Annual Meeting of the American Speech-Language-Hearing Association Miami Beach, November 16-19, 2006
The Effects of Age and Socioeconomic Status on the Use of Literate Language Features
Regina Lemmon & Hiram McDade Department of Communication Sciences & Disorders
University of South Carolina
ABSTRACT
Currenton & Justice (2004) examined the oral narratives of low income preschoolers, ages 3, 4, and 5 years, for the presence of five literate language features believed to be important for the comprehension of decontextualized language (e.g., reading). The current investigation expanded the Currenton & Justice study by including children from both low and middle income homes. Data were analyzed with regard to the effects of age and income group. Results from our low income children were comparable to Currenton & Justice with a few exceptions. Income level was not a significant factor for any of the measures taken
INTRODUCTION Development of our first language (speech) occurs in face-to-face communicative settings, where utterances are spoken within the context of shared experiences occurring here and now. Because the physical context plays an important role in the meaning of their early utterances, speech is considered contextual language. Reading, in contrast, is a decontextualized process, since individuals are no longer processing sentences related to shared events with their communicative partners. Unlike speech, reading comprehension is heavily dependent upon an individual’s ability to process decontextualized language. In decontextualized language, the context needed for comprehension is provided within the language itself, through what have been termed Literate Language Features (Greenlaugh & Strong, 2001). Literate language features emerge in spoken language before the child becomes a reader, and are used increasingly as the individual develops the capacity to carry on conversations about events happening to other people at different times and places. Thus, all of the information needed to comprehend literate language features is contained in the linguistic code. The literate language features observed in children’s speech eventually become internalized and their use broadens with exposure to books, print (Wallach & Butler, 1994) and oral narratives, such as sermons, speeches or fairy tales (Currenton & Justice, 2004; Greenlaugh & Strong, 2001). The purpose of this study was to examine the differences between young children’s use of five literate language features (elaborated noun phrases, conjunctions, adverbs and mental/linguistic verbs) across three age groups (3-, 4-, and 5-years) and two socioeconomic levels (low income and middle income). Specifically, this study was an expansion of the work conducted by Currenton & Justice (2004), who examined the oral narratives produced by preschoolers from low income families. Their results revealed no significant differences in the use of literate language features by African American (AA) and Caucasian (C) children, but a significant age effect, with 5-year-olds scoring higher than 4-year-olds, who scored higher than 3-year-olds. The specific literate language features showing the greatest difference by age were the use of conjunctions and mental & linguistic verbs. Their study did not, however, determine how the performance of these low-income children compared to their middle-income peers. Do middle income preschoolers utilize (and presumably possess) a larger repertoire of literate language functions than their low income peers? Does the age effect observed in Currenton & Justice’s investigation hold true for middle class preschoolers as well? METHOD • Seventy-eight children, ages 3, 4, and 5 years (with 26 per age group), participated in this study. • Half of the children in each age group (13) came from low income homes (based on their qualification for free lunch). These participants were match by age with children who did not qualify for free or reduced lunch. This latter group was considered middle income. • Each of the 78 children were seen individually for 30 to 40 minute sessions, during which time they generated original oral narratives using the wordless picture book,Frog,Where Are You? (Mayer, 1969). • All narratives were tape recorded and orthographically transcribed. • The transcriptions were segmented into Communication Units or C-units (Loban, 1976). An example of C-unit segmentation is as follows, “I saw something green / but it wasn’t my frog.” • The mean length of communication units (MLCU) for each narrative was calculated by dividing the total number of morphemes by the number of C-units. • The presence of the five literate language features was marked and counted within each C-Unit (Justice & Ezell, 2002). • The usage rate for each literate language feature (i.e., mean number of features per clause) was then calculated (as per Currenton & Justice, 2004). • Language samples from 23 of the 78 participants were randomly selected to be analyzed by a second examiner for reliability purposes. Each utterance was independently segmented and scored. Reliability coefficients for the six measures ranged from .93 to .99 • A 2 X 3 ANOVA was used to compare participants’ performance across the two income levels (low and middle Income) and three different age groups (3, 4, and 5 years). Because of the multiple comparisons, a Bonferroni procedure was used to adjust the alpha level from .05 to .01 •
ROBINSON, CONYERS, MCDADE, & MONTGOMERY: Neighborhood Density & Word Regocnition RESULTS
Figure 1. Use of literate language features as a function of age (low income only): A comparison of the current study with Currenton & Justice (2004)
1.20 1.10 1.00 .90 .80 .70 .60 .50 .40 .30 .20 .10 .00
2
Figure 2. Use of Literate Language Features as a function of Age and Income
1.20 1.10 1.00 .90 .80 .70 .60 .50 .40 .30 .20 .10 .00 SENP CENP ADV CONJ MV&LV
Frequency per Clause
Lemmon 3 yr C&J 3 yr Lemmon 4 yr C&J 4 yr Lemmon 5 yr C&J 5 yr
Frequency of Usage
3yr Low Income 3 yr Middle Income 4 yr Low Income 4 yr Middle Income 5 yr Low Income 5 yr Middle Income
SENP
CENP
ADV
CONJ
M&LV
Literate Language Features
Literate Language Feature
Figure 4. Mean length of communication unit as a function of Age and Income
Figure 3. Mean length of communication unit as a function of age (income level collapsed)
7.0 6.0 5.0
7.0 6.0 5.0
Low Middle
MLCU
3.0 2.0 1.0 0.0
MLCU
4.0
4.0 3.0 2.0 1.0
3 yr
4 yr
5 yr
0.0
3 yr 4 yr 5 yr
Age Group
Age Group
• As with Currenton & Justice (2004), of the five literate language features examined, the highest usage rate observed for all three age groups was simple elaborated noun phrases, averaging one per C-unit (SD=.38). • Conjunctions occurred at the rate of one per two C-units (SD=.40) while adverbs occurred approximately one per three C-units (SD=.14). • Complex elaborated noun phrases occurred an average of one per ten C-units (M=.11, SD=.16) • Mental & linguistic verbs were the least frequent feature, occurring less than one per ten C-units (M=.08, SD=.08). • The main effect for income level was not significant MLCU or for any of the five literate language features tested (simple elaborated, noun phrase, complex elaborated noun phrase, adverbs, and conjunctions). • Of these five literate language features, age groups differed only for the use of complex elaborated noun phrase (F=4.27; df 2,72; p < .01). 5 year old children used CENP more frequently than the 3 and 4 year olds, who did not differ from each other. • Effect sizes for age group were 0.30 for simple elaborated noun phrases, 0.06 for conjunctions, 0.35 for adverbs, and 0.16 for mental & linguistic verbs. • The main effect for age was also significant for the variable mean length of communication unit – MLCU (F=4.72; df 2,72; p <.01). Specifically, the MLCU for 3 year olds (5.24) was shorter than that for the 4 and 5 year old children, who did not differ from each other (6.03 and 6.02, respectively). • There was no significant difference in the MLCU between low and middle income children for any of the ages (F=1.98; df 1,72; p >.1). CONCLUSIONS • Children from low and middle income homes sampled in this study (ages 3, 4, and 5 years) did not differ in their use of any of the five literate language features. • The MLCU for children from low income homes did not differ from that of middle income children across all three age groups (ES=.35). • 4 and 5 year old children spoke in longer clauses than the 3-year-olds, but did not differ from each other. This is in contrast to Currenton & Justice (2004), who observed MLCU differences between all three age groups. • The use of complex elaborated noun phrases was the only discriminating literate language feature, increasing in usage for the 5 year old group.
Currenton, S.M. and Justice, L.M. (2004). African American and Caucasian Preschoolers’ Use of Decontextualized Language: Literate Language Features in Oral Language Narratives. Language, Speech, and Hearing Services in Schools, 35, 240-253. Greenlaugh, K.S. and Strong, C.J. (2001). Literate Language Features in Spoken Language Narratives of Children with Typical Language and Children with Language Impairments. Language, Speech, and Hearing Services in Schools, 32, 114-125. Justice, L.M. & Ezell, H.K. (2001). Written language awareness in preschool children from low-income households: A descriptive analysis. Communication Disorders Quarterly, 22, 123-134. Loban, W. (1976). Language development: Kindergarten through grade twelve (Reearch Report No. 18). Champaign, IL: National Council of Teachers of English. Mayer, M. (1969). Frog, where are you? New York: Dial Books. Wallach, G.P. and Butler, K.G. (1994). Creating Communication, Literacy and Success. In Wallach, G.P. and Butler, K.G. (Eds.), Language Learning Disabilities in School-age Children and Adolescents. New York, NY: MacMillan.
REFERENCES