Preliminary study of the effect of Incremental Rehearsal with a morphological component for teaching Chinese character recognition

Abstract

The current study examined the effectiveness of Incremental Rehearsal (IR) for teaching Chinese character recognition using a single-case experimental design. In addition, a morphological component was added to standard IR procedures (IRM) to take into account the role of morphological awareness in Chinese reading. Three kindergarten students in Hong Kong who were learning Cantonese-Chinese were taught Chinese characters with IR and IRM over six weeks using two ABAB designs. The study found that both IR and IRM effectively increased retention and maintenance of Chinese characters.

Keywords

Chinese reading incremental rehearsal learning Chinese characters morphology for Chinese paried associative learning

It is becoming increasingly clear that a relatively large number of Chinese children struggle to learn to read, but reading intervention research conducted with English speakers does not generalize to Chinese because the two languages are fundamentally different (Lau & Chan, 2003). Chinese differs from English in its (a) logographic and morphosyllabic properties, and (b) analytical nature and semantic transparency on both the character and word levels. We will describe each of these properties below.

Logographic and morphosyllabic properties

Chinese has a logographic writing system, in which graphemes are symbols that embody meaning. The significance of the Chinese logographic writing system is illustrated by contrasting it with the phonetic writing system in English. The English alphabet is based on one-on-one mapping between its basic units of writing or graphemes (i.e., alphabetic letters) and basic units of sound (i.e., phonemes), which do not have semantic value on their own. In contrast, the basic units in written Chinese (i.e., characters) are linked to entire syllables consisting of onset, rime and tone. This morpheme-syllable-grapheme correspondence has been called the morphosyllabic property of Chinese language (DeFrancis, 1989).

Cantonese is a Chinese dialect, in which there are six basic tones for each syllable. Some syllables have three additional tones. Syllables have different meanings when pronounced in different tones. For example, ‘風 /fung1/’ (wind) and ‘縫 /fung4/’ (to sew) have the same onset (/f/) and rime (/ung/), but distinct tones and meanings. Furthermore, characters can share identical onsets, rimes and tones but have different orthographic structures and meanings. Returning to the example of ‘風 /fung1/’ (wind), some of its homophones include ‘封 /fung1/’ (to seal), ‘峰/fung1/’ (peak), ‘蜂/fung1’/ (bee), ‘豐/fung1/’ (bountiful) and ‘瘋 /fung1/’ (insanity). The average tonal syllable corresponds with approximately three regularly used characters of distinct meanings and orthographic presentation in Cantonese (Chow et al., 2008). In contrast, most English words typically have one homophone only, and triple-homophone words like ‘pair’, ‘pear’, and ‘pare’ are rare (Hino, Kusunose, Lupker, & Jared, 2013).

Analytic and semantic transparency

Mair (1991) proposed two strategies used by Chinese learners to cope with the aforementioned abundance of homophones in Chinese language, both of which are related to the high semantic transparency and analytic nature of Chinese language. First, homophones can be differentiated from each other by their unique orthographic forms. Second, the distinct semantics of homophones becomes more apparent when the homophone is presented in combination with other characters to form a multi-character word.

On the character level, there are strong links between orthographic structure and meaning. Among modern-day Chinese characters, 85% are phono-semantic characters, also called phonetic compounds (Perfetti & Tan, 1998). Phono-semantic characters are composed of a semantic radical and a phonetic radical (Shu, 2003). Semantic radicals may be related to the meaning of the whole character, such that characters sharing the same semantic radical may belong in the same semantic category. To illustrate, the radical for to jump, to run, and to kneel, as verbs related to the feet, share a common the semantic radical for foot on their left side.

Some semantic radicals are in fact pictographic characters when they are presented in isolation. Shu (2003) estimated a total of 190 semantic radicals in the Chinese writing system. Phonetic radicals have potential connections with how the whole character is pronounced. Like their semantic counterparts, some phonetic radicals can also stand alone as pictographic characters (Shu et al., 2003). There are approximately 1100 phonetic radicals in written Chinese (Shu, 2003).

Most phono-semantic characters have semantic radicals on the left and phonetic radicals on the right (Yan, Zhou, Shu, & Kliegl, 2012). There are many phono-semantic characters with semantic radicals unrelated to their meaning, deemed by scholars as ‘morphologically opaque’ (Shu, 2003, p. 279). The relevance of information provided by semantic radicals can vary from 100%, meaning that all of the characters containing that semantic radical belong to the semantic category denoted by the radical (Jin, 1985), to 65% (Fan, 1986).

