Customising Table Tennis Coaching for Novice Players with Physical Disabilities: A Case Study

Study Design

This is a longitudinal training study on four individuals with different physical disabilities. A double pre-test design (separated two weeks apart) was adopted to establish baseline references, given the impracticality of recruiting matched controls with comparable disabilities. Participants underwent eight sessions of one-to-one supervised intervention (60 minutes per session, twice per week) during which they were taught various table tennis skills using pedagogy guided by the CLA framework. A post-test and a retention test (four weeks after the post-test) were conducted to assess participants’ physical performance and psychological well-being.

Participants

This study was approved by the Nanyang Technological University Institutional Review Board (Reference number IRB-2024-316). Participants provided written informed consent form prior to taking part in the study. Participants completed a background survey to determine their eligibility for the study. The inclusion criteria were: (1) male or female between 16 and 60 years old; (2) has a physical impairment; and (3) novice table tennis players (i.e. less than three times per year). Participants were excluded if they have any of the following: (1) intellectual disabilities, (2) undergone arm, trunk, or leg surgery within the past three months; (3) muscular or skeletal injuries within the past three months requiring more than seven days of rest, or (4) injury, discomfort, or pain at the time of study.

Four participants (one male, three females) with varying types and severity of physical disabilities, of which two were wheelchair users, were recruited for the study (Table 1). Their physical impairments were self-report, including short statue, cerebral palsy, impaired muscle power and/or passive range of motion, hypertonia, athetosis, and paralysed waist down. All participants were novices in table tennis and did not receive any formal training. Two participants were involved in other sports (weightlifting, air pistol) while the other two had no prior experience in sports.

Test Protocols

Eligible participants were invited to undergo two pre-tests, eight supervised training sessions, a post-test and a retention test. The test items included functional reach, table tennis skills and well-being index across all pre-, post- and retention test sessions. Additional table tennis skill tests were conducted at the end of lessons 2, 4 and 6 to track the learning of skills over time. At post-test, a post-interventional survey was also administered to gather feedback of the training programme. Details of the test items were described below.

Functional Reach Test. The functional reach test has been used in para-table tennis research to quantify the maximum area a player’s playing arm can cover on the table (Kong & Ma, 2024; Lim et al., 2015; Tang et al., 2019). Being able to reach a larger area is advantageous because it enables the player to return angled balls more easily. In the present study, we standardised the starting position by setting distance between the player and the centre of the table as 30 cm. Participants used a chalk stick to draw three large semicircles — each as large as possible — with their playing arm (Figure 1a). The area enclosed in the semi-circle drawn was termed “sweep area” which represented the maximum area reachable by the participants when they move their arm. A top-view digital image of the drawings was taken (Figure 1b), covering the area of the half-table to allow for perspective correction using the DxO ViewPoint 3 software. From this image, the edge of each semicircle was manually traced and the encompassed “sweep area” was determined using ImageJ, a public domain Javi image processing programme (https://imagej.net/ij/). The average of three trials per session were calculated for each participant.

Figure 1

Table Tennis Skill Tests. Forehand and backhand drives were used indicate table tennis skill levels, as these two types of strikes were mostly frequently tested in para-table tennis research (Kong & Ma, 2024). To test the stroke accuracy, participants received a serve from the coach in which they have to hit it with the forehand or backhand stroke 30 times each into the opposite diagonal half of the table. They were instructed to target a small area marked by an A3-size paper placed at the corner of the table (Figure 2a, 2b). If the ball was returned on the A3-size paper, they would score two points. If the ball landed within the boundary of the opposite half of the table, they would score one point. Balls that did not clear the net, landed on the same side of the table, or fell outside of the table would not score any points. The set-up for both the forehand and backhand test was generally the same and only differed in terms of the placement of the A3-size paper. For each test session, the maximum scores would be 60 (30 balls × 2 points) for forehand drives and 60 (30 balls × 2 points) for backhand drives. A video camera was used to record the table skill tests to allow offline scoring of the performance afterwards. This allows the coach to focus on ball feeding and interaction with the player during the tests.

Figure 2

Well-being Survey. Participants were asked to complete the World Health Organisation-Five Well-Being Index (WHO-5) questionnaire. This instrument was chosen because it is short, comprising only 5 questions, and is a good measure of participants’ well-being (Topp et al., 2015). Each question has a raw score of 0 to 5, resulting in a total raw score ranging from 0 to 25. Following common practice, the raw score is multiplied by 4 to obtain a percentage score ranging from 0 to 100% for easier interpretation. A percentage score of 0 represents worst possible, whereas 100% represents the best possible quality of life. A percentage score below 50% is often used as a cut-off point for poor mental well-being, suggesting the need for further assessment.

Intervention

Participants underwent eight sessions of one-to-one supervised intervention in their preferred table tennis training venues located in the community. The target training frequency was twice per week over four weeks, though a few adjustments (1–3 times per week) were made to accommodate the participant’s schedule. Each training session last approximately 60 minutes. All training was conducted by the same coach who was an able-bodied physical education student teacher. The coach was a skilled table tennis player himself and has experienced in coaching able-bodied learners. He also received formal training on the instructional methods and strategies in physical education, including the principles of CLA and nonlinear pedagogy. He travelled to the different training venues and provided the sport equipment required for training.

The one-to-one training plan was guided by the CLA (Pinder & Renshaw, 2019), aiming to facilitate individuals with physical disabilities to learn and acquire new skill via manipulating constraints. For instance, a coach could introduce a task constraint by controlling the amount of spin on the ball, requiring the learner to adjust and return the ball accordingly to counter the spin, thereby encouraging the exploration of novel movement solutions (Galatti et al., 2019). Compared with taking reference from able-bodies players, CLA-informed coaching could provide more specific training tailored to the learning needs of individuals with physical disabilities. Pertaining to table tennis, the constraints can be related to individual (e.g., different physical impairments), task (e.g., rules and equipment) and environment (e.g., distraction from other players in the venue) factors.

The intervention began with an empathy phase, which is an important step for the coach to better understand the perspectives of the learners, informing subsequent practice design to best benefit the learners (Pinder & Renshaw, 2019). First, participants completed a short survey and engaged in a discussion with the coach about areas they wished to improve, perceived obstacles or challenges, and the assistance they required (Table 2). From the survey results, most of the perceived obstacles or challenges in learning table tennis were physical factors such as mobility, balance and fitness. All participants indicated that they would need most assistance or support with footwork and mobility. Next, the coach would observe the participants’ techniques and skill levels during the two pre-test sessions without providing corrective instructions. Based on the survey, discussion and observation, the coach would design a training plan that he deemed most suitable for the participant’s learning needs.

Since the participants were new to table tennis, they were introduced to both handshake grip and penhold grip, which are the two common methods of holding the racket. Participants were allowed to explore and choose which grip they were most comfortable with, and all of them preferred the handshake grip. Over the eight training sessions, the task difficulty and the variety of table tennis skills would progress based on the individual’s response through the use of manipulating constraints. Since the training plan and progression were highly individualised, details of the lesson-by-lesson activities were reported in Table 3. Emphasis was placed on the players’ self-correction without over-reliance on the coach’s direct external feedback. Below are some examples of the coaching strategies implemented:

While there was no fidelity checklist in this exploratory study, the coach would monitor the progress and adjust the lesson plans to motivate and teach participants how to improve their table tennis skills. After each lesson, the coach recorded details of all activities and documented any modifications to the lesson plan, including the reasons for any necessary adjustments.

Data Analysis

Given the case study design, descriptive individual and group data of the four participants were presented to show changes over time across all assessments. The outcome measures include functional reach, table tennis skills, and well-being measures. Additionally, the reliable change index (RCI) at post-test and retention test timepoints were determined to check whether the magnitude of the changes for a given participant was statistically reliable (Jacobson & Truax, 1991). The RCI was calculated in the following manner: First, the test-retest reliability coefficient r was determined using Intraclass Correlation Coefficient [ICC (2,1)] from the Pre-test 1 and Pre-test 2 datasets. The average value of the two pre-test scores was then used as the baseline reference. Next, the standard error of measurement (SEM) and standard error of the difference (S_diff) were calculated from the standard deviation (SD) of the baseline scores and r:

Lastly, the RCI between the score at baseline (x₁) and a later timepoint (x₂) was computed using the following formula:

We interpreted the RCI at 90% confidence threshold (±1.645) considering the exploratory nature of this study with heterogenous participants. If the RCI value exceeds the threshold, the observed changes would be considered statistically reliable.

Functional Reach

There was no reliable improvement (RCI < 1.645) in functional reach across all participants at post-test and retention test (Table 4). The test-retest reliability at baseline was moderate (ICC = 0.638). At Pre-test 1, the two wheelchair players exhibited the largest (P3) and smallest (P4) “sweep area” covered by their playing arm.

Table Tennis Skills

The test-retest reliability at baseline was good (ICC = 0.826) for the forehand and moderate for the backhand drives (ICC = 0.701). Across the entire intervention period, all participants generally improved in both forehand and backhand drive accuracies over time (Figure 3). Some statistically reliable changes in forehand/backhand drives (RCI > 1.645) were observed in all participants, either at post-test, retention test or both (Table 4). Relative to the baseline, forehand scores increased by 31% (from 19 to 25) at post-test and by 51% (to 28) at retention test. Similarly, backhand scores improved by 53% (from 20 to 31) at post-test and by 58% (to 32) at retention test.

Figure 3

Well-being Index

Self-reported well-being index demonstrated good test-retest reliability at baseline (ICC = 0.877), with WHO-5 score ranging from 56% to 84% at baseline (Table 4). Overall, this index remained fairly constant and seemed unaffected by the intervention. No statistically reliable changes were observed at post-test or retention test (RCI < 1.645).

Post-intervention Survey

The post-intervention survey was conducted to evaluate the participants’ perceptions and satisfaction with the intervention. All participants indicated “Yes” when asked “Will you continue to play table tennis in the future?” They enjoyed playing table tennis very much (n = 2) or quite a lot (n = 2), and perceived the training as effective (n = 3) or somewhat effective (n = 1). After completing the training, they rated their improvement as very much (n = 1) or quite a lot (n = 3). All participants were very satisfied or satisfied with their forehand, backhand and pushing skills. Most (3 out of 4) were very satisfied or satisfied with blocking and service, but one participant indicated neutral for these two skills. In the open question for suggestions, one participant expressed that it “will be great to play with other person with disabilities or train together.” Another participant provided feedback that he/she would need a quieter environment for more effective learning.

Functional Reach

Participants did not demonstrate reliable improvement in their functional reach after the training programme. Having a smaller “sweep area” makes it harder for participants to return the ball, especially those hit to wider angles. Functional reach ability may be a key factor in gameplay situations but played a lesser role in the stroke accuracy tests of the present study, in which the ball was consistently served by the coach to approximately the same area. The present study involved both standing and wheelchair players, with “sweep area” ranging from 0.56 to 0.93 m² (Table 4). These results are comparable to those reported in previous studies using a similar protocol to assess functional reach in para-table tennis athletes (Lim et al., 2015; Tang et al., 2019). Lim et al., (2015) reported smaller “sweep area” for Class 1 wheelchair players (0.56 ± 0.08 m²) who had more severe impairments and functional limitation compared with Class 2 wheelchair players (0.64 ± 0.04 m²). However, there are notable difference in the “sweep area” between Pre-test 1 and Pre-test 2 in two participants (P1 and P3, Table 4). The large between-day variation is consistent with functional reach test exhibiting the lowest test-retest reliability among other measures. This is despite efforts to standardise the start position as 30 cm from the table and to take the average of three trials.

It is interesting to note that wheelchair players, who had severe lower limb impairments, do not necessarily reach a lesser range than standing player as seen in Participant P3 who out-performed both P1 and P2. The functional reach ability also depends on equipment constraints such as wheelchair design and the use of assistive accessories. For instance, Tang et al. (2019)’s case studies on two wheelchair para-table tennis athletes showed an improvement in “sweep area” by 11.7% (from 0.72 to 0.80 m²) when using a new grip and 94% (from 0.62 to 1.2 m²) for a new Velcro harness with bungee cord. Although the present study focused mostly on individual and task constraints associated with the player, future work can investigate if a customised design modification to the player’s wheelchair may further enhance stability and the execution of various table tennis skills.

