Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2026 Jun 16;16:27573. doi: 10.1038/s41598-026-58439-5

The design of Malay sentences for measuring reading performance for the Radner-UKM Malay Reading Charts

Mohd Izzuddin Hairol 1,2,✉, Syafiqah Aqilah Mohd Shafiee 2, Ker Sin Cheng 2, Zainora Mohammed 2,3, Rogayah Razak 4, Sharanjeet Sharanjeet-Kaur 2,3, Wolfgang Radner 5,6,7
PMCID: PMC13538680  PMID: 42303798

Abstract

The Radner Reading Charts for measuring reading performance are available in 11 European languages plus Turkish. There is a need to develop a Malay version, as Malay is spoken by 240 million speakers in Southeast Asia. The study’s purpose is to create and validate Malay sentence optotypes comparable in number and length of words, difficulty, and syntactical structure, for constructing the Malay version of the Radner Reading Charts. Forty Malay sentences were constructed following the methods used for the German Radner Reading Charts. We measured reading time (rt), reading speed (rs), and reading errors for 120 participants (mean age = 22.09 ± 1.28 years). Sentences were statistically standardized by selecting those within rtmean±0.25 × SD and rsmean±0.25 × SD. Validity was analyzed by comparing the sentences’ reading time and reading speed with the 75-word Malay Early Reading Evaluation Text. We selected 24 equally matched sentences, with rsmean=179.14 ± 4.66 words per minute, rtmean=4.83 ± 0.13 s, and mean reading errors of 0.13 ± 0.04 words. The Malay sentences and standard text’s reading speeds were highly correlated (r = 0.81; p = < 0.001). Cronbach’s alpha coefficient was 0.979. The 24 Malay sentence optotypes are highly comparable, with high validity and reliability, and are incorporated in the Radner-UKM Malay Reading Charts for determining reading acuity and speed for young, visually normal adult Malay speakers.

Keywords: Reading performance, Reading speed, Reading acuity, Radner reading charts

Subject terms: Health care, Public health

Introduction

Reading ability contributes significantly to learning capacity, and consequently, children with reading difficulties are compromised in their future academic success1. Reading performance is also one of the major features in several quality-of-life assessments2,3, with reduced reading speed being correlated with reduced quality of life4,5. Furthermore, reading performance is an indicator with which to assess the effectiveness of rehabilitation in patients with low vision6,7. Therefore, assessment of reading performance, with a validated and standardized tool, is crucial for accurate diagnosis and rehabilitation in individuals with reading difficulties.

To date, six reading charts have been designed based on the standards of the Visual Function Committee of the International Council of Ophthalmology (ICO) and the EN ISO 8596 standard, where the calibrated print sizes are logarithmically progressing to allow comparable evaluations of reading acuity and reading speed (see8 for a review). An example is the Bailey-Lovie Word Reading Charts9 which uses unrelated words of different lengths at each size, to take into account that readability can affect word length in patients with age-related macular degeneration. In contrast, the MNRead Charts10 have English sentences arranged in three lines, where each sentence is characterized by its length, defined as 60 characters including spaces. The charts applied the concept of “standard-length word”11,12 where a standard-length word is defined as having six characters, including spaces. Thus, a 60-character sentence consists of ten standard-length words. The use of standard-length words minimizes the scoring differences that would occur in the chart’s 19 sentences, where some sentences have 13 short words, and others have 10 longer words.

For a reading chart, test item standardization and complexity are crucial, as they may affect a reader’s reading acuity and reading speed13,14. To address this need, the Radner Reading Charts were developed with a standard concept, ensuring that each stimulus in the chart, called the sentence optotype, is comparable in terms of lexical difficulty, syntactic complexity, word length, number of syllables, and position of words14. The concept produces minimal variation between sentences, with optimal constant geometric proportions to allow accurate, standardized measurements of reading acuity and speed at every viewing distance. Reading charts developed with this concept have high reliability and validity, test-retest reliability, as well as high inter-chart reliability15.

Of special relevance for our present research focus, is that reading charts with sentence optotypes, have been applied to 12 languages: Danish, Dutch, English, French, German, Hungarian, Italian, Portuguese, Romanian, Spanish, Swedish, and Turkish9,13–21(see22 for a review). These standardized Radner Reading Charts, developed with English, other European languages, and Turkish, as test items, allow comparable assessment of reading performance across languages.

Currently, there are no Radner Reading Charts in a non-European language except for Turkish. A suitable target language for the development of another non-European Radner Reading Charts is Malay, which is widely spoken in populous Southeast Asia. Malay is Malaysia’s official language, with about 20 million native speakers representing about 60% of the country’s population. Furthermore, in large parts of Southeast Asia, including Indonesia, Brunei, Singapore, southern Thailand and the southern Philippines, Malay and its dialects are spoken by about 220 million people.

It would be beneficial to develop test items in the Malay language, which are constructed in the same standardized manner as those in European languages, to achieve standardized measures of reading acuity and reading speed in Malay speakers. The Malay language has a highly transparent alphabetic orthography, with near-perfect and consistent grapheme-phoneme correspondences23.

