Learning to write letters: Examination of student and letter factors

https://doi.org/10.1016/j.jecp.2014.07.009Get rights and content

Highlights

  • We investigated critical factors in the development of letter-writing skills.

  • Seven factors were examined to account for student and letter level variance.

  • Differences between students accounted for large variation in letter-writing skills.

  • Most of the variability was accounted for by letter name knowledge and age.

  • Letter characteristics did not substantially influence learning to write letters.

Abstract

Learning to write the letters of the alphabet is an important part of learning how to write conventionally. In this study, we investigated critical factors in the development of letter-writing skills using exploratory item response models to simultaneously account for variance in responses due to differences between students and between letters. Letter-writing skills were assessed in 415 preschool children aged 3 to 5 years. At the student level, we examined the contribution of letter-name knowledge, letter-sound knowledge, and phonological awareness to letter-writing skills. At the letter level, we examined seven intrinsic and extrinsic factors in understanding how preschool children learn to write alphabet letters: first letter of name, letters in name, letter order, textual frequency, number of strokes, symmetry, and letter type. Results indicated that variation in letter-writing skills was accounted for more by differences between students rather than by differences between letters, with most of the variability accounted for by letter-name knowledge and age. Although significant, the contribution of letter-sound knowledge and phonological awareness was relatively small. Student-level mechanisms underlying the acquisition of letter-writing skills are similar to the mechanisms underlying the learning of letter sounds. However, letter characteristics, which appear to play a major role in the learning of letter names and letter sounds, did not appear to influence learning how to write letters in a substantial way. The exception was if the letter was in the child’s name.

Introduction

Competency in writing (and reading) is a necessary ingredient for children’s academic success (e.g., Graham & Perin, 2007). Conventional writing first begins with children learning to write the letters of the alphabet. The importance of letter-writing skills is supported by empirical evidence indicating that the ability to write letters is an excellent predictor of early spelling, a word-level writing skill (Puranik, Lonigan, & Kim, 2011). Furthermore, substantial evidence indicates that letter-writing fluency (i.e., writing letters of the alphabet under timed conditions) is an excellent predictor of written compositional quality and quantity in kindergarten and elementary school children (e.g., Graham et al., 1997, Jones and Christensen, 1999, Puranik and Al Otaiba, 2012). Thus, learning to write letters constitutes an important early skill in the process of learning to write conventionally.

Despite the importance of letter-writing skills, less is known about the factors that contribute to the acquisition of letter-writing skills than is known about the development of letter names and letter sounds. An increased understanding of the factors that contribute to letter-writing skills could have important implications not only for improving basic developmental theories about writing but also for improving educational practices. In this study, we investigated the contribution of letter-name and letter-sound knowledge and phonological awareness (PA)—established predictors of children’s reading skills—to children’s letter-writing acquisition. In addition to the student characteristics of letter knowledge and PA, research indicates that letter characteristics such as letter type (whether a letter is a consonant–vowel [CV] or vowel–consonant [VC] letter) also influences the development of letter-name and letter-sound knowledge (Justice et al., 2006, Kim et al., 2010, Treiman and Kessler, 2003). These letter characteristics also are likely to influence the development of letter writing. Hence, this study examined the contribution of both student-level factors and letter characteristics to the acquisition of letter writing in preschool children. We addressed this question using exploratory item response modeling that allowed for the correct partitioning of variation among student factors and letter characteristics.

A large body of research now exists demonstrating the important role of letter knowledge in the acquisition of literacy skills (e.g., Bond and Dykstra, 1967, McBride-Chang, 1999, Share et al., 1984, Stevenson and Newman, 1986). Knowledge of letter names measured in preschool is one of the best concurrent and longitudinal predictors of learning to read in an alphabetic writing system (e.g., Lonigan, Schatschneider, & Westberg, 2008; see also Foulin, 2005, for a review). Research indicates that children learn a letter sound more easily when they know the name of the letter than when they do not know the name of the letter (Ehri, 1987, Treiman et al., 1998). In addition to the facilitating letter-sound knowledge, some evidence indicates that knowledge of letter names facilitates early spelling (Treiman & Rodriguez, 1999). Letter names appear to provide a link between letters and print and help children to understand that the words they see on a page or spellings are not arbitrary strings of letters. In this study, we asked whether knowledge of letter names facilitated letter writing similar to its role in facilitating letter-sound knowledge and spelling. Letter writing represents a child’s attempt to retrieve the graphic shapes and names of letters, and as such it should be facilitated by letter-name and/or letter-sound knowledge. Studies examining emergent literacy skills in young children have indicated strong correlations between letter-name and letter-writing skills (rs = .80–.83; Molfese et al., 2006, Worden and Boettcher, 1990) and between letter-sound and letter-writing skills (r = .83; Worden & Boettcher, 1990).

Phonological awareness, which represents sensitivity to and ability to manipulate sound units in words, is related to letter-name and letter-sound knowledge (Burgess and Lonigan, 1998, Mann and Foy, 2003, McBride-Chang, 1999, Share, 2004) and is a very strong predictor of learning to read in an alphabetic language (Lonigan et al., 2000, Wagner and Torgesen, 1987, Wagner et al., 1994). Relations between PA and alphabet knowledge are relatively robust. PA continues to be a unique predictor of letter-name and letter-sound knowledge even after accounting for variance at the child level and controlling for environmental factors (Burgess and Lonigan, 1998, Lonigan et al., 2000, Torppa et al., 2006). Furthermore, studies indicate that the relationship between letter-name and letter-sound knowledge and PA may be reciprocal (e.g., Burgess and Lonigan, 1998, Foy and Mann, 2006, Wagner et al., 1994). Children use their PA abilities to extract the letter sound from its name (Foy and Mann, 2006, Share, 2004), and PA has a greater facilitative effect on learning a letter sound when the letter name is known compared with when the letter name is unknown (Kim et al., 2010). In addition to facilitating reading skills, research indicates that both alphabet knowledge and PA also support word spelling, an early writing task (Puranik et al., 2011). Thus, letter names, letter sounds, and PA were included as student-level predictors because these skills may also play an important role in the development of children’s letter-writing skills.

Finally, because preschoolers in this study ranged in age from 3 to 5 years, it was essential to include age as a student level factor. Accounting for children’s age was important for two reasons. First, given that writing requires some degree of fine motor control, younger children may have more difficulty with letter writing than older children even when other factors related to letter writing are equal. Previous studies have indicated differences in letter writing based on age (e.g., Puranik & Lonigan, 2011). Second, we expected that letter-name knowledge would be a significant contributor to letter-writing skill; however, letter name knowledge of 3-year-olds would be different from that of 5-year-olds.

Although researchers have tended to focus on variations in skills at the student level as contributors to early reading skills, findings of more recent research highlight the need to also account for variation at the letter level (e.g., Evans et al., 2006, Justice et al., 2006, Kim and Petscher, 2011, Kim et al., 2010, Phillips et al., 2012, Treiman et al., 2010). These letter-level factors could be intrinsic; that is, they are part of the essential nature of the letters themselves and might help to describe inter-letter differences. Alternatively, letter-level factors could be extrinsic; that is, they result from social/environmental influences of learning such as influences at home and school as well as print exposure (e.g., Justice et al., 2006). We examined the influence of four extrinsic factors previously shown to influence the learning of letter names, letter sounds, and letter writing (i.e., first letter of a child’s name, later letters in a child’s name, textual frequency, and letter order) and three intrinsic factors (i.e., letter type such as CV or VC letter, number of strokes in a letter, and letter symmetry).

Evidence indicates that children learn the names of the letters contained in their own names before other letters. This influence is likely because children are exposed repeatedly to their names at home and at school compared with other words (e.g., Bloodgood, 1999, Treiman and Broderick, 1998). Children also demonstrate an own-name advantage when learning to write, and substantial evidence indicates that it is the first word a child learns to write (Bloodgood, 1999, Both-de Vries and Bus, 2010, Puranik and Lonigan, 2011, Puranik and Lonigan, 2012, Treiman and Broderick, 1998). Although children’s names are the first words they learn to write, this bias is most profound for the initial letter of their names. Justice and colleagues (2006) reported that children were 11 times more likely to know the first letter of their names compared with other letters in their names. The primary reason for this finding is that it is the first letter they see and it is frequently capitalized, which improves saliency (Treiman & Kessler, 2003).

Whereas the influence of the first letter in a child’s name is the strongest, Justice and colleagues (2006) reported that children were 1.5 times more likely to know the other letters in their names compared with letters not in their names. Research examining writing in preschool children indicates that children tend to use the letters contained in their names more often than other letters when spelling words (e.g., Both-de Vries & Bus, 2010).

Another factor that is likely to influence the acquisition of alphabet knowledge is textual frequency. Frequency appears to be an important factor in spoken word learning (Goodman et al., 2008, Hoff and Naigles, 2002, Huttenlocher et al., 1991, Naigles and Hoff-Ginsberg, 1998, Rice et al., 1994). Unlike the literature on oral language, however, the influence of textual frequency of letters on written language is mixed. Treiman, Kessler, and Pollo (2006), who examined factors affecting the learning of letter names in U.S. and Brazilian preschool children (mean age of 4 years 9 months), reported two findings suggesting that textual frequency may influence the learning of letter names. First, children in both countries made more errors on letters that occur less frequently in print in their respective languages compared with letters that occur more frequently. Second, differences in letter frequencies between English and Portuguese resulted in different patterns of performance of children in the two countries. In another study, when Pollo, Kessler, and Treiman (2009) compared the spellings of the same group of children, they found that early spellings were not random; the letters children used in their early spellings also reflected the relative frequencies of the letters found in the language to which the children were exposed. Portuguese-speaking children used more vowels in their early spellings compared with their English-speaking peers. Pollo and colleagues concluded that this was a reflection of the fact that vowels are more prevalent in Portuguese words than in English words. Puranik, Petscher, and Lonigan (2013), who examined letter writing in preschool children, reported that 5 of the 10 letters that children found easiest to write (A, B, T, I, and P) had high textual frequencies; however, they also reported that the remaining 5 of the 10 letters that were easiest for children to write (O, L, X, H, and E) were not among the most frequently occurring uppercase letters in English. In this study, we attempted to clarify the role of textual frequency in the development of letter-writing skills.

In English, letters are generally iconic, meaning that a letter name generally includes the sound that the letter represents (Treiman & Kessler, 2003). Studies have shown that children tend to use the iconic properties of letter names to acquire letter-sound knowledge. For example, preschool and kindergarten children learn the sounds of the letters contained in their names in the salient initial position of the letter (i.e., CV letters such as t and b) before letters that have their sounds at the end of their names (i.e., VC letters such as m and s) or letters that do not contain the sounds in their names (i.e., no-sound letters [NSLs] such as h and y) and vowels (a, e, i, o, and u; Foy and Mann, 2006, Kim et al., 2010, McBride-Chang, 1999, Treiman and Kessler, 2003, Treiman et al., 1997, Treiman et al., 1998). Whereas letter type might not directly affect learning to write a letter like it does learning letter sounds, it may affect letter writing indirectly. For example, children are more likely to learn to write a letter if they have a name/label for it. Thus if the letter name for a CV letter is learned before a VC letter or a vowel, this might influence which letters children learn to write first.

Letters differ in the number and combination of individual strokes. Although previous research on examination of number of strokes in children learning the Latin alphabet has been mixed (e.g., Treiman & Kessler, 2011), stroke features may affect the number and combination of muscles, mostly of the wrist and fingers that must be employed and controlled during writing (Hulstijn and van Galen, 1983, van Mier and Hulstijn, 1993). In contrast, research with Chinese children indicates that visual complexity affects the performance of their writing; that is, children produce more characters that contain few strokes than characters that contain many strokes (Yin & Treiman, 2013). Thus, the number of strokes in a letter could be a potential factor that might affect letter-writing skills, especially when children are first learning to write.

Finally, letters differ based on symmetry. Symmetry or correspondence in shape, and relative position of parts on opposite sides of a dividing line or median plane, could influence letter writing. Children learning the Latin alphabet perform better on writing and copying tasks with symmetrical letter forms such as the letter H compared with asymmetrical letter forms such as the letter J (Treiman & Kessler, 2011). Similar findings have been noted for Chinese children (Chan and Louie, 1992, Yin and Treiman, 2013).

In this study, we examined the influence of student- and letter-level factors on preschool children’s letter-writing abilities. It is important to note that some of the specific letter-level factors tested in the current study are not necessarily new (i.e., some have been tested in prior work); however, the current study is the first to test several letter-level factors simultaneously. In addition, it is the first study to simultaneously examine the influence of student- and letter-level factors on the development of preschool children’s letter-writing skills. We hypothesized that letter name would be a significant student-level factor and, consistent with previous research, we expected letter writing skills to vary by age groups. Therefore, our analysis included examining letter-writing skills at different age groups. The four questions addressed in this study were as follows:

  • 1.

    How much of the variance in letter-writing skills is accounted for by student- and letter-level factors?

  • 2.

    What are the probabilities of writing a given letter correctly as a function of student- and letter-level factors at different ages?

  • 3.

    What are the probabilities of writing each individual letter correctly at different ages?

  • 4.

    What are the probabilities of writing each individual letter correctly at different ages conditional on letter-name knowledge?

Section snippets

Participants

Participants in this study were 415 preschool children between 3 and 5 years of age (M = 52.53 months, SD = 8.99). These children were part of two larger studies examining emergent writing skills in preschool children. They were recruited from 59 public and private preschool centers that were selected to represent children from a wide range of socioeconomic status (SES) backgrounds. Among the sample, 59 children came from high-SES centers (<25% of students receive subsidies), 192 children came from

Descriptive data

Descriptive data for the measures for each age group are presented in Table 1. The average score on the letter-writing task was 10.46 (SD = 9.94), with older preschoolers showing better performance compared with younger preschoolers. Large standard deviations for the letter-writing task relative to their respective means were noted for both 3- and 4-year-olds, which may be indicative of floor effects. The distribution of scores was examined at each age level; results indicated that 47% of

Discussion

In this study, we examined factors that could potentially contribute to children’s developing ability to write letters, a precursor skill to more complex aspects of writing. The results of this study revealed that the primary contributors to children’s letter-writing skills were age and letter-name knowledge. The proportion of variance in letter writing that was accounted for by student-level factors ranged from 81% to 91%, whereas the proportion of the variance in letter writing that was

Acknowledgment

Support for carrying out this research was provided in part by Grant R305A090622 from Institute of Education Sciences. The opinions expressed are those of the authors and do not represent views of the funding agencies.

References (56)

  • H. van Mier et al.

    The effects of motor complexity and practice on initiation time in writing and drawing

    Acta Psychologica

    (1993)
  • L. Yin et al.

    Name writing in Mandarin-speaking children

    Journal of Experimental Child Psychology

    (2013)
  • J.W. Bloodgood

    What’s in a name? Children’s name writing and literacy acquisition

    Reading Research Quarterly

    (1999)
  • G.L. Bond et al.

    The cooperative research program in first-grade reading instruction

    Reading Research Quarterly

    (1967)
  • A.C. Both-de Vries et al.

    The proper name as starting point for basic reading skills

    Reading and Writing: An Interdisciplinary Journal

    (2010)
  • L. Chan et al.

    Developmental trend of Chinese preschool children in drawing and writing

    Journal of Research in Childhood Education

    (1992)
  • L.C. Ehri

    Learning to read and spell words

    Journal of Literacy Research

    (1987)
  • M.A. Evans et al.

    Letter names, letter sounds, and phonological awareness: An examination of kindergarten children across letters and of letters across children

    Reading and Writing: An Interdisciplinary Journal

    (2006)
  • J.N. Foulin

    Why is letter-name knowledge such a good predictor of learning to read?

    Reading and Writing: An Interdisciplinary Journal

    (2005)
  • J.G. Foy et al.

    Changes in letter sound knowledge are associated with development of phonological awareness in pre-school children

    Journal of Research in Reading

    (2006)
  • J.K. Gilbert et al.

    Word and person effects on decoding accuracy: A new look at an old question

    Journal of Educational Psychology

    (2011)
  • J.C. Goodman et al.

    Does frequency count? Parental input and the acquisition of vocabulary

    Journal of Child Language

    (2008)
  • S. Graham et al.

    Role of mechanics in composing of elementary school students: A new methodological approach

    Journal of Educational Psychology

    (1997)
  • S. Graham et al.

    A meta-analysis of writing instruction for adolescent students

    Journal of Educational Psychology

    (2007)
  • E. Hoff et al.

    How children use input to acquire a lexicon

    Child Development

    (2002)
  • J. Huttenlocher et al.

    Early vocabulary growth: Relation to language input and gender

    Developmental Psychology

    (1991)
  • D. Jones et al.

    Relationship between automaticity in handwriting and students’ ability to generate written text

    Journal of Educational Psychology

    (1999)
  • Cited by (24)

    • Self-regulation and early writing: A longitudinal examination from preschool through first grade

      2022, Journal of Experimental Child Psychology
      Citation Excerpt :

      The maximum score was 52. This task has been used in previous studies (Puranik, Petscher, & Lonigan, 2013, 2014) with good concurrent validity (correlations ranging from.54 to.60 with the criterion measure of name writing; Puranik, Schreiber, Estabrook, & O’Donnell, 2014). Internal consistency reliability for the letter-writing task was high (Cronbach’s α =.98).

    • Prephonological spelling and its connections with later word reading and spelling performance

      2022, Journal of Experimental Child Psychology
      Citation Excerpt :

      This allowed us to test an alternative explanation for the positive monogram and digram correlations and high idiogram proportions that have been observed in previous studies—that the results reflect not graphotactic knowledge but rather children’s enhanced ability to physically produce common letters and letter groups and letters from their own name. Speaking to the plausibility of this alternative explanation, several studies show that U.S. 3- to 5-year-olds are better at writing letters from their own name than writing other letters (Puranik, Petscher, & Lonigan, 2014; Treiman & Broderick, 1998). According to other studies, at least children at the older end of this age range are also better at writing letters that are more frequent in their language than those that are less frequent (Puranik, Petscher, & Lonigan, 2013; Treiman & Kessler, 2011; but see Puranik et al., 2014).

    • Home literacy practices and preschool children's emergent writing skills: An initial investigation

      2018, Early Childhood Research Quarterly
      Citation Excerpt :

      The maximum score was 52. This task has been used in previous studies (Puranik, Petscher, & Lonigan, 2013; Puranik, Petscher, & Lonigan, 2014; Puranik, Schreiber, Estabrook, O’Donnell, 2014) with good concurrent validity (correlations ranging from 0.54 to 0.60 with criterion measure of name-writing; Puranik, Schrieber, Estabrook, & O’Donnell, 2014). Internal consistency reliability for the letter-writing task was high (Cronbach’s α = 0.98).

    • Case- and form-sensitive letter frequencies in children's picture books

      2020, Early Childhood Research Quarterly
      Citation Excerpt :

      In contrast, however, Puranik, Petscher, and Lonigan (2014) reported that letter frequency did not explain substantial variance in children's letter writing abilities after controlling for child-level characteristics such as age and letter-name knowledge. According to the authors, differences in what children are asked to write versus what they produce spontaneously might account for the mixed results (Puranik et al., 2014). When examining how the writing process is influenced by frequency, researchers have demonstrated that children write slowly and dysfluently when writing infrequent syllables (Kandel & Perret, 2014), low frequency words (Laishley, Liversedge, & Kirkby, 2014), and pseudowords (Kandel & Valdois, 2006; Maldarelli et al., 2015).

    View all citing articles on Scopus
    View full text