“Double-Sided Learning”: A Qualitative Study of Teacher Candidate Experiences in a Virtual Reading Intervention Clinical Experience

Abstract

The COVID-19 pandemic altered the ways that teacher educators could provide teacher candidates (TCs) with clinically-rich opportunities to build TCs’ skills and self-efficacy related to teaching reading. The purpose of this qualitative study was to investigate the experiences of 19 TCs who participated in a semester-long virtual reading tutoring program during the COVID-19 pandemic. We conducted semi-structured interviews, then analyzed data through inductive and deductive coding. Four themes emerged, centering (a) the impact of the pandemic on typical clinical experiences, (b) the application of coursework within this virtual clinical experience, (c) TCs’ perceptions of program elements that supported their implementation, and (d) TCs’ perceived benefit of the program for their own development as a teacher and for the reading outcomes of the children with whom they worked. We discuss implications for future work related to enhanced clinical experiences related to TCs’ implementation of reading interventions for striving readers.

Learning to read is a multifaceted, complex, and scientific endeavor. Consequently, so is teaching a child to read (Foorman et al., 2016; Moats, 1999; 2020; Petscher et al., 2020). In recent years, students’ relatively low levels of reading proficiency and flat progress across time (U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics [U.S. DOE, IES, NCES], 2022) provided an impetus for grassroot movements (e.g., Hanford, 2022) and related legislative calls to enhance the implementation of instructional practices aligned with the Science of Reading, such as North Carolina’s statewide literacy initiative related to the Science of Reading within the 2021 Excellent Public Schools Act.

Many of these initiatives have centered K-3 reading practices, given the fact that third grade reading achievement is a strong predictor of later reading achievement and high school graduation (Goldhaber et al., 2020). These initiatives have called for teachers to have deep and extensive knowledge about how to teach reading using evidence-based practices (EBPs) to a wide range of students, including striving readers and students with disabilities. Researchers have examined in-service teachers’ knowledge and skills related to reading (e.g., Pittman et al., 2019); however, to best support PK-12 students in the long term, it is equally as important to consider the training of future teachers (i.e., teacher candidates [TCs]). Education preparation programs that are clinically-rich serve as a starting point for building and bridging knowledge about teaching reading using EBPs (Hindman et al., 2020; Kim & Snow, 2021).

Building TCs’ Reading Expertise within Education Preparation Programs

Researchers have examined ways to address the experiences, knowledge, and skills of TCs, including their use of reading EBPs. For example, Hindman et al. (2020) provided an overview of education preparation programs and their alignment with the Science of Reading, as related to authentic experiences in the classroom. Hindman et al. (2020) couched this overview within an acknowledgement of TCs’ relatively low knowledge of reading-related instructional practices (e.g., Clark et al., 2017). In response to these challenges, Hindman et al. (2020) provided a few key implications for programs to support TCs in their reading instruction. For example, they described the ways that education preparation programs should be knowledge- and assessment- centered, such that TCs learn new content systematically and with opportunities to apply content in classrooms and receive feedback within authentic experiences through supports such as coaching (Hindman et al., 2020). They also noted that many clinical experiences are often too short in duration to provide enough time for TCs to meaningfully apply and reflect on their instructional practices.

Similarly, in their systematic review, Hudson et al. (2021) noted that building TCs’ reading-related knowledge and skills depends on the opportunities for TCs to receive explicit and ongoing instruction related to instructional practices in reading. These ongoing opportunities are strengthened when expert modeling is provided initially, along with individualized support and feedback as TCs apply their new learning in the context of classroom instruction, often through clinical placements and related clinical hours (Hindman et al., 2020; Hudson et al., 2021). By improving the quality of clinical experiences and how TCs engage in reading EBPs during these experiences, teacher educators may improve the quality of novice teachers and, consequently, K-3 student reading outcomes.

These calls for building TCs’ expertise in the world of literacy researchers have been echoed by leaders in the field of special education teacher preparation. For example, Leko et al. (2015) outline practices aligned with the science of learning that serve as guideposts for practice-based special education preparation, including deliberate, scaffolded practice opportunities, structured tutoring, coursework that embeds field experiences, coaching, and performance feedback with reflection. Similarly, in a literature review of recent articles published in Teacher Education and Special Education, Brownell et al. (2020) found coaching and/or performance feedback to be the most commonly studied practice, with studies focused on increasing use of EBPs and highlighting the importance of opportunities to practice with feedback. The authors also noted a need for future research in supporting the development of integrated, complex instructional skills, rather than discrete skills, which has been the primary focus to date. Thus, literacy and special education researchers alike support converging ideas about how to best build TCs’ skills within education preparation programs. At the center of calls in both fields lie enhanced clinical experiences.

Enhanced clinical experiences involve time working directly with students in ways that are aligned with course content as well as professional responsibilities as an educator. Research in enhanced clinical experiences has called attention to three essential components. First, enhanced clinical experiences offer sustained implementation of EBPs with consistent students across time (Theobald et al., 2022). Second, enhanced clinical models provide consistent opportunities to receive performance feedback and error correction (Cohen et al., 2021; Thoele & Sayeski, 2022). Last, enhanced clinical models provide properly scaffolded supports for the TCs, such as coaching (Cohen et al., 2021). Coaching support within an enhanced clinical model provides the opportunity for TCs to develop a relationship with a coach who explicitly models skills for the TC and provides descriptive performance feedback (Cohen et al., 2020; 2021). Implementing clinical models with these features tends to increase TCs’ self-efficacy, or their belief in their ability to impact student learning (Bandura, 1997), when implementing EBPs (Cohen et al., 2021).

Theoretical Framing

These clinical models can be situated, in part, within theoretical underpinnings that guide the key components that are relevant to the learning process inherent in educator preparation, particularly those centering enhanced clinical experiences. At the core of the framework is the importance of experiential learning, whereby learners learn when immersed in experiences and have opportunities to reflect on those experiences to transform thinking for future work, thereby building teacher expertise (Dewey, 1938; Glazier et al., 2017). According to Kolb (1984/2014), the power of experiential learning lies in the phases of a learning cycle that includes (a) concrete experiences (e.g., a clinical tutoring experience), (b) reflection (e.g., debrief conversations with coaches), (c) abstract conceptualization (e.g., discussion with peers about experiences), and (d) active experimentation (e.g., moving into student teaching to incorporate new learning into future practices). Concurrently, the model for developing TC expertise and enhancing application of EBPs in practice may be informed by Ross and Bruce’s (2007) theory of TC development, which bridges TC self-assessment through reflective processes with other external factors such as change agent input (e.g., coaches) to influence teacher efficacy and, consequently, TCs’ future teaching goals and effort.

The COVID-19 Pandemic’s Context

These strong theoretical and empirical calls for enhanced clinical experiences to improve TC outcomes serve as an important guidepost in the field of teacher education. However, in Spring 2020, teacher educators across the world were required to pivot, alter, and even halt the ways in which their education preparation programs provided TCs with clinical experiences. Because of the COVID-19 pandemic, TCs faced challenges completing clinical experiences across the country (American Association of Colleges for Teacher Education [AACTE], 2021). Although research on special education preparation programs has served as a guiding light for creating and evaluating clinical experiences prior to the COVID-19 pandemic (e.g., Brownell et al., 2020), there is little known about the ways in which altered clinical experiences influenced TCs in special education preparation programs. However, recent research has suggested that changes in clinical experience placements resulted in lower TC self-efficacy (Choate et al., 2021; VanLone et al., 2022). These results suggest a need to consider ways programs have attempted to pivot face-to-face clinical experiences in reading to virtual contexts, particularly those that may have promising evidence of maintaining qualities of enhanced clinical experiences that tend to increase TC self-efficacy despite the trying COVID-19 context (e.g., Curwen, 2020; McGrath, 2022).

Implementing such clinical models aligns with the need to address the impact of the COVID-19 pandemic on students’ reading achievement. According to the 2022 National Assessment of Educational Progress (NAEP) report, reading scores between 2020 (pre-pandemic) and 2022 showed the largest significant decrease since 1990 (U.S. DOE, IES, NCES, 2022). In a series of briefs, The Education Trust and MDRC (2021a; 2021b; 2021c) suggest three strategies for school leaders to consider related to student learning in the context of COVID-19: (a) expanded learning time, (b) targeted intensive tutoring, and (c) the importance of strong relationships. Notably, providing individualized and targeted tutoring in a skill-building curriculum has been a key recommendation to alleviate adverse effects on student learning (e.g., The Education Trust & MDRC, 2021b).

To maximize effectiveness, the targeted intensive tutoring described by groups such as The Education Trust and MDRC (2021b) can and should mirror the approaches to reading intervention proven effective for students with or at-risk for disabilities (Foorman et al., 2016; Petscher et al., 2020). Implementing interventions through the context of an enhanced clinical model may be one way to address the concurrent needs to build TCs’ capacities to teach reading and to provide needed reading interventions (Hoffman et al., 2019). However, many of these models were implemented in a physical school. Far less is known regarding enhanced clinical models conducted virtually. Understanding the feasibility and effectiveness of virtual enhanced clinical models is important, given the current and future reliance on virtual instruction and support during emergencies or in rural and remote contexts (Ault et al., 2020).

The Present Study

Thus, the purpose of the present study, a qualitative research study aligned with the quality indicators of qualitative research in special education (Brantlinger et al., 2005), was to investigate the experiences of participants in a semester-long virtual reading tutoring program that served as an impetus for an enhanced clinical experience for TCs. The work in the present study actively incorporated the first two phases of Kolb’s (1984/2014) learning cycle: concrete experiences and reflection. TC participants also expressed connections to the final two phases of Kolb’s (1984/2014) learning cycle (i.e., abstract conceptualization and active experimentation) in their structured interviews. Additionally, the work directly centers the external change agent input from coaches (Ross & Bruce, 2007). The following research questions (RQs) guided our investigation:

1) How did TCs characterize their experience participating in a semester-long virtual reading tutoring program in the context of the COVID-19 pandemic?

2) What program elements did TCs believe helped (programmatic support) or hindered (programmatic barrier) their preparation to deliver reading interventions to striving readers in first grade in a virtual context?

Method

In the following section, we first outline the virtual reading intervention context that served as the setting for the volunteer-based virtual clinical experience. We provide information about the reading intervention itself, its training, and the coaching provided. Then, we summarize the specific details for the present qualitative study.

Context: Virtual Reading Intervention

Restrictions related to the COVID-19 pandemic eliminated TCs’ typical early clinical experiences prior to student teaching during the 2020-2021 academic year. We piloted a virtual reading tutoring program to simultaneously address the need for alternative clinical experiences for TCs and the need for interventions to alleviate learning loss related to the COVID-19 pandemic (The Education Trust & MDRC, 2021b). At the beginning of each semester, we invited any TC enrolled in our university’s courses on assessment or reading methods in special education to volunteer as a tutor. Any TC who volunteered was offered extra credit in their reading methods course. TCs met over Zoom with the same one to three students two to three times a week for 15 weeks. Each session lasted 45 min. TCs who served as tutors delivered the intervention using Google Slides and “shared” (i.e., presented) their screen for their student(s) via the Zoom meeting.

Across the pilot year, the program supported 21 first graders (6-7 year-olds) at one elementary school in a large city in the southeastern United States. School partners identified each first grader as a striving reader whose universal screening data for reading from the end of Kindergarten and beginning of first grade indicated intensive needs in reading relative to grade level benchmark goals. During the fall semester, the first graders accessed the Zoom meeting from their home. During the spring semester, the school resumed in-person instruction and the first graders accessed the Zoom meeting from their classroom. The goal of the program was to provide support to a local school district during a global pandemic and allow for reactivity and flexibility in an unprecedented and constantly evolving instructional environment. We prioritized immediate implementation to address the simultaneous need for reading intervention and TC clinical experiences and chose not to measure implementation fidelity or student outcome data in this pilot year.