Among Malay speakers who are learning English as their second language, reading anxiety significantly impedes their comprehension of academic texts24. An additional concern is that many native Malay speakers consider reading materials in English to be only for academic purposes; thus, English may not be widely used outside academic settings25. As it is well recognized that reading performance may differ depending on the language used in the reading material and the mother tongue of the reader26,27 and multilingual speakers have different reading performances depending on their first language when they read a text in their second language28, we can infer that the use of an English-based reading chart to clinically measure the reading performance of native Malay speakers may not result in accurate outcomes28. Therefore, the cross-language differences between Malay and English must be taken into consideration when choosing Malay words for planning Malay reading assessment and intervention programs29.

Thus, there is a need to design a standardized tool for a range of reading and reading-related measures, which would be beneficial for early diagnosis of reading difficulties in Malay30. We are aware of three clinical reading charts that have been developed to measure Malay language reading performance: The Universiti Kebangsaan Malaysia Malay Language Related Word Reading Text Test, the Buari-Chen Malay Reading Chart, and the UiTM Reading Charts. The Universiti Kebangsaan Malaysia Malay Language Related Word Reading Text Test31 consists of two sets of five reading texts with fixed font sizes. The texts are storylines from school textbooks for children in lower and upper primary school (ages seven to 12). Each text is composed of seven to eight lines of sentences, with about 60 characters per line, including spaces.

A more recent development is the Buari-Chen Malay Reading Chart (BCMRC)32, where each stimulus contains six Malay words, arranged in two lines. The words are of one, two, or three syllables. The appropriateness of the contextual sentences was based on the criteria of “within the scope of grammar, composition and arrangement”32. The third Malay clinical reading chart is the UiTM Reading Charts. The charts contain 14 sentences where each sentence has five to ten standard-length words, with 60 characters per sentence33.

There are some limitations to these Malay reading charts. First, the texts in the Universiti Kebangsaan Malaysia Malay Language Related Word Reading Text Test are used to measure reading speed and reading rate, but not reading acuity. Second, although the BCMRC shows accepted reliability with the MNRead Charts32 and the UiTM Reading Charts have consistent repeatability33, a close inspection of both charts reveals that the sentence length, even for the same stimulus size, is not always similar, and the sentence length between two lines is not always uniform32,33.

To address these limitations, the first purpose of our study was to create sentence optotypes in the Malay language which are as similar as possible in terms of word count and length, position of words, syntactical structure, as well as in reading time (reading speed) and difficulty. Secondly, our study also investigated the validity and reliability of the Malay sentence optotypes for evaluating reading performance, including reading acuity, reading speed, reading error, and reading time. Lastly, the selected Malay sentence optotypes were incorporated in the newly developed, logarithmically scaled, Radner-UKM Malay Reading Charts designed for measuring reading acuity and reading speed simultaneously.

Methods

Study participants

Our study participants were 120 undergraduate students (88 females, 32 males) from Universiti Kebangsaan Malaysia with a mean age of 22.09 ± 1.28 years. Participants were recruited using social media and printed advertisements distributed on the university campus. The inclusion criteria included the ability to read and converse fluently in the Malay language and no known self-reported ocular or systemic pathology. Exclusion criteria included any known conditions that would lead to reading difficulties, such as dyslexia. Normal visual and binocular status were verified through standard optometric assessments, including distance visual acuity and binocular vision testing. All participants had distance visual acuity of at least 0.0 logMAR, without any monocular and binocular vision disorders such as amblyopia, suppression and strabismus. However, comprehensive ophthalmic imaging and detailed ocular pathology assessments were not performed.

This research received ethical approval from Universiti Kebangsaan Malaysia’s Research Ethics Committee (UKM PPI/111/8/JEP-2021-132). Informed consent was obtained from all participants following the ethical standards of Good Scientific Practice and the Declaration of Helsinki. All study procedures adhered to the Declaration of Helsinki for research involving human subjects.

Sentence optotypes

Construction of the short Malay sentence optotypes was a collaborative effort by five fluent Malay speakers: a clinical linguist researcher, three academicians specializing in vision health and rehabilitation, and an undergraduate optometry student. Forty-six short sentences in Malay were constructed following the methods used for the German version of the Radner Reading Charts to create ‘sentence optotypes’13. Each sentence consisted of a main clause followed by a relative clause. All sentences were comparable to each other in terms of lexical difficulty, syntactic complexity, word length, number of syllables and word position.

Each sentence had 14 words of a maximum of three syllables arranged in three lines. The first and second lines contained five words, while the third line had four words. Each line had 28 to 30 characters, including spaces. In a single sentence, each word appeared only once.

In line 1 (five words), the first word had two syllables and three letters. The second word was a verb of three syllables with eight letters. For the following three words, each had two syllables with four to six letters. In line 2 (five words), the first word had one syllable with three letters, followed by a word of three syllables with eight letters. This was followed by the relative clause consisting of three, one-to-two-syllable words with four to six letters each. In line 3 (four words), the first word was a two-syllable word with four to five letters, and the remaining words had three syllables with six to eight letters each.

Figure 1 shows two examples of Malay sentence optotypes (with their English translations in italics) constructed based on the rules described above.

Fig. 1.

Fig. 1

Open in a new tab

Two examples of Malay short sentences from the 40 sentences used in the study. The English translation is shown in italics but does not appear during the experiment.

Five to seven sentence optotypes were printed, in Times New Roman font with a type size of 12 points, per piece of plain white DIN A4 paper. The printed paper was placed on a reading stand and covered with a blank piece of paper before the test began. The distance from the reading stand to the participant was 40 cm, and the reading distance was always measured using a measuring tape before each participant began reading, to ensure the accuracy of the distance.

Practice sessions

Practice sessions were conducted binocularly before the actual experiment began. Reading performance was assessed under binocular viewing conditions to reflect natural visual behavior and enhance ecological validity. Reading is a binocular task that relies on coordinated vergence, accommodation, and eye movements. Monocular testing can disrupt these processes and artificially reduce performance. Additionally, binocular summation improves visual efficiency, including contrast sensitivity and letter recognition, which are critical for fluent reading. As the study aims to measure reading time in a visually normal young adult population (19–25 years), binocular assessment provides a more representative measure of real-world visual ability than monocular testing. Participants were asked to read six Malay sentence optotypes to familiarize themselves with the data collection process. At the start of a practice session, they were asked to uncover the paper to reveal one sentence at a time and read each of the six sentences accurately and aloud. The reading was recorded, and from the recording, any reading errors were noted by the experimenter by marking the words read incorrectly. Errors were counted even when immediately corrected by the participant. Reading time was measured using a digital stopwatch, starting when the participants read the first word in each sentence and ending when they read the last word in that sentence.

Data collection

In the actual experiment, participants were asked to read 40 sentence optotypes using the same protocol described above. The six sentences that were used in the practice session were not included in the actual experiment, and data from the practice sessions were not included in the data analysis. Reading time (rt), reading speed (rs), and reading errors were measured for the 40 sentence optotypes. Reading speed in words per minute (wpm) was calculated based on the number of words in a sentence (14 words) and the time needed to read the sentence (14 words × 60 s divided by reading time). Subjects were advised not to stop reading or repeat the part of the sentence if they made a mistake. The procedure was the same for all sentences. Reading errors were recorded by marking the text on a separate form. Errors that were immediately corrected were also noted. A standard luminance of 80 to 90 cd/ m² was used.

To assess the validity of the reading time, reading speed and reading errors obtained with the Malay sentence optototypes, we also measured reading time, reading speed and reading errors with the 75-word-long Malay Early Reading Evaluation Texts as the standard text using the same protocol. The texts were printed in Times New Roman font with a type size of 12 points and single spacing on white DIN A4 paper. The Malay Early Reading Evaluation Text materials are based on previously developed and validated Malay early reading assessment frameworks, which have demonstrated good reliability, construct validity, and normative data in Malaysian children30. These tools assess multiple components of reading, including decoding, fluency, and comprehension, supporting their use in evaluating reading performance. The Malay Early Reading Evaluation Texts were selected to ensure linguistic and cultural appropriateness for the study population. As the study aims to assess functional reading performance in a Malaysian cohort, the use of linguistically relevant continuous text provides a more representative measure of real-world reading ability, despite the lower level of psychophysical standardization compared to sentence optotype charts.

Design of the UKM-Radner Malay Reading Charts

The most equally matched sentence optotypes in reading speed and reading errors were included in the Malay version of the Radner Reading Charts. A sample of the new Radner-UKM Malay Reading Chart is shown in Fig. 2.

Fig. 2.

Fig. 2

Open in a new tab

Detail from the Radner-UKM Malay Reading Chart 1 (1 of 3) illustrating the concept of sentence optotypes. The original size was DIN A4 (210 mm × 297 mm). The print sizes are scaled logarithmically.

The Radner-UKM Malay Reading Charts consist of three charts. Each chart has 14 complete sentences, with print sizes ranging from 1.1 to -0.2 logRAD (logReading Acuity Determination according to ISO 7921:2024) progressing in 0.1 log unit intervals when viewed at a distance of 40 cm. The sentences for each chart were selected and arranged so that the readers do not encounter the same sentence twice within the same chart.

Testing the new UKM-Radner Malay Reading Chart

Fifty-five new participants (mean age: 22.44 ± 1.48 years) who fulfilled the same inclusion criteria were recruited, and we determined reading acuity (defined as the smallest print size of which a participant could read more than 80% of the words correctly) and reading speed for each sentence. Test procedures were performed in a similar setup as described above for the practice session. Only one of three charts was presented to each participant, and the chart selection was randomized. Reading acuity was defined as the smallest print size at which more than 80% of the words were read correctly. This criterion was adopted to ensure that the measured print size reflected functional and meaningful reading performance rather than isolated word recognition or guessing. The use of an accuracy threshold is consistent with established reading assessment approaches, including the RADNER and MNREAD reading chart methodologies, where reading acuity is determined based on the smallest print size that can still be read with acceptable accuracy and fluency.

Data was only collected for the purpose of establishing a standardized test chart for near in the Malay language. As such, data for other older age groups or patients with ocular diseases were not collected, as it was not the aim of the study.

Statistical analyses

For inclusion in the Radner-UKM Malay Reading Charts, the sentences were statistically standardized by selecting those within rtmean±0.25 × SD and rsmean ±0.25 × SD. Independent t-tests were used to compare reading speed and reading time obtained with the Malay Early Reading Evaluation Text and the selected Malay sentence optotypes. Results are reported as t(df), where t represents the test statistic, and df denotes the degrees of freedom for the analysis. Internal consistency was determined by calculating Cronbach’s alpha and the coefficient of selectivity of the Malay sentence optotypes. The cut-off level for statistical significance was set at p < 0.05, two-tailed. Statistical analysis was performed using IBM SPSS ver. 25, IBM Corporation, NY, USA.

Results

We have successfully selected 24 Malay sentence optotypes to be included in the Malay version of the Radner Reading Charts following the protocols described above. Consequently, the Radner-UKM Malay Reading Charts were developed, and reading acuity and reading speeds for different stimulus sizes were measured.

Characteristics of reading the sentence optotypes in Malay

For the 40 sentence optotypes in Malay, the overall mean reading time was 4.85 ± 0.86 s. The mean reading time for individual sentences is shown in Fig. 3. The overall mean reading speed was 178.53 ± 31.10 words per minute (wpm). The mean reading speed for individual sentence optotypes is shown in Fig. 4. The mean reading error was 0.13 ± 0.05 words.

Fig. 3.

Fig. 3

Open in a new tab

Mean reading time for the 40 sentence optotypes. Black circles indicate excluded sentences and white circles indicate selected sentences. Error bars represent ±1SD.

Fig. 4.

Fig. 4

Open in a new tab

Mean reading speed for the 40 sentence optotypes. Black circles indicate excluded sentences and white circles indicate selected sentences. Error bars represent ±1SD.

Selection of sentences for the Radner-UKM Malay Reading Chart

For inclusion in the Radner-UKM Reading Charts, the sentence optotypes were statistically standardized by selecting those within rtmean±0.25 × SD and rsmean ±0.25 × SD. Thus, to be selected for the Radner-UKM Malay Reading Chart, the reading time of a sentence optotype had to be within 4.64 to 5.07 s (interval = 4.85 ± 0.25 × SD), and the reading speed had to be within 170.76 to 186.54 wpm (interval = 178.53 ± 0.25 × SD).

Based on the specific reading time and reading speed intervals parameters, 24 of the 40 sentence optotypes met the inclusion criteria. These were sentences 2, 3, 5, 6, 7, 9, 10,11, 13, 14, 17, 18, 19, 21, 23, 24, 28, 29, 30, 31, 32, 34, 37 and 39. These sentence optotypes had a mean reading time of 4.83 ± 0.13 s, a mean reading speed of 179.14 ± 4.66 wpm, and a mean reading error of 0.13 ± 0.04 words. The remaining 16 sentences were excluded because their reading times and reading speed intervals did not fulfil the inclusion criteria.

Comparison between the new selected Malay sentence optotypes and the Malay Early Reading Evaluation Text

For the 75-word Malay Early Reading Evaluation Text, the results were: (i) mean reading time 26.36 ± 3.34 s, (significantly longer than that for the new selected sentence optotypes (4.83 ± 0.13 s, t(24)=-211.829, p < 0.001)); (ii) mean reading speed 182.47 ± 31.89 wpm, (not significantly different from that for the sentence optotypes (179.14 ± 4.66 wpm, t(24)=-0.847, p = 0.405)); and (iii) mean reading error for the text was 1.45 ± 0.52 words, (significantly higher than that for the selected sentence optotypes (0.13 ± 0.04 words, t(24)=-15.752, p < 0.001)).

The correlation between the reading speed of the 24 selected sentences optotypes and the Malay Early Reading Evaluation Text was strong and significant (r = 0.84, p < 0.01, n = 150). Reliability analysis of the reading speed of the sentence optotypes revealed an overall Cronbach’s alpha coefficient of 0.986 for all 40 sentences, and 0.979 for the 24 selected sentences. The corrected item-total correlation for the 24 sentences varied between 0.741 and 0.860.

Reading acuity and reading speed with the Radner-UKM Malay Reading Charts

The mean reading acuity obtained with the Radner-UKM Malay Reading Chart for 55 new participants was − 0.055 ± 0.07 logRAD. The mean reading speed was 176.48 ± 31.46 wpm for the largest print size (1.1 logRAD), and decreased to 143.55 ± 36.37 wpm for the third smallest print size (0.0 logRAD). The smallest print sizes (-0.1 and − 0.2 logRAD) were challenging to read; only 35 participants (63.34%) and four participants (7.27%) could read at least 80% of the words correctly at these print sizes, respectively. The mean reading speeds at -0.1 and − 0.2 logRAD were 115.27 ± 30.30 wpm and 76.84 ± 8.25 wpm, respectively (Table 1).

Table 1.

Mean reading speed in words per minute (wpm) obtained for each of 14 print sizes on the Radner-UKM Malay Reading Chart measured in 55 participants.