Motor Skill Development

Previous research have shown that para-table tennis players adopt different techniques compared with able-bodied controls in executing forehand and backhand drives (Kong & Yam, 2022; Yam et al., 2021). This indicated that the para-players likely adopted compensatory strategies to overcome their functional limitations such as reduced trunk mobility or limited joint range of motion. As such, conventional training methods designed for able-bodied players may not be suitable for individuals with physical disabilities. For example, footwork and mobility are critical for successful performance in able-bodied table tennis (Lam et al., 2019) but standard footwork training cannot be directly applied to wheelchair players or standing players with lower-limb impairments. There is currently no established framework to guide sport coaching for individuals with disabilities. To support the diverse learning needs of individuals with disabilities, one study proposed the concept of manipulating constraints in table tennis coaching (Galatti et al., 2019); however, it did not implement or evaluate an actual coaching intervention. The present study pioneered a longitudinal intervention to coach table tennis to individuals with physical disabilities. Their improvements in forehand and backhand drives provide empirical evidence that novice players with varying types and severities of physical disabilities can benefit from CLA-informed sport pedagogy in effectively learning how to play table tennis.

Kissow (2015) postulated that physical competence achieved through physical activity may serve to help people with physical disabilities to stay socially connected, establish an ability to participate in activities and stay independent. In the present study, all participants improved their forehand and/or backhand stroke accuracy after eight sessions of customised training catered to their needs (Figure 3, Table 4). Given the lack of other training intervention studies on para-table tennis, direct comparison with the literature is not possible. In a similar sport, tennis, Favoretto et al. (2020) conducted an eight-week adapted programme with adults with physical disabilities and reported significant improvements in tennis skills over time. These improvements were observed consistently across participants of varying ages and disability types, aligning with the findings of the present study.

It is important to acknowledge that motor learning abilities may vary across age groups, levels of maturity, and types of disability. The present study recruited participants across a wide age range (22–42 years) and included individuals with diverse impairments (e.g., short stature, cerebral palsy, paraplegia, and muscle weakness). Overall, the customised training implemented can be considered successful in addressing their individual needs. For example, Participant P1 who scored the lowest in forehand drive at pre-tests showed marked improvement in this skill during the intervention (Figure 3a, Table 4), eventually finishing with the highest scores at post- (score = 31, RCI = 5.1) and retention tests (score = 34, RCI = 5.8). Another example is Participant P4, who played on a wheelchair and was very weak in backhand drive at baseline. With customised coaching, she was able to improve her backhand skills to a similar level as the other three participants at the end (Figure 3b, Table 4) with reliable changes at post-test (RCI = 2.8) and retention test (RCI = 2.6). It is worth highlighting that for both cases, the learning of the skill progressed most sharply during the early intervention phase after lesson 2 and was sustained throughout and after the training programme. This observation parallels with the learning of the cueing movement in pool billiards, which is also an upper-limb dominant task, whereby novice participants improved most rapidly in the first 50 trials during a one-visit experiment of 200 repeated shots (Haar et al., 2020).

Many previous studies on para-table tennis used a robot to deliver balls at fast speeds to test the player’s ability to return the balls accurately (Galas et al., 2023; Kong & Yam, 2022; Van Biesen et al., 2010; 2012; Yam et al., 2021). Such tests were developed to test the technical proficiency of well-trained competitive players and not suitable for less proficient players. Considering the participants in the present study are novices to table tennis, we did not use a robot to deliver the shots but to have the coach feeding the balls to specific areas. This approach is easy to implement, requiring only to mark an A3-size target area at the corner. The performance can be recorded using a regular mobile phone or basic camera, without the need for specialised equipment. The scoring procedures are straight-forward (0, 1 or 2 points) and can be performed by almost anyone, even those without prior knowledge of table tennis. This simple skill test has worked well for our participants with different types of physical disabilities. The test results are also able to detect changes over time for both forehand and backhand techniques. Sport instructors can consider implementing similar skill tests to regularly monitor learners’ progression over time and to adjust the training plan as needed.

Psychological Factors

Individuals with disabilities often face multiple barriers, including limited access to coaching, financial constraints, restricted venue accessibility, and transportation difficulties, which make it challenging to engage in physical activity and maintain an active lifestyle (Kim et al., 2025; Nikolajsen et al., 2021). The present longitudinal study required strong commitment from the participants as the programme lasted over three to four months. It is heartening to see that all four participants, despite encountering some difficulties in transportation and accessing the training venues, have completed the entire study without any dropouts. Following the intervention programme, all participants were satisfied with the improvement and mastery of key table tennis skills including forehand, backhand and pushing. All participants found the training enjoyable and effective, and expressed a strong desire to continue playing in the future. This positive feedback suggests that the customised training programme has been successful in cultivating interest in sport participation and enabling people with disabilities to enjoy sport.

The findings also provide evidence that a carefully adapted one-to-one training programme is a promising approach to introduce sports to novice learners. If such effort could be scale up to encourage active lifestyles via sport participation, it will be beneficial for promoting good health and well-being among individuals living with disabilities (Brown et al., 2014; Krahn, 2011).

The WHO-5 scores, which indicated one’s quality of life, varied among the participants but no scores fell below 50%, a suggested cut-off for poor mental well-being (Topp et al., 2015). While the WHO-5 is short and easy to administer, it may not be the most suitable tool for detecting changes in well-being following an exercise intervention. Alternative instruments that are more specific to sport or disability contexts may be more relevant.

Understanding learner’s preference and needs are critical in the design and planning of sport training programme, especially for individuals with disabilities (Pinder & Renshaw, 2019). During the empathy phase of the study, most participants shared concerns related to individual physical challenges such as quickness, balance, footwork, mobility and balance. After the participants experienced the training programme, they provided interesting and invaluable inputs on other aspects beyond physical concerns. Instead of receiving training from the coach alone, one participant would like to play or train together with other persons with disabilities. This reflects that social interaction and peer support are important for sport participation. The present study rented community training venues that were convenient for the participants; however, these shared spaces were also used by others during the training sessions. One participant mentioned the need for a quieter environment to support more effective learning. This feedback highlights the importance of carefully considering environmental constraints emphasised by the CLA. Individuals with disabilities may be more easily distracted by nearby activities such as balls unintentionally entering the training area from adjacent tables. The presence of onlookers may also cause discomfort or self-consciousness, which could affect learners’ confidence and engagement (Lingsom, 2012; Yeam, & Brooke, 2016). Future inclusive sport programmes should therefore consider not only physical barriers but also social and environmental factors to better support learning for people living with disabilities (Mears et al., 2024).

Limitations

There are several limitations to the present study. First, the sample size was small with different disability types, and there was no control group of a comparative pedagogy. As such, the present study was unable to determine whether the observed gains are attributable to disability type, CLA principles or simply practice. Extending from the present work, future studies could compare different inclusive pedagogy approaches on a larger sample size with a baseline comparison of participants’ physical independence such as Functional Independence Measure and the Barthel Index.

Second, the small magnitude of changes in the WHO-scores may be attributed to the individual differences in response to intervention, small sample size, as well as the time needed to observe noticeable enhancements (Smith & Merwin, 2021). Nevertheless, the post-intervention survey results highlight that participants perceived the programme positively, which is crucial for nurturing a lifelong active lifestyle among individuals with disabilities (Aitchison et al., 2022).

Third, we did not analyse the movement techniques or ball strategies associated with the table tennis strokes. In motor skill acquisition of racket sports, kinematic information can be useful to understand different movement strategies (Lee et al., 2014).

Fourth, the scoring criteria of the table tennis skill tests may be over-simplistic. Participants aiming for the smaller area at the corner for 2 points may miss only marginally by putting the ball outside of the table, resulting in 0 point. Demarcating the table with more zones or quantifying amount of deviation from the target may allow better discrimination.

Lastly, this is an exploratory study focusing on the sport of table tennis, which may limit the generalisability of the findings to other sports.