Tutoring Intervention, Training, and Coaching

We trained TCs to deliver the Road to Reading intervention (RTR; Blachman & Tangel, 2008), which provides instruction in foundational reading skills. RTR was developed for students requiring additional skill development in phonemic awareness and phonics. In a RTR lesson, students (a) learn new letter sounds with embedded practice; (b) build words with manipulative letter cards, linking together words through word chains in which only one grapheme changes (e.g., mat → mash involves changing the grapheme “t” to the grapheme “sh”); (c) read decodable words and high frequency words; (d) read decodable books; and (e) spell practiced sounds, words, and sentences through dictation practice.

RTR includes a systematic scope and sequence that builds to increasingly challenging phoneme/grapheme correspondences. For example, students start with the most frequent consonant sounds and short vowel sounds with a focus on CVC words (e.g., mat). Once students have mastered CVC words, lessons move onto digraphs (e.g., “sh” in ship), CVCe words (e.g., “a_e” in maze), vowel teams (e.g., “ea” in team), r-controlled vowels (e.g., “er” in fern), and final consonant-le syllable patterns (e.g., “-ple” at the end of purple). Each step of the RTR lesson involves structured and explicit instruction in the skill, with a teacher model, many opportunities to respond, and immediate/specific feedback. The approach to literacy instruction used in RTR follows recommendations for teaching foundational reading skills to early elementary students, particularly for striving readers and students identified with reading-related disabilities (Foorman et al., 2016; Petscher et al., 2020).

In training, TCs first watched a video, developed by our research team, that provided explicit instruction in each component of RTR. Next, TCs watched a model lesson in which one of the coaches delivered one RTR lesson with a group of first-graders. Last, TCs practiced an RTR lesson with a fellow TC prior to implementing with their assigned first grader(s).

TCs were supported by seven coaches who were faculty members or doctoral students in reading education or special education. Each coach supported between one and three TCs. Coaches observed each lesson for the first five weeks (i.e., 10 lessons) and observed between 10 and 20 lessons thereafter, depending on how much support was needed. Coaches provided behavior-specific feedback to the TCs on principles of explicit reading instruction and positive behavior supports. Depending on the TC/coach relationship, coaching support occurred (a) before sessions and/or when a tutor was planning lessons; (b) during lessons via direct modeling, scaffolded support, or private Zoom chat; or (c) after observation via a quick debrief within the TCs’ virtual breakout room, during a separate Zoom meeting with the TC, or via email.

Participants

At the end of the 2020-2021 academic year, we recruited participants for the present study using purposive sampling, a non-probability sampling that is used to target specific groups of participants with similar characteristics (Onwuegbuzie & Collins, 2007). We used purposive sampling to appropriately recruit participants who were TCs and had experienced the volunteer-based, enhanced clinical experience during the COVID-19 pandemic. We obtained permission from the university’s Institutional Review Board prior to recruitment, and analyzed data after the completion of the courses.

We interviewed 19 TCs (86.4% of all TCs who volunteered to tutor in the pilot year of this program). To best ensure confidentiality of participants’ identities, we assigned each participant a pseudonym. TCs were undergraduates either majoring in special education (n = 15; 79%) or dual majoring in elementary education and special education (n = 4; 21%). All dual majors provided tutoring in the Fall 2020 semester. All special education majors provided tutoring in the Spring 2021 semester. One TC provided tutoring across both semesters. Most TCs used she/her pronouns (n = 15; 79%), though one TC used she/they and three TCs used he/him. Most TCs were also traditional students who had enrolled in an undergraduate program immediately after completing high school and/or received financial support from parents (n = 14; 74%). Five TCs reported they were nontraditional students. When asked about their home community type, TCs reported urban (n = 9), rural (n = 7), and suburban (n = 3) communities. Ten TCs were White (53%), five were Black (26%), and four (20%) held other cultural identities, including (a) Arab, (b) White and Latinx, (c) Black and Latinx, and (d) Asian and Latinx. Participants varied in their years of experience supporting students with disabilities (M = 6 years; range: <1–20 years), through roles such as camp counselors or paraeducators. All participants reported between 0 and 1 year experience providing reading instruction. See Table S1 for complete participant demographics and pseudonyms.

Research Team

Our research team included two faculty members, three doctoral students, and two undergraduate research assistants who were majoring in psychology and part of the university’s undergraduate research program. We held ongoing discussions of our positionalities as the lens through which we conducted our research (Brantlinger et al., 2005; Trainor & Graue, 2014). The first faculty member (first author) held a Ph.D. in Special Education with a research area centering intensive intervention in reading. Prior to obtaining her doctorate, she served as a direct support professional for children and adults with developmental disabilities for 5 years and taught fourth grade in a suburban, Title I, public elementary school for 3 years. The second faculty member (third author) held a Ph.D. in Curriculum and Instruction (Literacy emphasis). Prior to obtaining her doctorate, she served as a preschool teacher, an English/Language Arts teacher, and a middle/high school reading specialist. Both faculty members and two of the doctoral students served as coaches in the program during both semesters. Our team featured a range of experience in working with elementary-age students within the context of reading intervention, from 20 years (third author) to less than 1 year (undergraduate research assistant). The majority of our team identified as White and without disabilities.

Procedures

At the end of the spring semester and upon receiving approval from the University Institutional Review Board, we emailed TCs and invited them to complete a 30-min interview about their experience, following up seven days later if needed. As an incentive, we provided a $5 gift card to each TC who completed an interview.

Interview Procedures and Member Checking

At the time of interviews, the interviewer (second author) was a doctoral student in special education who served as a coach in the program during both semesters. As program staff, he was perceived as an insider by TCs, which allowed him to quickly move to asking specific questions about TCs’ experiences (see Table S2 for interview protocol). Interviews were conducted via Zoom, recorded, and transcribed. Prior to the start of each interview, the interviewer thanked the TC for their participation and stated the purpose of the interview and the study. Then, the interviewer asked questions that addressed the research questions (see Table S2) using a semi-structured interview protocol (Kvale & Brinkmann, 2014). The interviewer asked follow-up questions to elicit examples or clarification on previous statements (Brantlinger et al., 2005). After approximately 30 min, the interviewer asked the participant if they had any more experiences to share that were not covered in answers to the previous question. Last, the interviewer thanked the TC for their time and ended the Zoom call. Interviews lasted between 11 min 59 sec and 32 min 33 sec (M = 24 min 10 sec). The fourth author (a doctoral student in special education) verified transcripts by listening to the audio recording and editing the transcript to ensure accuracy. As a content expert, the fourth author corrected content-specific phrases not captured by the transcription software (e.g., “phonemes”).

During each interview, the interviewer created field notes to use in member checking and validating interview content. After the interview, the interviewer emailed the TC a list of supports and barriers the TC described during their interviews from field notes (Brantlinger et al., 2005). The interviewer requested participants confirm the interview summary, add missing content, or correct the summary. Eighteen participants responded and confirmed the accuracy of quotes and codes without suggestions or revisions. One participant added to their summary.

Data Analysis

After verifying interview content, we unitized, or segmented, each interview to ensure coding of the same segmented units of analysis across transcripts and coders (Reed et al., 2018). We unitized, or segmented, transcripts by individual ideas expressed within quotes, breaking apart sentences when necessary. This process resulted in 1377 codable units across 19 transcripts (M = 72.58 units per transcript; range: 34–107 units per transcript).

Inductive Coding

Next, we used a triple-round coding process, pairing inductive structural coding with deductive sub-coding (Saldaña, 2013) to identify codes present within transcripts and answer the research questions. To begin this process, the first, second, and fourth authors read three transcripts independently and created a list of codes that described participants’ experiences. Next, we met to discuss emergent structural codes and came to a consensus on a structural codebook to use in the second round of coding. We created structural codes to broadly organize participant experiences across categories of content. Within each structural code, we generated subcodes to further characterize participants’ experiences, including the program characteristics that helped or hindered their experience. As we analyzed data, we shared our methods and findings with a colleague unaffiliated with the data collection who held expertise in qualitative research. This served as a form of investigator triangulation (Brantlinger et al., 2005).

After generating structural codes and subcodes, we met to discuss and come to a consensus. Our use of this iterative process across researchers served as a form of investigator triangulation during codebook development (Brantlinger et al., 2005). We identified four structural codes. First, rationale for volunteering represented reasons that contributed to the specific TCs’ decision to volunteer for the RTR program. Second, tutor characteristics were internal or external factors or actions that prepare teachers for instruction. Third, student characteristics were student factors or actions demonstrated during the RTR program. Finally, program characteristics included factors or resources specifically related to the RTR program and corresponding coaching supports.

Within these structural codes, 19 subcodes emerged. We generated a codebook with definitions of structural codes and subcodes. The codebook contained coding rules to guide coders. For example, within the rationale structural code, we added a coding direction in the codebook to not code a unit as rationale “if the tutor mentioned a reason that did not contribute to their decision to volunteer.” See Figure S1 for a map of structural codes and subcodes.

After creating the codebook, we imported unitized transcripts and subcodes into Taguette (https://app.taguette.org/), an open source qualitative coding software. Undergraduate research assistants (sixth and seventh authors) coded each transcript initially, with at least one doctoral student or faculty member serving as a second coder for intercoder agreement. We used the highlighting tool within Taguette to code each unit within each transcript. Some units represented multiple subcodes in one and therefore multiple codes. In all, the units represented 1699 codes (rationale: n = 69; tutor characteristics: n = 589; student characteristics: n = 175; program characteristics: n = 623). In some instances (n = 30), we felt ideas communicated within units were not represented by extant subcodes. When this happened, we coded the unit as “other.” In other instances, we tagged units with “no code” (n = 179) when the unit represented statements that were off topic, fillers, or continuations from other units. In Table S3, we provide a list of all subcodes, subcode definitions, frequency counts across all coded units, and an example quotation to illuminate the nature of each subcode, derived through our iterative approach to data analysis.

Reliability

After coders finished coding each transcript, we exported units as a comma separated value (CSV) file to check reliability. We used Recal to check intercoder reliability (http://dfreelon.org/utils/recalfront/). We set an alpha of .70 as the cutpoint for reliability (Krippendorff, 2004). When the alpha was less than .70, we analyzed alphas of individual subcodes to determine subcodes where coders were most unreliable. We met as a team to provide additional training to undergraduate research assistants on subcode definition, coding rules, and discuss discrepancies. Then, we repeated this process with the same transcript until the overall alpha was at least .70. Individual alphas of transcripts ranged between .702 and .725.

Deductive Coding

Next, to address RQ2, we applied a deductive coding method to all units tagged with a subcode in the Program Characteristic structural code (i.e., Coaching, Lesson Materials, Instructional Context, Multiple Students, Planning Documents, Training). We coded each unit as either a support, barrier, neutral characteristic, or suggestion. We defined supports as aspects of the program characteristic that helped TCs as a developing teacher. Barriers were aspects of the program characteristic that hindered TCs as a developing teacher. A neutral characteristic was a statement about what the TC did related to the program characteristic, but without any indication of it helping or hindering them as a support or barrier. Last, suggestions were ways to change the program characteristic in the future. TCs described the nature of the program characteristic as a support in the majority of units related to the Coaching (73.9% of applicable units), Lesson Materials (75.9%), Planning Documents (67.3%) and Training structural codes (62.2%). In contrast, TCs more frequently described the nature of the program characteristic as a barrier for the Instructional Context and Multiple Students structural codes (71.2% and 45.2% of the units for each structural code, respectively). See Figure S2 for a visualization of the frequency and proportion of units within each subcode where TCs described characteristics as supports, barriers, neutral characteristics, or suggestions.

Themes

Last, we reviewed and compared structural codes, subcodes, and follow-up codes to synthesize themes that emerged. We considered barriers and suggestions across our subcodes as a way to identify potential sources of disconfirming evidence for our themes (Brantlinger et al., 2005). From the four emergent themes, we generated thick descriptions, drawing upon detailed participant quotations to portray themes and major findings that addressed our research questions.

Findings

In this study, we examined TCs’ experiences in a semester-long virtual reading intervention clinical opportunity, in which we trained TCs in a reading intervention and provided coaching support as they implemented the intervention with striving readers in first grade. Across interviews, TCs who volunteered as a tutor reported similar experiences. Four themes emerged from our data analysis: (a) the COVID-19 impact on TCs’ professional preparation, (b) the aspects of experiential learning afforded by the tutoring program and coaching, (c) the power of having EBP materials to support and structure tutoring sessions, and (d) the perceived impact TCs felt the tutoring program had on their student’s reading skills and their own teaching. In this section, we will expand upon each of these themes.

“I’m Not Out There In Schools Right Now Getting Those Experiences”

The first theme centers on the way that TCs consistently described the impact they felt the COVID-19 pandemic had on their preparation as future teachers. This impact was most directly related to their perceived ability to practice and apply skills in classroom environments, which TCs identified as a core facet of preparation that had been stolen from them due to the pandemic. For example, when describing her decision to volunteer for the program, Vicki stated:

Last semester and this semester have been just not the best, because everything's at home, everything's on my computer. I haven't really been able to go to schools or anything like that…I haven't been able to interact with a kid through COVID really at all, until now.

TCs who chose to volunteer for this program felt the pressure of being later in their program. As juniors, they tended to be acutely aware of how different their experience was to the “typical” experience through the education preparation programs offered at their university. Prior to COVID-19, educator preparation involved many opportunities for early clinical experiences where TCs could apply practice coursework in authentic contexts with children. The COVID-19 context took away these more authentic early clinical experiences prior to student teaching. Many TCs expressed self-doubt as a result of these lost experiences. They reflected on how replacement activities left them wanting more, such as when Shelley stated, “I wanted to experience it and I wanted to see it work for a kiddo in a real life situation instead of watching videos or case studies or reading case study reports or whatever. I wanted to see it.”

That’s not to say self-doubt is not a challenging part of even typical undergraduate experiences. However, navigating beyond that self-doubt and feeling confident in career choices can be made easier through authentic experiences to build knowledge, skills, and self-efficacy, as exemplified through Amber’s reflection on her own self-doubt, which motivated her decision to volunteer for the tutoring experience:

In college you kind of have some doubts about your major from time to time. So like you always hit a certain point where you're like, “Is this what I can really see myself doing?” Which is what I hit. And basically, just seeing how much the student showed a lot of progress really let me know that this is what I'm meant to be doing. And that, even though I might've hit, like, a rough patch, that doesn't mean that I'm going to stay in that rough patch forever. It just gave me like a little booster to let me know that I can do anything I set my mind to, even if I don't believe in myself at that time.

For these TCs, COVID-19 came at the time in their preparation program in which they start taking the more advanced teaching methods courses and typically depend upon clinical experiences the most. In this way, the impact of COVID-19 on TCs' own education preparation program experience directly influenced their decision to participate in the program. Moreover, their participation shaped the ways they would later view their experiences in the program when implementing the reading intervention with coaching support to first graders.

“I’m Not Just Learning How To Do Some Of Those Things, I’m Actually Doing Them.”

With the worry of COVID-19’s impact on their ability to interact with students at the forefront of their minds, TCs described how the virtual clinical program provided opportunities to practice skills they would not have otherwise had. Specifically, TCs viewed the tutoring program as an opportunity for applied practice of pedagogical skills such as classroom management and reading instruction in a more authentic environment. Reflecting on this chance to walk the (virtual) walk, Brenda described:

We're required to do clinicals, but this is an outside thing…You're not just watching what other teachers do…I feel like this prepared me more than anything yet. And that's kind of saying a lot, but more than anything yet at [college]. Because I'm not just learning how to do some of those things, I'm actually doing them. Seeing where I can mess up and seeing how students will actually react and practicing that behavior management…Just from my experience, it developed me from learning the content, learning what I'm supposed to do, and transformed to actually doing it.

In addition to describing their overall experience, several TCs described how the program provided opportunities to reflect on teaching practices. For example, Penny described how she would often debrief with her “poor roommates” about all the ways she wanted to try to add in additional supports or small-scale adaptations to her instruction in response to her tutoring experiences to better meet the needs of her students, based on what she saw was working and aspects of her teaching practices she wanted to improve. In this way, the tutoring experience lit an excitement for teaching and day-to-day, data-based teaching decisions. Some TCs reflected on how the program provided opportunities to apply content learned in other courses to a clinical setting. This reflection was particularly true for TCs taking their reading methods course concurrent to the tutoring experience, in which TCs identified how aspects of reading development and instruction were enacted in the context of the RTR intervention. Additionally, this integration of content with clinical experiences contributed to a deeper understanding of the content and its applicability to TCs’ future careers, as Brad illustrated:

Well, I feel more comfortable with it now…and I feel like I could do it to more than one student now, which is probably the goal, right? Because you're doing reading tutoring, but when you're teaching, you're going to be teaching it to a whole bunch of people… I feel like now I could apply that out, which I would never be able to do at the beginning.

In this reflection, Brad hints at how having the opportunity for such practice-based learning and application lead to confidence in their own teaching abilities. TCs’ perceived growth in self-efficacy did not manifest overnight, however. Even with the training procedures we had in place, which were sufficient for many TCs, several TCs suggested future iterations of the program include more training and rehearsals prior to starting with their children. However, many TCs described how initial feelings of anxiety at the beginning of the program subsided as confidence developed through the program. For example, Prudy relayed, “I got more confident every single week, and, like, knowing that…I had support I can lean on if I needed it…I think that helped and, like, I just became more confident every week, as we did it.”

This transformation in self-efficacy for TCs was matched by newfound appreciation for just how much goes into teaching reading, something some TCs acknowledged they did not realize at the start. At times, this appreciation stemmed from the realization that approach to teaching reading has shifted from their own elementary years, such as when Wilbur stated: “I wasn’t even taught [through phonics-based instruction]. I mean when I look back at my school, it was a miracle that I really picked up on reading…So, I basically relearn how to read through teaching reading.” For Wilbur, and other TCs who volunteered for the tutoring program, course content on research-based reading strategies/processes led to new knowledge and skills that supported the work with and was enhanced by the applied practice during tutoring sessions.

“I Didn’t Have To, Like, Start From Scratch.”

In part, the course content and clinical application integration was perceived by TCs as impactful because of the programmatic components and virtual resources provided as part of the tutoring experience. Those programmatic components and resources included training supports, implementation supports, and coaching support during lesson planning, implementing, and debriefing. As Maybel noted, “I was overwhelmed at first… but how organized [the tutoring materials were], that was really helped. So, I didn’t have to start from scratch...It helped me know what I needed to do.” The sense of comfort from having those resources made the largest barrier identified by the TCs - the virtual context itself, including the challenges related to the unpredictability of and added layers of distractions from students’ home and school contexts - easier or more palatable to overcome.

The first programmatic component the TCs described was the training in the reading intervention prior to work with students. This training involved watching asynchronous training videos, completing synchronous Zoom-based practice sessions, and virtually observing a coach implementing the intervention with a group of students. Brad described:

The training at the beginning was a good support. I liked that a lot. As opposed to getting thrown in…we got to watch [our professor] teach the whole thing. And then she was kind of following the script a little bit as well and so I was able to see how it looked across the entire 45 minutes, which was really, really helpful. So that was a good support.

Similar to Brad, Polly also mentioned the training and resources as a support, describing how she felt those resources set her up for success, given the way, “The videos to watch and all of the resources…It was a really good balance between us being instructed on how to implement the program, but then also giving us autonomy to tailor it to our student.” Later, Polly noted that even with feeling set up for success, she engaged in additional practice sessions to build her confidence even more prior to implementation:

Before starting the whole session, I would take like 15 minutes just to look over my slides and just even practice by myself just so I don't fumble on anything and everything goes smooth. So just taking that small time of your day to just practice on your own has definitely really helped. I know when I first started I've practiced almost every day or I would have my brother pretend to be a student to help me out…just a little role-playing just so I would know what to say and do when that time comes.

Instructional materials and program structure consistently emerged as resources TCs believed facilitated their implementation of the intervention. These instructional materials and program structure included a scope and sequence of skills to introduce across lessons, Google Slide templates for the RTR lesson sequence, scripts to support implementation of each lesson component, and sample planned lists to use across those components, such as word lists for each skill, word building chains, and sentences to dictate. TCs described how these resources allowed them to focus on implementing lessons and building relationships rather than planning or writing lessons. This is best exemplified by Shelley’s comment: “The scripts, I think reducing that cognitive load…just not thinking about too much of what you have to say and getting that natural feel for it as you move along and get to know your kid. That’s really helpful.” Building off that sentiment, Shelley later described how having these resources freed her up instructionally to intentionally plan for greater alignment to her students’ interests, strengths, and needs.

Most frequently, however, TCs described coaching as the support they leaned on the most. The relationships they built with their coaches were meaningful and authentic. Tammy, for example, described:

And honestly, I felt like I've gotten closer with [my coach] with doing the tutoring. I feel like we've had a lot of meaningful conversations...And so I really appreciated that. I just love talking because normally I don't really talk in class ‘cause sometimes I get intimidated speaking up and forming those connections with the professor…Again, it's a learning experience. So it's really nice to just have [a coach], because it's almost like a mentor where you can just learn from them, with all their experiences.

That sense of having a coach as a cheerleader of sorts, who built up confidence while also providing scaffolded support, arose in other interviews as well, such as when Tracy stated, “My [coach] helped a lot by saying I did a good job. No matter, even if I did the worst that I’ve ever done. She found a way to say that I did something good.” This sense of having someone in their corner was important to TCs, particularly in early lessons, when TCs’ big dreams for those lessons may have been confronted with reality that did not quite align with expectations. These moments were crucial for tutor/coach relationships. Tracy, for example, shared:

I have always considered myself a good leader, so I was very excited to do so…and then, when I did the first lesson that I did and it didn't go how I planned for it to go -- because my script was messy and it was down and it was not working -- I was very upset like, “this was not how I had planned for it to go!” And [my coach] was just very encouraging. She was like, “well, you did a lot of good things.” So that helped, and then I figured out a better way to hold the script and practice more.

That tutor/coach relationship drove interactions and supported TCs as they grew in their perceived knowledge, skills, and self-efficacy. TCs noted how this structure of the coaching supported them until they did not need the coaching to the same degree anymore, such as when Polly discussed:

I really appreciate the fact that the support that I was given from [my coaches] was given to me in the way that I'm taught to give support to kids in my class or the kid that I'm tutoring. So the explicit instruction and the behavior-specific praise and all of those supports where you don't want to just sit there and tell them all the things that they're doing wrong. I feel like you guys led by example in the way that I was given feedback, in the way that I was coached.

Some TCs described how they wished their coach had provided less coaching support initially so they could troubleshoot more at the beginning, such as when Shelley noted, “I wish I kind of waited until after the lesson for her to give me feedback.” However, those same TCs characterized the coaching support through other components such as lesson planning and data-based decision-making as helpful. For example, Shelley stated, “I would send [my coach] my lesson, the slides template or whatever. And I would just ask her if this looked good and she would tell me if maybe change this high-frequency word to something that’s related to the Bob books or something like that.”

Importantly, an unintended support of the tutoring program was the sense of community it created among the participating TCs, who often leaned on each other as a tutoring “corp” of sorts. This sense of community resonated in Brenda’s reflection that, “Being able to say, “how’s your kid doing? Do you feel like you’re doing this right?” Those conversations with [other tutors], and the support from them saying, “I did this. Try that.”, was very helpful.” Though unintended, this sense of community served as another resource to turn to and reflect on the experiences, to reflect on teaching practices and solidify new learning. For one TC, this community became so important that she volunteered to tutor in both semesters to be able to continue to grow her own skills, while also supporting her classmates engaging in the experience for the first time. Through this community and the other programmatic supports described, TCs found ways to confront the challenging and constantly evolving instructional context during the 2020-2021 school year to build their teaching knowledge and skills.

“It’s A Kind Of, Like, A Double-Sided Learning With This Reading Tutoring.”

These supports also contributed to the perceived impact TCs felt the program had on themselves and their students. Despite the ways the program could have been improved, TCs described observing growth in their students’ academic skills and their teaching competencies. Some TCs characterized the mutual benefit of the experience in terms of their improved capacity to provide instruction. For example, Mickey opined:

I would encourage everyone to do this reading tutoring. It is definitely a blessing to you, and as you are blessing to the student. It's a kind of, like, a double-sided learning with this reading tutoring. You're getting to help a student. The student is helping you put all of the stuff that you learn throughout the course together in your hand and in practice.

