INTRODUCTION

Reading is among the most fundamental literacy skills acquired during the early years of elementary school as it provides the foundation for learning across all subject areas. Early reading is more than just recognizing letters. Learners need to connect letters with sounds, combine phonemes into syllables, and produce meaningful words. Reading competence as the initial competence in the Merdeka Curriculum of Indonesia is conceptualized as the core learning outcome for first-grade students. It is expected to be developed by having the students learn accurate decoding and comprehension as the prerequisites for their further academic achievement. Students who do not learn these basic skills in the early grades are likely to have continuing problems learning throughout their school years. Beginning readers are introduced to meaningful consonant-vowel (CV-CV) patterned words like mata, bola, kuda, and sapu. The words consist of open syllables that are relatively easy to segment and blend, making them suitable to introduce systematic decoding skills before students move on to more complex syllable structures. Successful reading of CV-CV words depends on mastery of a number of interrelated cognitive processes such as phonological awareness, grapheme-phoneme correspondence, decoding and blending . Deficits in any of these processes can have a major impact on reading acquisition in the early stages of literacy development.

The challenge becomes more pronounced for students who are at risk of being slow learners. Students with SLD generally need more direct, structured, and repeated instruction and tend to process information more slowly than their peers without SLD. While they may not have been formally diagnosed with intellectual disabilities, they often have trouble with letter–sound correspondence, blending syllables into meaningful words, maintaining attention to instruction, and remembering newly learned reading skills. Hence, they need instructional strategies that are tailored to their learning characteristics rather than traditional classroom practices. The result of the preliminary observation at SDN 007 Malinau Selatan Hilir found two students of the first grade shown the characteristics related to the risk of slow learning. The baseline assessment revealed that the students’ reading accuracy of meaningful CV–CV patterned words was merely an average of 44% which was far below the established mastery criterion of 80%. The students would confuse consonant sounds, relied heavily on letter-by-letter decoding, and had considerable difficulty in blending syllables into words. The results of this study indicate a need for immediate instructional intervention to address the root causes of their reading difficulties.

In theory, early reading success is the result of systematic instruction in phonological awareness, decoding and blending skills that ultimately result in automatic word recognition. Ehri (2020) explains that beginning readers build meaningful reading through the repeated associations of graphemes, phonemes, and semantic representations. Moats (2020) reminds us that explicit phonics instruction is the bedrock of effective decoding instruction. Students who have difficulty with grapheme–phoneme correspondence and blending require systematic phonics instruction; they do not just need to see printed words. In practice, however, early reading instruction in many classrooms still relies heavily on traditional approaches where teachers introduce words via textbooks or the blackboard and have students imitate pronunciation. This instruction often does not explicitly teach letter–sound relationships, use multimodal engagement, provide structured repetition, and give corrective feedback immediately. Because of these drawbacks, it is less effective to use with slow learners who need intensive instructional scaffolding with visual, auditory, and kinesthetic learning experiences.

Multisensory phonics instruction supported by Smart TV was used in this study to address these instructional needs. The instructional approach combines explicit phonics teaching with segmenting, blending, structured repetition and gradual visual–auditory prompting. The reason for choosing Smart TV is that it allows for the simultaneous presentation of text, images, animations, and audio on a large interactive display. This enables teachers to give repeated demonstrations and immediate feedback while keeping the attention of students. Smart TV is not simply a technological device but an instructional medium, which offers the opportunity for systematic phonics instruction to be taught consistently. Previous research has shown that phonics instruction and digital learning media can be effective individually for improving beginning reading skills. However, there are still few studies on the integration of multisensory phonics strategies such as segmenting, blending, repetition and visual-auditory prompting using Smart TV in a Classroom Action Research design among first-grade students at risk of slow learning. The present study is to describe the implementation of Smart TV-assisted multisensory phonics instruction and to examine its effectiveness in improving the ability to read meaningful CV–CV patterned words among first-grade students at risk of slow learning at SDN 007 Malinau Selatan Hilir because of the gap.

METHOD

This study applied Classroom Action Research (CAR) based on the model developed by Kemmis, McTaggart, and Nixon (2014). CAR was chosen because it allows teachers to improve their teaching through a cycle of planning, action, observation, and reflection that is repeated in response to real problems in the classroom. The study was conducted to overcome the reading difficulties of first-grade students who are at risk of being slow learners. The implementation of Smart TV-assisted multisensory phonics instruction and its continuous improvement were done in two cycles of action. This research was conducted at SDN 007 Malinau Selatan Hilir, North Kalimantan in the second semester academic year 2025/2026. Participants were two first grade students identified as being at risk for slow learning based on their baseline reading assessments, classroom observations, and consultation with the classroom teacher and parents. The “at risk of being slow learners” label was based on academic indicators and classroom observations, not formal psychological diagnoses.

The intervention was implemented in two cycles, each cycle consisting of four instructional meetings and a subsequent evaluation session. Each cycle included planning, action, observation and reflection. The instructional intervention included systematic phonics instruction, such as letter-sound correspondence, segmenting, blending, repeated practice, visual-auditory prompting, prompt fading, and multisensory learning activities. The Smart TV served as a teaching tool for delivering synchronized visual and auditory stimuli, enabling teachers to offer explicit phonics instruction, repeated demonstrations, and interactive learning experiences. The results of reflection in Cycle I led to an improvement of the instructional design for Cycle II. Data collection involved reading performance tests, classroom observations and documentation. In the preliminary stage, at the end of Cycle I and at the end of Cycle II, reading performance was evaluated with a rubric that measures reading fluency, accuracy of pronunciation and reading independence. Student learning activities and teacher instructional performance were assessed by means of observation sheets during the intervention. Expert judgment was used to validate all research instruments prior to data collection. The data were analyzed descriptively quantitatively and qualitatively by comparing students’ reading achievement, learning activities, teacher performance, patterns of reading errors and the level of instructional support needed for the preliminary condition, Cycle I and Cycle II. A participant was considered to have succeeded in the study when they could read meaningful CV–CV patterned words independently with a minimum reading accuracy of 80%.

RESULT AND DISCUSSION

Description of the Preliminary Condition (Pre-Cycle)

Prior to the instructional intervention, a baseline measure was administered to assess the participants’ ability to read meaningful CV–CV patterned words. The results indicated Participant RP read with 45% accuracy and Participant JA read with 43% accuracy, giving the average reading accuracy of 44%. The results suggest that both participants were reading well below the predetermined success criterion of 80% independent reading accuracy. The baseline assessment observations showed that the participants were experiencing similar challenges in reading. There is a suggestion that RP did not yet have a stable letter-sound correspondence as it often confused the sounds of several consonant letters. In contrast, JA tended to read by sounding out individual letters and was not able to blend syllables into meaningful words fluently. Also, both participants found it difficult to concentrate during traditional classroom instruction which was mainly based on textbooks and blackboards. These findings indicate that their reading problems were not limited to low reading accuracy but extended to decoding, blending, and sustained attention on task during instruction. The preliminary findings provided the basis for designing the instructional intervention implemented in this study. To address the identified learning needs, Smart TV-assisted phonics instruction was introduced, integrating letter–sound correspondence, segmenting, blending, repeated practice, visual–auditory prompting, prompt fading, and multisensory learning activities. This instructional approach was designed to provide explicit, systematic, and repetitive learning experiences that would help students establish stable grapheme–phoneme connections, improve their ability to blend syllables into meaningful words, and gradually reduce their dependence on teacher support during reading activities.

Cycle I Results

The Cycle I intervention was implemented over four instructional meetings on 13, 15, 20, and 22 April 2026 and the reading performance assessment was administered during the fourth meeting. The intervention procedures were based on the intervention framework in Table 3.2 and included systematic phonics instruction, segmenting, blending, repeated practice, and visual–auditory prompting supported by Smart TV. The main goal of Cycle I was to build students’ letter–sound correspondence and to enhance their capacity to blend syllables into meaningful words following the CV–CV pattern.

Table 1. Reading Performance of Participant RP in Cycle I

Table 1 indicates that Participant RP scored an average of 67% on the reading test, Participant JA scored an average of 57% and the overall average was 62%. Both participants showed significant improvement in reading performance after the multisensory phonics instruction assisted by Smart TV compared with the preliminary condition (44%). However, the average reading accuracy was still below the preset mastery criterion of 80% that indicated the intervention did not yet reach the intended level of success at the end of Cycle I.

Table 2. Reading Performance of Participant JA in Cycle I

Analysis of individual components of the assessment revealed that both participants performed relatively better when reading meaningful words with visual support. The lowest scores were consistently observed when reading meaningful words without picture support. Participant RP scored 60% on this task and Participant JA scored 50%, which indicates that both students relied heavily on outside visual cues and teacher support in order to decode. The results indicate that automatic recognition of meaningful words was not yet mature, especially in the absence of visual context. The results are compatible with the theoretical framework of Chapter II. Repeated establishment of grapheme-phoneme connections gradually develops automatic word recognition (Ehri, 2020). In Cycle I, it seemed that both participants were still in the first phase of this developmental process, where letter–sound correspondence and blending skills were just beginning to develop. Thus, decoding was slow and effortful and students continued to need visual–auditory prompts and teacher guidance to read meaningful CV–CV patterned words correctly. Student learning activities also demonstrated progressive improvement through the four instructional sessions of Cycle I. As shown in Table 3, the average level of student participation increased from 51% in the first meeting to 82% in the fourth meeting, which indicates the increasing engagement with the multisensory phonics activities delivered through Smart TV.

Table 3. Student Learning Activities Across Meetings

Average student learning activity was 66%, with Participant RP at 74% and Participant JA at 59%. Teacher instructional performance is rated as good, with an average implementation score of 83%, and has been steadily improving from 73% in the first meeting to 94% in the last meeting of the cycle. These results indicate that the instructional procedures were implemented with fidelity and the students’ engagement increased during all instructional procedures . However, the intervention did not lead to adequate improvement in independent reading accuracy to reach the desired success criterion . The reflection at the end of Cycle I showed that the intervention had positive initial effects but needed further refinement. Accordingly, the study continued to Cycle II, which included several improvements, such as more intensive phonics, segmenting and blending practice, systematic implementation of prompt fading, expansion of meaningful CV–CV vocabulary closely related to students’ daily experiences, and enhanced multisensory instruction with individualized reinforcement. These changes were expected to help students develop better decoding skills and become more automatic and independent in their word recognition.

Results of Cycle II

The second cycle was performed in four instructional meetings on 4, 6, 11, and 13 May 2026. The reading performance assessment was performed in the fourth meeting using the same instrument as in the first cycle to ensure an objective comparison. Instructional improvements in this cycle included more practice with phonics, segmenting and blending, systematic use of prompt fading, more meaningful CV–CV vocabulary and increased multisensory learning activities.

Tabel 4. Reading Accuracy Results in Cycle II – Participant RP

Table 5. Reading Accuracy Results in Cycle II – Participant JA

As can be seen in the tables 4 and 5 Participant RP’s average reading accuracy was 90% which falls under very good while Participant JA scored 81% in the good category. The combined mean reading accuracy of both participants was 85.5%, which exceeds the predetermined success criterion of 80%. Thus, the instructional intervention was successful and the classroom action research was ended at the end of Cycle II. Also student learning activities improved significantly in Cycle II. The average level of student participation increased from 66% in Cycle I to 93.49% in Cycle II, and reached 100% in the fourth meeting. Likewise, the teaching performance of the teachers increased from an average of 83% in Cycle I to 97.40% in Cycle II. Table 4.6 provides a summary of teacher instructional performance across meetings.

Tabel 6. Summary of Teacher Instructional Performance Across Meetings

The Cycle II improvements suggest that the changes made after the reflection phase of Cycle I (more intensive phonics, segmenting and blending practice, systematic prompt fading, increased exposure to meaningful CV–CV vocabulary, and enhanced multisensory instruction) contributed to strengthening students’ decoding skills, and helping to promote greater reading independence.

Tabel 7. Summary of Reading Accuracy Improvement from the Preliminary Condition to Cycle II

As can be seen in Table 7, the average of reading accuracy increased gradually from 44% in the preliminary condition, to 62% in Cycle I and to 85.5% in Cycle II. This improvement shows that the use of Smart TV-assisted multisensory phonics instruction significantly improved the reading ability of first-grade students at risk of being slow learners in reading meaningful CV–CV patterned words. The quantitative results were further illustrated in Figure 4.1 showing the reading accuracy of each participant and Figure 4.2 showing the trends of student learning activities and teacher instructional performance over all instructional meetings.

Discussion

The results of this study indicate a gradual improvement of the reading ability of the two participants through the intervention cycles. Average reading accuracy increased from 44% in the preliminary condition to 62% in Cycle I and to 85.5% in Cycle II, which was higher than the predetermined mastery criterion of 80%. This pattern of incremental gains indicates that the improvements in reading did not happen immediately following the introduction of the instructional medium but rather as a result of systematic development of decoding skills and automatic word recognition through repeated instructional practice. The findings support the theoretical framework proposed in this study which suggests that beginning reading develops progressively through explicit and systematic instructions. Results obtained in Cycle I showed measurable improvement, but still did not meet the success criterion. As explained by Ehri (2020), automatic word recognition is preceded by the effortless recognition of words, which emerges from the repeated formation of grapheme-phoneme correspondences. In Cycle I, participants had experienced only four instructional sessions and systematic repetition and visual-auditory prompting were still in the early stages of strengthening letter-sound correspondence and blending skills. Therefore, both participants still relied heavily on teacher support, particularly when reading meaningful CV–CV patterned words without picture support, where Participant JA achieved only 50% accuracy. These findings support the findings of Torgerson et al. (2019) who argue that the effectiveness of phonics instruction is a function of not only the instructional approach but also the consistency and quality of its implementation. Results of teacher observation in Cycle I (83%) also show that several components of instruction such as individualized reinforcement and systematic monitoring of reading errors had not been implemented optimally yet. This limits the effectiveness of the intervention overall.

The intervention in Cycle II met the success criterion defined after several instructional refinements based on the reflection made after Cycle I. Both participants were able to read more independently with more intensive phonics instruction, increased opportunities for segmenting and blending practice, systematic prompt fading, and more exposure to meaningful CV–CV vocabulary. Reading independence was significantly improved. The scores of the participants were 80%–90% (Cycle II) compared to 50%–70% (Cycle I). These results are consistent with the findings of Goldfeld et al. (2022) who emphasize that early reading intervention needs to be explicit and systematic and delivered with adequate intensity and frequency to be effective. Moreover, the enhancement of the teacher’s instructional performance from 83% in Cycle I to 97.40% in Cycle II is in line with the findings of Dilgard, Hodges, and Coleman (2022) which found that teachers’ ability to systematically implement phonics instruction consistently contributes directly to successful reading intervention outcomes. One of the important findings of this study is related to the role of Smart TV in the instructional process. According to the conceptual framework of the introduction, the improved reading performance should not be credited to the Smart TV as a technology device. Instead, the improvement was attributed to the systematic phonics instruction, such as letter-sound correspondence, segmenting, blending, repeated practice, and visual-auditory prompting, which was successfully facilitated through Smart TV. The technology supported students’ development of decoding through the provision of repeated demonstrations and multi-sensory learning experiences that included synchronized presentation of text, images, and audio. This interpretation aligns with Birsh (2011) and Jones and Christensen (2023), who maintain that technology enhances reading instruction when it acts as a medium to support evidence-based instructional strategies, rather than being the intervention itself.

The improvement of reading performance was also followed by the significant increase of students learning activities (66% to 93.49%) and teacher instructional performance (83% to 97.40%). These results indicate that higher levels of student engagement were associated with higher levels of instructional fidelity during phonics, segmenting, blending, and repeated reading activities. However, in accordance with the research design, these process variables should be viewed as supporting indicators that explain how the intervention produced positive outcomes rather than the primary criteria for determining instructional success. Overall, the findings indicate that the increase in reading meaningful CV–CV patterned words was a function of the interaction of explicit and systematic phonics instruction and an instructional medium that consistently facilitated its implementation. Findings showed that Smart TV helped to provide multisensory instruction, repeated practice and visual–auditory prompting but the primary factor responsible for the improvement was the structured phonics intervention. All indicators of success were achieved in Cycle II, so the classroom action research was stopped after the second cycle.

CONCLUSION

The findings of this study show that the Smart TV-assisted phonics instruction that integrates systematic phonics, segmenting, blending, repeated practice, visual–auditory prompting, and multisensory learning activities could effectively improve the reading ability of first-grade students at risk of being slow learners to read meaningful CV–CV patterned words. The reading accuracy increased from 44% in the preliminary condition to 62% in Cycle I and 85.5% in Cycle II, exceeding the predetermined mastery criterion of 80%. The improvement was followed by increased student learning activities and teacher instructional performance . This indicates that the success of the intervention was primarily due to the consistent implementation of explicit and systematic instructional strategies with Smart TV as a supporting instructional medium rather than a sole determinant of learning outcomes. Therefore, teachers should be encouraged to use structured phonics-based instruction supported by appropriate audiovisual media, schools should be encouraged to strengthen teachers’ capacity to implement evidence-based early reading interventions, and future studies should be encouraged to involve larger samples, longer intervention periods, and more diverse educational settings to further examine the effectiveness and sustainability of this instructional approach.