Phonetic radicals are more unreliable sources of information than semantic radicals, with only 26.3% of all phono-semantic characters containing phonetic radicals that accurately connote their pronunciation (Fan, Gao, & Ao, 1984). The percentage of reliable phonetic radicals is even lower among high-frequency phono-semantic characters, at 18.5% (Zhu, 1987). The awareness of the functions and positions of various components within a character is called orthographic knowledge or character morphology (Nagy et al., 2002; Packard et al., 2006). Orthographic knowledge has been reported as a predictor of Chinese character reading and sentence comprehension for first, third, and fifth grade children learning to read Chinese (Ho, Ng, & Ng, 2003). Additionally, Ho and colleagues (2003) reported significant, large and positive correlations between first- and third-grade students’ knowledge of phonetic radicals and their reading performance of Chinese characters (r = 0.87) and words (r = 0.73). A significant, small, positive correlation was also found between a measure of semantic-radical knowledge and scores on character reading (r = 0.33) and sentence comprehension (r = 0.34).

A Chinese speaker who is able to decipher the meaning of new words by its constituting morphemes is said to have morphological awareness. Morphological awareness is defined as conscious knowledge of morphemic structure within word compounds, and the ability to apply that pattern to manipulate or reflect on morphemes in the formation and comprehension of new words (Carlisle, 1995). Morphological awareness contributes substantially to Chinese literacy development (McBride-Chang, Chow, Zhong, Burgess, & Hayward, 2005). McBride-Chang, Shu, Zhou, Wat, and Wagner (2003) found that morphological awareness accounted for unique variance in Chinese character recognition among kindergarten and second grade students, even when the effects of phonological awareness, age, speeded naming vocabulary, and speed of processing were controlled. Morphological awareness significantly predicted second graders’ Chinese reading performance in Hong Kong (β = 0.23) and Beijing (β = 0.27), but measures of phonological awareness did not (β = .18 for Hong Kong; β = 0.16 for Beijing; McBride-Chang et al., 2005).

Chinese language and reading

Given the logographic nature of Chinese, visual skills may play an important role in the acquisition and recognition of Chinese characters (Tong, McBride-Chang, Wong, Shu, Reitsma, & Rispens, 2011). Paired-associative learning (PAL; Ho, Chan, Tsang, Lee, & Chung, 2006) may be an effective approach to learning Chinese characters because the connections between orthographic form and pronunciation are arbitrary in Chinese, unlike phonetically based English orthography. Chinese characters cannot be decoded like English words, but instead a visual stimulus (i.e., character) triggers the retrieval of a verbal response (i.e., pronunciation), to which it is associated from long term memory. Performance on measures of visual pairing was the most powerful predictor of reading ability in Chinese for students from Taiwan and Hong Kong, but did not predict reading performance of British students learning English (Li, Shu, McBride-Chang, Liu, & Xue, 2009).

Other interventions emphasize both character- and word- level morphology. Packard et al. (2006) provided Mandarin-speaking, first grade students in Beijing with training in character- and word- morphology by (a) explicitly analysing the functions of semantic and phonetic radicals within characters with students, (b) drawing attention to the semantic contributions of individual characters within multi-morpheme words, and (c) raising awareness of other words containing the same character. Students in the intervention group outperformed their peers in two of the three measures of morphological awareness, and morphological awareness accounted for unique variance in student growth and level in character-writing tasks. The centrality of visual-verbal PAL and morphological awareness to Chinese reading, and the malleability of character- and word-level morphology, are important factors to consider in effective Chinese reading interventions.

Incremental Rehearsal

Given the importance of visual-verbal pairing in learning Chinese character recognition, interventions that strengthen the association between characters and their pronunciations may be effective. Incremental Rehearsal (IR; Tucker, 1988) is a flashcard drill method that may achieve visual-verbal pairing for Chinese character recognition because it involves presentation of visual stimuli as prompts for verbal responses. In IR, known (K) and unknown (U) material are typically rehearsed in a ratio of 9:1. In each cycle of IR, one unknown stimulus (U1) is presented and taught. The student is asked to respond to U1, and then to a known stimulus (K1). After that, the stimuli are rehearsed in the sequence of U1, K1, K2, and then U1, K1, K2, K3, and so on, such that one more known stimulus is presented and responded to by the student after each presentation of K1. One cycle of IR is complete when the presentation of U1 is followed by K1 to K9. This is when K9 is removed from the end of the deck, and a new unknown stimulus, U2, is added as the first stimulus. The interventionist then presents, teaches and requests the student’s response to U2, and then rehearses U1 and K1 to K8 as known stimuli. The cycle repeats until the target number of unknown stimuli has been taught.

IR has been shown to be effective for teaching various subject matter to students of various ages, disability statuses and linguistic backgrounds. Previous research found strong effects for kindergarten children’s fluency in learning letter sounds (Peterson et al., 2014; Rahn et al., 2015; Volpe, Burns, DuBois, & Zaslofsky, 2011) and has consistently been shown to be effective with English language learners (Matchett & Burns, 2009; Peterson et al., 2014; Rahn et al., 2015). Petersen-Brown and Burns (2011) provided definitions and example sentences when teaching vocabulary with IR, which led to increased retention and generalization. Given that the use of semantic cues aided the retention of English words in this study, incorporating character- and word-level morphology may also enhance recognition of Chinese characters.