That mutual, “double-sided learning” became the primary motivator for many TCs. Although we did not collect formal measures of TC or student outcomes, TCs were able to reflect on the lessons through which their student progressed across the semester and the formative data from each session, such as the number and types of errors within lesson components. Based on those reflections, TCs often inferred impact on students’ academic skills. Simultaneously, TCs often perceived impacts on aspects of their students’ affective domain as well. For example, Polly described:

But this semester with my kiddo I have, if nothing else, just his confidence has grown. He was already a pretty outspoken little dude when I got him, but just seeing his confidence grow, in the beginning of this semester, he would [have] a little bit more trepidation when he was sounding out words or sounding out letters, sounds and things. Now, he'll literally get on the mic and he'll be like, "You can't stump me today. You're not going to be able to get me." So it's like a little game that we have now too.

In many ways, the perceived gains in student confidence contributed to TCs’ perceptions of their own growth in confidence, as previously noted. Several TCs described how the program structure and support allowed them to develop this instructional proficiency and confidence. Additionally, TCs noted the ways the program uncovered important understandings of their own teacher identity and their future work with children after graduation. For example, Jesse relayed, “I learned that I am capable of doing things like this even though I’m not a teacher yet; I’m getting better and better at actually delivering instruction.” Similarly, LouAnn noted the impact of the program was, “Confidence and that I can do this. This is why I’m here. I can actually educate children who are not my own…I feel I can make a difference.”

This sentiment of increased confidence and the connection to the ability to impact children was echoed across TCs. TCs sometimes attributed “double-sided learning” to affinity shared with students. For example, Mickey reflected:

Whatever the sounds were for the week, I made sure that I looked at videos that were provided by the professors, to make sure that I was saying that sounds right. That helped a lot with my student. And also, since my student and I have kind of similar backgrounds - English is her [second language] - Whenever she struggles for something specific, I can see it in myself, and … think through the things that would help her in that specific instance, which I thought was really, really cool.

Overall, the process of learning how to deliver reading instruction gave TCs the opportunity to see they were capable of teaching, and allowed them to grow confidence in themselves. TCs described the growth they observed in their student(s) alongside growth they noticed in themselves. The tutoring clinical experience provided opportunities for students and TCs to develop skills and confidence as they formed mutually-beneficial relationships.

Discussion

Our work provided an opportunity for TCs to receive training in and implement a reading intervention to first graders in need of supplemental reading instruction during the COVID-19 pandemic. We supported TC implementation through ongoing coaching support. During an unpredictable and constantly changing situation, we had no precedent for how to effectively develop a system that could work for TCs and students alike. To the best of our ability, we piloted a virtual tutoring program and related supports that we believed would help develop TC knowledge and skills related to reading interventions, which in turn would impact learners in our local schools, because we knew that something was better than nothing.

RQ1: How Did TCs Characterize Their Experience Participating in the Program?

We captured perceptions of TCs who served as reading tutors through structured interview. We described their experiences in the program, including the program strengths and opportunities for growth they identified. Overall, four themes emerged from the TC interviews: (a) COVID-19’s impact on educator preparation, (b) experiential learning through tutoring and coaching, (c) the benefit of structured intervention materials, and (d) the impact on TCs and students. First, TCs felt that the immediate impact of the COVID-19 pandemic on the typical clinical experiences they would have received in classrooms. Second, because of the ways that COVID-19 prevented many opportunities for classroom experiences, TCs described how their participation in the tutoring program provided a context for them to implement coursework content with a child, albeit in a virtual context. Third, TCs shared how the intervention manual, the explicit, systematic, and structured approach to teaching reading, and the related technological tools/resources allowed them to focus on implementation and growth rather than planning and creating materials. Last, TCs conveyed the belief that the program benefited both the TCs and students involved.

These four themes mirror the theoretical underpinnings that guided the development of our model for this virtual reading program with coaching support. Through their interviews, TCs expressed a strong understanding of how the COVID-19 pandemic had taken their early experiences in classrooms with students away from them and the impact the pandemic was having on their preparation to be a teacher (Theme 1). Because of this loss of experience, TCs expressed a desire to learn through concrete experiences and application – the first, “concrete experiences” phase of the experiential learning cycle (Kolb, 1984/2014) and closely related to the second theme from this study. As indicated by the third theme of this study, TCs also underscored the importance of having the opportunity to reflect on their practices, such as the conversations they were able to have with their coaches, which aligns with the second, “reflection” phase of the experiential learning cycle (Kolb, 1984/2014). As Ross and Bruce (2007) note in their theory of TC development, having such an external factor (e.g., a coach) serving as a change agent is pivotal to impact teacher self-efficacy above and beyond TCs’ own self-assessment. Finally, though we did not actively and intentionally plan our model to embed the third and fourth phases of Kolb’s (1984/2014) experiential learning cycle (“abstract conceptualization” and “active experimentation”, respectively), TCs frequently reflected on how the perceived impact of the program on themselves and their students (Theme 4) was most noticeable in moments outside of the tutoring context itself. For example, TCs noted a collaborative peer network to bridge across tutoring experiences and coursework and collectively deepen their understanding even more (abstract conceptualization). Similarly, TCs often described how their increased confidence allowed them to view themselves as teachers, capable of generalizing this work into new contexts such as student teaching. These descriptions are perhaps a sign of readiness for active experimentation, the final phase of Kolb’s learning cycle.

RQ2: What Program Elements Did TCs Believe Helped or Hindered Their Preparation to Deliver Reading Interventions to Striving Readers in First Grade?

Although we did not measure program efficacy or implementation fidelity, TCs consistently described the perceived benefit of the program for them and their students. In part, the perceived benefits TCs felt from the program may be related to the ways in which we aligned the program with current calls for education preparation programs, such as the need for initially receiving explicit instruction (Hudson et al., 2021) and, after training, applying coursework content with ongoing feedback in authentic experiences (Brownell et al., 2020; Hindman et al., 2020; Hudson et al., 2021; Leko et al., 2015). Additionally, through our enhanced clinical model, our TCs engaged in repeated interaction with EBPs across the same students and received immediate and consistent feedback through coaching support (Cohen et al., 2020; 2021; Theobald et al., 2022; Thoele & Sayeski, 2022). As Hindman et al. (2020) noted, rehearsal and application of reading EBPs may require field experiences that occur in durations longer than afforded by traditional clinical experiences. To this end, the TCs in our virtual tutoring program implemented intervention multiple times a week for an entire semester, which ensured an experience lasting a long enough duration for deeper application and reflection.

Although TCs frequently mentioned observing their students develop reading skills, they most often articulated how they themselves developed skills, including confidence, self-efficacy, knowledge of reading EBPs, and use of positive behavior supports. These data do not suggest program effectiveness, but illustrate the perceived impact of the program. TCs viewed most of the elements included in the program structure as supportive to their growth, including training, the intervention manual and corresponding resources, coaching support, and an extended program duration. These reflections echoed Hindman et al. (2020), who theorized that having scaffolding supports such as these may begin to bridge the Science of Reading research into practice for TCs during their preparation program and in their initial years as novice teachers. The barriers TCs described were typically environmental, such as noisy classrooms or homes and sporadic student attendance, as well as internal, such as their lack of confidence prior to starting the program. TCs occasionally shared how they overcame barriers and offered suggestions for program improvement. According to TCs, the virtual reading intervention provided opportunities for meaningful rehearsal even with the undesirable, pandemic-related context within which the program was implemented.

Future Research and Limitations

Our work during this pilot year of the reading tutoring program and the subsequent findings from the TC interviews served as a foundation for future iterations of our own development of this enhanced clinical experience. The preliminary findings from this study provide an overview of TCs’ perceptions of the tutoring program. These data allowed us to understand how TCs characterized their experiences and, broadly, what program elements TCs believed served as programmatic supports or barriers that helped or hindered their preparation to deliver reading EBPs. Through future analyses of these interview data, we will analyze the characteristics of the TC experience more deeply, by examining the ways that each programmatic element impacted the TC experience across the different ways that the programmatic element may have been implemented. For example, TCs received coaching through pre-session check ins, on-the-spot coaching during sessions, and post-session debriefs. As analyzed in this study, the data did not allow us to consider the nuances of the TC experiences. We believe those nuances offer important considerations for future work conceptualizing and implementing enhanced clinical experiences.

Unfortunately, given that our work was situated within the context of the constantly evolving height of the COVID-19 pandemic, we did not prioritize collecting implementation fidelity data, formal K-3 student outcome data, or additional TC outcomes related to implementing reading EBPs or self-efficacy. Future researchers examining enhanced clinical experiences should consider collecting these data. First, obtaining fidelity data could provide coaches additional sources of data to inform their primary coaching points each session. Having such data would also complement the TC-reported gains in knowledge, skills, and self-efficacy, as a way to increase our understanding of the effect of such an enhanced clinical experience on TC outcomes. For example, with such data, future researchers could investigate the effect of coaching support on TCs' ability to implement EBPs with fidelity. Second, collecting formal student data could serve as a way to assess the effect of the reading EBP on student outcomes. With such data, future researchers could consider the effect of different models of coaching on TC outcomes, with the potential to explore the cascading effect of these experiences on student reading achievement.

Although the height of the COVID-19 pandemic has passed, this work and the related future work remains relevant to special education for a few key reasons. First, online tutoring remains an ever-present instructional delivery modality (Singer, 2021). One affordance of the pandemic was an increased attention to the infrastructure required to provide such tutoring remotely. With this infrastructure now in place, virtual programs remain and serve as potential ways to accelerate learning in a post-COVID-19 world. Virtual coaching programs may also be an avenue through which educator preparation programs may provide more equitable access to high quality instructional coaching for even remote clinical placements. Of note, the trainings and materials provided during our tutoring program can be used in face-to-face settings, therefore our findings related to the TC experiences and the programmatic supports/barriers may also have applicability and transferability across instructional settings and coaching contexts (e.g., Rowland et al., 2023; Wang et al., 2023).

Finally, though we attempted to maximize the credibility of our analyses through adherence to Brantlinger et al.’s (2005) guidelines, we recognize that our analyses are limited to the scope of the interview context itself. We only conducted interviews at a single time point, and TCs are likely to have shared what was most salient in their mind at the time of the April/May 2021 interview, nearly a semester after the clinical experience for the Fall 2020 TCs and at the end of the clinical experience for the Spring 2021 TCs. Having multiple interviews over time would help tell the story and tension of the perceived supports versus barriers more comprehensively (Denzin & Lincoln, 2017). Future research that considers the ways these perceived supports and barriers evolve across time, both within and beyond the semester-long experience, would deepen the understanding of the TC experience.

Conclusion

More than ever, there has been an urgent call for increased supplemental reading intervention support to mitigate learning opportunity gaps resulting from the COVID-19 pandemic (The Education Trust & MDRC, 2021a; 2021b; 2021c). For example, the Coronavirus Aid, Relief, and Economic Security (CARES) Act of 2020 (PL 116-136) earmarked funds for the explicit purpose of implementing summer and supplemental programs that address the needs of students at greatest risk of poorer outcomes, such as students with a lower socio-economic status (SES) and students with disabilities. The work reported in this study provides preliminary evidence for one way education preparation programs and teacher educators may consider adapting practices for an enhanced clinical model to support TCs’ developing skills and provide necessary supplemental instruction to striving readers. This model could be applied in a virtual context or translated back to a face-to-face context. The initial perceptions shared by this relatively small number of TCs suggest that virtual contexts need not be a barrier to providing TCs with opportunities to practice and receive feedback implementing EBPs to benefit their preparation and, perhaps, to their future students.

Supplemental Material

Supplemental Material - “Double-Sided Learning”: A Qualitative Study of Teacher Candidate Experiences in a Virtual Reading Intervention Clinical Experience

Supplemental Material for “Double-Sided Learning”: A Qualitative Study of Teacher Candidate Experiences in a Virtual Reading Intervention Clinical Experience by Samantha A. Gesel, Jared H. Stewart-Ginsburg, Erin K. Washburn, Corinne R. Kingsbery, Jocelyn H. Lev, Rachel Brooks, and Hayley Owen in Journal of Special Education Technology