged 3.43 and ranged fi-om 2.6 to 4.6. sessions, she delivered 87.5% of the models and tests correctly. Discussion Tutor 2 delivered appropriate error correction The USDOE reported that only 226,000 out of 2 procedures 87.5% of the time online and 96.6% of the million students (8.9%) nationwide who qualified for time face-to-face. Tutor 3 correctly gave error supplemental tutoring services actually received those correction 80% of the time online and 66% of the time services (USDOE, 2005). One way to provide these face-to-face. Tutor 4 correctly gave error correction supplemental tutoring services is to enhance classroom 98.5% online and 100% face-to-face. instruction with live systematic, comprehensive, and Table 5 displays the rate of reinforcement from explicit online reading tutoring. There are several each tutor during both online and face-to-face implications for students who receive online tutoring. conditions. Researchers observed and recorded both Eirst, rural locations can gain access to trained and specific academic reinforcement and general praise for supervised tutors who deliver scientifically proven purposes of supervised feedback. The overall rate of methods of instruction. Second, approximate reading praise delivered by tutors online was 2.78 times per min. strengths and needs can be gauged by delivering (range 1.15-3.95). The overall rate of praise delivered assessments online. Third, students receive one-on-one face-to-face 2.68 times per min. (range 1.80-3.25). high paced instruction. Eourth, intensive practice and Online, specific academic praise ranged from .3 to 4 error correction is delivered as 1 to 2 program lessons praise statements per min., and the mean was 2.15. delivered in a 50 min. period. Eifth, students are able Eace-to-face, specific academic praise ranged fi-om 1 to to build relationships around reading with college-aged

Rural Special education Quarteriy 2011 50(5) 25 students, and, in turn, pre-service teachers gain real the mouse" or place the mouse behind the monitor, or world teaching experience. Sixth, parents, students, to sit up tall with hands in lap. Given that the students administrators, and tutors self report that the responded quickly and appropriately, the tutors interacdon across a distance delivery system is posidve, delivered appropriate reinforcement. and students make gains in reading. Last, tutors are The second obstacle faced was the quality of instructed in delivering research-based instruction, sound. When delivering any type of instruction, clear supervised, and given feedback on performance, and all sound is essendal to discriminate the discreet sounds instructional decisions are derived from ongoing students make when learning to read. The researchers assessment of student skills. found that any background noise (e.g., teacher The implications for schools are also numerous. delivering instrucdon, other computers, hallway noise) With a collaborative effort, are able to would degrade the sound picked up from the deliver reduced-cost tutoring to public school students. microphone. One way to solve the sound issue was to This tutoring is perhaps less expensive than it would cost use noise-canceling headsets. In addition to to hire tutors locally and train them to deliver microphone sensidvity, tutees found it hard to hear on instruction. In addition, public schools are gaining their end using only a single ear headset. The access to trained and supervised tutors who can deliver researchers found that a duel ear headset with a noise- Tier 2 and 3 Rtl intervenrions. As with all types of canceling microphone was the soludon for clear sound intervendons, there is an expense the school must incur. both on the tutors' end and tutees' end. It is important The technology needed is an upfront cost schools or to note that tutors did not use the audio capabilides parents must provide, as well as monthly Internet fees. included in the Adobe Connect environment. They For some parents, the cost of technology and Internet used a voice over Internet protocol (VOIP) system in connecdon may be too much. However, some parents place of Adobe Connect audio capabilities to avoid report in-home, online tutoring allows them greater dropped voice packets. The combination of video and flexibiUty and convenience in addressing their students' audio through Adobe Connect created a restriction in needs. Rather than taking their students to a central computer processing and Internet bandwidth usage. locadon to get tutoring, they can simply turn on their The use of VOIP alleviated the computer processing computer and tutoring is delivered to their home office. and bandwidth restriction by splitting the computer In addirion, parents no longer need to pay for other processing demands into two systems (i.e., VOIP and miscellaneous costs associated with face-to-face tutoring PC). (e.g., automobile fuel, dme, babysitter). Online tutoring The third obstacle was voice tone and volume. increases parents' and teachers' involvement in reading Occasionally, a tutored student was shy and did not interventions by providing them access and weekly talk with strong tone or volume. In this case, the updates of student progress. In a typical online home researchers already had the appropriate microphone tutoring session, tutors will debrief the parents at the and headset. The next step was to shape the student to end of die lesson to share data and inform the parents speak with stronger tone or increased volume. This was on how the tutoring session went. done by modeling appropriate tone and volume. After Several obstacles were encountered in this pilot tutors modeled the correct response, they asked the study. The first and most vexing was the absence of student to respond until they were able to get physical proximity of tutors and tutees and the impact appropriate levels of volume and tone. The tutor used this lack of proximity exerted toward initiating and speech exercises with this student prior to each sustaining instructional control. It became clear that tutoring session to make sure the student could be managing a student's learning behavior who is 3000 heard clearly. miles away may be difficult and cannot be Fourth, the transition from face-to-face tutoring to accomplished without physical proximity. Managing online tutoring was difficult for inexperienced tutors. student behavior was addressed in several ways. First, The Adobe Connect environment is fairly user-friendly. we used high quality, high paced instruction with However, the setup is different when compared to a several opportunities for student responding, face-to-face session. Tutors have to manipulate the reinforcement, and error correcdon. Second, we used a screen environment rather dian turn a page to deliver "you/me" game to manage behavior udlizing a token the appropriate student stimuli. Further, tutors only economy. Tutors allowed students 5 min. after each see what is on a camera. Idendfying discrete behaviors tutoring session to play a game or hsten to music that could lead to problems are difficult to see if the condngent on obtaining more points than the teacher. tutor is busy manipulating the student's stimuli and Third, prior to each tutoring session, tutors gave providing reinforcement or error correction for students explicit instruction on appropriate tutoring student responses. Tutors reported a high learning behaviors. For example, a tutor might have told a curve since the virtual environment was so different. student to adjust the camera for full viewing, to "Park However, when looking at the tutoring behaviors

24 Rurol Speciol educotion Quarterly 2011 30(3) online and face-to-face, overall rates of reinforcement, file also compressed the text, making it difficult to see. error correction, and instructional pace were similar One way to get around this obstacle was to enlarge the across both conditions. Tutors who were well trained original PDF file and then convert the file. Once tutors in delivering the direct instruction had fewer problems were able to enlarge the original high quality scan and transitioning to online tutoring than those who had then convert it to flash paper, students were able to see never been trained previously in DI. Another the text clearly. technique used to smooth the administration of Future researchers should take advantage of tutoring was to develop procedures for the beginning technological developments to explore other methods and ending of each tutoring session. This allowed for delivering tutoring services to students in need tutors to follow the same procedures each session and living in rural areas. Additional research also needs to eased transitions from assessment to instruction. be done in online instructional arrangements to The fifth and final obstacle the researchers determine evidence-based practices of managing encountered was the visual clarity of stimuli used. They student behavior. Finally, a comparison should be made had to scan materials into a PDF format and convert to determine the efficacy of delivering instruction those files into a flash paper format. The compression through an online system versus in a traditional of original scans yielded poor results as the compressed classroom setting.

References Adams, G. L., & Engelmann, S. (1996). Research on Direct In- Engelmann, S., Camine, L., Johnson, G. (1999). Corrective reading struction: 25years beyond DLSTAR. Seattle, WA: Educational Achievement decoding level A: Word attack basics. Columbus, OH: Science Research Systems. Associates (SRA). American Federation of Teachers (AFT). (1999). Building one the Engelmann, S., Meyer, L., Camine, L., Becker, W., Eisele, J., & best, learning from what works: Five Promising remedial reading interven- Johnson, G. (1999). Corrective reading decoding level B\ Decoding strate- tion programs. Washington, DC: Author. gies. Columbus, OH: Science Research Associates (SRA), Barker, S., Gersten, R., & Keating, T. (2000). When less may be Engelmann, S., Meyer, L., Johnson, G., & Camine, L. (1999). Cor- more: A 2-year longitudinal evaluation of a volunteer tutoring program rective reading decoding level C: Skill applications. Columbus, OH: Sci- requiring minimal training. Reading Research Quarterly, 55(4), 494-519. ence Research Associates (SRA). Bernard, R. M., Abrami, R C, Lou, Y., Borokhovski, E., Wade, A., Felton, R. H. (1993). Effects of instruction on the decoding skills of Wozney, L., Wallet, RA., Fiset, M., & Huang, B. (2004). How does dis- children with phonological-processing problems. Jotirnal of Learning tance education compare with classroom instruction? A meta-analysis of Disabilities, 26, 583-589. the empirical literature. Review of Educational Research, 74, 379-439. Fletcher, J. M., & Lyon, G. R. (1998). Reading: A research-based Beth-Marom, R., Saporta, K., & Caspi, A. (2005). Synchronous vs. approach. In W. M. Evers (Ed.), What's wrong in America's classrooms? asynchronous tutorials: Factors affecting students' preferences and choices. (pp. 49-90). Stanford, CA: Hoover Institution Press. Journal of Research on Technology in Education, 37(3), 245-262. Forbush, D. E., Stenhoff, D., M., Vasquez, E., Furzland, M., & Blackman, B. (1996). Preventing early reading failure. In S. C. Cramer Alexander, M. (2007). Evaluation of an online tool for assessing compe- & W. Ellis (Eds.), Learning disability: Lifelong issues (pp.113-118). Balti- tence in achievement testing. Teacher Education and Special Edtication, more, MD: Paul C. Brookes. ,30(3), 142-154, Blackman, B. (1997). Early intervention and phonological aware- Good, R, H,, Simmons, D. C, & Kame'enui, E. J. (2001). The ness: A cautionary tale. In Blackman (Ed.), Foundations of reading acqui- importance of decision-making utility of a continuum of fluency-based indicators of foundational reading skills for third-grade high states out- sition and dyslexia: Implications for early intervention (pp. 347-379). comes. Scientific Studies of Reading, 5(3), 257-288. Mahwah, NJ: Lawrence Erlbaum. Greenwood C. R., Carta, J. J. & Kamps, D. (1990). Teacher-medi- Branzburg, J, (2004). A tour of online tutors. Retrieved from http:// ated versus peer-mediated instruction: A review of educational advan- wwvv.teachlearning.com/story/march2004 tages and disadvantages. In H. C. Foot, M. M. Morgan, & R. H. Shute Campbell, J. R., Hombo, C. M., & Mazzeo, J. (2000). NAEP1999 (Eds.), Children helping children (pp. 177-205). New York, NY: John trends in academic progress: Three decades ofstttdent performance. Wash- Wiley and Sons. ington, DC: U.S. Department of Education. Greenwood, C. R., Delquadri, J. C, & Hall, R. V. (1989). Longitu- Campbell, M. L. (1988). Corrective Reading program evaluated with dinal effects of classwide peer tutonn^. Journal of^Educational Psychology, secondary students in San Diego. ADI News, 7(4), 15-17. Si, 151-160. Catts, H. W., Fey, M. E., Tomblin, J. B., & Zhang, X. (2002). A Greenwood, C. R., Dinwiddie, G., Bailey, V., Carta, J. J,, Dorsey, longitudinal investigation of reading outcomes in children with language D., Köhler, F., Nelson, C, Rotholz, D., & Schulte, D. (1987). Field rep- impúrmcMs. Journal of Speech, Language, and Hearing Research, 45,1142- lication of classwide peer tutoring. Journal of Applied Behavior Analysis, 1157. 77,521-538. Collins, B. C, & Galyon Keramidas, C. (2006). Planning and deliv- Gregory, R. P., Hackney, C, & Gregory, N. M. (1982) Corrective ering online instruction. In B. L. Ludlow, B. C. Collins, & R. R. Menlove reading programme: An evaluation. British Journal of Educational Psy- (Eds.), Online instruction for distance education delivery: Preparing spe- chology, 52(1), 33-50. cial educators in and for rural areas. Victoria, BC: Trafford. Haisley, F, B.,Tell, C. A., & Andrews, J. A. (1981). Peers as tutors in Delquadri, J. C, Greenwood, C. R., Stretton, K., & Hall, R. V. the mainstream: Trained "teachers" of handicapped adolescents./oKr««/ (1983). The peer tutoring spelling game: A classroom procedure for in- of Learning Disabilities, 14, 224-226. creasing opportunity to respond and spelling performance. Education and Holdsworth, P (1984-85). Corrective Reading tested in U.K. ADI Treatment of Children, 6, 225-239. News, 4{2), I, 4-5.

Rural Special education Quarteriy 2011 30(3) 25 Individuals with Disabilities Education Act, 20 U.S.C. § 1400-1487 No Child Left Behind Act of 2001, Pub. L. No 107-110, 115 Stat. (1997,2004) 1425 (2002) Johnson, D. W., & Johnson, R. T. (1984). The effects of intergroup Odom, S. L., Brandinger, E., Gersten, R., Homer, R. H., Thomp- cooperation and intergroup competition on ingroup and outgroup cross- son, B., & Harris, K. R. (2005). Research in special education: Scientific handicap relationships. The Journal of Social Psychology, 124, 85-94. methods and evidence-based practices. Exceptional Children, 71(2), 137- Kasendorf, S. J., & McQuaid, P. (1987). Corrective Reading evalu- 148. ation study. ADI News, 7(1), 9. Richard, A. (2005). Supplemental help can be hard to find for rural Lyon, G. R. (2001). Measuring success: Using assessments and ac-students. Education "Week, 2.5(14), 1-3. counta-bility to raise student achievement. Subcommittee on education re- Schacter, J. (1999). Reading programs that work: A review of pro- form committee on education and the workforce U.S. House of Repre- grams for pre-kindergarten to 4" grade. [Online]. Available: http:// sentatives Washington, DC [Online]. Available: http://www.nrrforg/ www.mfforg/pubs/ME279.pdf Iyon_stateiiient3-01 .htm Snow, C, Burns, S.,i& Griffen, P (Eds.). (1998). Preventing read- Maheady, L., & Harper, G. (1987). A classwide peer tutoring pro- ing difficulties in young children. Washington, DC: National Academy Press. gram to improve the spelling test performance of low-income, third- and Spellings, M. (2007). Building on results: A blueprint for strengthen- fourth-grade students. Education and Treatment of Children, 14, 177- ing the No Child Left Behind Act. Washington, DC: U.S. Department of 189. Education. Retrieved Juné 15, 2007 from http://wwvv.ed.gov/policy/ Marchand-Martella, N. E., Blakely, M., & Schaefer, E. (2004). As- elsec/leg/nclb/buildingonresults.pdf pects of schoolvvide implementations. In N. E. Marchand-Martella, T.A. Stein, M., & Kinder, D. (2004). Reading. In N. E. Marchand- Slocum, & R. C. Martella (Eds.), Introduction to direct instruction (pp. Martella, T. A. Slocum, & R. C. Martella (Eds.), Introduction to direct 304-334). Boston, MA: Pearson/Allyn & Bacon. instruction [pp. 100-139). Boston, MD: Pearson/Allyn & Bacon. Meyer, K. A. (2002). Qtiality in distance education: Eocus on on-line Thomson, B. (1992). A field report: Specific leaming disabilities learning. San Francisco, CA: Jossey-Bass. Corrective Reading pilot study 1989-90. ADI News, 11(2), 11-13. National Assessment of Educational Progress. (2007). The nation's Thome, M. T. (1978). Payment for reading: The use of the correc- report card. Washington, DC: U.S. Department of Education. tive reading scheme' with junior maladjusted boys. Remedial Education, National Reading Panel (2000). Report of the National Reading 13(2), 87-90. Panel: Teaching children to read: An evidence-based assessnient of the scien-Torgesen, J. K. (1997). The prevention and remediation of reading tific research literature on reading and its implications for reading instruc-disabilities: Evaluating what we know from research. Journal of Academic tion. Jessup, MD: National Institute for Literacy. Language Therapy, 1, 11-47. National Research Council. (1998). Preventing reading difficulties United States Department of Education. (2005) Supplemental edu- in young children. Washington, DC: National Academy Press. cational services non-regulatory guidance. Washington, DC: U.S. Depart- Nelson, J. R., Johnson, A., & Marchand-Martella, N. (1996). Ef- ment of Education. fects of direct instruction, cooperative learning and independent learning Watkins, C. L., & Slocum, T. A. (2004). The components of Direct practices on the classroom behavior of students with behavioral disorders: Instruction. In N. E. Marchand-Martella, T. A. Slocum, & R. C. Martella A comparative analysis. Journal of Emotional and Behavioral Disorders, 4,(Eds.), Introduction to direct instruction (pp. 28-65). Boston, MA: 53-62. Pearson/ Allyn & Bacon.

26 Rural Special Cducatian Quarterly 2011 30(3) Copyright of Rural Special Education Quarterly is the property of ACRES 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.