Printer Friendly

Effects of early intervention on hearing impaired children.

Effects of Early Intervention on Hearing Impaired Children

One of the current trends in services for exceptional children is early intervention. Based on a meta-analysis of 74 studies, Casto and Mastropieri (1986) concluded that early intervention produces moderate gains in the cognitive, linguistic, and academic development of a variety of handicapped populations, although there is little evidence of gains in social and personal development. Increased hours of programming are associated with greater gains. Few studies of handicapped children have investigated whether gains are retained, although studies of disadvantaged populations suggest that benefits might dissipate over time. In addition there is little evidence that "earlier is better"; in fact, Casto and Mastropieri's analysis showed that larger effect sizes occurred with interventions beginning after, rather than before, 36 months of age. Casto and Mastropieri also concluded that programs that involved direct instruction by parents produced significant gains, but were no more effective than those that did not.

There are few data on the effectiveness of early intervention with hearing-impaired children. Summarizing the research prior to 1980, Moores (1982, p. 240) concluded that there is little evidence of lasting benefit. Two recent studies have obtained contradictory results. Seewald, Ross, Giolas, and Yonovitz (1985) found no relationship between preference for auditory versus visual stimuli and either age at which hearing aid usage was initiated or age of first program. White and White (in press), however, found better expressive and receptive language among hearing-impaired children who began their education early. There have been no studies of the effect of increased hours of programming on hearing impaired children, and no studies of the effect of early intervention on social development. The one study of direct parent instruction found no effect (Bodner-Johnson, 1985).

The purpose of the present study was to investigate the effects of age of intervention, program intensity, and direct parent instruction on the language and social development of children with severe and profound hearing losses. The data were drawn from a larger study of the development of hearing-impaired children (Musselman, Lindsay, & Wilson, 1988).

METHOD

Description of the Sample

The population was defined as all children in Ontario with severe or profound hearing losses who were 3, 4, or 5 years of age in the first year of data collection. Parental permissions were obtained from 159 (80%) of the total population of 202. Excluded from the analysis were 14 children with two deaf parents, 21 children whose loss was acquired after 12 months of age, and 6 children who subsequently moved out of the Province. The resulting sample contained 118 children and is broadly representative of the general population of hearing-impaired children with hearing parents, who are prelingually impaired.

Design

The study was longitudinal. Children were first tested when they were 3, 4, or 5 years of age and again in 3 or 4 years. The time of final data collection was scattered across children so as to equalize, as much as possible, the number of children in each age group.

The sample included children from all preschool programs in Ontario, representing the most common service-delivery models: hospital-based clinic programs, home visiting programs, and day school and residential school programs. The first two provide instruction on an individual basis, and the latter two, on a group basis. Both auditory/oral and total communication programs were represented. All programs employed qualified teachers of the hearing-impaired and had access to a full range of audiological, educational, and related support services.

Independent Measures

Children's hearing threshold levels (HTLs) were calculated as the average threshold in the better ear at 500, 1000, and 2000 Hz (American National Standards Institute, ANSI, 1969). The performance section of the Wechsler Intelligence Scale for Children--Revised (WISC-R) intelligence test was administered in the final year. Parent interviews provided information on the educational history of the child, mothers' and fathers' education, and type of instruction provided by parents. The socioeconomic status of each family was coded using Blishen's index, which ranks occupations on the basis of education and income (Blishen & Roberts, 1976).

Dependent Measures

The study included measures of receptive and expressive spoken language, receptive language in the child's educational mode (spoken language alone or simultaneous method), and receptive and expressive language in the mother's mode. A measure of social development was also obtained. The spoken language measures were administered only in the final year because measurable levels of performance were not expected to be evident in the first year. All other measures were administered in both years. The specific measures are briefly described below.

Spoken Language: Reception of Spoken Language (comprehending simple, spoken phrases). The measure used was the 12-item short form of the Assessment of Children's Language Comprehension (ACLC) by Foster, Giddan, and Stark (1973), which has a picture choice format. For this study, the 12 items were presented with the mouth area covered so that only auditory cues were available.

Spoken Language: Production of Spoken Words (producing words in isolation). A test was developed based on previous work by the authors which had identified the words most commonly known by young deaf children (Keeton, Lindsay, & Reich, 1979; Reich, Keeton, & Lindsay, 1981). Each of 24 words was represented by a simple picture. Children were shown the pictures one by one, and received one point for each label that was semantically correct and intelligible.

Receptive Language. The screening test of the Language Assessment Battery (LAB) was used to assess children's comprehension of language in their primary mode of instruction (Keeton, Lindsay, & Reich, 1980). The screening test consists of 41 sentences arranged in order of difficulty, each sentence representing a particular syntactic structure of English. Examples are: agent/action (e.g., The dog is walking.); agent/action/patient (e.g., The boy throws a cookie.); plurals (e.g., The father eats cookies.); and disjunction (e.g., Make the man or the girl stand up.). Items in the test were presented in either spoken language alone or simultaneous method. Simultaneous presentations followed the system of manual English used in the child's class. Children demonstrated comprehension by manipulating objects in a doll's house. A child's score was the number of sentences correctly enacted.

Mother-Child Communication. Mothers were directed to present a set of referential communication tasks to their child. The tasks spanned a range of difficulty from single words to short phrases and stories. Children indicated comprehension through object identification, picture selection, or picture sequencing. Mothers were encouraged to use whatever mode or modes of communication they normally used at home. A child's receptive score was the number of items correct, out of a possible 42. There was also an expressive task, in which mothers asked children to label objects. A child's expressive score was the number of labels that were recognizable, out of a maximum of 24.

Social Development. The Development Profile by Alpern and Boll (1972) was used as a core around which an instrument was developed to measure social development. Sixty-six items were selected from the Alpern and Boll scale which reflected normal social development and which were independent of language ability. These items were supplemented with 88 new items derived from Gesell's (1949) work to provide adequate discrimination throughout the age range. The scales cover the following areas of development: self-help (eating, toileting, dressing, bathing), social relations (sleeping, play and pastimes, personhood, peer relations), and social comprehension (personal space, comprehension, responsibility, functional sequences). A child's score was the total number of items passed.

Characteristics of the Sample

All children in the study had HTLs of at least 65 dB. The overall sample average was 99.9 dB (SD = 14.0). The average age of onset was 1.4 months (SD = 3.3). The average WISC-R performance score was 108.0 (SD = 17.7). The children averaged 51 months of age in the first year and 81 months in the final year of the study.

Age of Intervention. Age of intervention was collapsed into 6-month intervals. Almost half of the sample (48%) had been enrolled in an educational program prior to 24 months of age. Ninety-one percent of the sample were enrolled prior to 36 months of age. All children in the study were aided, having generally received their aid shortly before beginning training.

Program Intensity. The programs in the study varied in intensity. Children in individual programs (i.e., hospital, home-visiting, and itinerant programs) generally received specialized service for relatively few hours per week compared to children who attended classroom programs. Classroom programs themselves varied in the number of days per week and the number of hours per day, depending on the age of the child and the services available. Program intensity was represented for each child by the cumulative number of hours of specialized training. By the first year of the study, the children averaged 608 hours of specialized programming (SD = 660), and by the final year, 1999 hours (SD = 1425).

Parent Instruction. The extent and nature of parents' involvement in their children's education was explored via an in-depth interview, using a flexible but highly structured series of probes. Most parents reported that they attempted to help their children with language development at home (96% in the first year and 92% in the final year). Some described the use of general language stimulation techniques, such as intensive verbal interaction, experiential enrichment, and educational games. This stimulation category comprised 20% of the parents in the first and 39% in the final year. Some parents reported that they provided direct instruction to their child at home. Parents were considered to be providing direct instruction if they described specific teaching objectives and procedures that they followed on a systematic basis under the guidance of a trained teacher of the hearing-impaired. In the first year, 74% of the families met these criteria, dropping to 51% in the final year.

Amount of direct parent instruction was coded into five categories. Children whose parents provided no instruction or who used stimulation techniques, were assigned a value of zero on this variable. Children receiving direct parent instruction were assigned a value of 1 to 4, reflecting the number of hours of instruction received per week, as follows: (a) 0.5 - 2.5 hr; (b) 3.0 - 5.5 hr; (c) 6.0 - 8.5 hr; and (d) 9 + hr. Information on the amount of direct instruction was obtained in the first and final years as well as at the midpoint of the study. For each child, the category value at each of these three points in time was accumulated to represent the total amount of direct parent instruction. This varied from 0 to 4 in the first year and 0 to 12 in the final year.

Analysis

A log transformation of the expressive spoken language measure and a square root transformation of the receptive language measure was used to improve the distribution of these variables, which were not normally distributed. Each dependent variable was subjected to a multiple regression analysis with age, HTL, IQ, age of intervention, program intensity, and direct parent instruction as independent variables. A forced entry model was used, with the independent variables entered one at a time in the order listed. In the case of transformed variables, confirmatory analyses were also conducted on the original scores.

For the receptive language, mother-child, and social development measures, analyses were conducted on the increase in scores between tests. The score in the final year served as the dependent variable and the number of months between tests was added to the list of independent variables. The value entered for direct parent instruction reflected the time spent in direct instruction between test periods, rather than the cumulative amount of time.

The original study included measures of expressive spoken language that were derived from a sample of connected speech. Due to the limited spoken language of the children, these measures did not provide adequate discrimination and are not reported. Analysis of these measures confirmed the general pattern of findings.

Two-by-two (2 x 2) chi-square analyses were used to assess the relationship between parents' instructional style (direct instruction or stimulation) and children's performance (high or low) on each dependent variable. Residual scores from multiple regression analyses of age, HTL, and IQ provided the basis for assigning children to either a high-performance group (i.e., those with residual scores of 0 and above) or low-performance group (i.e., those with residual scores below 0). Those few children whose parents provided neither direct instruction nor stimulation were excluded from these analyses. To analyze the effect of parent instructional style on the increase in scores, children were assigned to performance groups on the basis of the residual increase in scores. Children whose parents provided direct instruction throughout the entire interval between test periods were assigned to the direct instruction group; those whose parents provided stimulation during all or part of the interval were assigned to the stimulation group.

RESULTS

Spoken Language

The regression analysis of receptive spoken language is summarized in Table 1. For each independent variable, the table shows the change in [R.sup.2] associated with the entry of the variable, the significance of the change, the simple and partial correlations, and the value and significance of t in the final equation.

The table shows that age had a significant relationship to receptive spoken language in the final equation. HTL and IQ were associated with significant effects both at initial entry and in the final equation. The effect associated with age of intervention was not significant, either at initial entry or in the final equation. Program intensity was associated with a significant, but negative effect. The effect associated with direct parent instruction was not significant, either at initial entry or in the final equation.

The analysis of expressive spoken language reveals a similar pattern. Age, HTL, and IQ were associated with significant effects, with no effect for age of intervention. Program intensity was associated with a significant, negative effect at initial entry, but not in the final equation. There was, however, a significant and positive effect associated with direct parent instruction in the final equation (t = 2.75, p = .007).

The chi-square analyses of parent instructional style revealed no significant differences between the stimulation and direct instruction groups on either spoken language measure.

Receptive Language

Table 2 summarizes the analysis of the receptive language measure in the first year. Age, HTL, and IQ were associated with significant effects. Age of intervention was associated with a significant effect at initial entry. Inspection of the simple and partial correlations reveals that age of intervention is a suppressor variable. Although the simple correlation between age of intervention and the dependent variable was small, the partial correlation was sufficiently large to make a significant contribution to the equation. Neither program intensity nor direct parent instruction was associated with significant effects. Following entry of direct parent instruction into the equation, the contribution of age of intervention was reduced to marginal significance, due to the presence of a weak negative correlation between age of intervention and direct parent instruction (r = -.19).

Analysis of the receptive language measure in the final year again found significant effects associated with age, HTL, and IQ. Significant effects were not associated with either age of intervention or program intensity. There was, however, a marginally significant, negative effect associated with direct parent instruction (t = -1.80, p = .075).

Analysis of the increase in scores between years showed no effects associated with either age of intervention, program intensity, or direct parent instruction. Chi-square analysis of parent instructional style revealed no significant differences in either the first or final year or the increase in scores between years.

Mother-Child Communication

Regression analysis of the receptive mother-child communication measure found significant effects associated with age and IQ in both years. HTL was associated with significant effects at initial entry in both years, but not in the final equation. Age of intervention, program intensity, and direct parent instruction were not significantly related to the receptive communication score in either year or the increase in scores between years. The chi-square analysis showed no significant effects associated with parent instructional style.

Regression analysis of the expressive mother-child communication measure showed significant effects associated with age in both years. HTL and IQ were associated with significant effects only in the first year; this may reflect the presence of a ceiling effect in the final year, as the mean score was 22.0 out of a possible 24 (SD = 2.43). There were no significant effects associated with age of intervention, program intensity, or direct parent instruction in either year. The chi-square analysis showed no significant relationship associated with parent instructional style.

Neither age of intervention nor direct parent instruction was significantly related to the increase in expressive mother-child communication scores between years. There was a significant, positive effect associated with program intensity on the increase in scores (t = 2.12, p = .038). Chi-square analysis of parent instructional style showed that the majority of children receiving direct instruction fell into the high-gain group (26/37), while the majority of those receiving stimulation (29/55) fell into the low-gain group ([X.sup.2] = 4.34, p = .037).

Social Development

Analysis of the social development measure revealed significant effects associated with age in both years, and with IQ in the final year. HTL was not significantly related to social development in either year. There were no significant effects associated with age of intervention, program intensity, or direct parent instruction in the first year, the final year, or the increase in scores between years. Chi-square analysis of parent instructional style revealed that the majority of children receiving stimulation had high scores in the final year (31/44), while the majority of those receiving direct instruction (29/57) has low scores ([X.sup.2] = 4.65, p = .031).

DISCUSSION

Age of Intervention

Age of intervention was significantly related to the receptive language measure in the first year. The relationship in the final year was not significant, and no significant relationships were found between age of intervention and any other measure.

Taking the age of the sample into consideration, it is possible to reconcile the findings of this study with those from other studies of hearing-impaired children. White and White (in press) obtained favorable results associated with age of intervention, but the children in their study were younger at the time of testing than most of the children in the present study. Using an older sample than the present study, Seewald et al. (1985) found no effects associated with age of intervention. Thus the findings from studies of hearing-impaired children are consistent with an hypothesis that early intervention produces short-term gains.

It is important to keep in mind that most children in the sample were enrolled in an educational program prior to 3 years of age. This study, therefore, concerns the impact of intervention beginning in infancy compared with intervention beginning in the preschool period. Different results might have been obtained with a different distribution on this variable.

Program Intensity

Program intensity was positively related to the increase in expressive mother-child communication. There were no other positive effects. The results, therefore, do not provide strong support for the value of intensified programming, a finding that differs from the general conclusions of Casto and Mastropieri (1986).

Significant negative relationships were found between program intensity and the two measures of spoken language. It can be hypothesized that these findings reflect the reciprocal relationship that exists between program intensity and degree of integration. To explore this possibility, children were divided into two groups: integrated children were those who had spent more than half of their total instructional time in a regular classroom; segregated children were those who had spent at least half of their time in a segregated classroom. In the final year, when the measures of spoken language were obtained, 39% of the sample fell into the integrated group, and 61% into the segregated group.

These two groups were compared using analysis of variance with age, HTL, and IQ as covariates. The integrated group was found to score considerably higher than the segregated group on both measures of spoken language. On receptive spoken language, the adjusted mean score of the integrated group was 7.4, compared with 5.0 for the segregated group (F(1,97) = 14.75, p [is greater than less] .000). On expressive spoken language, the integrated group had an adjusted mean of 14.6, compared with 10.0 for the segregated group (F(1,97) = 9.78, p = .002). There were no significant differences between these two groups on any other measure. Given the small size of the integrated group, it is likely that these results reflect the selective placement of children with better spoken language into integrated settings, rather than the outcome of integrated placement.

Parent Instruction

Direct instruction by parents was positively related to the expressive spoken language measure and the increase in expensive mother-child communication. There were no other positive relationships between direct parent instruction and outcomes. In the final year, there was a marginally significant and negative relationship between direct parent instruction and receptive language, and a significant, negative relationship between direct parent instruction and social development. This pattern of results does not provide support for the value of direct parent instruction, a finding which is consistent with the conclusions of Casto and Mastropieri for a general population and that of Bodner-Johnson for a hearing-impaired population.

Conclusion

The study failed to obtain evidence of lasting gains associated with intervention during infancy, intensified programming, or direct instruction by parents, findings that are in general agreement with previous literature. Given this pattern of findings, we must ask why early intervention continues to receive widespread support within the educational community.

A number of hypotheses might be advaned. As has been suggested for disadvantaged populations (Bronfenbrenner, 1974), it is possible that early intervention accelerates children's acquisition of formal language and test-taking skills, an advantage that is soon dissipated once the remainder of the population enters school. Alternatively, it is possible that long-term benefits do result from early intervention, but that the measures used in research have not been sufficiently sensitive to detect them.

A second hypothesis is suggested by findings from the present study showing that age of intervention was negatively related to the family's socioeconomic status (r = -.25, p = .004), mothers' years of education (r = -.16, p = .040), and fathers' years of education (r = -.32, p = .01). Similar relationships occurred between these variables and direct parent instruction. Thus children of higher status and better educated parents tended to receive intervention earlier, and their parents were more likely to provide direct instruction. The study found that these family characteristics were not significantly related to the dependent variables. It might be hypothesized, however, that a halo effect operates whereby educators subjectively evaluate children from advantaged backgrounds more favorably than those from less advantaged families.