Print size
(logRAD)		 Number of participants able to read at least 80% of the words correctly (n) Reading speed (wpm)
(mean ± SD)
1.1 55 176.48 ± 31.46
1.0 55 174.80 ± 32.52
0.9 55 179.29 ± 22.36
0.8 55 170.62 ± 25.95
0.7 55 161.45 ± 25.44
0.6 55 166.71 ± 35.16
0.5 55 169.89 ± 33.13
0.4 55 161.12 ± 34.02
0.3 55 156.78 ± 30.24
0.2 55 154.32 ± 29.15
0.1 55 163.55 ± 36.37
0.0 55 143.55 ± 36.37
-0.1 35 115.27 ± 30.30
-0.2 4 78.84 ± 8.25
Open in a new tab

Discussion

In this study, we report on 24 sentence optotypes in the Malay language that were developed to be as similar as possible to each other (in terms of word count, word length and position, syntactic structure, and lexical difficulty), to be included in the Malay version of the Radner Reading Charts. An important aspect of the Radner Reading Charts concept is sentence standardization. The 24 selected Malay sentence optotypes had a high Cronbach’s alpha coefficient of 0.979, and corrected item-total correlation ranging between 0.741 and 0.860, which was well above 0.3, the threshold established by34,35 indicating that the sentence optotypes were statistically similar and acceptable.

To ensure that the sentence optotypes were as comparable as possible from the perspective of sentence readability, we also employed two additional parameters, namely reading time and reading speed. In previous studies, very short reading time and reading speed intervals have been shown to be adequate for selecting sentence optotypes13,16,18,20,21. Therefore, we followed this approach, and in our present study, we set the value as an overall mean of +/-0.25 SD16. The 24 Malay sentences optotypes read within these intervals were selected for inclusion in the newly developed Radner-UKM Malay Reading Charts. Indeed, it has been demonstrated that reading charts using equally selected sentence optotypes represent accurate tools for determining reading acuity and speed22.

In our present study, the structure and characteristics of the Malay sentence optotypes were kept as constant as possible to control for text difficulty across sentence optotypes. In 1976, Carver introduced a model for analyzing text difficulty. He suggested that the average word length - calculated by dividing the total number of characters, including spaces, by the number of words - increases as text passages become more difficult11,12. Carver has demonstrated that using this concept to calculate an overall mean word length reduces the variation in reading speed across a wide range of text difficulty levels. He introduced the term “standard-length word”, which is defined as a word with six character spaces. However, this concept applies only to English and does not account for the complexity of the content of a text. For example, a paragraph intended for medical experts may have a low standard word length but still be more difficult to understand than the word length would suggest.

The standard-length word concept has been used in the MNRead Charts10, where 60 characters, including spaces, are considered equivalent to 10 words of six characters, including spaces. However, the actual word length and number of words per sentence differ and vary between sentences. Our present study shows that standard-length Malay words applied in single sentences take a significantly shorter reading time and have fewer reading errors than when reading the long, standard Malay texts, whilst having comparable reading speed14,36–39.

Malay reading charts, such as the BCMRC, also use short six-word sentences and six unrelated words as the test stimuli32. However, there are several differences in definitions and methodological constructs between the Malay sentences or groups of words in the BCMRC and the Malay sentence optotypes selected in the present study. The first difference relates to the construction of the words used as the test stimuli. In the BCMRC, the sentences’ physical lengths are not always constant even for the same print size; moreover, for a single sentence arranged in two lines, the length of the half-sentence in the second line is not always uniform. In contrast, the Malay sentence optotypes we selected have a similar number of words and characters, and a similar number of syllables and physical length. The sentence optotype’s physical length is a crucial parameter to be considered when constructing reading test items for reading performance analyses, as reading speed for longer-structured Malay sentences is slower than for shorter-structured Malay sentences40. The second difference relates to sentence construction. The sentences we present have a main clause followed by a relative clause. This represents a still simple, but more complex, text construction than main clauses only, and thus is closer to real reading conditions. In our opinion, the reliability and validity of the 24 selected Malay sentence optotypes are therefore optimised for analysing reading performance with single sentences, and consequently for measuring reading acuity16.

Reading performance can be measured by several parameters, such as reading error41,42, reading comprehension43, reading rate44,45, reading acuity46, and reading speed47,48. The Radner-UKM Reading Charts provide a standardized Malay-language reading assessment tool capable of measuring established reading performance parameters, including reading errors, reading rate, reading acuity, and reading speed. The primary contribution of the Radner-UKM Reading Charts lies in the development of a standardized Malay-language version for assessing established reading performance parameters, including reading acuity, reading speed, reading rate, and reading errors13,14. The charts do not measure reading comprehension, as they were designed to investigate the functional properties of the eye and not the brain. The Malay version of the Radner Reading Charts may serve as a useful tool for reading assessment among young, visually normal Malay-speaking adults, although further validation in broader age groups and older populations with ocular diseases is required before wider clinical application. Nevertheless, reading performance is influenced not only by ocular factors but also by higher-level visual and cognitive processing, and standardized reading charts may therefore provide additional insight into visual–cognitive function in certain clinical contexts. Nonetheless, the four reading parameters obtained with the Radner Reading Charts provide detailed information about reading performance. For example, the changes in reading speed observed with the Radner-UKM Malay Reading Charts (Table 1) are likely to be due to changes in print size rather than the length of the text14,36 as the other construct aspects were kept constant. Thus, in the case of a fixed print size, deviations in reading speed from values observed in visually normal adult Malay speakers are more likely attributable to factors such as cognitive, oculomotor, attentional or sensorimotor interactions37–39 rather than the physical construct of the test stimuli.

This study has several limitations. One limitation was that during the measurement of reading speed, there was a possibility of head movements while reading. Ideally, a head rest and a chin rest would limit head movements, but these features would introduce an unnatural reading posture, and thus are likely to affect the measured reading parameters. In our present study, variations in head movements were minimized by observing the participants, and ensuring that their reading distance was consistent while they read the sentences aloud. Head movements were also unlikely to have significantly affected reading speed measurements due to the participants’ normal visual function and the short reading material. However, monitoring head movements may be relevant in future studies involving patients with visual field defects who use compensatory reading strategies. The second limitation of the present study is that it relates to the use of the 75-word Malay Early Reading Evaluation Text as the comparison reference. As the linguistic characteristics and standardisation process of the 75-word Malay Early Reading Evaluation Text were not comprehensively described in the literature, its suitability as a direct comparison reference may be uncertain. Differences in lexical or syntactic complexity between the texts could have influenced reading performance outcomes independently of chart design. Future studies should therefore incorporate more rigorously standardised Malay reading materials or conduct formal linguistic equivalence analyses when comparing reading charts. The third limitation of the present study was that it evaluated internal consistency but did not assess test–retest reliability or measurement stability over time. Future studies should include repeated testing sessions and intraclass correlation coefficient analyses to further establish the reliability of the chart. The fourth limitation of the present study is that comprehensive ophthalmic examinations and ocular imaging were not performed to objectively confirm the absence of ocular pathology. Visual and ocular health status were determined primarily through participant history and basic visual screening procedures. Therefore, the possibility of undetected subclinical ocular abnormalities cannot be completely excluded. The present validation was conducted exclusively among young, visually normal Malay-speaking adults. Therefore, the findings should not be generalized to older individuals, children, or patients with ocular diseases without further clinical validation.

Nevertheless, the selection of the Malay sentence optotypes and the measurement of reading speed with the Radner-UKM Malay Reading Charts have added important data for internationally comparable reading tests. The addition of Malay as the thirteenth language, applying the sentence optotype principle, may support standardized measurement of reading performance among Malay-speaking readers. The present study was limited to young adults with normal visual function and therefore does not provide information regarding age-related changes in reading performance or the applicability of the chart in ocular disease populations. In addition, the relationship between corrected visual acuity and reading acuity was not specifically investigated. Since reading performance may be influenced by aging and various ocular conditions, future studies should evaluate the performance of the Radner-UKM Malay Reading Charts in individuals with visual impairment, including retinal disease, glaucoma, low vision, and other ophthalmic conditions, as well as across different age groups. Such investigations would further establish the chart’s clinical utility in ophthalmology.

Conclusion

The present study presents the concept of sentence optotypes for the Malay language. The 24 selected Malay sentence optotypes have been incorporated into the Malay version of the Radner Reading Charts, the Radner-UKM Malay Reading Charts. The charts, consisting of valid and reliable test stimuli, may be useful for assessing reading performance in young, visually normal Malay-speaking readers.

Acknowledgements

Special thanks to the volunteers who participated in this study. We also thank Sarah L Poynton for proofreading the manuscript.

Author contributions

All authors contributed to the study’s conception and design. All authors read and approved the final manuscript. Material preparation was performed by all authors. Data collection was performed by S.A.M.S. and K.S.C. Data analysis was performed by all authors. Supervision of data collection was conducted by S.S., Z.M., R.A.R., and M.I.H. The first draft of the manuscript was written by M.I.H. and S.A.M.S. Previous versions of the manuscript were commented on by S.S., Z.M., R.A.R., M.I.H., AND W.R.

Data availability

All datasets are publicly available at https://doi.org/10.6084/m9.figshare.25516348.v1.

Declarations

Competing interests

Wolfgang Radner receives royalties for the Radner Reading Charts. The other authors have no potential conflict of interest to disclose.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Hernandez, D. J. Double Jeopardy: How Third Grade Reading Skills and Poverty Influence High School Graduation. Annie E Casey Foundation (2011). https://eric.ed.gov/?ID=ED518818
  • 2.Leske, D. A., Hatt, S. R., Liebermann, L. & Holmes, J. M. Evaluation of the adult strabismus-20 (AS-20) questionnaire using Rasch analysis. Invest. Ophthalmol. Vis. Sci.53, 2630–2639 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Sabri, K., Knapp, C. M., Thompson, J. R. & Gottlob, I. The VF-14 and psychological impact of amblyopia and strabismus. Invest. Ophthalmol. Vis. Sci.47, 4386–4392 (2006). [DOI] [PubMed] [Google Scholar]
  • 4.Hazel, C. A., Petre, K. L., Armstrong, R. A., Benson, M. T. & Frost, N. A. Visual function and subjective quality of life compared in subjects with acquired macular disease. Invest. Ophthalmol. Vis. Sci.41, 1309–1315 (2000). [PubMed] [Google Scholar]
  • 5.Owsley, C., McGwin, G., Lee, P. P., Wasserman, N. & Searcey, K. Characteristics of low-vision rehabilitation services in the United States. Arch. Ophthalmol.127, 681–689 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Xiong, Y. Z., Calabrèse, A., Cheong, A. M. Y. & Legge, G. E. Reading acuity as a predictor of low-vision reading performance. Invest. Ophthalmol. Vis. Sci.59, 4798–4803 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Gall, C. & Sabel, B. A. Reading performance after vision rehabilitation of subjects with homonymous visual field defects. PM R. 4, 928–935 (2012). [DOI] [PubMed] [Google Scholar]
  • 8.Radner, W. Reading charts in ophthalmology. Graefe’s Archive for Clinical and Experimental Ophthalmology vol. 255 1465–1482 Preprint at (2017). [DOI] [PMC free article] [PubMed]
  • 9.Bailey, I. L. & Lovie, J. E. The design and use of a new near-vision chart. Optom. Vis. Sci.57, 378–387 (1980). [DOI] [PubMed] [Google Scholar]
  • 10.Mansfield, J. S. & Legge, G. E. Lawrence Erlbaum Associates, Mahwah NJ,. The MNREAD acuity chart. in Psychophysics of Reading in Normal and Low Vision (ed. Legge, G.) 1–32 (2007). 10.1201/9781482269482-10 [DOI]
  • 11.Carver, R. P. Word length, prose difficulty, and reading rate. J. Lit. Res.8, 192–203 (1976). [Google Scholar]
  • 12.Carver, R. P. Reading Rate: A Review of Research and Theory (Academic, 1990).
  • 13.Radner, W. et al. A new German Reading Chart for the simultaneous evaluation of reading acuity and reading speed. Klin. Monbl Augenheilkd. 213, 174–181 (1998). [DOI] [PubMed] [Google Scholar]
  • 14.Radner, W. et al. The validity and reliability of short German sentences for measuring reading speed. Graefe’s Archive Clin. Experimental Ophthalmol.240, 461–467 (2002). [DOI] [PubMed] [Google Scholar]
  • 15.Radner, W., Radner, S. & Diendorfer, G. A new principle for the standardization of long paragraphs for reading speed analysis. Graefe’s Archive Clin. Experimental Ophthalmol.254, 177–184 (2016). [DOI] [PubMed] [Google Scholar]
  • 16.Radner, W. & Diendorfer, G. English sentence optotypes for measuring reading acuity and speed - The English version of the Radner Reading Charts. Graefe’s Archive Clin. Experimental Ophthalmol.252, 1297–1303 (2014). [DOI] [PubMed] [Google Scholar]
  • 17.Calossi, A., Boccardo, L., Fossetti, A. & Radner, W. Design of short Italian sentences to assess near vision performance. J. Optom.7, 203–209 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Maaijwee, K., Mulder, P., Radner, W. & Van Meurs, J. C. Reliability testing of the Dutch version of the Radner Reading Charts. Optom. Vis. Sci.85, 353–358 (2008). [DOI] [PubMed] [Google Scholar]
  • 19.Munch, I. C., Jørgensen, A. H. R. & Radner, W. The Danish version of the Radner Reading Chart: Design and empirical testing of sentence optotypes in subjects of varying educational background. Acta Ophthalmol.94, 182–186 (2016). [DOI] [PubMed] [Google Scholar]
  • 20.Alió, J. L. et al. Design of short Spanish sentences for measuring reading performance: Radner-Vissum test. J. Cataract Refract. Surg.34, 638–642 (2008). [DOI] [PubMed] [Google Scholar]
  • 21.Rosa, A. M. et al. Development of the Portuguese version of a standardized reading test: The Radner-Coimbra Charts. Arq. Bras. Oftalmol. 79, 238–242 (2016). [DOI] [PubMed] [Google Scholar]
  • 22.Radner, W. Toward an internationally accepted standard for reading charts. Prog Retin Eye Res101, (2024). [DOI] [PubMed]
  • 23.Borleffs, E., Maassen, B. A. M., Lyytinen, H. & Zwarts, F. Cracking the code: The impact of orthographic transparency and morphological-syllabic complexity on reading and developmental dyslexia. Front. Psychol. . (2019) 10.3389/fpsyg.2018.02534 (2019) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zin, Z. M. & Rafik-Galea, S. Anxiety and academic reading performance among Malay ESL learners. J. Pan-Pacific Association Appl. Linguistics. 14, 41–58 (2010). [Google Scholar]
  • 25.Muhamad, S. N., Azmi, M. N. L. & Hassan, I. Reading interest and its relationship with reading performance: A study of english as second language learners in Malaysia. Humanit. Social Sci. Reviews. 7, 1154–1161 (2019). [Google Scholar]
  • 26.Jamaludin, K. A., Alias, N., Khir, M., DeWitt, R. J., Kenayathula, H. B. & D. & The effectiveness of synthetic phonics in the development of early reading skills among struggling young ESL readers. School Eff. School Improv.27, 455–470 (2016). [Google Scholar]
  • 27.Jamaludin, K. A., Alias, N. & Johari, R. Research and trends in the studies of phonological knowledge and reading development: A review on selected journals. MOJES: Malaysian Online J. Educational Sci.2, 27–36 (2018). [Google Scholar]
  • 28.Winskel, H. Learning to read in multilingual Malaysia: a focus on Bahasa Melayu, Tamil and Chinese. GEMA Online J. Lang. Studies20, (2020).
  • 29.Lee, L. W., Low, H. M. & Mohamed, A. R. A comparative analysis of word structures in malay and English Children’s stories. Pertanika J. Social Sci. Humanit.21, 67–84 (2013). [Google Scholar]
  • 30.Lee, J. A. C. et al. An Early Reading Assessment Battery for Multilingual Learners in Malaysia. Front. Psychol.11, 1700 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Omar, R., Bauri, N., Knight, V. F. & Mohammed, Z. Pembangunan Ujian Teks Bacaan Perkataan Berkait Bahasa Melayu Universiti Kebangsaan Malaysia. Jurnal Sains Kesihatan Malaysia. 13, 51–56 (2015). [Google Scholar]
  • 32.Buari, N. H. & Chen, A. H. Buari-Chen Malay reading chart (BCMRC): Contextual sentence and random words 2-in-1 design in Malay language. Pertanika J. Sci. Technol.25, 135–150 (2017). [Google Scholar]
  • 33.Buari, N. H., Yusof, N. H., Mohd-Satali, A. & Chen, A. H. Repeatability of the Universiti Teknologi MARA Reading Charts. Bangladesh J. Med. Sci.14, 226–240 (2015). [Google Scholar]
  • 34.Cronbach, L. J. Coefficient alpha and the internal structure of tests. Psychometrika16, 297–334 (1951). [Google Scholar]
  • 35.Streiner, D. L. & Norman, G. R. Health Measurement Scales: A Practical Guide to Their Development and Use. Health Measurement Scales: A Practical Guide to their Development and Use (Oxford University Press, 2008). 10.1093/acprof:oso/9780199231881.001.0001 [DOI]
  • 36.Stifter, E. et al. Reliability of a standardized reading chart system: Variance component analysis, test-retest and inter-chart reliability. Graefe’s Archive Clin. Experimental Ophthalmol.242, 31–39 (2004). [DOI] [PubMed] [Google Scholar]
  • 37.Gavril, L., Roșan, A. & Szamosközi, Ș. The role of visual-spatial attention in reading development: a meta-analysis. Cogn. Neuropsychol.38, 387–407 (2021). [DOI] [PubMed] [Google Scholar]
  • 38.Paterson, K. B. et al. Effects of normative aging on eye movements during reading. Vis. (Switzerland). 4, 7 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Broadway, J. M., Franklin, M. S. & Schooler, J. W. Early event-related brain potentials and hemispheric asymmetries reveal mind-wandering while reading and predict comprehension. Biol. Psychol.107, 31–43 (2015). [DOI] [PubMed] [Google Scholar]
  • 40.Jufri, S., Buari, N. H. & Chen, A. H. Text structures affect reading speed. Social Manage. Res. J.13, 118–131 (2016). [Google Scholar]
  • 41.Douglas, G., Grimley, M., Hill, E., Long, R. & Tobin, M. The use of the NARA for assessing the reading ability of children with low vision. Br. J. Visual Impairment. 20, 68–75 (2002). [Google Scholar]
  • 42.Khalid, N. M., Buari, N. H. & Chen, A. H. Comparison of oral reading errors between contextual sentences and random words among schoolchildren. Int. Educ. Stud.10, 47–55 (2016). [Google Scholar]
  • 43.Savolainen, H., Ahonen, T., Aro, M., Tolvanen, A. & Holopainen, L. Reading comprehension, word reading and spelling as predictors of school achievement and choice of secondary education. Learn. Instr. 18, 201–210 (2008). [Google Scholar]
  • 44.Legge, G. E., Pelli, D. G., Rubin, G. S. & Schleske, M. M. Psychophysics of reading-I. Normal vision. Vis. Res.25, 239–252 (1985). [DOI] [PubMed] [Google Scholar]
  • 45.Brysbaert, M. How many words do we read per minute? A review and meta-analysis of reading rate. J. Mem. Lang.109, 104047 (2019). [Google Scholar]
  • 46.Colenbrander, A. Reading acuity-an important parameter of reading performance. Int. Congr Ser.1282, 487–497 (2005). [Google Scholar]
  • 47.Mansfield, J. S., Legge, G. E. & Bane, M. C. Psychophysics of reading. XV: Font effects in normal and low vision. Invest. Ophthalmol. Vis. Sci.37, 1492–1501 (1996). [PubMed] [Google Scholar]
  • 48.Nárai, Á., Amora, K. K., Vidnyánszky, Z. & Weiss, B. Predicting reading speed from eye-movement measures. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) vol. 12663 LNCS 453–466 (2021).

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

All datasets are publicly available at https://doi.org/10.6084/m9.figshare.25516348.v1.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES