G Model
JCD-5656; No. of Pages 12 Journal of Communication Disorders xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Communication Disorders
Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months Carolyn D. Sotto a,*, Erin Redle b, Dakshika Bandaranayake c, Jean Neils-Strunjas d, Nancy A. Creaghead c a
The University of Cincinnati, Department of Communication Sciences & Disorders, 3202 Eden Avenue #340, Cincinnati, OH 45267-0379, United States Cincinnati Children’s Hospital Medical Center and the University of Cincinnati, Cincinnati, OH, United States c The University of Cincinnati, Cincinnati, OH, United States d Armstrong Atlantic State University, Savannah, GA, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 June 2013 Received in revised form 24 January 2014 Accepted 9 February 2014 Available online xxx
Purpose: Language develops at variable rates in young children, yet markers for different developmental trajectories, have not been identified. Production of fricatives in words may be one marker because they are later developing sounds and contribute to syntactic production. We examined whether children who produced fricatives in words by 18 months had better vocabulary and grammar scores at 18, 24, and 30 months than children who did not. Method: The expressive language skills of 37 toddlers who did and did not produce fricatives in words by 18 months of age were compared at 18, 24 and 30 months of age. Results: Expressive vocabulary scores and the use of grammatical markers were significantly better for children who produced fricatives by 18 months than for those who did not. This effect was consistent across scores at 18, 24, and 30 months. The total number of consonants and total number of fricatives produced at 18 months did not significantly predict expressive vocabulary scores. Conclusions: Because the children who produced fricatives by 18 months demonstrated better expressive language skills than their peers who did not produce fricatives by 18 months, the early production of fricatives by toddlers may be a marker of a faster expressive language developmental trajectory. Learning outcomes: Readers will: (1) describe the relationship between early phonological development and early expressive lexical development, (2) describe the role of perceptual and motor development in speech sound the acquisition, and (3) describe the potential relationship between the production of fricatives and expressive language development. ß 2014 Elsevier Inc. All rights reserved.
Keywords: Fricatives Speech and language development Vocabulary Toddler
Early and accurate identification of language impairment is critical for children’s social, emotional and academic development. There is abundant research regarding the negative effects of language impairment on literacy and school success and the need for early intervention to assure that children are prepared for school (Aram & Nation, 1980; Brinton & Fujiki, 2004; Catts, 1993; Catts, Fey, Zhang & Tomblin, 1999; Hall & Tomblin, 1978). Early markers that identify children who
* Corresponding author. Tel.: +1 513 558 8528. E-mail address: [email protected] (C.D. Sotto). http://dx.doi.org/10.1016/j.jcomdis.2014.02.003 0021-9924/ß 2014 Elsevier Inc. All rights reserved.
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 2
C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
may be at risk for a delayed or disordered trajectory of expressive language and markers that differentiate late talkers from those who will require services for language delay are thus highly important. The rate and pattern of speech and language development is highly variable among young children (Bernthal, Bankson, & Flipsen, 2013; Sander, 1972; Smit, 1986). To be most useful, it is important that early markers be easily identifiable by the various professionals who assess the development of young children. Predictors of language development that have been identified as relevant at or before 18 months include: joint attention (Brooks & Meltzhoff, 2005; Watt, Wetherby, & Shumway, 2006), play (Rescorla & Goossens, 1992; Watt et al., 2006; Weismer, Murray-Brance, & Miller, 1994), comprehension (Buschmann et al., 2009; Thal, Tobias, & Morrison, 1991; Watt et al., 2006), vocabulary (Rescorla & Schwartz, 1990; Rice & Bode, 1993; Weismer et al., 1994), word combinations (Rescorla & Schwartz, 1990), gesture development (Thal & Tobias, 1992, 1994; Watt et al., 2006), and speech development (Kuhl et al., 2008; McCune & Vihman, 2001; Mirak & Rescorla, 1998; Oller, Eilers, Neal, & Schwartz, 1999). Indicators of speech and language developmental trajectories related specifically to speech include early speech perception skills (Kuhl et al., 2008), age of onset of canonical babbling (Oller et al., 1999), number of phonemes available in the repertoire (McCune & Vihman, 2001), diversity of the consonant inventories and syllable shapes in early vocalizations (McCune & Vihman, 2001; Rescorla & Ratner, 1996), and the relationship among these variables (Whitehurst, Smith, Fischel, Arnold, & Lonigan, 1991). Collectively, these indicators have been identified as markers to assist with the identification of children who are at risk for expressive language delay, but many of them require extensive periods of observation or advanced training in speech and language analysis. The identification of indicators of language development which are readily identifiable without extensive training and/or testing could assist speech-language pathologists and other professionals in determining the potential course of language development. Fricatives may give us a novel framework for looking at the expressive language trajectory of an individual child. Developmental data suggests that this class of sounds is mastered later in development (Goldman, Fristoe, & Williams, 2000; Ingram, Christensen, Veach, & Webster, 1980; Shriberg, 1993) and this is believed to be because these sounds are phonologically more marked, require greater perceptual skills, and greater motor coordination (Ferguson, 1978; Ingram, 1976; Kent, 1992). Current research shows that the phonological, perceptual, and motor systems interact closely to form a foundation for early speech and language development (Kent, 2000; Kuhl, Conboy, Padden, Nelson, & Pruitt, 2005; Rescorla & Ratner, 1996). The ability to produce a fricative in a meaningful way at an early age may represent a minimum competence across the perceptual, motor, and linguistic systems that can positively influence the child’s further acquisition of the linguistic system, including the lexicon and morphosyntax (Davis & MacNeilage, 1995; Gildersleeve-Neumann, Davis, & MacNeilage, 2000; Kent, 2000; Moeller et al., 2007; Nittrouer, 2002). Perceptual and motor maturation appear to have a role in children’s relatively late acquisition of fricatives (Faber and Best, 1994). Children’s ability to perceive fricatives matures or modifies with age (Nittrouer, 2002; Wagner, Ernestus, & Cutler, 2006). The reported later development of fricatives in speech production may reflect differences in the perceptual characteristics of fricatives compared to other phonemes (Faber & Best, 1994; Kuhl, 1994, 2000, 2004; Kuhl et al., 2008; Werker & Tees, 1984). Fricative sounds are more difficult to parse from the speech stream as they are produced at a higher frequency with less intensity than other classes of sounds (Ferguson, 1978). Fricative production requires more precise motor control than the early developing sound classes, stops, nasals and glides (Kent & Murray, 1982). Accurate fricative production requires not only the precise placement of the tongue, but also an additional synchrony of all independent articulators to produce the force and control and the degree of movement needed to generate the noisy turbulence that characterizes fricatives (Kent, 1992). Children with more advanced motor control for word production may produce words earlier than peers without the same level of control (Hoff & Parra, 2011). From a linguistic viewpoint, markedness is considered a factor in the sequence of sound acquisition (Jakobson, 1968). Fricatives are considered more complex or more marked than stops, nasals, and glides. A natural constraint against production of fricatives, which are marked sounds, is related to their less frequent representation in languages although fricatives are frequent in English (Ingram, 1976; Shriberg & Kwiatkowski, 1980). They also appear later in the child’s phonetic repertoire (Barlow & Gierut, 1999; Ingram et al., 1980). Given that fricatives are more complex than other early developing sounds (stops, nasals, glides), their acquisition indicates greater phonological knowledge. The child’s acquisition of these more complex sounds can suggest that the child’s system is primed to acquire the more complex sound combinations required for production of words and sentences. Accurate production of fricatives may also support the production of some grammatical morphemes (plurals, possessives and verb forms) because they are realized by the /s/ and /z/ phonemes. Given that standardized expressive language measures (Wiig, Secord, & Semel, 2004) require a child to produce these responses, interpretation can be more accurate when a child’s post-vocalic phonetic inventory includes /s/ and /z/ (Brown, 1973; Petinou, Schwartz, Gravel, & Raphael, 2001). The relationship between early phonological development and early expressive lexical development has been described by several researchers (Schwartz & Leonard, 1982; Stoel-Gammon, 1991, 2011). Production of a new word requires semantic knowledge as well as an awareness of the articulatory movements required for that word’s production (Stoel-Gammon, 2011). Word learning is related to phonological development during early language acquisition; children are more likely to attempt to imitate and spontaneously produce words containing phonemes already within their repertoire (Paul & Jennings, 1992; Stoel-Gammon, 1991; Vihman & Greenlee, 1987). Schwartz & Leonard (1982) found that word learning prior to 18 months is related to the presence or absence of the phonemes included in the words, and Stoel-Gammon (1998) found that these general patterns of development continue beyond the early word learning period. Overall, toddlers with expressive
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
3
[(Fig._1)TD$IG]
Table 1 Fricative status, mean total number of consonants and mean total number of fricatives at 18 months for 37 subjects. Subject
Fric (+/ )
Consonant repertoire
Total consonants
Total fricatives
Pre-vocalic fricatives
Inter-vocalic fricatives
Post-vocalic fricatives
1M 2M 3F 4M 5F 6F 7M 8F 9F 10M 11M 12M 13F 14F 15F 16M 17M 18M 19F 20M 21F 22M 23M 24M 25F 26M 27M 28F 29F 30F 31M 32F 33F 34M 35F 36M 37M
+
/p, b, t, d, k, w, j, n, s, R/ – /b, d, k, g, w, j, m, l, f, v, s, R/ /b, d, k, g, m, h, f, s, R, ./ /p, b, t, d, k, w, j, m, h, f, s, z, R, tR/ /p, t, d, m, n, h, s, R, / /b, t, d, k, g, r, s, z, R, , tR/ /p, b, t, d, g, j, m, h, s, z, R/ /p, b, d, m, f, R/ /p, t, d, j, m, n, h, f, s, R/ /p, b, t, d, w, j, n, r, s, z, R, ð, tR/ /p, b, t, d, w, j, n, r, s, z, R, ð, tR/ /p, b, t, d, g, m, r, h, f, s, R, a/ /p, b, t, d, g, m, r, h, f, s, R, a/ /p, b, t, d, k, m, n/ /b, t, d, k, g, j, m, l, r, f, s, z, R, ./ /b, t, d, k, g, m, n, l, r, f, s, z, R, ./ /p, d, g, j, n/ /p, b, t, d, g, n/ /d/ /p, b, t, d, g, j, h, s, R, tR, ./ /t, d, g, m, n/ /b, t, d, m, n, h/ /t, d, n, h/ /b, d, g, m, n/ /b, d, m/ /p, b, t, d, w, f, s/ /p, b, t, d, k, m, n, h, s, z, R, ./ /p, b, d, k, g, w, m, f, R/ /b, t, d, w, n, h, f, s, z, R/ /b, t, d, g, w, h/ /b, t, d, k, g, l, h, s, R/ /p, b, d, k, g, w, m, n, f, s/ /b, t, d, w, m, n/ /p, t, d, k, n, l, s/ /p, b, t, d, k, n, s, z/ /p, b, d, n, f, s/
10 0 12 10 14 9 11 11 6 10 13 13 12 12 7 14 14 5 6 1 11 5 6 4 5 3 7 12 9 10 6 9 10 6 7 8 6
2 0 4 4 5 3 5 3 2 3 5 5 4 4 0 5 5 0 0 0 4 0 0 0 0 0 2 4 2 4 0 2 2 0 1 2 2
/s, R/ – /f, v, s, R/ /R, ./ /f, s, z, R, tR/ /s, R, / /s, z, , tR/ /R/ /f, R/ /s/ /s, R, ð, tR/ /s, R, ð, tR/ /f, s, R/ /f, s, R/ – /s, z, R, ./ /f, s, z, ./ – – – /tR, ./ – – – – – – /R, ./ – /f, z, R/ – – – – /s/ /s, z/ –
– – – – – – – – – – – – /a/ /a/ – – – – – – – – – – – – – – – – – /R/ – – – – /s/
/R/ – /f/ /f, s/ /f, s, R/ /s/ /s, R/ /s, z/ /f, R/ /f, s, R/ /s, z, R/ /s, z, R/ /f, s, R/ /f, s, R/ – /f, s/ /s, R/ – – – /s, R/ – – – – – /f, s/ /s, z, R/ /f, R/ /s/ – /s, R/ /f, s/ – /s/ /s, z/ /f/
+ + + + + + + + + + + + + +
+
+ + + + + + + + +
Fig. 1. Comparison of MBCDI: WS expressive vocabulary raw scores by fricative group and age.
language delay have been shown to exhibit smaller phonetic repertoires and less complex syllable shapes than typical peers (Paul & Jennings, 1992; Rescorla & Ratner, 1996). Conversely, children with larger vocabularies also tend to have larger (Stoel-Gammon & Dale, 1988, as cited in Stoel-Gammon, 2011) and more accurate phonemic repertoires (Smith, McGregor, & Demille, 2006). Additionally, children’s phonological memory is greater for phonemes they are able to produce (Gathercole, 2006; Storkel & Morrisette, 2002), also enhancing language learning. Children who do produce fricatives during early word learning may have the potential for greater growth in both short and long term vocabulary development. The above evidence demonstrates a relationship between speech sound development and language development, but the specific relationship between speech sound development and word production has been studied in regard to only a limited number of sound classes. There may also be clinical utility in examining the relationship between specific phonemes/sound classes in the phonetic repertoire and early expressive language skills (Smith et al., 2006). The established phonological
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 4
C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
system may also enhance the development of vocabulary and other language skills through clear communication exchanges and the ability to attach referents to phonetic forms (McCune & Vihman, 2001); however, the relationship between specific phonemes or classes of phonemes, such as fricatives, and the subsequent trajectory of expressive language is currently unknown. The purpose of this exploratory study was to determine if the production of fricatives in words may be a reliable indicator of a faster rate of expressive language development. The expressive language skills of young children as measured by the MacArthur-Bates Communicative Development Inventories: Words & Sentences (MBCDI: WS) (Fenson et al., 2000, 1993) who did and did not produce fricatives were compared over time through multiple assessments. The research question was: Is there a significant difference between the expressive language skills of children who do and do not produce fricatives prior to or at 18 months of age? We hypothesized that the expressive language skills of children who produced fricatives early should be significantly more advanced than those who did not. 1. Methods 1.1. Participants A total of 37 typically developing toddlers (17 females, 20 males) participated in the study. Children entered the study between 9 and 12 months of age and remained until the age of 30 months. A phone screening was completed when children were between 6 and 9 months to determine if they met the inclusion/exclusion criteria prior to enrollment. Exclusion criteria included the presence of a communication or developmental delay, hearing loss, chronic medical condition, or primary language other than English as reported by the parent or identified by the pediatrician. All participating children underwent an audiological evaluation including otoscopic examination and standard behavioral audiometric testing within 3 months of enrollment. Thirty-five of the children were of Caucasian descent and two were of Asian descent. Maternal education and socio-economic status was not directly assessed but US Census data (United-States-Census-Bureau, 2000) were used to determine that all children lived in ZIP codes with median household incomes above the threshold for poverty based on family size and income (Dalaker, 2001). The number of siblings in the families ranged from zero to 3 with a mean of 2.0 children per family. Of the participants, eleven were firstborn or an only child and nineteen were the youngest member of their family. Two sets of twins and 1 set of triplets participated in this study. 1.2. Assessment materials Expressive vocabulary and emerging syntax were selected as the representative language measures for the age of the children in the study (Heilmann, Weismer, & Hollar, 2005). The expressive vocabulary, sentence length and grammar of the children were assessed using the MBCDI: WS (Fenson et al., 1993, 2007). The MBCDI: WS is a parent checklist that assesses early language development including vocabulary and syntax in children between the ages of 16 and 30 months (Dale, 1991; Fenson et al., 1993, 2007; Heilmann et al., 2005). The MBCDI: WS utilizes the parent’s knowledge of their children’s communicative skills, and is a widely used tool in both the United States and other countries for studying early language (Dale, 1991; Moeller et al., 2007; Tsao, Liu, & Kuhl, 2004). It has shown good concurrent validity with other measures of expressive language (Heilmann et al., 2005). 1.3. Procedures As part of a larger study, speech samples were conducted in the children’s homes during 30–60 min visits between 9 and 18 months to determine the presence of fricatives. Affricates were also noted and were included as part of the fricative repertoire throughout the analysis. Affricates were not considered as a separate category and were included in the count of fricatives. A speech language pathologist holding the CCC-SLP and/or a trained graduate student in speech-language pathology observed the children in their homes. During the initial visit the primary caregiver completed a case history, which was reviewed for changes during subsequent visits. Speech samples were audio-taped using an analog voice recorder with a supplemental microphone while the speech-language pathologist and/or trained graduate student interacted with the child. An interactive play format using a common set of toys and objects (e.g. pretend French fries, pretend milkshake, cell phones, zebra, sock, shoes, toothbrush), books, and pretend play activities was designed to elicit fricatives in words. The examiner engaged the child in play with the materials and encouraged the child to talk about them, using the names of the objects and pictures in the book. During spontaneous speech production, the evaluating speech-language pathologist and/or graduate student transcribed fricative production online. Parents were asked to confirm the fricative repertoire observed by the researchers. If the parent reported the presence of a specific fricative in the child’s repertoire and the phoneme could not be elicited during the visit, the parent was given the option of recording the child’s speech and sending the tape to the clinician, or an additional visit was scheduled. If a parent reported that a child produced a fricative in a word and the researchers could not confirm that fricative, the fricative was not recorded as present in the child’s inventory. Vocalizations containing fricatives were categorized into one of the following contexts: isolation, babble (variegated or reduplicated), or word. Utterances were classified as a word, based on the ‘‘consistency of phonetic form, relationship of that form to an adult word, and range of contexts of use’’ (McCune
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
5
& Vihman, 2001, p. 674). Based on the presence (+) or absence ( ) of at least one fricative produced in any single word prior to or at 18 months, children were classified into Fricative (+) (11 males, 14 females) and Fricative ( ) (9 males, 3 females) groups. Additional data extracted from the 18-month speech sample included the total number of consonant phonemes produced in words, the number of fricatives produced in words, and the position of fricative production in words (prevocalic, intervocalic, postvocalic). Parents were then asked to complete the MBCDI: WS during visits at 18 (M = 18.09, SD = 0.58), 24 (M = 24.4, SD = 0.50), and 30 (M = 29.94, SD = .66) months of age. Parents completed Part I of the MBCDI: WS at the 18-month visit and both Parts I and II at the 24 and 30 month visits. Part I measures expressive vocabulary development through parent documentation of words the child produces via a checklist. Part II measures expressive syntactic development through a checklist of words that include grammatical markers, a checklist of multi-word combinations, and the parent’s documentation of the child’s three longest utterances. The investigators reviewed the written instructions for the MBCDI: WS with the parents. Parents either completed the form during the visit or completed the form within one day of the visit and then returned it via mail. 1.4. Data analysis The 18, 24, and 30 month MBCDI: WS scores for the following variables were independently calculated by a graduate clinician and a speech-language pathologist. From Part I the total number of expressive vocabulary items was determined. From Part II, M3L was calculated and the presence of plural, possessive, present progressive and past tense markers were noted. The speech-language pathologist and graduate student then compared each of their MBCDI: WS results to ensure scoring accuracy. Mean length of utterance (M3L) was calculated from the parental report of the 3 longest utterances. All data were entered into a Microsoft Excel spreadsheet. Accurate data entry was checked by one of the lead investigators of the study by comparing the scores to the data collection sheets. These scores were further entered into the MBCDI scoring program (Marchman, 2004) for additional analysis, and all data were finally entered into the SAS program for statistical analysis. Normal distribution was first confirmed in the group data via Q–Q plot inspection and the Shapiro–Wilk test prior to proceeding with group comparisons. To isolate study-specific effects, a mixed-effect model was first developed to determine which factors significantly contributed to MCBDI: WS scores across time. This model included fricative status at 18 months (+, ), visit time (18, 24, and 30 months), an interaction term for fricative status at 18 months and visit time (to account for longitudinal design), the total number of consonants at 18 months, and the total number of fricatives at 18 months. This approach was also used to determine if these same factors significantly contributed to the parent-reported M3L at 24 and 30 months. Categorization of the grammatical markers maintained the format of the parent report form with ordinal choices of often [2], sometimes [1], not yet [0]. Data regarding specific grammatical markers at 18 and 24 months were analyzed using the Wilcoxon 2 Sample test, and a Bonferroni correction of a/8 was applied due to the multiple comparisons. Using the Bonferroni correction, the p value of each independent analysis was compared to a = .0063. 1.5. Reliability Reliability for fricative status, number of different fricatives and position, and total number of consonants at 18 months, was calculated. Fourteen (14) audio samples from the visits were independently coded for presence/absence of fricatives, number of different fricatives, fricative position, and phonetic repertoire by a second investigator. Inter-rater reliability was 100% for fricative status and number of different fricatives, 100% for prevocalic position, 100% for intervocalic and 93% for postvocalic positions, and 86% for total number of consonants. All 37 audio samples were re-coded by the same investigator who originally analyzed them. Intra-rater reliability for fricative status, number of different fricatives and position, and total number of consonants was 100%. A second investigator counted the length of each utterance reported by the parent and calculated the M3L for 54% of the samples (n = 20/37), which resulted in 100% inter-rater reliability. Total scores for 54% of the samples (n = 20/37) on the MBCDI: WS at 18, 24, and 36 months were re-calculated by a second investigator, which resulted in inter-rater reliability of 95%. MBCDI: WS total scores for 40% of the samples at 18, 24, and 36 months (n = 15/37) were recalculated by the same investigator who originally scored them, which resulted in intra-rater of 100%. 2. Results Data regarding presence of fricatives and other consonants for the individual participants are presented in Table 1. Fig. 1 shows the MBCDI: WS mean scores of the Fricative (+) and Fricative ( ) groups at 18, 24 and 30 months. At 18 months the Fricative (+) group had a mean raw score of 116.44 (SD = 60.34), while the Fricative ( ) group had a mean raw score of 48.17 (SD = 37.16). At 24 months the Fricative (+) group’s mean raw score was 425 (SD = 138.44), and the M3L was 6.18 (SD = 2.71), while the Fricative ( ) group’s mean score was 214 (SD = 136.17), and the M3L was 2.18 (SD = 1.45). At 30 months, the Fricative (+) group’s mean score was 576.68 (SD = 62.48), and the M3L was 8.76 (SD = 3.43), while the Fricative ( ) group’s mean score was 468.96 (SD = 155.70), and the M3L was 5.15 (SD = 1.80). The results of the mixed-effect model indicated that the main effects for fricative status (+, ), visit time, and the interaction term were significantly related to the MBCDI: WS scores as shown in Table 2. Total number of fricatives at 18
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
6
Table 2 Main effect results for mixed-effect model of the relationship between variables and expressive vocabulary scores on the MBCDI: WS. df Fricative status Visit number Fricative status visit number Total number of fricatives Total number of consonants
(1, (2, (2, (1, (1,
33) 33) 33) 33) 33)
F
p
4.72 279.00 6.16 0.19 .70
.037*
JCD-5656; No. of Pages 12 Journal of Communication Disorders xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Communication Disorders
Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months Carolyn D. Sotto a,*, Erin Redle b, Dakshika Bandaranayake c, Jean Neils-Strunjas d, Nancy A. Creaghead c a
The University of Cincinnati, Department of Communication Sciences & Disorders, 3202 Eden Avenue #340, Cincinnati, OH 45267-0379, United States Cincinnati Children’s Hospital Medical Center and the University of Cincinnati, Cincinnati, OH, United States c The University of Cincinnati, Cincinnati, OH, United States d Armstrong Atlantic State University, Savannah, GA, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 June 2013 Received in revised form 24 January 2014 Accepted 9 February 2014 Available online xxx
Purpose: Language develops at variable rates in young children, yet markers for different developmental trajectories, have not been identified. Production of fricatives in words may be one marker because they are later developing sounds and contribute to syntactic production. We examined whether children who produced fricatives in words by 18 months had better vocabulary and grammar scores at 18, 24, and 30 months than children who did not. Method: The expressive language skills of 37 toddlers who did and did not produce fricatives in words by 18 months of age were compared at 18, 24 and 30 months of age. Results: Expressive vocabulary scores and the use of grammatical markers were significantly better for children who produced fricatives by 18 months than for those who did not. This effect was consistent across scores at 18, 24, and 30 months. The total number of consonants and total number of fricatives produced at 18 months did not significantly predict expressive vocabulary scores. Conclusions: Because the children who produced fricatives by 18 months demonstrated better expressive language skills than their peers who did not produce fricatives by 18 months, the early production of fricatives by toddlers may be a marker of a faster expressive language developmental trajectory. Learning outcomes: Readers will: (1) describe the relationship between early phonological development and early expressive lexical development, (2) describe the role of perceptual and motor development in speech sound the acquisition, and (3) describe the potential relationship between the production of fricatives and expressive language development. ß 2014 Elsevier Inc. All rights reserved.
Keywords: Fricatives Speech and language development Vocabulary Toddler
Early and accurate identification of language impairment is critical for children’s social, emotional and academic development. There is abundant research regarding the negative effects of language impairment on literacy and school success and the need for early intervention to assure that children are prepared for school (Aram & Nation, 1980; Brinton & Fujiki, 2004; Catts, 1993; Catts, Fey, Zhang & Tomblin, 1999; Hall & Tomblin, 1978). Early markers that identify children who
* Corresponding author. Tel.: +1 513 558 8528. E-mail address: [email protected] (C.D. Sotto). http://dx.doi.org/10.1016/j.jcomdis.2014.02.003 0021-9924/ß 2014 Elsevier Inc. All rights reserved.
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 2
C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
may be at risk for a delayed or disordered trajectory of expressive language and markers that differentiate late talkers from those who will require services for language delay are thus highly important. The rate and pattern of speech and language development is highly variable among young children (Bernthal, Bankson, & Flipsen, 2013; Sander, 1972; Smit, 1986). To be most useful, it is important that early markers be easily identifiable by the various professionals who assess the development of young children. Predictors of language development that have been identified as relevant at or before 18 months include: joint attention (Brooks & Meltzhoff, 2005; Watt, Wetherby, & Shumway, 2006), play (Rescorla & Goossens, 1992; Watt et al., 2006; Weismer, Murray-Brance, & Miller, 1994), comprehension (Buschmann et al., 2009; Thal, Tobias, & Morrison, 1991; Watt et al., 2006), vocabulary (Rescorla & Schwartz, 1990; Rice & Bode, 1993; Weismer et al., 1994), word combinations (Rescorla & Schwartz, 1990), gesture development (Thal & Tobias, 1992, 1994; Watt et al., 2006), and speech development (Kuhl et al., 2008; McCune & Vihman, 2001; Mirak & Rescorla, 1998; Oller, Eilers, Neal, & Schwartz, 1999). Indicators of speech and language developmental trajectories related specifically to speech include early speech perception skills (Kuhl et al., 2008), age of onset of canonical babbling (Oller et al., 1999), number of phonemes available in the repertoire (McCune & Vihman, 2001), diversity of the consonant inventories and syllable shapes in early vocalizations (McCune & Vihman, 2001; Rescorla & Ratner, 1996), and the relationship among these variables (Whitehurst, Smith, Fischel, Arnold, & Lonigan, 1991). Collectively, these indicators have been identified as markers to assist with the identification of children who are at risk for expressive language delay, but many of them require extensive periods of observation or advanced training in speech and language analysis. The identification of indicators of language development which are readily identifiable without extensive training and/or testing could assist speech-language pathologists and other professionals in determining the potential course of language development. Fricatives may give us a novel framework for looking at the expressive language trajectory of an individual child. Developmental data suggests that this class of sounds is mastered later in development (Goldman, Fristoe, & Williams, 2000; Ingram, Christensen, Veach, & Webster, 1980; Shriberg, 1993) and this is believed to be because these sounds are phonologically more marked, require greater perceptual skills, and greater motor coordination (Ferguson, 1978; Ingram, 1976; Kent, 1992). Current research shows that the phonological, perceptual, and motor systems interact closely to form a foundation for early speech and language development (Kent, 2000; Kuhl, Conboy, Padden, Nelson, & Pruitt, 2005; Rescorla & Ratner, 1996). The ability to produce a fricative in a meaningful way at an early age may represent a minimum competence across the perceptual, motor, and linguistic systems that can positively influence the child’s further acquisition of the linguistic system, including the lexicon and morphosyntax (Davis & MacNeilage, 1995; Gildersleeve-Neumann, Davis, & MacNeilage, 2000; Kent, 2000; Moeller et al., 2007; Nittrouer, 2002). Perceptual and motor maturation appear to have a role in children’s relatively late acquisition of fricatives (Faber and Best, 1994). Children’s ability to perceive fricatives matures or modifies with age (Nittrouer, 2002; Wagner, Ernestus, & Cutler, 2006). The reported later development of fricatives in speech production may reflect differences in the perceptual characteristics of fricatives compared to other phonemes (Faber & Best, 1994; Kuhl, 1994, 2000, 2004; Kuhl et al., 2008; Werker & Tees, 1984). Fricative sounds are more difficult to parse from the speech stream as they are produced at a higher frequency with less intensity than other classes of sounds (Ferguson, 1978). Fricative production requires more precise motor control than the early developing sound classes, stops, nasals and glides (Kent & Murray, 1982). Accurate fricative production requires not only the precise placement of the tongue, but also an additional synchrony of all independent articulators to produce the force and control and the degree of movement needed to generate the noisy turbulence that characterizes fricatives (Kent, 1992). Children with more advanced motor control for word production may produce words earlier than peers without the same level of control (Hoff & Parra, 2011). From a linguistic viewpoint, markedness is considered a factor in the sequence of sound acquisition (Jakobson, 1968). Fricatives are considered more complex or more marked than stops, nasals, and glides. A natural constraint against production of fricatives, which are marked sounds, is related to their less frequent representation in languages although fricatives are frequent in English (Ingram, 1976; Shriberg & Kwiatkowski, 1980). They also appear later in the child’s phonetic repertoire (Barlow & Gierut, 1999; Ingram et al., 1980). Given that fricatives are more complex than other early developing sounds (stops, nasals, glides), their acquisition indicates greater phonological knowledge. The child’s acquisition of these more complex sounds can suggest that the child’s system is primed to acquire the more complex sound combinations required for production of words and sentences. Accurate production of fricatives may also support the production of some grammatical morphemes (plurals, possessives and verb forms) because they are realized by the /s/ and /z/ phonemes. Given that standardized expressive language measures (Wiig, Secord, & Semel, 2004) require a child to produce these responses, interpretation can be more accurate when a child’s post-vocalic phonetic inventory includes /s/ and /z/ (Brown, 1973; Petinou, Schwartz, Gravel, & Raphael, 2001). The relationship between early phonological development and early expressive lexical development has been described by several researchers (Schwartz & Leonard, 1982; Stoel-Gammon, 1991, 2011). Production of a new word requires semantic knowledge as well as an awareness of the articulatory movements required for that word’s production (Stoel-Gammon, 2011). Word learning is related to phonological development during early language acquisition; children are more likely to attempt to imitate and spontaneously produce words containing phonemes already within their repertoire (Paul & Jennings, 1992; Stoel-Gammon, 1991; Vihman & Greenlee, 1987). Schwartz & Leonard (1982) found that word learning prior to 18 months is related to the presence or absence of the phonemes included in the words, and Stoel-Gammon (1998) found that these general patterns of development continue beyond the early word learning period. Overall, toddlers with expressive
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
3
[(Fig._1)TD$IG]
Table 1 Fricative status, mean total number of consonants and mean total number of fricatives at 18 months for 37 subjects. Subject
Fric (+/ )
Consonant repertoire
Total consonants
Total fricatives
Pre-vocalic fricatives
Inter-vocalic fricatives
Post-vocalic fricatives
1M 2M 3F 4M 5F 6F 7M 8F 9F 10M 11M 12M 13F 14F 15F 16M 17M 18M 19F 20M 21F 22M 23M 24M 25F 26M 27M 28F 29F 30F 31M 32F 33F 34M 35F 36M 37M
+
/p, b, t, d, k, w, j, n, s, R/ – /b, d, k, g, w, j, m, l, f, v, s, R/ /b, d, k, g, m, h, f, s, R, ./ /p, b, t, d, k, w, j, m, h, f, s, z, R, tR/ /p, t, d, m, n, h, s, R, / /b, t, d, k, g, r, s, z, R, , tR/ /p, b, t, d, g, j, m, h, s, z, R/ /p, b, d, m, f, R/ /p, t, d, j, m, n, h, f, s, R/ /p, b, t, d, w, j, n, r, s, z, R, ð, tR/ /p, b, t, d, w, j, n, r, s, z, R, ð, tR/ /p, b, t, d, g, m, r, h, f, s, R, a/ /p, b, t, d, g, m, r, h, f, s, R, a/ /p, b, t, d, k, m, n/ /b, t, d, k, g, j, m, l, r, f, s, z, R, ./ /b, t, d, k, g, m, n, l, r, f, s, z, R, ./ /p, d, g, j, n/ /p, b, t, d, g, n/ /d/ /p, b, t, d, g, j, h, s, R, tR, ./ /t, d, g, m, n/ /b, t, d, m, n, h/ /t, d, n, h/ /b, d, g, m, n/ /b, d, m/ /p, b, t, d, w, f, s/ /p, b, t, d, k, m, n, h, s, z, R, ./ /p, b, d, k, g, w, m, f, R/ /b, t, d, w, n, h, f, s, z, R/ /b, t, d, g, w, h/ /b, t, d, k, g, l, h, s, R/ /p, b, d, k, g, w, m, n, f, s/ /b, t, d, w, m, n/ /p, t, d, k, n, l, s/ /p, b, t, d, k, n, s, z/ /p, b, d, n, f, s/
10 0 12 10 14 9 11 11 6 10 13 13 12 12 7 14 14 5 6 1 11 5 6 4 5 3 7 12 9 10 6 9 10 6 7 8 6
2 0 4 4 5 3 5 3 2 3 5 5 4 4 0 5 5 0 0 0 4 0 0 0 0 0 2 4 2 4 0 2 2 0 1 2 2
/s, R/ – /f, v, s, R/ /R, ./ /f, s, z, R, tR/ /s, R, / /s, z, , tR/ /R/ /f, R/ /s/ /s, R, ð, tR/ /s, R, ð, tR/ /f, s, R/ /f, s, R/ – /s, z, R, ./ /f, s, z, ./ – – – /tR, ./ – – – – – – /R, ./ – /f, z, R/ – – – – /s/ /s, z/ –
– – – – – – – – – – – – /a/ /a/ – – – – – – – – – – – – – – – – – /R/ – – – – /s/
/R/ – /f/ /f, s/ /f, s, R/ /s/ /s, R/ /s, z/ /f, R/ /f, s, R/ /s, z, R/ /s, z, R/ /f, s, R/ /f, s, R/ – /f, s/ /s, R/ – – – /s, R/ – – – – – /f, s/ /s, z, R/ /f, R/ /s/ – /s, R/ /f, s/ – /s/ /s, z/ /f/
+ + + + + + + + + + + + + +
+
+ + + + + + + + +
Fig. 1. Comparison of MBCDI: WS expressive vocabulary raw scores by fricative group and age.
language delay have been shown to exhibit smaller phonetic repertoires and less complex syllable shapes than typical peers (Paul & Jennings, 1992; Rescorla & Ratner, 1996). Conversely, children with larger vocabularies also tend to have larger (Stoel-Gammon & Dale, 1988, as cited in Stoel-Gammon, 2011) and more accurate phonemic repertoires (Smith, McGregor, & Demille, 2006). Additionally, children’s phonological memory is greater for phonemes they are able to produce (Gathercole, 2006; Storkel & Morrisette, 2002), also enhancing language learning. Children who do produce fricatives during early word learning may have the potential for greater growth in both short and long term vocabulary development. The above evidence demonstrates a relationship between speech sound development and language development, but the specific relationship between speech sound development and word production has been studied in regard to only a limited number of sound classes. There may also be clinical utility in examining the relationship between specific phonemes/sound classes in the phonetic repertoire and early expressive language skills (Smith et al., 2006). The established phonological
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 4
C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
system may also enhance the development of vocabulary and other language skills through clear communication exchanges and the ability to attach referents to phonetic forms (McCune & Vihman, 2001); however, the relationship between specific phonemes or classes of phonemes, such as fricatives, and the subsequent trajectory of expressive language is currently unknown. The purpose of this exploratory study was to determine if the production of fricatives in words may be a reliable indicator of a faster rate of expressive language development. The expressive language skills of young children as measured by the MacArthur-Bates Communicative Development Inventories: Words & Sentences (MBCDI: WS) (Fenson et al., 2000, 1993) who did and did not produce fricatives were compared over time through multiple assessments. The research question was: Is there a significant difference between the expressive language skills of children who do and do not produce fricatives prior to or at 18 months of age? We hypothesized that the expressive language skills of children who produced fricatives early should be significantly more advanced than those who did not. 1. Methods 1.1. Participants A total of 37 typically developing toddlers (17 females, 20 males) participated in the study. Children entered the study between 9 and 12 months of age and remained until the age of 30 months. A phone screening was completed when children were between 6 and 9 months to determine if they met the inclusion/exclusion criteria prior to enrollment. Exclusion criteria included the presence of a communication or developmental delay, hearing loss, chronic medical condition, or primary language other than English as reported by the parent or identified by the pediatrician. All participating children underwent an audiological evaluation including otoscopic examination and standard behavioral audiometric testing within 3 months of enrollment. Thirty-five of the children were of Caucasian descent and two were of Asian descent. Maternal education and socio-economic status was not directly assessed but US Census data (United-States-Census-Bureau, 2000) were used to determine that all children lived in ZIP codes with median household incomes above the threshold for poverty based on family size and income (Dalaker, 2001). The number of siblings in the families ranged from zero to 3 with a mean of 2.0 children per family. Of the participants, eleven were firstborn or an only child and nineteen were the youngest member of their family. Two sets of twins and 1 set of triplets participated in this study. 1.2. Assessment materials Expressive vocabulary and emerging syntax were selected as the representative language measures for the age of the children in the study (Heilmann, Weismer, & Hollar, 2005). The expressive vocabulary, sentence length and grammar of the children were assessed using the MBCDI: WS (Fenson et al., 1993, 2007). The MBCDI: WS is a parent checklist that assesses early language development including vocabulary and syntax in children between the ages of 16 and 30 months (Dale, 1991; Fenson et al., 1993, 2007; Heilmann et al., 2005). The MBCDI: WS utilizes the parent’s knowledge of their children’s communicative skills, and is a widely used tool in both the United States and other countries for studying early language (Dale, 1991; Moeller et al., 2007; Tsao, Liu, & Kuhl, 2004). It has shown good concurrent validity with other measures of expressive language (Heilmann et al., 2005). 1.3. Procedures As part of a larger study, speech samples were conducted in the children’s homes during 30–60 min visits between 9 and 18 months to determine the presence of fricatives. Affricates were also noted and were included as part of the fricative repertoire throughout the analysis. Affricates were not considered as a separate category and were included in the count of fricatives. A speech language pathologist holding the CCC-SLP and/or a trained graduate student in speech-language pathology observed the children in their homes. During the initial visit the primary caregiver completed a case history, which was reviewed for changes during subsequent visits. Speech samples were audio-taped using an analog voice recorder with a supplemental microphone while the speech-language pathologist and/or trained graduate student interacted with the child. An interactive play format using a common set of toys and objects (e.g. pretend French fries, pretend milkshake, cell phones, zebra, sock, shoes, toothbrush), books, and pretend play activities was designed to elicit fricatives in words. The examiner engaged the child in play with the materials and encouraged the child to talk about them, using the names of the objects and pictures in the book. During spontaneous speech production, the evaluating speech-language pathologist and/or graduate student transcribed fricative production online. Parents were asked to confirm the fricative repertoire observed by the researchers. If the parent reported the presence of a specific fricative in the child’s repertoire and the phoneme could not be elicited during the visit, the parent was given the option of recording the child’s speech and sending the tape to the clinician, or an additional visit was scheduled. If a parent reported that a child produced a fricative in a word and the researchers could not confirm that fricative, the fricative was not recorded as present in the child’s inventory. Vocalizations containing fricatives were categorized into one of the following contexts: isolation, babble (variegated or reduplicated), or word. Utterances were classified as a word, based on the ‘‘consistency of phonetic form, relationship of that form to an adult word, and range of contexts of use’’ (McCune
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
5
& Vihman, 2001, p. 674). Based on the presence (+) or absence ( ) of at least one fricative produced in any single word prior to or at 18 months, children were classified into Fricative (+) (11 males, 14 females) and Fricative ( ) (9 males, 3 females) groups. Additional data extracted from the 18-month speech sample included the total number of consonant phonemes produced in words, the number of fricatives produced in words, and the position of fricative production in words (prevocalic, intervocalic, postvocalic). Parents were then asked to complete the MBCDI: WS during visits at 18 (M = 18.09, SD = 0.58), 24 (M = 24.4, SD = 0.50), and 30 (M = 29.94, SD = .66) months of age. Parents completed Part I of the MBCDI: WS at the 18-month visit and both Parts I and II at the 24 and 30 month visits. Part I measures expressive vocabulary development through parent documentation of words the child produces via a checklist. Part II measures expressive syntactic development through a checklist of words that include grammatical markers, a checklist of multi-word combinations, and the parent’s documentation of the child’s three longest utterances. The investigators reviewed the written instructions for the MBCDI: WS with the parents. Parents either completed the form during the visit or completed the form within one day of the visit and then returned it via mail. 1.4. Data analysis The 18, 24, and 30 month MBCDI: WS scores for the following variables were independently calculated by a graduate clinician and a speech-language pathologist. From Part I the total number of expressive vocabulary items was determined. From Part II, M3L was calculated and the presence of plural, possessive, present progressive and past tense markers were noted. The speech-language pathologist and graduate student then compared each of their MBCDI: WS results to ensure scoring accuracy. Mean length of utterance (M3L) was calculated from the parental report of the 3 longest utterances. All data were entered into a Microsoft Excel spreadsheet. Accurate data entry was checked by one of the lead investigators of the study by comparing the scores to the data collection sheets. These scores were further entered into the MBCDI scoring program (Marchman, 2004) for additional analysis, and all data were finally entered into the SAS program for statistical analysis. Normal distribution was first confirmed in the group data via Q–Q plot inspection and the Shapiro–Wilk test prior to proceeding with group comparisons. To isolate study-specific effects, a mixed-effect model was first developed to determine which factors significantly contributed to MCBDI: WS scores across time. This model included fricative status at 18 months (+, ), visit time (18, 24, and 30 months), an interaction term for fricative status at 18 months and visit time (to account for longitudinal design), the total number of consonants at 18 months, and the total number of fricatives at 18 months. This approach was also used to determine if these same factors significantly contributed to the parent-reported M3L at 24 and 30 months. Categorization of the grammatical markers maintained the format of the parent report form with ordinal choices of often [2], sometimes [1], not yet [0]. Data regarding specific grammatical markers at 18 and 24 months were analyzed using the Wilcoxon 2 Sample test, and a Bonferroni correction of a/8 was applied due to the multiple comparisons. Using the Bonferroni correction, the p value of each independent analysis was compared to a = .0063. 1.5. Reliability Reliability for fricative status, number of different fricatives and position, and total number of consonants at 18 months, was calculated. Fourteen (14) audio samples from the visits were independently coded for presence/absence of fricatives, number of different fricatives, fricative position, and phonetic repertoire by a second investigator. Inter-rater reliability was 100% for fricative status and number of different fricatives, 100% for prevocalic position, 100% for intervocalic and 93% for postvocalic positions, and 86% for total number of consonants. All 37 audio samples were re-coded by the same investigator who originally analyzed them. Intra-rater reliability for fricative status, number of different fricatives and position, and total number of consonants was 100%. A second investigator counted the length of each utterance reported by the parent and calculated the M3L for 54% of the samples (n = 20/37), which resulted in 100% inter-rater reliability. Total scores for 54% of the samples (n = 20/37) on the MBCDI: WS at 18, 24, and 36 months were re-calculated by a second investigator, which resulted in inter-rater reliability of 95%. MBCDI: WS total scores for 40% of the samples at 18, 24, and 36 months (n = 15/37) were recalculated by the same investigator who originally scored them, which resulted in intra-rater of 100%. 2. Results Data regarding presence of fricatives and other consonants for the individual participants are presented in Table 1. Fig. 1 shows the MBCDI: WS mean scores of the Fricative (+) and Fricative ( ) groups at 18, 24 and 30 months. At 18 months the Fricative (+) group had a mean raw score of 116.44 (SD = 60.34), while the Fricative ( ) group had a mean raw score of 48.17 (SD = 37.16). At 24 months the Fricative (+) group’s mean raw score was 425 (SD = 138.44), and the M3L was 6.18 (SD = 2.71), while the Fricative ( ) group’s mean score was 214 (SD = 136.17), and the M3L was 2.18 (SD = 1.45). At 30 months, the Fricative (+) group’s mean score was 576.68 (SD = 62.48), and the M3L was 8.76 (SD = 3.43), while the Fricative ( ) group’s mean score was 468.96 (SD = 155.70), and the M3L was 5.15 (SD = 1.80). The results of the mixed-effect model indicated that the main effects for fricative status (+, ), visit time, and the interaction term were significantly related to the MBCDI: WS scores as shown in Table 2. Total number of fricatives at 18
Please cite this article in press as: Sotto, C. D., et al. Fricatives at 18 months as a measure for predicting vocabulary and grammar at 24 and 30 months. Journal of Communication Disorders (2014), http://dx.doi.org/10.1016/j.jcomdis.2014.02.003
G Model
JCD-5656; No. of Pages 12 C.D. Sotto et al. / Journal of Communication Disorders xxx (2014) xxx–xxx
6
Table 2 Main effect results for mixed-effect model of the relationship between variables and expressive vocabulary scores on the MBCDI: WS. df Fricative status Visit number Fricative status visit number Total number of fricatives Total number of consonants
(1, (2, (2, (1, (1,
33) 33) 33) 33) 33)
F
p
4.72 279.00 6.16 0.19 .70
.037*
