CURRENT MANAGEMENT FOR WORD FINDING DIFFICULTIES BY SPEECH-LANGUAGE THERAPISTS IN SOUTH AFRICAN REMEDIAL SCHOOLS

Word finding difficulties frequently found in learners with language learning difficulties (Casby, 1992) are an integral part of SpeechLanguage Therapists’ management role when working with learning disabled children. This study investigated current management for word finding difficulties by 70 Speech-Language Therapists in South African remedial schools. A descriptive survey design using a quantitative and qualitative approach was used. A questionnaire and follow-up focus group discussion were used to collect data. Results highlighted the use of the Renfrew Word Finding Scale (Renfrew, 1972, 1995) as the most frequently used formal assessment tool. Lan­ guage sample analysis and discourse analysis were the most frequently used informal assessment procedures. Formal intervention pro­ grammes were generally not used. Phonetic, phonemic or phonological cueing were the most frequently used therapeutic strategies. The authors note strengths and raise concerns about current management for word finding difficulties in South African remedial schools, par­ ticularly in terms of bilingualism. Opportunities are highlighted regarding the development of assessment and intervention measures rele­ vant to the diverse learning disabled population in South Africa.


INTRODUCTION
Learners with language learning disabilities frequently ex perience word finding difficulties (Casby, 1992;German, 1998) as cited by German, 2005) Word finding difficulties (WFDs) are an integral aspect of the Speech-Language Therapist's management practices when working with a learner who has a language learning disability (LLD).Using The Royal College of Speech Therapists' definition of practice (1998), hereinafter, management refers to the assessment, formulating and implementing intervention pro grammes and decision making about discharge from therapy.Many factors impact on the complex management of developmen tal WFDs by South African Speech-Language Therapists (SLTs), including: controversies about the nature and impact of WFDs; the training of SLTs in WFDs; the role of the SLT in managing WFDs; the context in which services are provided; the model of service delivery in use; policy issues including inclusive educa tion; outcomes based education and language issues.Furthermore, Speech-Language Therapists (SLTs) have experienced frustration with the outcome of therapy for WFDs (Haynes, 1993).
"Learning disability is a general term that refers to a hetero geneous group o f disorders manifested by significant difficulties in the acquisition and use of listening, speaking, reading, writing, reasoning or mathematical abilities" (National Joint Committee on Learning Disabilities (1991, as cited by Owens, 2004).South Afri can learners with learning disabilities exhibit discrepancies be tween their intellectual potential (usually average to above aver age) and expected academic performance and are typically admit ted to remedial schools (S, Mohamed, personal communication, August, 24* 2004).Speech-language therapy services for learners with LLD and WFDs in the South African context are frequently provided in the context of remedial education.Within the current educational system in South Africa, learners with LLD (and possi ble WFDs) are educated at long and short-term remedial schools or within mainstream schools.SLTs' practice in different educational settings is shaped and influenced by current and changing educa tional policies such as Outcomes^ Based Education (Department of Education, 2001).Another policy influencing the practice of SLTs is inclusive education where the focus is on overcoming barriers in the system that prevent meeting learning needs by adapting the support systems in the classroom (Department of Education, 2001).Policies like this impact on service delivery models for learners with LLD and possible WFDs viz.individual versus group therapy, traditional pull-out system or classroom-based therapy.
WFDs in children with LD, who meet the criteria for ad mission into a remedial school, are pervasive in nature, and impact on their communicative and academic performance (Faust, Dimitrovsky & Shacht, 2003;Lemer, 2000).This takes the form of dif ficulties with confrontation naming, reading, and conversational discourse.Secondary behaviours such as hesitations, repetitions and re-formulations occur (Dockrell, Messer, George & Wilson, 1998) and may result in inaccurate messages being conveyed (Faust et al., 2003).It is therefore important for WFDs to be sys tematically identified, assessed and treated as early as possible in the learner's educational career (Faust et al., 2003).Speechlanguage therapy interventions for WFDs have been reported to be ineffective (Dockrell et al., 1998), and have thus not enabled the learner to achieve the academic and communicative competence required for progress in education.

The causes of Word Finding Difficulties
There is no unequivocal explanation for the presence of WFDs (Casby, 1992).Within a psycholinguistic framework, lim ited processing resources, inadequate or ill-defined semantic repre sentations, inadequate phonological specifications and naming speed deficits are possible contributing causes of WFDs (Dockrell et al., 1998).
Within the processing view, the two inter-related compo nents of the initial learning, organisation and storage of informa tion that impacts on the learning o f vocabulary (Easton, Sheach & Easton, 1997), and the ability to retrieve words (Macgregor & Leonard, 1989, cited in Easton et al., 1997) may impact on WFDs.Thus storage strength (i.e.how well a word is learned) and re trieval strength (i.e.how easily a word is retrieved) are important.
A debate exists about the types of word-knowledge viz.semantic The South African Journal o f Communication Disorders, Vol. 53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.) and phonological features of a word and the link between them in relation to storage and retrieval strength (Easton et al., 1997).
An additional view of the cause of WFDs is that of inade quate semantic specification (Easton et al., 1997).Semantic defi cits result in simple semantic networks, with limited options for associations for stimulus words, which may result in oral and writ ten WFDs (Casby, 1992).
A further causal view is that WFDs are attributable to an inadequate phonological representation (Easton et al., 1997).In adequate phonological representation could, according to a proc essing model used by Chiat and Hunt (1993), indicate difficulty at an input (phonological representation) or output (phonological ac cess) level.Naming difficulties may suggest difficulties in phono logical storage (Webster, 1994).Deficits in phonological access may present as difficulties in the naming of familiar words such as colours, digits, letters or objects (Clark, 1988 as cited by Webster, 1994).
However while it is not always easy to determine what the underlying cause of the child's WFD is (Easton et al., 1997), it is important to assume an interaction between the two processes of semantic representation and phonological specification.

Assessment for WFDs
According to German (1989a), a comprehensive assessment for WFDs should incorporate naming tasks, a comprehension task and the identification of word finding characteristics.A thorough assessment of WFDs is needed for bilingual learners (German, 1989a), which is of particular importance in the South African context.Learners acquiring English as a second language, particu larly in the school setting, should have their abilities assessed us ing tools sensitive to their cultural and linguistic backgrounds (Cheng, 1994;Montgomery, 1999 as cited by Apel, 2001).
The strategies for assessing WFDs may be formal and in formal (Wiig & Semel, 1984).Spontaneous language sample analysis is an informal tool suggested by Wiig and Semel (1984).A number of formal or standardised assessment procedures of Brit ish and American origin used by SLTs to assess word finding are highlighted in available literature, but may be of limited applica tion to the South African learner.Wiig and Semel's (1984) guide lines for selection of procedures for assessing WFDs highlights the need to have a range of assessment contexts and procedures to diagnose or to determine the nature, severity and impact of a WFD.
To provide an in-depth assessment of word finding abilities and difficulties, one has to select multiple procedures with a num ber of tasks which require naming in a variety of contexts (Wiig & Semel, 1984), such as: • Rapid naming upon confrontation of visual stimuli (such as the Renfrew Word Finding Scale 'Renfrew, 1972'Renfrew, , 1995)) German and Simon (1991), also highlight the importance of as sessing children's WFDs in discourse.While a battery of reliable and valid measures, which allow for the profiling of WFDs, is important, the limitations of formal assessment tools applied to the South African context need to be acknowledged.Furthermore, a shift from standardised testing to criterion-referenced and educator-produced tests is recommended (Department of Education Directorate: Inclusive Education, 2002)

Intervention
Word finding intervention is complex with no single treat ment approach available that is known to address all WFDs.For storage difficulties, therapy should identify and elaborate semantic features of words.For retrieval difficulties, therapy should aim at semantic categorisation or visual imagery (Wiig & Becker-Caplan, 1984as cited in German, 1992).When working with storage, one should aim to increase vocabulary, targeting common, frequent and meaningful words and those with concrete referents.The words should be taught in an orderly way, such as by using themes, and should highlight categorical and associative relation ships (Nippold, 1992).
Aims of intervention for WFDs (Nippold, 1992) include hierarchically enhancing naming accuracy, improving retrieval strength and increasing the use of strategies by increasing metacognitive and meta-linguistic awareness of factors that affect word finding ability, such as looking at meaning changes across contexts (Nippold, 1992).
The focus of intervention differs if a learner presents with WFDs in spontaneous language or discourse.Therapy should fo cus on increasing accuracy and fluency rate.A number of retrieval strategies, such as phonological, visual and semantic categories in discourse, need to be used and taught (German, 1992).Response to particular types of intervention also informs us of the nature of the learner's problem in relation to reading difficulties.Learners with single phonological core deficits respond well to phonologically based therapy, which aims to increase phonemic awareness.Learners with single processing speed deficits or a double-deficit (a deficit in both areas) are typically only partially served by such intervention.They would require emphasis on building fluency in reading and building automaticity in lower level processes serving these skills (Wolf & Segal, 1992).
From the above it is evident that working with learners with LLD and WFDs is a complex area, impacted on by factors such as changing educational policy.It is also an area where SLTs have experienced frustration with the outcome of therapy (Haynes, 1993).
Since WFDs are frequently found in learners with LLD (Casby, 1992) and are an integral part of the SLTs' management when working with LLD, the purpose of this study was to deter mine current management for WFDs in learners with LLD within the context of South African remedial schools, an undocumented area of investigation.Such investigation may highlight current strengths and difficulties in managing learners with WFDs and serve to guide SLTs toward the required shift in management within the South African educational setting.SLTs' opinions on the management of learners with LLD and WFDs'currently used in the various remedial educational institutions were important to as certain how these management practices meet the current needs of therapists and learners.

METHODOLGY Aim
The aim of the study was to determine SLTs' current as s e s s m e n t and intervention practice for word finding difficulties in the language learning disabled population in South African reme dial schools.Critical questions needed to be asked in order to meet the aim of the study.These included: 1. What is the current practice in terms of assessment proce dures used for learners with word finding difficulties?2. What is the current practice in terms of intervention pro cedures used for learners with word finding difficulties?

Research design
The nature of this research study was descriptive.The de sign employed could be further described as a descriptive survey as it allowed the researcher to describe characteristics of a sample at a particular point (Leedy, 1997).As it is important to collect data for qualitative research using a skilled individual rather than a single instrument, (Leedy, 1997), a questionnaire (quantitative) and a focus group discussion (qualitative) were used (McMillan & Schumacher, 2000).SLTs employed at remedial schools were par ticipants and it was their responses to the questionnaires that formed a significant basis of the research study.

Participant selection criteria
Participants needed to be SLTs working in South African government remedial schools.Furthermore, they needed to be reg istered with the Health Professions Council of South Africa and the South African Council of Educators.Participants needed to be willing to participate and learners with LLD and WFDs had to be part of their caseloads.
Figure 1 summarises the process of obtaining approval to conduct research and of participant selection.

Profile of participants
A summary about the SLT participants is provided in Table 1.L earners with LLD and W FDs had to be part o f their caseloads.
Participants needed to be willing to participate.

Reasons
The focus of the study was the man agem ent o f W FDs by SLTs in this particular context and SLT s were to be the participants.
This was to ensure that the partici pant SLTs had the necessary qualifi cations and registrations to work in this area.
This population was the focus of the research study.This is an im portant ethical consid eration.

Data collection methods
The data collection procedure involved two phases: ad ministration of the questionnaire (survey method) and follow-up focus group discussion.The focus group discussion is a strategy to obtain a better understanding of information obtained in the re search procedure (McMillan & Schumacher, 2000).Focus groups are unstructured interactions in a small group.Interactions are with group members and the group leader or researcher (Bowling, 1997).The questionnaire was administered following completion of a pilot study using four qualified SLTs with relevant experience who were not part of the study sample.

Description o f the Questionnaire
All questions were developed using the critical questions as a guide and were developed using relevant theory.A total of 23 questions were developed (refer to Appendix A).These questions were divided into three sections focussing on demographics, prac tice of WFDs and efficacy of intervention for WFDs.Nineteen of

APPROVAL from PRINCIPLES
Gauteng: 8 from 9 Kwazulu Natal: 11 from 12 Western Cape: 5 from 5 Questionnaires Sent Gauteng 62 Questionnaires Sent KwaZulu Natal Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.) the questions were close-ended questions.The questions fol lowed a Likert Scale format (Bailey, 1991).A statistical pro gramme, Statistical Package for the Social Sciences (SPSS) (SPSS Inc., 1995), was used to explore patterns identified using frequency counts and percentages.Four open-ended questions were included in the questionnaire allowing for richer data to be obtained: The responses to the open-ended questions were quali tatively analysed.All questionnaires were coded using letters and numbers on questionnaires to ensure anonymity and confi dentiality.

Focus Group Discussion Schedule
Information obtained from the questionnaire needed to be explored further to improve the quality and richness of the infor mation obtained (McMillan & Schumacher, 2000).This was achieved by developing a focus group discussion schedule which documented the open-ended questions to be investigated in an unstructured way (McMillan & Schumacher, 2000) using questions to probe further information where necessary.
Participants in the focus group discussion were from Kwa-Zulu Natal which allowed ease of access.Eight partici pants attended the single focus group discussion at which the researcher and a scribe were present.

Data analysis
Data analysis for this research embodied both quantita tive and qualitative analysis as the research tools used for data collection characterised both types of research.According to Vithal and Jansen (1997) data analysis includes three steps, sum marised as ( 1) scanning and cleaning the data, ( 2) organizing the data and (3) representing the data.
These steps and how they were applied to this study are presented in Table 2.

Table 2: Description of the 70 Participant SLTs
The research procedure involved a number of pivotal phases.Each phase was dependant on the completion of the previous phase.A summary of the research procedure, high lighting the phases is presented in Figure 2.

Ethical considerations
Using guidelines outlined by Leedy (1997) certain ethical issues were considered for this research study.Potential partici pants were informed in writing about the research study.Informa tion included an ethical clearance number (03140A), who was un dertaking the research, why it was being undertaken and who fi nanced it.A letter to possible participants included a consent form, (informed consent) indicating willingness to participate! in the re search.Letters to participants highlighted that they could withdraw from the study at any time, or refuse to participate without any pen alty.Confidentiality in terms of the participants, both SLTs and remedial schools, was guaranteed.All the information recorded in this research document was coded in order to uphold this.Further more, the participants were assured that once the study was com pleted, audiocassettes used for the recording of the focus group discussions would be destroyed.Results of this study were pro vided to the participants.A summary of the findings and the impli cations thereof, were posted, faxed or e-mailed to the individuals, institutions, Superintendent Generals o f the provinces and all the relevant school principals that participated.

Reliability and validity
For the purpose of this study, a pilot study was conducted to Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)refine the questionnaire, with the aim of increasing reliability.Ecological validity was evaluated by the extent to which findjngs could be generalised beyond the find ings of the present study (McMillan & Schumacher, 2000).Strategies used to in crease validity in the context of this re search included varied data collection strategies (McMillan & Schumacher, 2000) (for this study, a questionnaire and focus group discussion).This resulted in the tri angulation of data across techniques, where consistency of findings by different data collection methods was checked (Bailey, 1991).
It was important to be aware of the Hawthorne effect, where participants may have acted differently since they were par ticipants in a research project (McMillan & Schumacher, 2000).Furthermore, acknowl edging bias, particularly in descriptive re search, was important since it is inevitable (Leedy, 1997)

RESULTS
Results from the questionnaire, com pleted by 70 SLTs in South African reme dial schools, and from the focus group dis cussion, involving SLTs in the greater Dur ban area in KwaZulu Natal follow.The re sults are presented and discussed in two main sections: assessment and intervention practice.

Assessment Procedures
Results indicated that participant SLTs usually used both standardised and informal assessment procedures when as sessing for WFDs (question 13)."Other" assessment procedures used by only 8% of participants, included subjective measures, speaking to parents and other members of staff and observation.A summary of re sults obtained is presented in Figure 3.

Standardised Assessments
Participants were asked to indicate which standardised assessment tools they used in assessing WFDs in learners with LLD (question 14).It was found that the Renfrew Word Finding Scale (Renfrew, 1972(Renfrew, , 1995) ) was by far the most fre quently used test to assess WFDs.
Eighty three percent of the partici pants who answered this question used either the Renfrew Word Finding Scale (RWFS) only or used it together with some other test(s).The results are summa rized in Figure 4.
The use of the RWFS as a tool for assessing WFDs was further explored in the focus group discussion to determine participants' views of this tool for identi fying WFDs.Participants at the focus group discussion offered views including that the RWFS depended on an intact vo cabulary, identified only severe WFDs, needed to be used with another tool and did not effectively test for WFDs, as indi cated by the following excerpt:

"...I d o n 't believe i t 's m ore effective than
ju st even a spontaneous language sam ple charting the W FD.... "

Informal Measures Used In the Assess ment o f Learners with LLD and WFDs
On the questionnaire, participants were asked to indicate which informal measures were used to identify WFDs (question 15).Language sample analysis (LSA) or discourse analysis (DA) was the most frequently used informal measure to identify WFDs.Thus 77 % of the partici pants in the sample who answered the question used either LSA or DA only or together with some other measure(s).These results are summarised in Figure 5.

SLTs View o f Assessment Procedures fo r WFDs
The results showed that 45 % of participant SLTs in the sample disagreed that assessment procedures were relevant to the South African population but 58 % agreed that the procedures assisted in identifying WFDs.Thirty four percent disagreed and 31 % remained neutral re garding assessment procedures providing sufficient detail to provide treatment goals.
The results suggest that almost fifty percent of the participants working with WFDs in LLD were aware that the formal tools used were not relevant to the South African population.1.The m easures listed under "other" include subjective m easures, speaking to parents and other m embers of staff, observation and a com bination of standardised tests and inform al measures.2. The calculation of the mean is based on allocating 0 for a "never" response, 1 for a "som e tim es" re sponse, 2 for a " usually" response and 3 for an "always" response.Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)The second area of focus was cur rent management practice in terms of in tervention procedures used for learners with LLD and WFDs (question 16).More than 60% of the participants in the sample did not use any formal interven tion programme.Of those who did use a formal programme, a variety of pro grammes were given, with no particular one being used by the majority of schools in the sample.The results obtained are summarized in Figure 6.
The findings suggested that pho netic cueing combined with some "other" strategy was the informal therapeutic strategy most used in working with LLD children with WFDs (question 17).Ninety one percent of the participants who answered the question used either phonetic cueing only or together with some other strategy or strategies.Other informal strategies included for example: categorizing, associations, mind map ping, cloze tasks and visual cueing.A summary of the results indicating the therapeutic strategies used in interven tion, is presented in Figure 7.

Administering Therapy
Participants provided either group, individual or both individual and group therapy to learners with LLDs and WFDs (question 10).A summary of results is presented in Figure 8.
Using the Chi Square Test of Association, no significant differences between the provision of individual ver sus group therapy were noted.'However,qualitatively, some differences were identified.This issue was explored dur ing the focus group discussion and re vealed the following about intervention for learners with WFDs.Three I variables were identified as influencing' whether SLTs provided intervention for WFDs individually or in a group.These vari ables were the SLT's caseload, the popu lation of learners at the school, and deci sions of school management.Participants in the focus group discussion -indicated that the greater the caseload, the more likely group intervention was provided.Further, the more severe,a learner's diffi culty the less likely they were to receive group therapy.In terms of caseload, two thirds of the participants in the sample saw 10 or--less learners with LLD and WFDs per week.These results are sum marized in Figure 9.

SLTs' Views on Intervention Procedures Used fo r WFDs
An equal number of participants in the sample (36 %) agreed or were neutral in terms of intervention procedures being relevant to the South African population of learners with WFDs and LLDs (question 21).Sixty one percent of the participants agreed that intervention pro cedures were useful in addressing WFDs.

Continuing Professional Development in the Area o f WFDs
There was an overwhelming re sponse (100%) where all participant SLTs in the sample wanted WFDs to be covered in continuing professional development programmes.Due to this overwhelming response, continuing professional develop ment as an issue was further explored dur ing the focus group discussion.The results are discussed in the discussion of current practice in terms of intervention proce dures.

Current Assessment Practice for WFDs in Learners with LLD
An in-depth and thorough assess ment for WFDs in learners with LLD with multiple procedures and a number of tasks in a variety of contexts is recommended (Wiig & Semel 1984, Wolf & Segal 1992, German, 1989a).Despite this, the RWFS as a single tool was by far the most fre quently used standardised assessment tool for identifying WFDs.A number of possi bilities for this assessment practice exist.These include training in WFDs, knowl edge bases, available resources and poli cies.Knowledge from undergraduate training may be related to limited choice of assessment procedures for WFDs since two-thirds of participant SLTs responded that word finding was not an area covered in training.On the other hand, the avail ability of resources in the South African context could impact on the choice of assessment tools used.SLTs may be re stricted in purchasing standardised proce dures specifically for WFDs like the Test of Word-Finding (TWF) (German, 1989b) due to availability, high costs or other practical reasons.Standardised assess ment tools are generally expensive.The RWFS is relatively cost effective.It is also quick to administer (Renfrew, 1995), making it functional for SLTs in the re medial school environment with large caseloads (Stockman, 2000).
In assessment, the literature stresses the importance of identifying the type and possible cause of the WFD.This influences the intervention selected.While some participant SLTs described WFDs according to semantic deficits, retrieval deficits and general characteris tics of WFDs, the descriptions did not seem to match their assessment practices.These focussed mainly on assessing ex pressive vocabulary using the RWFS.The purpose of the RWFS is to identify if a child is able to name a word (Renfrew, 1995), not to assess word finding specifi cally.The use of the RWFS could there fore impact on service delivery, since it provides limited information about WFDs.In comparison German's Test of Word Finding (1989b), assesses accuracy of word finding, word comprehension, speed of naming, use of gestures and extra verbalizations.Augmenting assessment with qualitative observation of how the learner responds (as recommended by Renfrew, 1995), provides additional, valuable but partial information about WFDs.Furthermore, Renfrew (1995), when commenting on studies informing revisions to the RWFS test, notes that its use with a South African population is problematic (Pahl & Kara, 1992).Using tests not standardised on the population with which you work impacts on the va lidity of findings (Apel, 1999).SLTs in South African remedial schools need to be aware of this limitation regarding the in terpretation of formal test findings when assessing WFDs.
Using vocabulary tests, such as the RWFS, with multicultural populations often provides a negatively biased view of linguistic competence (Stockman, 2000).Using standardised tests not normed for our population highlights ethnocentricity.With the shift in South African educa tional policies, criterion-referenced and educator-produced tests are recommended by the Department of Education (Department of Education Directorate: Inclusive Education, 2002).Consistent with the Department of Education's rec ommendation for criterion-referenced tests, Terrell and Terrell (1993) suggest the use of criterion-referenced assess ments with individuals from diverse cul tural backgrounds.SLTs assessing WFDs should consider devising criterion-refer enced and educator-produced tests for use in the context of South African remedial schooling.
The most frequently used informal assessment procedures for identifying WFDs in learners with LLD were lan guage sample analysis (LSA) or discourse analysis.The use of LSA as an informal assessment procedure to determine how WFDs create difficulties for the learner with a LLD is recommended by Wiig and Semel (1984).Using LSA with the learner with LLD is particularly relevant when formal tests are not relevant to the popula tion with whom we work (Pahl & Kara, 1992).However, it is important to bear in mind whether the LSA is in the learner's first language.Apel (1999) suggests that SLTs use context-based assessment methods such as assessing communication skills in the classroom.Such an approach would be consistent with a more consulta tive or collaborative role (Schoeman, 2003).

Current Intervention Practice, espe cially Procedures Used for WFDs in Learners with LLD
The findings of the study revealed that most participant SLTs in the sample did not use formal programmes for inter vention.This is consistent with the lim ited availability of specific programmes The South African Journal o f Communication Disorders, Vol. 53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.) for WFDs (German, 1992).Possible reasons for predominant use of informal programmes are similar to those discussed under as sessment for WFDs namely: knowledge of available pro grammes, limited training, and limited resources.It is possible that even if resources were not limited, intervention programmes developed in other countries using unfamiliar stimulus material would be inappropriate to South African learners with WFDs.Using informal programmes may be an advantage in that SLTs can plan and develop appropriate interventions for the South Af rican context.
Most participant SLTs agreed that the formal intervention procedures they used for WFDs in learners with LLD were use ful.Although participant SLTs found the programmes useful, the programmes were not specifically developed for targeting WFDs.Therefore, a very specific aspect of WFDs, which may not be the learner's area of difficulty in word finding, may be addressed, raising questions about the effectiveness of intervention.
The most frequently used informal therapy strategy for WFDs was phonetic, phonemic or phonological cueing.These findings are supported by literature recommending these strate gies for intervention with WFDs (Wiig & Semel, 1984).How ever, SLTs need to be aware that response to a specific type of intervention informs us of the nature of the learner's problem.For example, if the response to phonological based therapy is good, the difficulty is phonologically based (Wolf & Bowers, 2000).Adopting such an approach to treatment implies that diag nostic therapy is relevant for WFDs.Intervention strategies would need to change if the child does not respond to selected strategies.SLTs in remedial education in South Africa should also incorporate into intervention for WFDs a number of other skills.These include: phonological segmentation and imagery (Wing, 1990), self-cueing, semantic intervention and phonologi cal awareness intervention (Wittmann, 1996), combined semantic and phonological treatment approaches (Easton et al., 1997) and discourse based intervention (Stiegler & Hoffman, 2001).How ever, since there is little evidence to support the efficacy of treat ment approaches for learners with WFDs (Casby, 1992), SLTs need to monitor the effects o f their treatment (Wittmann, 1996).This will assist in determining the most appropriate treatment for the learner with LLD and WFDs.
The predominant languages in which intervention for WFDs was provided were English and Afrikaans.A concern arises for learners who speak other official languages of South Africa and for whom the language of instruction is English or Afrikaans.When working with learners who use more than one language, a differential diagnosis is necessary.The aim is to de termine if the difficulties are related to a first language learning problem, a problem with acquiring a second language or if the learner is in a normal process of learning an additional language (Miller, 1984).SLTs should question which the optimal lan guage to provide therapy for WFDs in, is (Miller, 1984).
Consistent with the findings of Haynes (1993), participant SLTs in South African remedial schools seem to be providing strategies for WFDs but it did not appear that consideration was given to the general aims of intervention outlined by German (1992).Since therapy for WFDs is complex and there is no single intervention approach that addresses all WFDs, the general aims of intervention as outlined by German (1992) can be used to guide such intervention.These aims are to enhance naming accu racy, improve retrieval strength and develop meta-cognitive and meta-linguistic strategies to improve WFDs and to increase accu racy and fluency in spontaneous language.Consistent with the findings o f Haynes (1993), SLTs in South African remedial schools need to extend their understanding of language process ing so that therapy for WFDs can be addressed in a broader man ner.The issues raised by participants about institutions' demands and experience for example, become secondary to providing the most effective and accountable service (Baum, 1998), for learn ers in remedial schools.

Caseloads of SLTs for Learners with LLD with WFDs
It is difficult to account for the low caseload of learners with LLD and WFDs reported by the participants.Compared to the prevalence figure of 49 % of all grade four and five learners with LD having WFDs (German, 1998cited by German, 2005), this low caseload of less than ten learners with LLD and WFDs is of concern.As WFDs are the domain of the SLT, this may imply that the WFDs of some learners are not being identified.If the WFDs of learners in remedial schools which focus on reducing barriers to learning (Schoeman, 2003), and aim to return learners to mainstream schools, are not being identified and treated, these learners are possibly not receiving the help that they need.
Other possible reasons for this low caseload of learners with WFDs receiving intervention from SLTs may be low priori tisation given to WFDs in comparison to other difficulties experi enced by learners in remedial schools.Furthermore, limited availability of procedures appropriate to the South African con text may be another reason for small caseloads of learners with WFDs for SLTs in remedial schools.

Limitations of the Study
A number of limitations in both the implementation and results of this study are noted.Since the sample was limited to SLTs in South African remedial schools, the results obtained provide information specific to this population.This means that the findings cannot be generalised to other populations of SLTs.
The focus group discussion was only completed in KZN for logistical reasons.If these focus group discussions had been extended to other regions, the results may have provided a valu able comparison o f findings.Some responses on the focus group discussion differed from the findings of the questionnaire, possi bly because participants were given more opportunity to explain or expand on certain issues.
The questionnaire used a predominantly close-ended ques tion format with SLTs ticking the appropriate response.Such a format was selected to facilitate quick completion of; the ques tionnaire (Bailey, 1991).However, in selecting such a', question naire format, more detailed responses were limited to i questions posed in the focus group discussion.This may have had an im pact on the quality of the information obtained.
It is recommended that biased or leading questions be avoided (Leedy, 1997).Although this was a criterion used for developing the questionnaire for this study, examples of tests were provided for specific questions.Generally the responses to those specific questions used the examples provided.Richer in formation may have been obtained with open-ended questions.

CONCLUSION
From this study, the following conclusions can be drawn.In terms of assessment of WFDs in South African remedial schools, there were concerns, strengths and opportunities for change.It seems that the vast majority of participant SLTs in South African remedial schools who typically assess WFDs use Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol. 53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)standardised tests.This is despite the tests b ein g biased against our population and limited in their application for comprehensive assessm ent of WFDs in learners with LLDs.Reasons for this concerning assessment practice may include limited training in WFDs and limited resources.It is however encouraging that some participant SLTs in remedial schools were incorporating informal assessment measures into their standard test battery.Us ing informal strategies reduces test bias but the manner of obtain ing the language sample and the mother tongue of the learner are crucial considerations.Using LSA, for example, is a useful strat egy to observe WFDs in discourse, but LSA used with another tool such as the RWFS is not enough.Furthermore, SLTs need to include teachers in the assessment of WFDs.Teachers can pro vide information about the impact of WFDs on classroom per formance.As SLTs follow policy changes in education and shift from standardised testing to criterion-referenced and educatorproduced tests (Department of Education Directorate: Inclusive Education, 2002), assessment practices for WFDs will become more applicable to our context.
In terms of intervention for WFDs by SLTs in South Afri can remedial schools, there were also concerns, strengths and op portunities for change.O f concern was the finding that, consis tent with Haynes (1993), some participant SLTs were only pro viding strategies for WFDs.While this is an important interven tion principle, it is not sufficient (German, 1989a, German, 1992, Wiig & Semel, 1984, Wolf & Bowers, 2000).It seems that cur rent service delivery models typify a "pull out" model with lim ited collaboration with teachers.This is problematic in that it is not in line with changes in educational policy.Other concerns include small caseloads for WFDs in learners with LLD and the language of instruction differing from the mother tongue of the learner receiving intervention.The predominant use of informal programmes in intervention for WFDs may be an advantage in that SLTs can plan and develop appropriate interventions for the South African context.Again, in the context of changing educa tional policy, an opportunity exists for reviewing and extending current intervention practices for WFDs.
With the changing education system for learners in reme dial schooling, SLTs face many challenges.One of these is to re visit the management of WFDs in learners with LLD.SLTs in remedial education should embrace the challenges by reviewing and improving existing management practices.They should be encouraged to build on existing J strengths to serve the learners with LLD whom they serve.Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)

APPENDIX B Focus Group Discussion Interview Schedule
Thank you for participating today.My role is as facilitator to discuss some of the responses to the questionnaires on practice for WFDs.The discussion is informal.Based on the responses obtained, I would like to explore a few issues.It is important for me to briefly discuss some of the reasons why I selected WFDs as a topic to research: • It has always been a significant part of my caseload within a remedial school environment • I have not always been sure which assessments, interventions, strategies or therapy outcome measures to use when dealing with WFDs and have sometimes experienced frustration at the benefit of intervention • With the proposed changes for service delivery in education, I have queries about how practice for WFDs will be af fected 100 questionnaires were sent to remedial schools in South Africa.73 were returned and 69 were used for analysis.(Four were not analysed as the therapists did not work with WFDs or had no experience in the area).
1.I would like to start by asking how you define WFDs? (Depending on responses, use this as adjusted probe ques tion: Do you for example consider it to be a processing deficit or a storage deficit? 2.
In terms of caseload, it was found that all of you provide therapy for WFDs in English.Some of the children with WFDs in your caseloads speak English as a second language.What impact does this have on your assess ment and intervention.What patterns do you encounter, if any?

3.
In terms of formal assessments, the RWFS was by far the most frequently used tool for identifying WFDs.How do you feel about the use of the RWFS?What do you feel about its "fit" to your definition of WFDs?How do you feel about its appropriacy and effectiveness in the assessment of WFDs? 6 .Within a school, the various SLTs differed as to whether they used standardised versus informal assessment pro cedures and in' their assessment and treatment approaches.Can you provide possible reasons for this?

7.
Participants frequently responded that their undergraduate training in WFDs was inadequate but generally thera pists said WF therapy is sometimes or usually effective.Why do you think this is? 8.
Every participant said "Yes" to WF being a topic of interest for CPD.Why is this?What other solutions for in creasing knowledge in this area do you have?Who would you like to see doing a WFD workshop and how do you think it could be done to be of most value to you? i 9.
Many therapists seemed to have a small caseload for WFDs and some gave it low priority in terms of interven tion.How significant do you think the impact of WFDs is on academic performance?10.How do you see WF therapy fitting into the proposed changes in education such as an increased role for the therapist as consultant and less role in providing intervention to children?

11.
Are there any other issues you would like to raise?
The South African Journal o f Communication Disorders, Vol. 53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.) for participant selectionThe South African Journal o f Communication Disorders,Vol.53, 2006 f rem edial schools in SA was obtained from SAUVSE.• Perm ission from regional S uperintendent G enerals to perform the research was obtained.P H A S E 2 • Rem edial schools w ere contacted to determ ine if SLTs w ere em ployed there.• Perm ission from school principals to perform the research w as obtained.P H A S E 3 • P ilot study • A lterations to the questionnaire w ere made.P H A S E 4 • Letters w ere distributed to potential participants inform ing them o f the research aim.• Letters o f consent to participate w ere sen t to participants.• The revised questionnaire was adm inistered to SLTs in rem edial schools.• Participants w ere rem inded that questionnaires w ere due for subm ission.P H A S E S • Q uestionnaires w ere scanned to determ ine i f they m et the criteria for data analysis.• Statistical data analysis was com pleted.P H A S E 6 • R esponses on questionnaires w ere used to identify trends to develop the interview schedule.• A focus group discussion was held in a selected region.P H A S E 7 • Research and research findings w ere w ritten up.• Feedback to education authorities and participants regarding the research findings was com pleted.

Figure 2 :
Figure 2: Summary of Research Phases by the language medium o f the school.The majority o f SLTs at dual medium schools provide therapy in English and Afrikaans.SLTs at single medium schools provide therapy in the language o f the school Procedure Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol.53, 2006

Figure 5 :Figure 6 :Figure 7 :Figure 8 :Current
Figure 5: Informal measures used for assessment of WFDs (Key: L SA =language sam ple analysis.DA=discourse analysis) Gateway under licence granted by the Publisher(dated 2012.)

Figure 9 :
Figure 9: Number of learners with learning disabilities as well as with WFDs seen per week informal measures you use to identify word finding difficulties in language learning disabled children.Please list in order of frequency of use, e.g. 1. Language sample analysis.formal programmes/ and or activities do you use in your work with word finding difficulties in language learning disabled children.Please list in order of frequency of use e.g. 4. Lingui Systems Auditory Association Activity Booklet.Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol.53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)SECTION C -TICK the appropriate box.19.Do you feel that your therapy for word finding difficulties in language learning disabled children is effective?is your opinion on the management procedures used for word finding difficulties in children with language learning disabilities?relevant to the South African Context Intervention procedures are useful in addressing word finding difficulties Therapy outcome measures are useful in identifying the benefit o f therapy The South African Journal o f Communication Disorders, Vol.53, 2006 Reproduced by Sabinet Gateway under licence granted by the Publisher (dated 2012.)Ingrid de Rauville, Sandhya Chetty and Jenny Pahl 22. Did your training prepare you to adequately manage language learning disabled children with word finding difficulties?word finding difficulties in language learning disabled children adequately prepared me for practice in this area 23.Is word finding an area that you would like to be addressed for Continuing Professional Development?YES NO Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol.53, 2006 in terms of practice specifically regarding individual versus group therapy.Based on your experience in schools, what information can you provide about this pattern? 5. Do you find intervention needs change from year to year?Can you explain this?