Current study

The success of IR in teaching various subject matter and the link between visual-verbal PAL for Chinese character recognition suggest that IR may be an effective intervention for character recognition. Thus, the current study preliminarily examined the effectiveness of IR, with and without character- and word-level morphology, in increasing recognition of Chinese characters among kindergarten students.

Method

Participants and setting

Data-collection occurred in one kindergarten classroom in Hong Kong, China that used Cantonese as the language of instruction for Chinese language. Cantonese is the most widely spoken Chinese dialect in Hong Kong (Lee & Leung, 2012). Many schools require students to attain standards of proficiency in reading, writing and speaking either Cantonese- or Mandarin-Chinese (Wong & Rao, 2004). Kindergartens in Hong Kong differ from those in the United States in that they contain three grades, K1, K2, and K3, with students from 3- to 6-years-old.

The kindergarten that participated in the current study had a total of 281 students across three grades (K1, K2, and K3) for four classes per grade and thus a total of 12 classes. Approximately 33% of the students received tuition subsidies through the Kindergarten and Child Care Centre Fee Remission Scheme, provided by the Hong Kong government to families with financial need. The school was not able to provide statistics on the student body’s ethnic composition, but the principal stated that the large majority of students were either from Hong Kong or mainland China, with no more than 2% of students of Southeast Asian descent.

Participants of this study were selected because they had been taught the Chinese language with Cantonese as the medium of instruction, as opposed to other Chinese dialects such as Mandarin. Teachers at the participating kindergarten were asked to nominate K3 students who had had a minimum of two years of formal education using Cantonese as the language of instruction for Chinese language, but who continued to struggle learning the characters (i.e., knew less than 50% of characters that had been taught). However, each student also had to demonstrate sufficient success in learning Cantonese-Chinese on a school-based assessment. Every semester, the kindergarten teachers would rate children on their abilities in reading, writing, speaking, and listening in Cantonese-Chinese. Each ability is rated on a scale of 1 to 4 (1 = in need of improvement, 2 = basic mastery, 3 = proficient, 4 = excellent), such that the maximum rated score was 16. A score of 12 or higher was needed to be included in the study so that the difficulties learning the characters were not due to language delays. All of the nominated students were 5- to 6-years-old, had received at minimum two years of Chinese language instruction in Cantonese and could communicate fluently in Cantonese. They included two Chinese boys and one Chinese girl.

The first participant, Tiffany (pseudonym), was a female who was 5-years and 10-months old. She was Hong Kong-Chinese and spoke Cantonese as her first language. Tiffany’s received a rating of 16 out of 16 for her Chinese language abilities on the most recent school-based rating of Chinese language. She attended 95% of school days during the spring semester.

The second participant was named Brendan (pseudonym), and was a 6-year and 4-month old male. He was Hong Kong-Chinese and spoke Cantonese as his first language. His Chinese language abilities in reading, writing, speaking, and listening received a total rating of 15.5 out of 16. He attended 97% of school days during the spring semester.

Finally, William (pseudonym) was a male who was 6-years and 4-months old. He was Hong Kong-Chinese and spoke Cantonese as his first language. He received ratings in Chinese language abilities of 16 of 16. He attended school 99% of the school days during the spring semester.

Materials

Characters used as known items and those to be taught were selected from the Chinese Character Database of the Research Centre for the Humanities Computing of the Chinese University of Hong Kong (Kwan, 2003). The preliminary pool of characters included 12 pictographic characters as potential known characters and 57 phono-semantic characters as potential unknown characters. Each of the selected pictographic and phono-semantic characters were printed on 3 x 5 inch index cards, using a portrait (vertical) orientation, to be used during assessment and instruction.

Known items

Pictographic characters were chosen as potential known characters because they contained few (between two and four) strokes and were not visually complex. They were also identified by the kindergarten teachers as part of the Chinese language curriculum in K1 and K2. Before administering the intervention, the characters were presented to each participant, who was asked to read them aloud. Nine pictographic characters that the student read with correct pronunciation and tone within three seconds were identified to use as knowns, which are listed in Table 1.

Table 1.

Known, pictographic characters with number of strokes and pronunciation.

Known character	Number of strokes	Pronunciation
人	2	jan4
口	3	hau2
女	3	neoi5
山	3	san1
水	4	seoi2
手	4	sau2
火	4	fo2
日	4	jat6
月	4	jyut6

Unknown items

There were three criteria for the selection of phono-semantic characters to be taught in the study. First, selection was based on conformity to orthographic structure of semantic radicals on the left and phonemic radicals on the right. Second, all the chosen phono-semantic characters had phonetic radicals that were pronounced identically to the whole character. Finally, all characters had frequencies of 500 or below according to an online data-base for Chinese character frequency from the Chinese University of Hong Kong (Ho & Kwan, 2001), to reduce the likelihood of students learning characters outside of the instructional sessions.

The unknown phono-semantic characters for each participant are displayed in Table 2. The known phono-semantic characters (pronounced correctly within three seconds) were discarded and 24 characters were selected from the unknown characters to form four lists of six characters each, to be used over four weeks of instruction. The lists were compiled such that, (a) semantic radicals of characters did not repeat across lists, to prevent semantic radicals learned in one condition from generalizing to another, and (b) the number of strokes in characters of each list did not differ by more than four strokes, so that participants were unlikely to use the visual complexity of characters, rather than its orthographical structure to identify characters. The ranges of number of strokes varied by list. The lowest range was 8–12 strokes, and the highest range was 11–15 strokes. In brief, while the nine known pictographic characters were the same for the participants, the lists of unknown phono-semantic characters were individualized. Each student was taught six unknown characters in each condition.

Table 2.

Lists of unknown, phono-semantic characters for each participant, with semantic radicals and number of strokes.

	Tiffany	Brendan	William
List 1	愉, 惱, 塘, 境, 傅, 停	愉, 惱, 塘, 境, 傅, 停	惺, 惱, 境, 增, 傅, 傑
Semantic radicals	心, 土, 亻	心, 土, 亻	心, 土, 亻
Number of strokes	11–13	11–13	11–15
List 2	味, 喉, 枝, 植, 招, 指	味, 喉, 枝, 植, 招, 指	焗, 炸, 溪, 淚, 蜘, 蜻
Semantic radicals	口, 木, 手	口, 木, 手	火, 水, 虫
Number of strokes	8 –12	8 –12	9–13
List 3	焗, 炸, 溪, 淚, 蜘, 蜓	焗, 炸, 溪, 淚, 蛛, 蜂	誇, 誠, 榴, 植, 捉, 指
Semantic radicals	火, 水, 虫	火, 水, 虫	言, 木, 手
Number of strokes	9–13	9–13	9–13
List 4	紋, 紛, 徑, 律, 脖, 胰	紋, 紛, 徑, 律, 脖, 胰	紋, 紛, 徑, 律, 脖, 胰
Semantic radicals	系, 彳, 肉	系, 彳, 肉	系, 彳, 肉
Number of strokes	9–11	9–11	9–11

Conditions

Participants were taught Chinese characters with IR with or without an additional morphological component for three times each week, approximately 15 minutes each time, in an unused classroom at the kindergarten. Both conditions were administered in Cantonese and are described below.

Incremental Rehearsal (IR)

At the start of each intervention session, the researcher placed the card with the one unknown phono-semantic character in front of cards with nine known, pictographic characters. The researcher then said ‘This character is pronounced ___. How do you pronounce this character?’. If the student pronounced the character correctly, the researcher replied ‘Good job’. If the student pronounced it incorrectly, the researcher corrected the student and had him or her pronounce the character again until correct pronunciation and tone was achieved. When the student correctly pronounced the unknown character, he or she was then shown and asked to read to the unknown character and the first known character. Next, the student read the unknown character, followed by two known characters. This process continued until the unknown character and all nine known characters had been read by the student. The researcher then removed the last known character from the stack of flashcards and added a second unknown character to the top of the stack. The researcher then taught the second unknown character in the same manner described above. Two unknown characters were taught each session. Throughout the procedure, students were given verbal reinforcement for correct responses and immediate corrective feedback if they mispronounced the words or were not able to respond correctly within three seconds after the character had been presented.

IR with a morphological component (IRM)

The intervention with the morphological component was implemented as described above, with only one exception. After the student read the character correctly from the initial presentation and modeling, the researcher covered the phonetic radical on the right to direct the student’s attention to the semantic radical. The researcher then told the student what the semantic radical of the character was, how it was related to the meaning of the character (character morphology) and illustrated this semantic link by providing an example, multi-character word that contained the character being taught (word morphology). If the student gave an answer to either questions about the semantic radical or words containing the character being taught, they were given affirmative or corrective feedback.

Then the student was asked, ‘How do we pronounce this character?’ again and was given feedback according to their correct or incorrect response. Instruction continued identically to the IR condition, until the next new character is introduced using the IRM script outlined above.

Dependent measures

At the beginning of each session, students’ retention of characters taught in the previous week was assessed. The researcher presented the characters that were taught the previous week in random order one at a time to the student, asking that they read the characters aloud. If the student responded with correct pronunciation and tone within three seconds of presentation, then the character was counted as retained. If the student did not respond within three seconds, or read the character with either an incorrect pronunciation or incorrect tone, then the character was identified as not retained. No feedback was given to the student throughout this testing. The dependent measure of the study was the number of characters recognized within one list (six characters).

At the last assessment session in the sixth week, participants were tested once for maintenance of all the characters that had been taught in both intervention conditions by again presenting the characters in random order. As illustrated in Table 3, maintenance was assessed four weeks after instruction of List 1 (using IR), three weeks after instruction of List 2 (using IRM), two weeks after instruction of List 3 (using IRM) and one week after instruction of List 4 (using IR).

Table 3.

Instruction and Assessment Schedule.

Week	Baseline Assessment	Intervention	Test for Retention
1	Pre-test to identify knowns and unknowns in first session, Baseline data collection for List 1	None	None
2	Baseline data-collection for List 2	IR for List 1	None
3	Baseline data-collection for List 3	IRM for List 2	Test retention of List 1
4	Baseline data-collection for List 4	IRM for List 3	Test retention of List 2
5	None	IR for List 4	Test retention of List 3
6	None	None	Test retention of List 4, Test maintenance of all lists in last session

Note. IR = Incremental Rehearsal; IRM = Incremental Rehearsal with Morphological Instruction.

Procedures

The study occurred across a total of six weeks. Week 1 consisted of baseline assessments, the interventions occurred in weeks 2 through 5, and the maintenance assessment occurred in week 6. The pre-test of all potential known and unknown characters occurred during the first session. Each participant was also tested for baseline recognition of the six characters from the first list on three assessment sessions during Week 1. In the second week, List 1 was taught using IR, and baseline data for List 2 were collected. After that, retention of List 1 was tested, List 2 was taught using IRM and baseline recognition of List 3 was measured, and so on. The last week consisted of data-collection for retention of List 4 and maintenance of all four lists. Each list went through one week of baseline data-collection (three sessions), one week of instruction (three sessions), then one week of retention data collection (three sessions).

Experimental design

The research questions were investigated with two ABAB single-case designs. This design was selected as it could compare the effectiveness of two interventions against baseline. The independent variable was the method of instruction, which included Incremental Rehearsal (IR) and Incremental Rehearsal with a morphological component (IRM). Single-case designs are a class of experimental methodology that has been utilized for decades in a number of disciplines including psychology and education, and is ideal to show that a change in the treatment directly leads to a change in outcome (Riley-Tillman & Burns, 2009). The focus of single-case designs is on internal validity, with limited implications for external validity, which is appropriate for initial studies in lines of inquiry in order to tightly control conditions to enhance conclusions about causality. However, additional group-level research, or multiple replications of the single-case design, is usually then needed to understand the generalizability of the conclusions.

Fidelity of implementation

To ensure that the interventions were implemented with fidelity and reliable data-collection, 22% (n = 15) of the intervention sessions were observed by a doctoral student from the Department of Psychology of the University of Hong Kong. The observer was given separate implementation checklists for the IR and IRM conditions, which contained nine and 11 core features of the interventions, respectively. The number of items observed was divided by the total number of items and multiplied by 100%. The results indicate that the researcher implemented the interventions with 99% fidelity in the 15 sessions.

Inter-observer agreement was also calculated during these 15 (22%) assessment sessions. The observer and researcher independently recorded participants’ responses to the characters as correct (correctly stated within three seconds) or incorrect (not correctly stated or correctly stated after three seconds). The number of characters receiving the same rating from both the observer and the researcher was divided by the total number of characters tested during the observations. This was then multiplied by 100% to give a percentage of agreement between the observer and researcher. In this study, the inter-observer agreement was 95%.

Results

The study was a preliminary examination of the effectiveness of Incremental Rehearsal (IR) in teaching kindergarten children to identify Chinese characters. The number of characters retained by each participant throughout the six weeks of the study are displayed in Figure 1. The means and Standard Deviations for the data and the number maintained at the final assessment session are displayed in Table 4.

Figure 1.

Number of characters recognized for all three students.

Table 4.

Mean number of characters retained and maintained per condition.

	1 Week After Instruction			Maintenance
Student	IR (SD)	IRM (SD)	Total (SD)	IR	IRM
Tiffany	3.00 (2.19)	4.00 (1.10)	3.50 (1.73)	12	26
Brendan	3.00 (0.63)	4.17 (1.47)	3.58 (1.24)	27	21
William	5.33 (1.03)	4.17 (1.47)	4.75 (1.36)	33	33

Note. IR = Incremental Rehearsal; IRM = Incremental Rehearsal with Morphological Instruction.

Tiffany’s graph in Figure 1 shows no overlap between baseline and intervention data for either of the conditions, which suggests that both conditions were effective in increasing Chinese character recognition. There was a downward trend in the data as the study progressed for Tiffany given that she retained five of the characters in List 1 and List 2, but retention decreased to three for List 3 and one for List 4. She retained a slightly higher average number of characters onr week after instruction when taught by IRM (M = 4.00; SD = 1.10) than by IR (M = 3.00; SD = 2.19). Likewise, at the final assessment session, Tiffany’s maintenance of characters taught by IRM (26) also exceeded that of IR (12).

Brendan’s graph in Figure 1 shows immediate increase in recognition of characters after each administration of IR and IRM. Although there was some variation in percentage of characters retained, ranging from two to six, the variation was not systematic between the conditions. On average Brendan retained 3.00 (SD = 0.63) of the characters taught by IR and 4.17 (SD = 1.47) of those taught by IRM. Brendan recognized more characters in the final maintenance assessment with IR (27) than IRM (21).

William showed immediate increases in recognition of characters after instruction with IR and IRM. Furthermore, his retention of characters was high across both conditions, with nine of the 12 data points being 4 or higher. Like Tiffany and Brendan, William’s response to instruction in the two conditions did not contain any systematic differences. Although retention of characters taught by IR (M = 5.33, SD = 1.03) was slightly higher than that of IRM (M = 4.33; SD = 1.47) the week after instruction, this small difference disappeared by the final assessment session, where 33 of the 36 of characters were maintained in both conditions.

Discussion

This study was the first attempt to test the effectiveness of IR for instruction beyond English language and mathematics and also to add to the intervention studies for Chinese reading. The results across all three participants supported the effectiveness of IR in increasing Chinese character recognition. Each time IR was administered, participants’ character recognition rose immediately to levels above baseline, which was replicated across character lists and across participants. These results are consistent with the existing empirical evidence for the effectiveness of IR (e.g., Peterson et al., 2014; Rahn et al., 2015), but expanded the scope of previous IR research.

Although we did not experimentally compare the two IR interventions, there did not seem to be an obvious advantage of one intervention over the other. A potential explanation for the lack of observed differential effect between the two treatment conditions could be each student’s developmental stage. Ho, Chang, Lee, Tsang, and Luan (2004) proposed three sequential stages of Chinese reading acquisition, namely the Logographic Stage, the Cipher Stage and the Orthographic Stage. The Logographic Stage is characterized by children’s perception of Chinese characters as whole visual stimuli linked with sounds. When students have been exposed to more characters and can abstract out orthographic regularities in the position and pattern of strokes, they enter the Cipher Stage. At the final, Orthographic Stage, students are able to process whole characters or character parts (e.g. radicals) holistically. Among these stages, IR matches the developmental skills of the Logographic Stage, whereas IRM, particularly its character-morphology component, may be more compatible with students in the Orthographic Stage. Therefore, it may be that students in this study were in the Logographic Stage. Additional research is needed to examine the effects of IR for students in different stages of developing Chinese reading.

Implications for practice

As it is the first study to investigate the effectiveness of IR for Chinese character recognition among kindergarten children, results of this study should be considered preliminary and need replication. With the preliminary evidence in the study, educators and practitioners may consider using IR to supplement other empirically-based instructional methods for teaching Chinese, such as dialogic reading (e.g., Chow et al., 2008). Although the efficiency of implementing this as part of the general curriculum is not feasible, the simplicity of procedures, readily available materials and short time required for implementing IR mean that parents and teachers may administer the intervention with kindergarten students who are struggling to retain newly learned characters.

The current study, and the numerous studies of IR with English reading, suggest that IR could be used to help students learn paired associations in various languages. IR could probably be used with some confidence to teach words with other languages that are logographic (e.g., Japanese and Korean), or alphabetic such as most romance languages (e.g., Italian, French, and Spanish). However, additional research in those specific languages is needed.

Limitations and future research

Several limitations should be noted in the interpretation of results, they also suggest improvements that can be made in future research. First, the participants occasionally did not know the example word used for word-level morphology in IRM, which compromised the examination of IRM’s effects. The researcher’s response to these situations was to provide a different example word, the potential effect of which is unknown. A second limitation is that the current study did not measure retention and maintenance of semantic radicals and example words of characters taught by IRM. Without evidence that the contents of morphology instruction had been retained and maintained, it would be difficult to ascertain the functional relationship between the morphological component and character recognition, even if students had retained significantly higher percentages of characters in IRM relative to IR.

The current study can also be expanded by measuring not just retention and maintenance of Chinese characters, but also generalization. Drawing on the research literature in English instruction, IR has been found effective not only in the acquisition and maintenance of sight words presented in isolation, but also in their generalization, measured in the ability to read those words in context of a sentence or passage (Burns; 2007; Petersen-Brown & Burns, 2011). Future researchers could replicate the study, but also add in opportunities to read the characters in context to perhaps increase generalization and have a stronger effect on reading development.

While this study examined the effectiveness of a variation of IR by incorporating character- and word- morphology components, there are other modifications of IR worth considering in future research. Many of these come from existing literature on IR for other subject areas. One example is to adapt IR with a vocabulary component to use in Chinese (Petersen-Brown & Burns, 2011). This may involve presenting multi-morpheme words instead of isolated characters, and then verbally presenting the definition of the word as well as having students generate an example sentence with correct usage of the Chinese word. Moreover, the utility of IR can be studied in writing as well as reading Chinese characters, given that Konrad (2012) successfully adapted IR for spelling English words.

Conclusion

The current study was the first to use IR to teach Chinese characters to kindergarten students learning Cantonese-Chinese as their first language and resulted in support for the effects of doing so. The generalizability of results cannot be established empirically with the use of single-case design studies. Therefore, future research should replicate the current study and examine the effects of IR for Chinese reading using experimental group designs. Given the unique nature of Chinese language, the additional research seems warranted.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biographies

Elena Kwong, MA is a third-year school psychology graduate student at the University of Minnesota. Her current research interests include language acquisition and reading interventions.

Matthew K. Burns, is the Associate Dean for Research for the College of Education and a Professor of School Psychology at the University of Missouri. He publishes research regarding the use of assessment data to determine individual or small-group interventions (skill-by-treatment interaction), response to intervention, academic interventions, and facilitating problem-solving teams. Dr Burns served on the faculty of the University of Minnesota for ten years and Central Michigan University for five years.

References

Burns

M. K.

(2007) Comparison of drill ratio and opportunities to respond when rehearsing sight words with a child with mental retardation. School Psychology Quarterly 22: 250–263. https://dx-doi-org.web.bisu.edu.cn/10.1037/1045-3830.22.2.250 .

Carlisle

J. F.

(1995) Morphological awareness and early reading achievement. In: Feldman

L. B.

(ed.) Morphological aspects of language processing, Hillsdale, NJ: Erlbaum, pp. 189–209.

Chow

B.W.-Y.

McBride-Chang

Cheung

Chow

C.S.-L.

(2008) Dialogic reading andmorphology training in Chinese children: Effects on language and literacy. Developmental Psychology 44: 233–244. https://dx-doi-org.web.bisu.edu.cn/10.1037/0012-1649.44.1.233 .

DeFrancis

(1989) Visible speech: The diverse oneness of writing systems, Honolulu: University of Hawaii.

Fan, K. Y. (1986, September). Graphic symbol of the Chinese character. Paper presented at the Symposium of Chinese Character Modernization. Beijing, China.

Fan, K. Y., Gao, J. Y., & Ao, X. P. (1984). Pronunciation principles of the Chinese character and alphabetic writing scripts. In Beijing: National Commission of Chinese Character Reform, Chinese character reform 3 (pp. 19–22). Beijing: Beijing: National Commission of Chinese Character Reform (in Chinese).

Hino

Kusunose

Lupker

S. J.

Jared

(2013) The processing advantage and disadvantage for homophones in lexical decision tasks. Journal of Experimental Psychology: Learning, Memory, and Cognition 39: 529–551. https://dx-doi-org.web.bisu.edu.cn/10.1037/a0029122 .

C. S.-H.

Chan

D. W.

Tsang

S.-M.

Lee

S.-H.

Chung

K. K. H.

(2006) Word learning deficit among Chinese dyslexic children. Journal of Child Language 33: 145–161. https://dx-doi-org.web.bisu.edu.cn/10.1017/s0305000905007154 .

C. S. H.

Chan

D. W. O.

Lee

S. H.

Tsang

S. M.

Luan

V. H.

(2004) Cognitive profiling and preliminary subtyping in Chinese developmental dyslexia. Cognition 91: 43–75. https://dx-doi-org.web.bisu.edu.cn/10.1016/s0010-0277(03)00163-x .

10.

Ho, C. S. H., Ng, T. T., & Ng, W. K. (2003). A “radical” approach to reading development in Chinese: The role of semantic radicals and phonetic radicals. Journal of Literacy Research, 35, 849–878. https://dx-doi-org.web.bisu.edu.cn/10.1207/s15548430jlr3503_3.

11.

Ho, S.W. & Kwan, T.W. (2001). Hong Kong, mainland China and Taiwan: Chinese character frequency: A trans-regional, diachronic survey. Retrieved May 1, 2013 from Chinese University of Hong Kong, Production Unit Research Centre for Humanities Computing. http://humanum.aRT.cuhk.edu.hk/Lexis/chifreq.

12.

Jin, J. H. (1985). Properties of Chinese characters. In Beijing National Commission of Chinese. Character Reform Chinese character reform 5 (pp. 13–15). Beijing: Beijing National Commission of Chinese Character Reform (in Chinese).

13.

Konrad, M. (2012). Examining the efficiency of incremental rehearsal oral and written procedures for spelling. (Doctoral dissertation) The Ohio State University, Ohio.

14.

Kwan, T. W. (2003). Chinese character database: With word-formations phonologically disambiguated according to the Cantonese dialect. Retrieved May 1, 2013 from The Chinese University of Hong Kong, Production Unit- Research Centre for Humanities Computing. http://humanum.aRT.cuhk.edu.hk/Lexis/lexi-can/.

15.

Lau

K. L.

Chan

D. W.

(2003) Reading strategy use and motivation among Chinese good and poor readers in Hong Kong. Journal of Research in Reading 26: 177–190. https://dx-doi-org.web.bisu.edu.cn/10.1111/1467-9817.00195 .

16.

Lee

K. S.

Leung

W. M.

(2012) The status of Cantonese in the education policy of Hong Kong. Multilingual Education 2: 1–22.

17.

Shu

McBride-Chang

Liu

H. Y.

Xue

(2009) Paired associate learning in Chinese children with dyslexia. Journal of Experimental Child Psychology 103: 135–151. https://dx-doi-org.web.bisu.edu.cn/10.1016/j.jecp.2009.02.001 .

18.

Mair

V. H.

(1991) Building the future of information processing in East Asia: Demands facing linguistic and technological reality. In: Mair

V. H.

Liu

(eds) Characters and computers, Amsterdam: IOS Press, pp. 1–8.

19.

Matchett

D. L.

Burns

M. K.

(2009) Increasing word recognition fluency with an English language learner. Journal of Evidence-Based Practices for Schools 10: 194–206.

20.

McBride-Chang

Chow

B.W.-Y.

Zhong

Y.-P.

Burgess

Hayward

(2005) Chinese character acquisition and visual skills in two Chinese scripts. Reading and Writing: An Interdisciplinary Journal 18: 99–128. https://dx-doi-org.web.bisu.edu.cn/10.1007/s11145-004-7343-5 .

21.

McBride-Chang

Shu

Zhou

Wat

C. -P.

Wagner

R. K.

(2003) Morphological awareness uniquely predicts young children’s Chinese character recognition. Journal of Educational Psychology 95: 743–751. https://dx-doi-org.web.bisu.edu.cn/10.1037/0022-0663.95.4.743 .

22.

Nagy, W. E., Kuo-Kealoha, A., Xinchun, W., Wenling, L., Anderson, R. C., & Xi, C. (2002). The role of morphological awareness in learning to read Chinese. In: Chinese children’s reading acquisition (pp. 59–86). USA: Springer.

23.

Packard

J. L.

Chen

Gaffney

J. S.

Anderson

R. C.

(2006) Explicit instruction in orthographic structure and word morphology helps Chinese children learn to write characters. Reading and Writing 19: 457–487. https://dx-doi-org.web.bisu.edu.cn/10.1007/s11145-006-9003-4 .

24.

Petersen-Brown

Burns

M. K.

(2011) Adding a vocabulary component to incremental rehearsal to enhance retention and generalization. School Psychology Quarterly 26: 245–255. https://dx-doi-org.web.bisu.edu.cn/10.1037/a0024914 .

25.

Peterson

Brandes

Kunkel

Wilson

Rahn

N. L.

Egan

McComas

(2014) Teaching letter sounds to kindergarten English language learners using incremental rehearsal. Journal of School Psychology 52: 97–107. https://dx-doi-org.web.bisu.edu.cn/10.1016/j.jsp.2013.11.001 .

26.

Perfetti

C. A.

Tan

L. H.

(1998) The time-course of graphic, phonological, and semantic activation in Chinese character identification. Journal of Experimental Psychology: Learning, Memory, and Cognition 24: 1–18. https://dx-doi-org.web.bisu.edu.cn/10.1037//0278-7393.24.1.101 .

27.

Rahn

N. L.

Wilson

Egan

Brandes

Kunkel

Peterson

McComas

(2015) Using incremental rehearsal to teach letter sounds to English language learners. Education and Treatment of Children 38(1): 71–91. https://dx-doi-org.web.bisu.edu.cn/10.1353/etc.2015.0000 .

28.

Riley-Tillman

T. C.

Burns

M. K.

(2009) Evaluating educational interventions: Single-case designs for measuring response to intervention, New York, NY: Guilford.

29.

Shu

(2003) Chinese writing system and learning to read. International Journal of Psychology 38: 274–285. https://dx-doi-org.web.bisu.edu.cn/10.1080/00207590344000060 .

30.

Shu

Chen

Anderson

R. C.

Xuan

(2003) Properties of school Chinese: Implications for learning to read. Child Development 74: 27–47. https://dx-doi-org.web.bisu.edu.cn/10.1111/1467-8624.00519 .

31.

Tong

McBride-Change

Wong

A. M. Y.

Shu

Reitsma

Rispens

(2011) Longitudinal predictors of very early Chinese literacy acquisition. Journal of Research in Reading 34: 315–332. https://dx-doi-org.web.bisu.edu.cn/10.1111/j.1467-9817.2009.01426.x .

32.

Tucker, J. A. (1988). Basic flashcard technique when vocabulary is the goal. Unpublished teaching materials, University of Tennessee at Chattanooga. Chattanooga, TN.

33.

Volpe

R. J.

Burns

M. K.

DuBois

Zaslofsky

A. F.

(2011) Computer-assisted tutoring: Teaching letter sounds to kindergarten students using incremental rehearsal. Psychology in the Schools 48: 332–342. https://dx-doi-org.web.bisu.edu.cn/10.1002/pits.20557 .

34.

Wong, N. C. M., & Rao, N. (2004). Preschool education. In M. Bray, & R. Koo (Eds.), Education and society in Hong Kong arid Macau: Comparative perspectives on continuity and change (pp. 15–34). Hong Kong: The University of Hong Kong, Comparative Education Research Central/Kluwer Academic Publishers.

35.

Yan

Zhou

Shu

Kliegl

(2012) Lexical and sublexical semantic preview benefits in Chinese reading. Journal of Experimental Psychology: Learning, Memory, and Cognition 38: 1069–1075. https://dx-doi-org.web.bisu.edu.cn/10.1037/a0026935 .

36.

Zhu, Y. P. (1987). Analysis of cuing functions of the phonetic in modern China. Unpublished paper, East China Normal University (In Chinese).