A final hypothesis is that some types of intervention during infancy do produce long-term gains, or do produce gains in some children. An important limitation of the present study is the fact that qualitative aspects of programming and parent instruction were not considered. The results, therefore, reflect the effectiveness of general educational practice, rather than the effects of best practice. The lack of general effects for these variables suggests that the quality of programming is critical, and that general practice needs to be improved before it achieves its objectives on a broad scale. Future research should attempt to identify the parameters of effective early intervention and the type of children who do benefit.

REFERENCES

Alpern, G. D., & Boll, T. J. (1972). Developmental profile. Aspen, CO: Psychological Development Publications.

Blishen, B. T., & Roberts, H. A. (1976). A revised socioeconomic index for occupations in Canada. Canadian Review of Sociology and Anthropology, 13, 5-15.

Bodner-Johnson, B. (1985). Families that work for the hearing-impaired child. The Volta Review, 87, 131-139.

Brofenbrenner, U. (1974). A report on longitudinal evaluations of preschool programs. (Vol. II). Is early intervention effective? (Department of Health, Education and Welfare, OHD 73-0025). Washington, DC: U.S. Government Printing Office.

Casto, G., & Mastropieri, M. A. (1986). The efficacy of early intervention programs: A meta-analysis. Exceptional Children, 52, 417-424.

Foster, R., Giddan, J. J., & Stark, J. (1973). Assessment of children's language comprehension. Palo Alto, CA: Consulting Psychologists Press.

Gesell, A. L. (1949). Child development: An introduction to the study of human growth. New York: Harper & Row.

Keeton, A., Lindsay, P., & Reich, C. (1979). The language assessment battery (LAB): Final report. Toronto: The Ontario Institute for Studies in Education.

Keeton, A., Lindsay, P., & Reich, C. (1980). The language assessment battery: Teacher's manual (rev.). Toronto: The Ontario Institute for Studies in Education.

Moores, D. F. (1982). Educating the deaf: Psychology, principles, and practices (2nd ed.). Hopewell, NJ: Houghton Mifflin.

Musselman, C. R., Lindsay, P. L., & Wilson, A. K. (1988). An evaluation of current trends in preschool programming for deaf children. American Journal of Speech and Hearing Disorders, 53, 71-88.

Reich, C., Keeton, A., & Lindsay, P. H. (1981). The language assessment battery for hearing-impaired children. The Association of Canadian Educators of the Hearing-Impaired (ACEHI) Journal, 7, 155-163.

Seewald, R. C., Ross, M., Giolas, T. G., & Yonovitz, A. (1985). Primary modality for speech perception in children with normal and impaired hearing. Journal of Speech and Hearing Research, 28, 36-46.

White, S. J., & White, R. E. C. (in press). The effects of hearing status of the family and age of intervention on receptive and expressive oral language skills in hearing-impaired children. In H. Levitt, N. McGarr, & D. Geffiner, (Eds.), Speech, language and communication skills of deaf children: ASHA Monographs. Rockville, MD: American Speech and Hearing Association.

CAROL REICH MUSSELMAN is Associate Professor, and ANNE KEETON WILSON is Associate Professor and Chair, and PETER H. LINDSAY is Associate Professor, Department of Special Education, The Ontario Institute for Studies in Education, Toronto, Ontario, Canada.
COPYRIGHT 1988 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1988 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Musselman, Carol Reich; Wilson, Anne Keeton; Lindsay, Peter H.
Publication:Exceptional Children
Date:Nov 1, 1988
Words:4290
Previous Article:Special education teacher preparation: a synthesis of four research studies.
Next Article:Current disciplinary practices with handicapped students: suspensions and expulsions.
Topics:


Related Articles
The many faces of early intervention.
Mathematics achievement of hearing impaired adolescents in different placements.
Play behavior of hearing impaired children: integrated and segregated settings.
Promoting social interaction between young children with hearing impairments and their peers.
Early cognitive development and American Sign Language.
Depression may impair child development.
South City News: Hearing is on agenda at meeting; ASTON.
Using electropalatographic feedback to treat the speech of a child with severe-to-profound hearing loss.
Language Partners: Building a Strong Foundation.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters