Die gebruik van onsinwoordherhaling, regtewoordherhaling, sinsherhaling en getalherhaling in die indentifisering van kinders met spesifieke taalgestremdheid

Gagiano, Salomé (2013-12)

Thesis (MA)--Stellenbosch University, 2013.

ENGLISH ABSTRACT: As South Africa is a culturally and linguistically diverse country, there is a great need for an accurate evaluation instrument that can identify children with specific language impairment (SLI) early, regardless of their language group. The language tests that are available in South Africa are mostly available in only English or Afrikaans (see Penn 1998). In the literature four repetition tasks have been identified as potential markers of SLI: nonword repetition (Coady and Evans 2008:11), sentence repetition and digit repetition (Ziethe, Eysholdt and Doellinger 2013:1) as well as real word repetition (Dispaldro, Benelli, Marcolini and Stella 2009:941). The purpose of this study was to compare the sensitivity of these four potential markers with each other in an attempt to compile an accurate measuring instrument of SLI for Afrikaans and English. The study also aimed to collect baseline data with isiXhosa-speaking children. The participants consisted of five groups of five-year-olds: The first three groups consisted of 20 typical developing (TD) Afrikaans-, 20 TD English-, and 20 TD isiXhosa-speaking children. The fourth and fifth group consisted of five Afrikaans-speaking and five English-speaking children with SLI. All four repetition tasks were performed with both Afrikaans- and both English-speaking groups; only digit repetition and nonword repetition were performed with the isiXhosa-speaking group. Refined and graded items were (i) selected from longer lists of items that were used in the pilot study, and (ii) based on the results of the pilot study, were recorded on a CD in a recording studio. This recording served as method of presentation of the items to each participant individually during the main study. There were no statistically significant differences between the performance of the TD Afrikaans- and the TD English-speaking groups. The TD isiXhosa-speaking group, however, performed significantly better on the nonword repetition task but worse on the digit repetition task than the other two TD groups. Both groups with SLI performed poorly, and significantly poorer than their corresponding TD group, in all four of the repetition tasks. For the Afrikaans-speaking children, sentence repetition followed by real word repetition distinguished best between participants with and without SLI. Sentence repetition, followed by nonword repetition, was shown to be most sensitive in distinguishing TD children from children with SLI in the English group. The percentage sounds/words correctly repeated per word/sentence, for both the nonword repetition task and the real word repetition task, was more accurate than the raw scores for the Afrikaans- and English-speaking groups. Some items did prove to be more sensitive for the difference between the performance of the TD and the SLI groups in Afrikaans and English. These more sensitive items may be appropriate for inclusion in a screening tool for SLI for use with five-year-olds. In conclusion, the results of this study confirm all four repetition tasks to be markers of SLI. Sentence repetition and nonword repetition were more sensitive markers of SLI in English-speaking five-year-olds, with sentence repetition and real word repetition being the more sensitive markers of SLI in Afrikaans-speaking five-year-olds. The isiXhosa-speaking group performed similarly to the Afrikaans- and English-speaking groups in the nonword repetition- and digit repetition tasks. Based on these results, recommendations are made regarding the content and presentation mode of a screening test comprising repetition tasks for the identification of possible SLI in Afrikaans- and English-speaking children.

AFRIKAANSE OPSOMMING: Aangesien Suid-Afrika ‘n kultureel en linguisties diverse land is, is daar ‘n groot behoefte aan ‘n metingsinstrument wat kinders met spesifieke taalgestremdheid (STG) , uit alle taalgroepe, vroegtydig en akkuraat kan identifiseer. Die taaltoetse wat wel in Suid-Afrika gebruik kan word, is bykans slegs in Afrikaans of Engels beskikbaar (kyk Penn 1998). In die literatuur is daar vier herhalingstake geïdentifiseer wat potensiële merkers van STG kan wees: onsinwoordherhaling (Coady en Evans 2008:11), sinsherhaling en getalherhaling (Ziethe, Eysholdt en Doellinger 2013:1) asook regtewoordherhaling (Dispaldro, Benelli, Marcolini en Stella 2009:941). Die doel van hierdie studie was om die sensitiwiteit van hierdie vier potensiële merkers met mekaar te vergelyk in ‘n poging om ‘n akkurate metingsinstrument van STG vir Afrikaans en Engels op te stel. Basislyndata oor herhaling deur isiXhosasprekende kinders is ook in hierdie studie ingesamel. Die deelnemers het uit vyf groepe vyfjariges bestaan. Die eerste drie groepe was 20 tipies ontwikkelende (TO) Afrikaans-, 20 TO Engels-, en 20 TO isiXhosasprekendes. Die vierde en vyfde groep het uit vyf Afrikaanssprekende- en vyf Engelsprekende kinders met STG bestaan. Al vier herhalingstake is met die twee Afrikaans- en die twee Engelssprekende groepe uitgevoer; slegs getalherhaling en onsinwoordherhaling is met die isiXhosasprekende groep uitgevoer. Verfynde en gegradeerde items geselekteer (i) uit die langer itemlyste gebruik in die loodsstudie en (ii) op grond van die loosstudieresultate, is op CD opgeneem in ‘n klankateljee, en dié opname is gebruik as die metode van aanbieding van die items aan elke deelnemer individueel tydens die hoofstudie. Daar is geen statisties beduidende verskil tussen die prestasie van die TO Afrikaans- en TO Engelssprekende-groepe gevind nie. Die TO isiXhosa-sprekende groep het egter beter as hierdie twee groepe gepresteer in onsinwoordherhaling maar swakker in getalherhaling. Die twee STG-groepe het beide swak, en beduidend swakker as hul ooreenstemmende TO-groep, op alle herhalingstake gevaar. Vir Afrikaanssprekendes het sinsherhaling, gevolg deur regtewoordherhaling, die beste tussen die deelnemers met en sonder STG onderskei, terwyl die mees sensitiewe take vir die Engelssprekende deelnemers sinsherhaling gevolg deur onsinwoordherhaling was. Persentasie klanke/woorde korrek herhaal per woord/sin was in die geval van onsinwoordherhalings- en regtewoordherhalingstake vir beide Afrikaans en Engels meer akkuraat as wat roupunttellings was. Sommige items in elk van die herhalingstake was beide vir Engels en Afrikaans meer sensitief vir die verskil tussen die prestasies van die TO- en STG-groepe. Hierdie meer sensitiewe items mag toepaslik wees vir insluiting in ʼn siftingstoets vir STG vir gebruik met vyfjariges. Opsommend: Die resulate bevestig al vier herhalingstake as merkers van STG. Sinsherhaling en onsinwoordherhaling is meer sensitiewe merkers van STG vir Engelssprekende vyfjariges, met sinsherhaling en regtewoordherhaling wat meer sensitiewe merkers van STG vir Afrikaanssprekende vyfjariges is. Die isiXhosasprekende groep het soortgelyk aan die Afrikaans- en Engelssprekende groepe ten opsigte van onsinwoordherhaling en getalherhaling presteer. Op grond van die resultate word aanbevelings gemaak vir die inhoud en aanbeidingswyse van ‘n siftingstoets bestaande uit herhalingstake vir die identifisering van moontlike STG in Afrikaans- en Engelssprekende kinders.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/85717
This item appears in the following collections: