Outcomes of Different Speech and Language Goal Attack ... · PDF fileTyler et al.: Outcomes of Different Interventions 1079 that was similar to that achieved by the phonological intervention

Tyler et al.: Outcomes of Different Interventions 1077Journal of Speech, Language, and Hearing Research • Vol. 46 • 1077–1094 • October 2003 • ©American Speech-Language-Hearing Association1092-4388/03/4605-1077

Ann A. TylerKerry E. LewisAllison Haskill

Leslie C. TolbertUniversity of Nevada,

Reno

Outcomes of Different Speechand Language Goal AttackStrategies

The purpose of this study was to assess phonological and morphosyntacticchange in children with co-occurring speech and language impairments usingdifferent goal attack strategies. Participants included 47 preschoolers, ages 3;0(years;months) to 5;11, with impairments in both speech and language: 40children in the experimental group and 7 in a no-treatment control group.Children in the experimental group were assigned at random to each of 4different goal attack strategies: (a) in the phonology first condition, childrenreceived a 12-week block of phonological intervention followed by 12 weeks ofwork on morphosyntax; (b) the morphosyntax first condition was the same asphonology first, with the order of interventions reversed; (c) the alternatingcondition involved intervention on phonology and morphosyntax goals thatalternated domains weekly; and (d) the simultaneous condition addressedphonological and morphosyntactic goals each session. Data were collectedpretreatment, after the first intervention block, and posttreatment (after 24 weeks).For the control group, data were collected at the beginning and end of a periodequivalent to 1 intervention block. Change in a finite morpheme composite andtarget generalization phoneme composite was assessed. Results showed thatmorphosyntactic change was greatest for children receiving the alternatingstrategy after 24 weeks of intervention. No single goal attack strategy wassuperior in facilitating gains in phonological performance. These results providepreliminary evidence that alternating phonological and morphosyntactic goalsmay be preferable when children have co-occurring deficits in these domains;further research regarding cross-domain intervention outcomes is necessary.

KEY WORDS: intervention, phonology, morphosyntax, goal attack,generalization

One of the largest subgroups of children with language disor-ders has both phonological and morphosyntactic impairments(Conti-Ramsden & Botting, 1999; Rapin & Allen, 1983, 1988;

Snow, 1998; Wolfus, Moscovitch, & Kinsbourne, 1980). Phonological andmorphosyntactic impairments co-occur in 35% to 77% of clinic samplesof children with speech or language disorders (Bishop & Edmundson,1987; Paul & Shriberg, 1982; Shriberg & Austin, 1998; Shriberg &Kwiatkowski, 1982, 1994; Shriberg, Kwiatkowski, Best, Hengst, &Terselic-Weber, 1986; Shriberg, Tomblin, & McSweeny, 1999; Tallal, Ross,& Curtiss, 1989). Despite this rate of comorbidity, relatively little isknown about how much emphasis the domains of phonology andmorphosyntax should receive during intervention (Camarata, 1998). Notonly must clinical decisions be made about goals within each domain,

1078 Journal of Speech, Language, and Hearing Research • Vol. 46 • 1077–1094 • October 2003

but a strategy for sequencing goals across domains mustbe selected.

A goal attack strategy refers to the way in whichmultiple goals are approached or scheduled. Fey (1986)described three different goal attack strategies: (a) a ver-tical strategy in which one goal at a time is focused onuntil some predetermined level of accuracy is achieved,(b) a horizontal strategy in which several goals are re-peatedly targeted within every session, and (c) a cycli-cal strategy in which several goals are targeted, eachfor a specified time period independent of accuracy, andthe sequence is repeated. Weiss (2002) suggested thatspeech-language pathologists (SLPs) must consider anumber of variables, such as one’s theory of languagelearning, how specific goals may interact with the strat-egy, and individual characteristics of the child in select-ing a goal attack strategy.

Goal attack strategies have received limited atten-tion in speech and language intervention research. Inan investigation of goal attack strategies in phonologi-cal intervention, Tyler, Edwards, and Saxman (1987)found vertical and cyclical strategies to be essentiallyequivalent in effecting phonological change (see alsoGierut, 1998). In the only investigation of a cross-do-main goal attack strategy, Tyler and Sandoval (1994)found that the 2 children who received a combination ofphonological and language interventions showed markedimprovements in both phonology and language. In gen-eral, though, the relative efficacy of different goal at-tack strategies involving multiple domains has not beenevaluated.

There is, however, a body of research focusing ongeneralization across domains, when only phonologicalor morphosyntactic goals were targeted and effects onthe nontarget domain were measured (Bopp, 1995; Duder,Camarata, Camarata, Koegel, & Koegel, 1998; Fey etal., 1994; Fey & Stalker, 1986; Hoffman, Norris, &Monjure, 1990; Matheny & Panagos, 1978; Tyler, Lewis,Haskill, & Tolbert, 2002; Tyler & Sandoval, 1994; Tyler& Watterson, 1991; Wilcox & Morris, 1995). First, withregard to the effects of phonological intervention onchange in morphosyntax, there is evidence suggestingthat phonological intervention facilitated improvementin grammatical markers subject to surface-level inter-actions with phonological forms (Bopp, 1995; Fey &Stalker, 1986; Tyler & Sandoval, 1994). Tyler andSandoval (1994), for example, treated final consonantdeletion/cluster reduction in 2 children. Both childrenshowed an increase in production of plural, regular pasttense, possessive, and/or third person singular mor-phemes. In contrast, the 2 children in this study whoreceived an indirect narrative intervention made no suchimprovements in phonetically complex morphophonemicforms; although mean length of utterance (MLU) did

increase. Fey and Stalker (1986) reported on a child whodisplayed an idiosyncratic phonological error patternaffecting morphophonemic forms and improved gram-matical morpheme productions as a result of phono-logical intervention. Duder et al. (1998) studied 12 chil-dren who were assigned randomly to two types ofphonological intervention or a control group. Despitetreatment-related gains in phonology, treatment groupsshowed no statistically significant gains in MLU or per-cent complex sentences as compared to the control group.Differences in findings from Duder et al. and smallersample studies (Bopp, 1995; Fey & Stalker, 1986; Tyler& Sandoval, 1994) may be due in part to differences inintervention targets and the specificity of measures used:Probes of morphophonemic structures were used in thesmaller studies, whereas Duder and colleagues used theglobal measure of MLU.

Similarly, results from studies examining the effectsof a variety of language interventions on phonologicalperformance are conflicting. Fey et al. (1994) examinedthe effects of morphosyntactic intervention on changein phonology in a group study of 25 children with mod-erate to severe morphosyntax and phonological impair-ments. Children were randomly assigned to a cliniciantreatment group, a parent treatment group, or a delayedtreatment (control) group. Goals focused on morphemes,such as copula and auxiliary be, as well as pronounsand conjunctions and grammatical operations for ques-tions and negatives. The two treatment groups madelarge gains in grammar after 5 months of a focusedstimulation intervention. In comparison to the controlgroup, the treatment groups made no significant gainsin phonology, as measured by percentage of consonantscorrect (PCC; Shriberg & Kwiatkowski, 1982).

In another group of studies, various language in-terventions appeared to facilitate change in phonology(Hoffman et al., 1990; Matheny & Panagos, 1978; Tyleret al., 2002; Wilcox & Morris, 1995). Matheny andPanagos (1978), using the highly structured Montereyprograms to teach different sentence forms (Gray &Ryan, 1973), observed gains in articulation, althoughthey were not as great as were those in syntax. Wilcoxand Morris (1995) used standardized tests to examinethe effects of a language-focused curriculum for childrenwith speech and language impairments and foundgrowth in phonology was greater than that of normallydeveloping peers. Recently, Tyler et al. (2002) investi-gated the efficacy and cross-domain effects of both amorphosyntax and a phonological intervention for 27preschoolers. Tyler et al. found both interventions wereeffective in facilitating change in the targeted domainafter 12 weeks, in comparison to a control group that didnot receive intervention. In addition, the morphosyntaxintervention led to cross-domain change in phonology

Tyler et al.: Outcomes of Different Interventions 1079

that was similar to that achieved by the phonologicalintervention. Thus, for children who received morpho-syntactic intervention, the amount of phonology changewas significantly greater than that observed for the con-trol group. For children who received phonology inter-vention, the amount of phonology change was signifi-cantly greater than that observed for the control group,but morphosyntactic change was no different.

Interpretation of findings regarding the effects oflanguage interventions on phonology is complicated bydifferent interventions, designs, and dependent mea-sures. Although there were similarities between the ex-perimental control and larger participant numbers inthe two most rigorous studies (Fey et al., 1994; Tyler etal., 2002), there were other notable differences. For ex-ample, Fey et al. (1994) used a focused stimulation in-tervention that involved no direct elicitation of targetproductions from the children, whereas Tyler et al. (2002)had an elicited production component in addition to fo-cused stimulation. Further, intervention targets in theTyler and colleagues study were finite morphemes,whereas Fey and colleagues targeted additional syntac-tic and grammatical operations. Finally, measures ofchange used for phonology were different in the two stud-ies: Fey et al. used the global measure of PCC and Tyleret al. used a measure of target and generalization pho-neme accuracy. All of these variables may help to explainthe difference between the findings of the two studies.

To summarize, results from research on the cross-domain generalization effects of morphosyntactic orphonological intervention for children with impairmentsin both domains have unclear clinical implications. Fur-ther, there have been no investigations of combined pho-nological and morphosyntactic intervention using dif-ferent goal attack strategies. It is not known what typeof goal attack strategy would result in greater gains inboth phonology and morphosyntax. As a first step, itseems important to compare the three types of goal at-tack strategies applied to goals from multiple domains.For example, treating speech and morphosyntactic do-mains vertically in blocks, simultaneously within activi-ties, or alternately in a cyclical fashion are just some ofthe many options available. A vertical goal attack strat-egy in which a block of intervention focusing on mor-phosyntax is followed by a block focusing on phonologyshould result in early gains in morphosyntax. Similarly,a vertical strategy with phonology first should fosterearly gains in phonology. An early focus on morpho-syntax, however, may lead to greater overall gains inthat domain because development of language is moreprotracted than development of phonology and languagemay benefit from a sort of incubation period (Tyler etal., 2002). A simultaneous strategy that is essentiallyhorizontal because it involves focus on both speech and

morphosyntax goals within every session would seemadvantageous due to its continual focus on interactingdomains. Such an approach might, however, overwhelmand confuse the child because the focus is mixed andunclear. In contrast, a goal attack strategy where pho-nological and morphosyntactic goals are alternatedweekly may lead to significant gains in both domainsbecause the focus on a particular domain is obvious forthe week in which it is targeted and similar to the focusachieved in a block intervention. The cyclical nature ofan alternating strategy also capitalizes on the child’srole in the gradual acquisition process. The purpose ofthis investigation was to assess phonological andmorphosyntactic change in children with co-occurringspeech and language impairments using different goalattack strategies. The following research questions wereposed:

1. In comparison to a no-treatment control, do the fourgoal attack strategies result in significant changein phonological and morphosyntactic performanceafter 12 weeks of intervention?

2. Which goal attack strategy produces greater changein phonology and morphosyntax after 24 weeks ofintervention?

a. A vertical (block) focus on morphosyntax followedby phonology.

b. A vertical (block) focus on phonology followed bymorphosyntax.

c. A weekly alternating focus on both phonology andmorphosyntax.

d. A simultaneous focus on both phonology andmorphosyntax.

MethodParticipants

Participants included 47 preschoolers, ages 3;0(years;months) to 5;11, with impairments in both speechand language development: 40 children in the experi-mental group and 7 in a control group. All children hadreceived speech-language evaluations and were identi-fied as potential participants through review of theirevaluation results in consultation with the evaluatingSLP. Children in the experimental group were enrolledin speech-language services in early childhood programsin Washoe County School District, Reno, NV. For thesechildren, speech-language services consisted of partici-pation in one of the four experimental interventions. Thecontrol group consisted of children who had been placedon waiting lists for speech-language services. Parentsof the children initially placed on waiting lists were con-tacted 12 weeks after the initial evaluation. If a child


had not been enrolled in services since the initial evalu-ation, the parents were asked to allow the child to be re-evaluated and to allow the initial and re-evaluation datato be used for the present investigation. Consequently,children were not assigned at random to the controlgroup.

Selection criteria for children in both experimentaland control groups included (a) documentation of expres-sive language scores at least 1 SD below the mean onthe Preschool Language Scale–3 (PLS-3; Zimmerman,Steiner, & Pond, 1992) or the Clinical Evaluation of Lan-guage Fundamentals–Preschool (CELF-P; Wiig, Secord,& Semel, 1992), or MLU in morphemes greater than 1.5SDs below the mean based on Leadholm and Miller’s(1993) normative data; (b) documentation of speech per-formance at least 1 SD below the mean on the Bankson-Bernthal Test of Phonology (BBTOP; Bankson &Bernthal, 1990); (c) documentation of nonverbal cogni-tive functioning within 1.5 SDs of the mean on the Co-lumbia Mental Maturity Scale (CMMS; Burgemeister,Blum, & Lorge, 1972); (d) normal hearing, as indicatedby pure-tone screening; (e) normal functioning on oralmotor assessment (Robbins & Klee, 1987); and (f) neu-rological, behavioral, and motor skills reported withinnormal limits in assessment results.

Receptive performance was not included in the se-lection criteria, and some children had receptive scoreswithin the normal range. Children also could qualifyas participants if they had standard scores within thenormal range on the PLS-3 but considerable morpho-syntactic deficits, as evidenced by an MLU that wasgreater than 1.5 SDs below the mean. The PLS-3 doesnot have a strong emphasis on morphosyntax, and assuch may not accurately identify all children with mor-phosyntactically based language impairments. Therewere 7 participants whose expressive PLS-3 scores fellwithin the normal range but whose morphosyntacticdeficits were documented by an MLU greater than 1.5SDs below the age mean. Further, these participantswere referred for the study by their school SLP becausethey had been identified to receive or were receivingintervention focused on morphosyntax. All 47 partici-pants met the inclusion criteria, as described, and groupmeans for each inclusion variable are shown in Table1. As a final check on group similarity, separate one-way analyses of variance (ANOVAs) by group proce-dures were applied for each inclusion variable. Resultsindicated that there were no significant group differ-ences for any of the inclusion variables, Fs(4, 42) =0.344–3.75, p = .07–.846.

Children received one of four interventions at oneof six sites. Each intervention was carried out at mul-tiple sites, and interventions were assigned to sites atrandom. Each intervention represented one of four

different goal attack strategies that focused on both pho-nology and morphosyntactic goals: (a) phonology firstfocused on only phonology for the first 12-week block,followed by a 12-week block of morphosyntactic inter-vention; (b) morphosyntax first focused on only morpho-syntax goals for the first 12-week block, followed by a12-week block of phonology intervention; (c) alternat-ing intervention involved a focus on phonology or mor-phosyntactic goals that alternated weekly; and (d) si-multaneous intervention had an integrated focus onboth phonology and morphosyntactic goals in every ses-sion. The random assignment of samples to interven-tions resulted in 10 children assigned to each of the blockstrategies, 11 to the alternating strategy, and 9 to thesimultaneous strategy.

Strategy OverviewThe morphosyntax and phonological interventions

were designed specifically for this study. These inter-ventions were scheduled in blocks for the morphosyntaxfirst and phonology first strategies and alternatedweekly in the alternating strategy. The simultaneousintervention designed for this study involved the samecomponents as the separate morphosyntax and phonol-ogy interventions but were meshed within activities.Morphosyntactic goals addressed primarily finite mor-phemes, and phonological goals addressed both segmen-tal and syllable structure forms. Regardless of interven-tion type, four goals for phonology and four goals formorphosyntax were selected for each child and sched-uled in cycles. Thus, progression from one goal to thenext was not criterion-based, but time-based. In themorphosyntax first and phonology first interventionstrategies, one goal was targeted during each week in a4-week cycle and then the sequence (cycle) was repeatedtwice. Thus, each child received three cycles of inter-vention (12 weeks) focusing on his or her speech ormorphosyntactic goals, followed by 12 weeks of focus onthe other domain—24 weeks in total (see Figure 1). Inthe alternating strategy, four goals in each domain wereselected for each child and alternated over the course of8 weeks. For example, in the first week, MorphosyntaxGoal 1 was targeted and in the second week, PhonologyGoal 1, and so on, as shown in Figure 1. This 8-weekcycle was performed a total of three times to equal 24weeks. In the simultaneous strategy, there were also fourgoals for each domain, with one morphosyntactic andone phonology goal combined within activities for eachsession during each week in a 4-week cycle. This cyclewas then performed six times for a total of 24 weeks(see Figure 1).

Children were enrolled in programs housed in fourelementary schools and the university clinic, working withfour certified SLPs and four graduate student interns.


All children received two intervention sessions per week,one 30-min individual and one 45-min group session withthe same type of intervention applied in both sessions.Groups consisted of no more than 3 children. Both ses-sions were provided by graduate student interns un-der supervision of the early childhood or universityprograms’ SLPs. Interns attended a training sessionin which they viewed videotapes of intervention proce-dures and were provided with a comprehensive manualexplaining the procedures and containing instructionsfor their implementation. To further ensure reliableimplementation of the intervention strategies acrosssites, data collection forms and duplicate sets of materi-als were provided with specific instructions regardingtheir use (i.e., scripts for clinician input, number of mod-els/elicitations). Also, each site was visited each semes-ter by the first author to ensure correct implementa-tion of procedures. It should be noted that graduate

student interns changed on a semester basis, so overthe 24-week intervention participants worked with twodifferent graduate interns. Further, there were typicallyfour to six interns applying each strategy because chil-dren receiving the same strategy attended differentschools. The multiple numbers of trained clinicians work-ing with children assigned to the same strategy was con-sidered an inherent protection to the threat of clinicianbias in intervention. Finally, although children from allstrategies had absences, there were no marked differ-ences across the strategies in the number of group orindividual sessions attended. The morphosyntax firstgroup attended an average of 43 (range = 37–47) of thescheduled 48 sessions, the phonology first group at-tended an average of 40 (range = 35–46), the alternat-ing group attended an average of 42 (range = 40–48),and the simultaneous group attended an average of 42(range = 37–46).

Table 1. Means and standard deviations for selection criteria variables for each intervention strategygroup.

Age Receptive Expressive W. C. P.Strategy (months) CMMS Language Language MLU Inventory Inventory Inventory

Language firstM 51.9 97.9 88.1 77.6 2.61 69.1 68.5 70.5SD 6.0 9.8 10.8 11.0 0.58 4.65 3.84 5.66Min 42 83 73 58 2.00 65 65 65Max 58 113 103 95 4.00 76 76 79

Phonology firstM 48.9 105.9 86.8 76.5 2.58 70.0 68.5 71.6SD 5.4 9.3 13.5 14.2 0.59 6.86 6.04 9.42Min 43 91 68 50 2.00 65 65 65Max 58 117 108 97 3.00 82 83 90

AlternatingM 48.6 104.4 84.8 80.4 2.85 71.0 69.0 70.0SD 5.0 7.8 16.1 9.3 0.51 5.81 4.49 5.42Min 42 91 61 64 2.00 65 65 65Max 60 119 115 97 4.00 83 76 80

SimultaneousM 55.8 103 84.4 73.4 2.77 67.6 67.2 68.2SD 9.5 10.2 20.6 15.24 0.96 5.77 5.63 5.78Min 43 89 50 50 1.00 65 65 65Max 68 124 127 93 4.00 82 82 82

ControlM 47.6 108.9 88.6 77.3 2.84 67.7 66.4 67.1SD 6.0 10.1 8.6 11.0 0.57 3.68 3.36 3.18Min 42 97 79 53 2.00 65 65 65Max 57 124 100 84 4.00 75 74 73

Note. CMMS = standard scores from the Columbia Mental Maturity Scale; Receptive and Expressive Language =standard scores from the Preschool Language Scale–3 or the Clinical Evaluation of Language Fundamentals–Preschool; MLU = mean length of utterance; W. Inventory = Word Inventory, C. Inventory = Consonant Inventory,P. Inventory = Phonological Process Inventory standard scores from the Bankson-Bernthal Test of Phonology.


Data Collection and AnalysisAll children participated in pre- and posttesting

procedures administered in their preschool or at theUniversity of Nevada Speech and Hearing Clinic at thebeginning, middle, and end of intervention periods. Datacollection sessions were held 2–4 weeks prior to the on-set of intervention (Sample 1), at midyear after 12 weeksof intervention (Sample 2), and 2 weeks after 24 weeks

of intervention (Sample 3). For the children in the con-trol group, data collection sessions were held at the be-ginning and end of a 12-week period.

Pre- and posttreatment measures were obtainedfrom analysis of a spontaneous language sample and asingle word citation sample obtained from the BBTOP,which was supplemented with 15 additional words (pri-marily nouns) to ensure that the 24 consonants occurred

Figure 1. Cycle configurations for each goal attack strategy.


a minimum of three times each in initial and final wordpositions. Samples were audio- and videotaped using aPanasonic SVHS camcorder and a Marantz PMD 230or 430 audio recorder with one or more external lapelmicrophones. All samples were collected by researchassistants in small, quiet rooms in the children’s schoolsor in the university clinic. Each task and analysis pro-cedure is described below.

LanguageSpontaneous language samples of at least 200 ut-

terances were obtained from conversations between aresearch assistant and child that centered around aPlaymobil™ house and accessories. Research assistantsalso prompted the children to produce narratives by look-ing at books in the Carl series by Alexandra Day. Natu-ralistically based examiner scripts were used whenneeded to create at least three obligatory contexts foreach of Brown’s (1973) 14 grammatical morphemes. Forexample, to obligate the irregular past tense, the clini-cian “broke” a window on the house and asked the child,“What happened?”

Child and examiner utterances were transcribed andchild utterances were coded using the guidelines fromthe Kansas Language Transcription Database manual(Howe, 1992) to enable subsequent analysis with theSystematic Analysis of Language Transcripts program(SALT; Miller & Chapman, 2000). SALT was used todetermine MLU in morphemes and Brown’s (1973) stagefor all participants. Additionally, SALT was used to findeach instance of correct, incorrect, and omitted gram-matical morphemes; percentage correct usage for eachmorpheme was derived by dividing the total number ofcorrect usages by the total number of obligatory con-texts. A finite morpheme composite (FMC; Bedore &Leonard, 1998; Rice, Wexler, & Cleave, 1995) was calcu-lated by determining percentage correct usage of thefollowing finite morphemes: regular past tense -ed, thirdperson singular regular -s, contractible and uncontrac-tible copula be verbs, and uncontractible and contract-ible auxiliary be verbs. Children with specific languageimpairment (SLI) typically perform poorly on this mea-sure; FMC was the dependent language measure usedin this study (Bedore & Leonard, 1998; Rice et al., 1995).

PhonologyThe BBTOP, a standardized test of speech sound

production, was used to document phonological impair-ment as well as to elicit a sample of single words in whicheach of the 24 consonants occurred at least three timesin the initial and final word positions, as permitted byEnglish phonotactics. Broad transcriptions were madeon-line during administration of the BBTOP by gradu-ate research assistants. These transcriptions were then

checked and modified from audiotape replay by a seniorresearch assistant and the first author. Transcriptionswere entered into the Interactive System for Phonologi-cal Analysis (ISPA; Masterson & Pagan, 1993), a com-puter analysis program that generates quantitative datasuch as percentages of phonological process occurrence,frequency of occurrence of phones in the phonetic in-ventory, and PCC. The mean PCC for the 47 partici-pants in this study was 58% (range = 27%–84%). Thedependent phonology measure used in this study, thetarget generalization composite (TGC), was a percent-age reflecting the accuracy of target and generalizationsounds selected for each child from the total number ofopportunities for these sounds, in the positions targeted,on the BBTOP. Each participant’s phonology goals andgeneralization targets are shown in the Appendix. Forexample, if a child had final /f/, initial /k/, final /S/, and /s/clusters as targets, the total number of these sounds andof the generalization targets final /v/, initial /g/, initial /f,v/, final /k, g/, and initial /S/ that were produced correctlywas divided by the total number of opportunities for thesounds on the BBTOP. On average, the number of targetand generalization sounds from which the composite wascalculated was 32, with a range from 17 to 53.

ReliabilityLanguage Samples

For orthographic transcription of samples, the thirdauthor, who was blind to group assignment of the par-ticipants, completed the initial and second passes throughthe language samples and acted as an expert coder. Onthe first pass, child utterances were transcribed, and onthe second pass, examiner utterances and grammaticalcodes for the child utterances were added and generalrevisions were made. Next, trained speech-languagepathology graduate student research assistants listenedto audio recordings of the samples and made correctionsto the transcripts. The investigator and assistants thendiscussed discrepancies and tried to reach a consensusas to the appropriate form to include in the transcript.The investigator who completed the initial transcrip-tion and coding then served as an expert coder and madethe final determination for cases in which a consensuscould not be achieved. Percentage of agreement for tran-scription reflects the number of transcription discrep-ancies between the investigator and research assistantsprior to consensus; it was calculated by dividing thenumber of discrepancies in transcription of child utter-ances by the total number of words in the sample andmultiplying by 100. Transcription agreement exceeded90% across 120 samples (40 participants at three sam-pling points) and ranged from 94% to 100%. Percentageof agreement for the identification of obligatory contextswas 98% and ranged from 97% to 100%.


Phonology SamplesBroad transcriptions were made online during ad-

ministration of the BBTOP by graduate research assis-tants who were blind to group assignment of the par-ticipants. These transcriptions were then checked andmodified from audiotape replay by two trained transcrib-ers, a senior research assistant and the first author. Todetermine interjudge reliability for the two trained tran-scribers, 20% of the BBTOP samples were retranscribedby the transcriber who had not performed the originaltranscription. Point-to-point reliability was calculatedbased on each judge’s transcription of each consonant.Segmental transcriptions that were identical (exclud-ing diacritics) were coded as agreements. The overallmean for speech transcription agreement was 90% across24 samples, with a range from 73% to 97%.

Goal Selection and InterventionStrategiesLanguage Goals

Language goals addressed morphosyntactic struc-tures and phonological goals addressed both segmentaland syllable structure forms (see Appendix for goals foreach participant). Goals for each child were selectedbased on analysis of the initial speech and languagesamples and consideration of the primary and sharedneeds of group members. Children grouped togethershared the same goals in both the individual and groupsessions. Goal selection was based on needs of the ma-jority of children in a group. The aim was to have nomore than one goal that was considered low priorityfor each individual child. Morphosyntactic targets wereselected for language goals because overall they weremore problematic for the children in this study thanwere other types of language targets (e.g., sentencestructure). Targeted morphemes were used with 0%–50% accuracy in the initial language sample and werefrom adjacent stages in Brown’s (1973) stage sequencefor grammatical morpheme acquisition. It should benoted that the need to have similar goals for all chil-dren in a group sometimes superceded the 0%–50% cri-terion. Thus, if copula is was a primary goal for 2 of thechildren in a group, then the third child also was as-signed that goal, regardless of that child’s pre-interven-tion accuracy level. Preference was given to finite mor-phemes (e.g., regular past tense -ed, third personsingular regular (3SR), copula and auxiliary be verbs).Three of the four finite morphemes represented by theFMC were intervention targets for 36 of the 40 children.The remaining 4 children had two finite morphemes asgoals. Although both the auxiliary and copula is weregoals for children in some of the strategies when others

had only copula, there are numerous reports of gener-alization from the copula to auxiliary is and vice versa(Hegde, 1980; Hegde, Noll, & Pecora, 1979; Hughes &Carpenter, 1985; Leonard, 1974). Because goals wereprimarily copula and auxiliary is, past tense -ed, and3SR, morphophonemic targets consisted of clusters suchas /ts, ks, bz, nz, dz, pt, st, ft, bd, nd, zd, mpt/.

Phonology GoalsPhonological goals included sounds from adjacent

categories in Shriberg’s (1993) developmental sequence(e.g., early 8, middle 8, late 8). For example, target com-binations such as /f, s, ‰/ and /l, S, k/ represented themiddle 8 (/k, f, ‰/) and late 8 (/s, S, l/) categories. Phonol-ogy goals were selected with consideration of the follow-ing: phonetic inventory and positional constraints, pro-cesses used > 35%–40%, sound classes affected, andword/syllable structure limitations. Each child had fourgoals, with one of those focused on cluster targets forcluster reduction/simplification error patterns. For mostparticipants there were three cluster exemplars (range =2–4) for the cluster goal, but the number of stimulusitems was equal to that of other goals. Phonological tar-gets for the majority of children consisted of initial /s/clusters, an initial affricate or liquid, velars, and—lessoften—a final fricative. Typically, one or two of the single-ton targets were missing from the child’s inventory ormissing from the inventory in the position it was taught.Occasionally, for an individual child, a fifth high-prioritytarget was included in the individual session only. Be-cause this study involved morphophonemic forms po-tentially impacted by the phonological processes offinal consonant deletion and final cluster reduction, thechildren’s BBTOP responses were analyzed for the ap-plication of these two processes. The children appliedthese processes in the 69–72 possible contexts with amean frequency of 12% (range = 0%–43%). It should benoted that only 6 children (1, 11, 13, 19, 25, and 28)deleted final consonants or reduced final clusters morethan 20% of the time. Thus, when targets /f, s, z, S, ‰/ werein the final position they were selected as vehicles toeliminate stopping, depalatalization, or backing/glottalreplacement, but not final consonant deletion. Althoughfinal /s/ was a target for 5 children, 2 had correct pro-ductions, but other group members had /s/ as a goal forstopping or glottal replacement. Six children had final/z/ as a target, but again, 2 had accurate productions.Other group members (n = 3) had the target for stop-ping or backing; only 1 child deleted final /z/. Generali-zation targets were identified based on the goals for eachchild. These targets were for cognates of trained soundsand for trained sounds in the untrained (initial or final)position if those sounds were not already produced cor-rectly by the child.


Intervention ProceduresMorphosyntax

Language intervention procedures involved auditoryawareness activities, focused stimulation activities, andelicited production activities (Camarata, Nelson, &Camarata, 1994; Cleave & Fey, 1997; Fey, Cleave, Long,& Hughes, 1993; Nelson, 1989). These language activi-ties were centered around the themes of animals, food,and water. Detailed written scripts were created for allactivities for each session to ensure reliable implemen-tation of the intervention across clinicians (Haskill,Tyler, & Tolbert, 2001). An outline of a morphosyntacticintervention script is provided in Table 2. Each differ-ent type of activity was implemented in every group andindividual session, and progression from one goal tothe next occurred each week, regardless of individualchildren’s performance. Auditory awareness activitieswere designed to heighten children’s awareness of themorphosyntactic targets in the context of children’s booksand songs that were read and sung in each session. Fo-cused stimulation activities were designed to providechildren with multiple models of target structures in a

naturalistic communicative context. They involved re-casts and expansions of children’s utterances and op-portunities to use target forms in response to contextu-ally relevant questions or prompts. When a child omitteda target morpheme or mispronounced it, the productionwas simply recast during focused stimulation activities.

Elicited production activities were implementedwith the goal of eliciting 20–30 productions of each tar-get morpheme. These activities were sequenced hierar-chically by level of support from Cycle 1 to Cycle 3. InCycle 1, the most clinician support was provided usingforced choice tasks, in Cycle 2 clinician support was nei-ther maximal nor minimal for cloze tasks, and in Cycle3 the least amount of clinician support was given usingpreparatory sets. Forced choice tasks obligated the pro-duction of a morpheme by providing the child with thechoice of two responses, both of which contained the tar-get (“The man jumps or runs?”). For cloze tasks, the cli-nician began an utterance and paused prior to the tar-get form to give the child an opportunity to produce it(“What does the man do? He ____.). Preparatory setsinvolved techniques whereby the clinician indirectly

Table 2. Example intervention scripts for phonology, language, and simultaneous interventions.

Type of Phonology: Language: Third person Simultaneous:activity Final /f/ singular regular (3SR) Initial /l/ & copula be

Auditory awareness Read The Three Little Pigs, Discusses Brown Bear story using Clinician-generated story embedded with thein which final /f/ words are 3SR forms language and phonology targetsemphasized (HUFF, PUFF, (“Brown Bear SEES a blue horse…”).WOLF) Song: “Brown Bear SEES all his friends”

Sound description, “Fan” sound concept Identify and describe the /l/soundconceptualization (picture of fan)and identification LONG sound (sort long and

short spaghetti)

Elicited production Feed final /f/ pictures to a Cloze-task level: Elicit production of initial /l/ (forced choice level)puppet while naming them Re-enact Brown Bear story. & focus stimulation on copula: Discuss materials(24–32 productions per Clinician: What does Brown Bear needed for lollipop animal activity.child) do here? He Clinician: That IS a lollipop. IS the lollipop LITTLE

<pause>__ or LARGE?Child: SEES a bird. Child: LITTLE.

Clinician: Those ARE scissors. ARE the scissorsLONG or LARGE?Child: LONG.

Focused stimulation/ Complete a dramatic play Art activity: Decorating animal pictures Focus stimulation on initial /l/ and elicit productionnaturalistic activity chef activity, focusing on the Clinician: John TAPES feet on the dog. of copula BE (forced choice level): Act out clinician-

words LOAF, PUFF, and Steve MAKES a blue horse. generated story using lollipop animal puppets.FLUFF Clinician: Who is the LLAMA? Mary IS or John IS?

Child: Mary IS.Clinician: Who is the LION? Steven IS or John IS?Child: Steven IS.

Phonological Odd-one-out identification: Odd-one-out identification: (LION, LAMB, LLAMA,awareness (HUFF, PUFF, star) sun)


demonstrated for the child how to use target forms andthen gave the child a turn to form his or her own similarproduction at the sentence level. If a child respondedincorrectly at any level of support during the elicitedproduction tasks, he or she was provided with a correctmodel and a request for imitation.

Although the intervention was largely naturalisticin nature, session scripts were written to provide a pri-mary focus on a specific target morpheme; therefore, con-versational contexts were designed to ensure a greaterlikelihood the children would hear multiple models of,and have opportunities to produce, the target morpheme.To determine the number of responses per session, 37.5%of the children’s data records were tabulated. An aver-age of 75–80 models were provided of each morphemein each script, and children produced an average of 23occurrences of the targets.

PhonologyPhonological intervention procedures involved a

combination of techniques that were both clinician-di-rected and child-centered. The intervention included fourcomponents: (a) auditory awareness activities designedto heighten children’s awareness of target sounds anddirect their attention to the sounds’ auditory–acousticattributes; (b) conceptual activities designed to developchildren’s awareness of the differences and similaritiesbetween target sounds and their contrasts; (c) produc-tion practice activities, both drill play and naturalistic,designed to help establish production of a new sound, tofacilitate practice of that sound in communicative con-texts, and to increase awareness of the success/failurein communicating an intended message; and (d) onephonological awareness activity designed to stimulatepreliteracy skills by increasing awareness of the speechsound system. Although each intervention session in-volved activities for each of the four components, aware-ness and conceptual activities were brief, lasting only 5–10 min of the session, while the majority of the session’stime focused on production practice. Auditory awarenessactivities involved listening to word lists and books inwhich the text contained the target sound used repeat-edly through rhyming, alliteration, or a repetitivephrase. Conceptual activities were adapted from thoseused in Metaphon (Howell & Dean, 1994) and requiredthe children to discuss and practice sounds in contrast-ing classes. The goal of production practice activities wasto elicit 24–32 target productions per session. Drill playactivities involved direct elicitation using phonetic place-ment, shaping, and cueing techniques. In these activi-ties, clinicians responded to incorrect responses by pro-viding a model and eliciting an imitation. In naturalisticactivities, responses were not explicitly required, butchildren were given the opportunity to produce targets,

and their productions were recast in the conversationalcontext. The phonological awareness activity involvedrhyme detection or initial sound identification, depend-ing on whether the target was in the word final or ini-tial position, respectively. Print was included on picturestimulus cards to support literacy skills. An outline of aphonology intervention script is provided in Table 2.

As a check on the fidelity of the phonological inter-vention with respect to elicitation of the required num-ber of target responses in production practice activities,40% of the participants’ individual data records withineach group were reviewed. Children in the morphosyntaxfirst strategy averaged 26 productions (range = 16–51),children in the phonology first strategy averaged 27productions (range = 17–50), children in the alternat-ing strategy averaged 33 productions (range = 18–60),and children in the simultaneous strategy averaged 25(range = 14–37) productions of their target sounds persession, indicating that they met the required 24–32target sound productions per session. It should be notedthat sometimes this number was not achieved in groupsessions due to turn taking and decreased opportuni-ties to respond.

AlternatingThe procedures described above for the morpho-

syntax and phonology interventions were also used inthe alternating strategy, but the goals from each domainwere alternated weekly.

SimultaneousSimultaneous intervention procedures involved au-

ditory awareness activities, focused stimulation activi-ties, elicited production activities, and phonologicalawareness activities, all similar to those described abovein the Intervention Procedures section. First, morpho-syntactic and phonology goals to be combined withinactivities were selected so there was as little overlap inword structure as possible. For example, if a bound fi-nite morpheme was the morphosyntactic target, then aphonology target involving word initial, and not final,position was selected. Some examples of target combi-nations are as follows: initial /g/ and 3SR, final /f/ andirregular past tense, and final /‰/ and copula be. Thesecombinations were never presented to the childrenwithin the same stimulus word. Rather, a separate setof stimulus words was developed for each target. Whenboth targets were the focus of an activity, stimuli forone of the two received focused stimulation, whereas theother target was the focus for production. A typical ses-sion included these activities: (a) introduction to the tar-get sound, (b) auditory awareness for both targets pro-vided through reading a book, (c) focused stimulationon the morphosyntactic target and elicited production


of the phonology target within an activity, (d) focusedstimulation on the phonology target and elicited pro-duction of the morphosyntactic target within an activ-ity, and (e) phonological awareness activity for the pho-nology target. An outline of a simultaneous interventionsession is provided in Table 2.

Statistical AnalysisThe first research question, which addressed the

effectiveness of the four different goal attack strategiesafter 12 weeks, was assessed using analysis of covari-ance (ANCOVA) procedures (SPSS, 1988). Morphosyn-tactic and phonology change in the four intervention strat-egies was compared to that found in the no-treatmentcontrol group. The dependent measures were FMCchange for morphosyntax and TGC change for phonol-ogy. Initial levels of these variables were statisticallyanalyzed for group differences. ANOVA procedures re-vealed there were no significant by-group differenceswith respect to initial levels of TGC, F(4, 42) = 0.240, p= .914, and FMC, F(4, 42) = 0.894, p = .476. ANCOVAprocedures were also used to determine which of thefour intervention strategies produced greater change inphonology and morphosyntax, as addressed by the sec-ond research question. The initial phonology and mor-phosyntax levels were used as covariates in the corre-sponding analyses. In addition, the magnitude of theeffect for statistically significant follow-up comparisonswas calculated using Cohen’s d statistic (Cohen, 1988).

ResultsThe first research question asked whether the four

goal attack strategies produced a significantly greaterchange in morphosyntax and/or phonology after 12 weeksof intervention than was observed for the no-treatmentcontrol group. This intervention versus control compari-son was made at 12 weeks.

Morphosyntactic Change inIntervention Versus Control Groupsat 12 weeks

Morphosyntactic scores at the first sampling pointwere used as a covariate. The data met the assumptionof parallelism of slopes with respect to the covariate, asevidenced by a nonsignificant F ratio for the Covariate ×Dependent Variable interaction. The covariate adjust-ment did not alter the statistical significance status ofthe data; statistically significant differences were foundfor the morphosyntactic data both before and aftercovariate adjustment. The ANCOVA revealed a statisti-cally significant difference in morphosyntactic changeacross the five groups, F(4, 41) = 3.6, p = .015.

Follow-up t tests revealed that the goal attack strat-egies of two of the intervention groups (morphosyntaxfirst and alternating) produced significantly greater mor-phosyntactic change than was found in the control group(see Table 3). Change in the alternating group was sig-nificantly greater than in the control group, t(16) = 3.0,p = .003, d = 1.53. Change in the morphosyntax firstgroup was also significantly greater than in the controlgroup, t(15) = 1.82, p = .037, d = .95.

Phonology Change in InterventionVersus Control Groups at 12 Weeks

A second ANCOVA was applied to assess phonologychange across the five groups. Phonology scores at thefirst sampling point were used as a covariate. The datamet the assumption of parallelism of slopes with respectto the covariate, as evidenced by a nonsignificant F ra-tio for the Covariate × Dependent Variable interaction.The covariate adjustment did not alter the statisticalsignificance status of the data; nonsignificant resultswere found both before and after covariate adjustment.The ANCOVA results revealed that, with respect to pho-nology change, there was no significant main effect for

Table 3. Means and standard deviations for percentage change in morphosyntax (FMC) from Sample 1 to2 and from Sample 1 to 3 (after 24 weeks of intervention) for each goal attack strategy.

FMCTime 1 FMC1–2 FMC1–3

Strategy M SD M SD M SD

Morphosyntax first 46.9 24.9 14.4 14.2 18.1 20.8Phonology first 37.7 26.6 6.2 14.8 11.1 17.0Alternating 43.8 17.9 23.0 20.4 34.1 18.4Simultaneous 31.2 27.4 8.5 12.4 19.2 13.8Control 49.9 17.0 –0.07 12.2 NA

Note. FMC = finite morpheme composite.


group, F(4, 41) = 1.4, p = .25 (see Table 4). However,based on the directional hypothesis that the interven-tion groups would produce greater change than the no-treatment control group, those planned comparison t testswere computed (Rosenthal, Roznow, & Rubin, 2000). Theresults of the planned comparisons revealed that threeof the four intervention strategies produced phonologychange that was significantly greater than that foundfor the control group: morphosyntax first group, t(15) =1.62, p = .05, d = .85; alternating group, t(16) = 1.84, p =.03, d = .94; simultaneous group, t(14) = 2.29, p = .01, d =1.22. Change in the phonology first group was not foundto be statistically significant, t(15) = 1.56, p = .06. How-ever, the obtained effect size for this comparison waslarge (d = .81).

The second research question addressed differencesin morphosyntactic change and phonological changeacross the four strategies after 24 weeks of interven-tion. Separate ANCOVAs were applied for morpho-syntactic and phonological change, with Sample 1 FMCsand TGCs, respectively, used as covariates.

Morphosyntactic Change AcrossIntervention Strategies at 24 Weeks

After 24 weeks, differences in morphosyntacticchange across the four experimental groups were assessedusing ANCOVA procedures. The data met the assump-tion of parallelism of slopes with respect to the covari-ate, as evidenced by a nonsignificant F ratio for the Co-variate × Dependent Variable interaction. After 24 weeksof intervention, results revealed a significant group dif-ference for morphosyntactic change, F(3, 35) = 4.17, p =.013 (see Table 3). Of the four intervention strategies,the alternating strategy produced the greatest morpho-syntactic change in comparison to the phonology first strat-egy, t(19) = 3.37, p = .0018, d = 1.55; the morphosyntaxfirst strategy, t(19) = 2.32, p = .026, d = 1.06; and the si-multaneous strategy, t(18) = 2.41, p = .02, d = 1.13. Gains

observed for the morphosyntax first, phonology first, andsimultaneous strategies did not differ from one another.

Phonology Change Across InterventionStrategies at 24 Weeks

After 24 weeks of intervention, results revealed nosignificant difference in phonology change across the fourintervention strategies, F(3, 35) = 0.12, p = .95. The datamet the assumption of parallelism of slopes with respectto the covariate, as evidenced by a nonsignificant F ra-tio for the Covariate × Dependent Variable interaction.Mean percentages for phonological change in the mor-phosyntax first, phonology first, alternating, and simul-taneous groups were 35.4, 35.5, 29.7, and 36.7, respec-tively (see Table 4).

DiscussionMorphosyntactic Change

To date, there have been no investigations of theeffects of different goal attack strategies in targetingboth phonological and morphosyntactic impairments,although many preschool children on the caseloads ofpracticing SLPs have these co-occurring deficits. In thepresent study of goal attack strategy outcomes, everyeffort was made to reduce threats to internal validity byusing a control group design and random assignment ofparticipants to experimental groups. Ethical consider-ations precluded assigning children at random to thecontrol group and, as such, any time the control groupis not assigned at random, some caution must be takenin interpreting results from treatment versus controlcomparisons. Results showed that after 12 weeks of in-tervention, morphosyntactic change was significantlygreater for the alternating strategy as well as the mor-phosyntax first strategy than it was for the no-interven-tion control group, demonstrating the efficacy of themorphosyntactic intervention used in this study. After

Table 4. Means and standard deviations for percentage change in phonology (TGC) from Sample 1 to 2and from Sample 1 to 3 (after 24 weeks of intervention) for each goal attack strategy.

TGCTime 1 TGC1–2 TGC1–3

Strategy M SD M SD M SD

Morphosyntax first 25.5 11.82 19.3 15.1 35.4 30.2Phonology first 28.9 23.2 18.7 22.4 35.5 29.5Alternating 29.6 18.0 21.2 23.8 29.7 18.1Simultaneous 22.8 22.3 26.5 22.5 36.7 19.9Control 25.0 12.8 3.3 7.1 NA

Note. TGC = target generalization composite.


24 weeks of intervention, morphosyntactic change wasgreatest for children receiving the alternating goal at-tack strategy in comparison to all other strategies. Thechange in FMC resulting from this strategy was morethan a standard deviation greater than that of any ofthe other strategies (all ds > 1.0). For the children withco-occurring phonological and morphosyntactic impair-ments in the present study, the associated large effectsizes suggest that the alternating strategy (as opposedto other strategies or external variables) best explainsthe variability in morphosyntactic change. This is espe-cially significant because difficulty with finite morphemeshas been shown to set atypical language learners apartfrom typical ones (Bedore & Leonard, 1998; Rice & Wexler,1996). Furthermore, these morphemes are known to beespecially resistant to intervention for children with lan-guage impairments (Goffman & Leonard, 2000).

It might be hypothesized that the alternating strat-egy was superior for morphosyntactic change due to itsearly focus on morphology, as well as a sustained focuson that domain every other week over the 24-week pe-riod, thus capitalizing on the gradual nature of the learn-ing process. This explanation is supported by the betterperformance of the morphosyntax first group in com-parison to the other groups at 12 weeks. In addition,phonology outcomes at 12 weeks were best for the alter-nating, simultaneous, and morphosyntax first strategies,all of which had a focus on morphology from the verybeginning. Leonard (1981) suggested that initial train-ing on linguistic forms might heighten a child’s abilityto observe the speech of others and extract various forms,such as new sounds. The simultaneous strategy alsoinvolved an early focus on morphosyntax, as well as re-peated cycling of goals over 24 weeks, but did not yieldsuperior morphosyntactic change. The multiple focus onphonological and morphosyntactic targets within activi-ties in the simultaneous strategy may have been over-whelming for the children for whom both the phonologi-cal and morphosyntactic systems were compromised.

The present results show that the morphosyntax firstand alternating strategies produced significant gains inmorphosyntactic and phonological performance in com-parison to the control group after 12 weeks, supportingprevious research in which language interventions fa-cilitated change in phonology (Hoffman et al., 1990;Matheny & Panagos, 1978; Tyler et al., 2002; Wilcox &Morris, 1995). This finding diverges from that of Fey etal. (1994), who observed no gains in phonology follow-ing a successful morphosyntactic intervention. There aremany differences in the interventions and in the out-come measures that could explain this difference. Ourpattern of results, however, is consistent with the claimthat clinicians may need to elicit and reinforce produc-tions of complex morphophonemes to achieve significant

gains in phonology with intervention on morphosyntax.This component was present in all of our strategies butwas absent from the focused stimulation approach usedby Fey and colleagues.

Phonological ChangeThere were no significant differences in phonologi-

cal change between the different goal attack strategiesafter either 12 or 24 weeks of intervention. In compari-son to the no-treatment control group, however, phono-logical change was greater for the morphosyntax first,alternating, and simultaneous strategies. Although thecomparison of change for the phonology first group andthe control group did not reach statistical significance(p = .06), the associated effect size was large (d = .81).Rosenthal et al. (2000) cautioned against presuming nodifference exists when data fail to reach statistical sig-nificance but have large effect sizes. In an omnibus testsuch as the ANCOVA, a larger sample size would likelyhave resulted in a statistically significant differencebetween change in the phonology first and control groups.

Despite the significant gains on the phonology mea-sure at 12 weeks, no single goal attack strategy wassuperior in facilitating gains in phonological perfor-mance after 24 weeks. Unlike morphosyntax, which maybenefit from an early goal attack focus, it may not mat-ter when phonology receives primary focus. It seemsmore likely, however, that by requiring correct produc-tions of finite morphemes such as 3SR, the morpho-syntactic intervention led to improvements in produc-tion of final clusters and other less complex aspects ofphonology (Gierut & Champion, 2001). In essence, fo-cus on cluster production in the morphosyntactic inter-vention may have “washed out” any differential effectsof block, simultaneous, or alternating configurations onphonological performance. Further research is neededto determine possible effects of treating finite mor-phemes on the phonological system.

Finally, although the phonological intervention inthe present study appeared to have an effect on phonol-ogy at 12 weeks, it did not have the cross-domain effecton morphosyntax that others have observed (Bopp, 1995;Fey & Stalker, 1986; Tyler et al., 2002; Tyler & Sandoval,1994). This is most likely because the phonological in-tervention did not focus on final /s, z/ or cluster formsthat are required for precise production of finite mor-phemes such as past tense -ed and 3SR. Further, only6 children in this study applied final consonant dele-tion/cluster reduction more than 20% of the time. Thus,there were few children whose morphological difficul-ties could have been affected by their phonological limi-tations and for whom it might be hypothesized that workon phonology targets would result in morphological gains


similar to the child in Fey and Stalker’s (1986) study. Avertical goal attack strategy in which phonology istreated to a criterion accuracy, not unlike the block strat-egy in the present study, might be preferred when themorphological problem is due to phonological factors.

Clinical ImplicationsIn the present study, phonology and morphosyntactic

goals were targeted simultaneously, alternately on aweekly basis, or vertically in blocks (12 weeks) for chil-dren with co-occurring phonological and morphosyn-tactic deficits. Although it may seem desirable to targetdomains of speech and morphosyntax simultaneouslyor separately in blocks, the results reported here pro-vide preliminary evidence that alternating phonologicaland morphosyntactic goals on a weekly basis may lead togreater morphosyntactic change. Phonological outcomesfrom the different strategies in this study are less clear.The lack of a difference for phonological performance bystrategy along with the potential cross-domain effects ofthe morphosyntactic intervention suggests it is possiblethat providing morphosyntactic intervention alone for24 weeks could be just as successful as the alternatingstrategy. A 24-week morphosyntactic intervention anda 24-week phonological intervention would need to becompared experimentally before concluding that the al-ternating strategy leads to the greatest morphosyntacticchange.

Children were randomly assigned to the differentstrategies without the consideration of a number of vari-ables that Weiss (2002) suggested may be important inselection of a goal attack strategy. Perhaps most relevantof these, to the results of the present study, is the inter-relatedness of goals across the two domains. That is, towhat extent is the child’s morphological problem due tophonological factors and to what extent is the speechproblem independent of the morphological problem?Ultimately, the selection of a strategy may be determinedby the extent to which a child’s specific goals in phono-logical and morphosytactic domains are interrelated. Thenecessary individual analyses were not completed, so itis not known if a strategy other than the alternatingone may be more effective for children whose profile, forexample, showed interrelated goals across domains orone domain that was relatively more impaired than an-other. It also is not known how slight differences in tar-gets across groups may have influenced the results. Thelarge standard deviations for morphosyntactic and pho-nological change indicate that individual responses tothe intervention strategies were highly variable. Thus,one type of intervention may have been very effectivefor some children and less effective for others. Differenttypes of errors and starting levels may have contrib-uted to different amounts of generalization learning

across participants. Further research regarding the in-teraction of different error patterns, intervention tar-gets, and outcome measures is necessary to clarifymechanisms by which intervention focused on morpho-syntax may cause change in phonology and vice versa.This would help in determining whether an alternatingstrategy is always the preferred strategy for childrenwith co-occurring phonological and morphosyntacticimpairments or whether certain conditions make otherstrategies equally or more efficacious.

AcknowledgmentsSupport for this research was provided by National

Institute on Deafness and Other Communication DisordersGrant DC03358 to the University of Nevada, Reno. Apprecia-tion is extended to Kellie Paul for preparing the phonologicalintervention; to Veronica Spurlock, Holly Anderson, JenniferHodges, Carissa Welch, Lisa Steigman, Chelsea Nork, KelliePaul, and Holly Willet for data collection, transcription, andreliability coding. We also wish to express our sincere thanksto the children, their parents, and the Washoe County SchoolDistrict administrators Joy Erickson and Jen Lamb, theSLPs, and the teachers who participated in this research.

ReferencesBankson, N., & Bernthal, J. E. (1990). Bankson-Bernthal

Test of Phonology. Austin, TX: Pro-Ed.

Bedore, L. M., & Leonard, L. B. (1998). Specific languageimpairment and grammatical morphology: A discriminantfunction analysis. Journal of Speech, Language, andHearing Research, 41, 1185–1192.

Bishop, D. V. M., & Edmundson, A. (1987). Language-impaired 4-year-olds: Distinguishing transient frompersistent impairments. Journal of Speech and HearingDisorders, 52, 156–173.

Bopp, K. (1995, April). The effects of phonological interven-tion on morphosyntactic development in preschool childrenwith phonological and morphosyntactic disorders.Unpublished master’s thesis, University of BritishColumbia, Vancouver, British Columbia, Canada.

Brown, R. (1973). A first language. Cambridge, MA:Harvard University Press.

Burgemeister, B., Blum, B., & Lorge, I. (1972). ColumbiaMental Maturity Scale (3rd ed.). New York: The Psycho-logical Corporation.

Camarata, S. (1998). Connecting speech and language:Clinical applications. In R. Paul (Ed.), Exploring thespeech-language connection (pp. 299–318). Baltimore:Brookes Publishing.

Camarata, S., Nelson, K., & Camarata, M. (1994).Comparison of conversational-recasting and imitativeprocedures for training grammatical structures in childrenwith specific language impairment. Journal of Speech andHearing Research, 37, 1414–1423.

Cleave, P. L., & Fey, M. (1997). Two approaches to thefacilitation of grammar in children with language


impairments: Rationale and description. AmericanJournal of Speech-Language Pathology, 6(1), 22–32.

Cohen, J. (1988). Statistical power analysis for the behav-ioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Conti-Ramsden, G., & Botting, N. (1999). Classification ofchildren with specific language impairment: Longitudinalconsiderations. Journal of Speech, Language, and HearingResearch, 42, 1195–1204.

Duder, C., Camarata, S., Camarata, M., Koegel, R., &Koegel, L. (1998, November). No free lunch? Languagechange during phonological treatment of children.Presented at the annual convention of the AmericanSpeech Language Hearing Association, San Antonio, TX.

Fey, M. E. (1986). Language intervention with youngchildren. San Diego, CA: College-Hill Press.

Fey, M. E., Cleave, P. L., Long, S., & Hughes, D. L. (1993).Two approaches to the facilitation of grammar in childrenwith language impairment: An experimental evaluation.Journal of Speech and Hearing Research, 36, 141–157.

Fey, M., Cleave, P. L., Ravida, A. I., Long, S. H., Dejmal,A. E., & Easton, D. L. (1994). Effects of grammarfacilitation on the phonological performance of childrenwith speech and language impairments. Journal of Speechand Hearing Research, 37, 594–607.

Fey, M. E., & Stalker, C. H. (1986). A hypothesis-testingapproach to treatment of a child with an idiosyncratic(morpho)phonological system. Journal of Speech andHearing Disorders, 51, 324–336.

Gierut, J. A. (1998). Treatment efficacy: Functionalphonological disorders in children. Journal of Speech andHearing Disorders, 41, S85–S100.

Gierut, J. A., & Champion, A. H. (2001). Syllable onsets II:Three-element clusters in phonological treatment. Journalof Speech, Language, and Hearing Research, 44, 886–904.

Goffman, L., & Leonard, R. (2000). Growth of languageskills in preschool children with specific languageimpairment: Implications for assessment and interven-tion. American Journal of Speech-Language Pathology, 9,151–161.

Gray, B., & Ryan, B. (1973). A language program for thenonlanguage child. Champaign, IL: Research Press.

Haskill, A. M., Tyler, A. A., & Tolbert, L. C. (2001).Months of morphemes: A theme-based cycles approach.Eau Claire, WI: Thinking Publications.

Hegde, M. (1980). An experimental-clinical analysis ofgrammatical and behavioral distinctions between verbalauxiliary and copula. Journal of Speech and HearingResearch, 23, 864–877.

Hegde, M. N., Noll, M. J., & Pecora, R. (1979). A study ofsome factors affecting generalization of language training.Journal of Speech and Hearing Disorders, 44, 301–320.

Hoffman, P. R., Norris, J. A., & Monjure, J. (1990).Comparison of process targeting and whole languagetreatment for phonologically delayed preschool children.Language, Speech, and Hearing Services in Schools, 21,102–109.

Howe, M. L. (1992). Transcription and coding manual foranalysis of language samples: Kansas Language Tran-script Database. Unpublished manual, University ofKansas, Lawrence.

Howell, E., & Dean, J. (1994). Treating phonologicaldisorders in children: Metaphon—theory to practice (2nded.). London: Whurr.

Hughes, D. L., & Carpenter, R. L. (1985). Short-termsyntax learning and generalization effects. Child Lan-guage Teaching and Therapy, 5, 204–222.

Leadholm, B. J., & Miller, J. F. (1993). Language sampleanalysis: The Wisconsin guide. Milwaukee: WisconsinDepartment of Public Instruction.

Leonard, L. (1974). A preliminary view of generalization inlanguage training. Journal of Speech and HearingDisorders, 39, 429–436.

Leonard, L. (1981). Facilitating linguistic skills in childrenwith specific language impairment. Applied Psycholinguis-tics, 2, 89–118.

Masterson, J., & Pagan, F. (1993). Interactive System forPhonological Analysis [Computer software]. San Antonio,TX: The Psychological Corporation.

Matheny, N., & Panagos, J. M. (1978). Comparing theeffects of articulation and syntax programs on syntax andarticulation improvement. Language, Speech, and HearingServices in Schools, 9, 57–61.

Miller, J., & Chapman, R. (2000). SALT: SystematicAnalysis of Language Transcripts [Computer software].Madison: University of Wisconsin.

Nelson, K. (1989). Strategies for first language teaching. InM. Rice & R. Schiefelbusch (Eds.), The teachability oflanguage (pp. 263–310). Baltimore: Brookes Publishing.

Paul, R., & Shriberg, L. D. (1982). Associations betweenphonology and syntax in speech-delayed children. Journalof Speech and Hearing Research, 25, 536–547.

Rapin, I., & Allen, D. A. (1983). Developmental languagedisorders: Nosologic considerations. In U. Kirk (Ed.),Neuropsychology of language, reading, and spelling (pp.155–184). New York: Academic Press.

Rapin, I., & Allen, D. A. (1988). Syndromes in developmen-tal dysphasia and adult aphasia. In F. Plum (Ed.),Language, communication, and the brain (pp. 57–75). NewYork: Raven Press.

Rice, M., & Wexler, K. (1996). Toward tense as a clinicalmarker of specific language impairment in English-speaking children. Journal of Speech and HearingResearch, 39, 1239–1257.

Rice, M., Wexler, K., & Cleave, P. (1995). Specific lan-guage impairment as a period of extended optionalinfinitive. Journal of Speech and Hearing Research, 38,850–863.

Robbins, J., & Klee, T. (1987). Clinical assessment oforopharyngeal motor development in young children.Journal of Speech and Hearing Disorders, 52, 271–277.

Rosenthal, R., Rosnow, R. L., & Rubin, D. B. (2000).Contrasts and effect sizes in behavioral research: Acorrelational approach. Cambridge, U.K.: CambridgeUniversity Press.

Shriberg, L. D. (1993). Four new speech and prosody-voicemeasures for genetics research and other studies indevelopmental phonological disorders. Journal of Speechand Hearing Research, 36, 105–140.

Shriberg, L. D., & Austin, D. (1998). Comorbidity of


speech-language disorders: Implications for a phenotypemarker for speech delay. In R. Paul (Ed.), Exploring thespeech-language connection (pp. 73–117). Baltimore: PaulH. Brookes.

Shriberg, L. D., & Kwiatkowski, J. (1982). Phonologicaldisorders III: A procedure for assessing severity ofinvolvement. Journal of Speech and Hearing Disorders,17, 256–270.

Shriberg, L. D., & Kwiatkowski, J. (1994). Developmen-tal phonological disorders: I. A clinical profile. Journal ofSpeech and Hearing Research, 37, 1100–1126.

Shriberg, L. D., Kwiatkowski, J., Best, S., Hengst, J., &Terselic-Weber, B. (1986). Characteristics of childrenwith phonologic disorders of unknown origin. Journal ofSpeech and Hearing Disorders, 51, 140–161.

Shriberg, L. D., Tomblin, J. B., & McSweeny, J. L.(1999). Prevalence of speech delay in 6-year-old childrenand comorbidity with language impairment. Journal ofSpeech, Language, and Hearing Research, 42, 1461–1481.

Snow, D. (1998). Prosodic markers of syntactic boundariesin the speech of 4-year-old children with normal anddisordered language development. Journal of Speech,Language, and Hearing Research, 41, 1158–1170.

SPSS. (1988). SPSS-X user’s guide (3rd ed.). Chicago: Author.

Tallal, P., Ross, R., & Curtiss, S. (1989). Familial aggrega-tion in specific language impairment. Journal of Speechand Hearing Disorders, 54, 167–173.

Tyler, A. A., Edwards, M. L., & Saxman, J. H. (1987).Clinical application of two phonologically based treatmentprocedures. Journal of Speech and Hearing Disorders, 52,393–409.

Tyler, A. A., Lewis, K. E., Haskill, A., & Tolbert, L. C.(2002). Efficacy and cross-domain effects of a morpho-syntax and a phonology intervention. Language, Speech,and Hearing Services in Schools, 33, 52–66.

Tyler, A. A., & Sandoval, K. T. (1994). Preschoolers withphonological and language disorders: Treating differentlinguistic domains. Language, Speech, and HearingServices in Schools, 25, 215–234.

Tyler, A. A., & Watterson, K. H. (1991). Effects of phono-logical versus language intervention in preschoolers withboth phonological and language impairment. ChildLanguage Teaching and Therapy, 7, 141–160.

Weiss, A. L. (2002). Planning language intervention foryoung children. In D. K. Bernstein & E. Tiegerman-Farber(Eds.), Language and communication disorders in children(pp. 256–314). Boston: Allyn & Bacon.

Wiig, E. H., Secord, W., & Semel, E. (1992). ClinicalEvaluation of Language Fundamentals–Preschool. SanAntonio, TX: The Psychological Corporation.

Wilcox, K. A., & Morris, S. E. (1995). Speech outcomes ofthe language-focused curriculum. In M. Rice & K. Wilcox(Eds.), Building a language-focused curriculum for thepreschool classroom: A foundation for lifelong communica-tion (pp. 171–180). Baltimore: Paul H. Brookes.

Wolfus, B., Moscovitch, M., & Kinsbourne, M. (1980).Subgroups of developmental language impairment. Brainand Language, 10, 152–171.

Zimmerman, I., Steiner, V., & Pond, R. (1992). PreschoolLanguage Scale–3. San Antonio, TX: The PsychologicalCorporation.

Received July 30, 2002

Accepted March 11, 2003

DOI: 10.1044/1092-4388(2003/085)

Contact author: Ann A. Tyler, PhD, University of Nevada,Department of Speech Pathology & Audiology, Nell J.Redfield Building/152, Reno, NV 89557.E-mail: [email protected]

Appendix (p. 1 of 3). Participants’ morphosyntax goals, phonology goals, andgeneralization targets.

Morphosyntax Phonology GeneralizationChild goals goals targets

A4 3SR, -ed, ptirr, copula be Initial: /l, S, fa/ & /st, sn, sk/ Initial: /k, g, ‰, ¸, v/Final: /‰, k/ Final: /¸, S, f, v, l/

A5 3SR, -ed, ptirr, copula be Initial: /l, S/ & /st, sn, sk/ Initial: /‰, ¸/Final: /‰/ Final: /S, l/

A6 3SR, -ed, ptirr, copula be Initial: /l, S, sa/ & /st, sn, sk/ Initial: /z/Final: /‰/ Final: /s, z, S, l/

A8 3SR, -ed, ptirr, copula be Initial: /‰, k, S/ & /sn, sp, sk/ Initial: /¸/Final: /k, S, ‰, ¸/

A10 3SR, -ed, ptirr, copula be Initial: /‰, k, S/ & /sn, sp, sk/ Initial: /g, ¸/Final: /k, S, ¸/

A15 3SR, -ed, ptirr, copula be Initial: /l, S/ & /sl, st, sn/ Initial: /‰, ¸/Final: /‰/ Final: /l, s, z, S, ¸/




A16 3SR, -ed, ptirr, copula be Initial: /l, S/ & /sl, st, sn/ Initial: /‰, ¸/Final: /‰/ Final: /l, ¸/

A17 3SR, -ed, ptirr, copula be Initial: /l, S/ & /sl, st, sn/ Initial: /‰, ¸/Final: /‰/ Final: /l, S, ¸/

A18 -s, 3SR, -ed, copula be Initial: /k, sl, fl/ Initial: /g, z, f, v/Final: /f, s/ Final: /v/

A20 -s, 3SR, -ed, copula be Initial: /k, sl, fl/ Initial: /s, z, f, v/Final: /f, s/ Final: /v, s, z/

B1 3SR, -ed, ptirr, copula be Initial: /s, ‰/ & /sn, sk/ Initial: /¸, z/Final: /f/ Final: /s, z, ‰, ¸/

B2 -ed, ptirr, ’s, auxiliary be Initial: /l/ & /kl, fl/ Initial /v, ¸/Final: /f, ‰/ Final: /l/

B3 3SR, -ed, ’s, auxiliary be Initial: /l/ & /kl, fl/ Initial: /‰, f, v, s, z/Final: /f, ‰/ Final: /¸, v/

B7 3SR, -ed, ptirr, copula be Initial: /s, ‰/ & /sn, sk/ Initial: /¸, v/Final: /f/ Final: /S, ‰, ¸/

B9 3SR, -ed, ptirr, ’s Initial: /g, la/ & /sl, fl, st/ Initial: /v/Final: /fa, s/ Final: /l, z/

B11 3SR, -ed, ’s, copula be Initial: /g/ & /sl, fl, st/ Initial: /k/Final: /f, s/ Final: /g, k, f, z, S/

B12 3SR, -ed, ptirr, copula be Initial: /g, ‰/ & /sl, st, sn/ Initial: /S, ¸/Final: /S/ Final: /‰, ¸/

B13 3SR, -ed, ptirr, copula be Initial: /g, fa, ‰a/ & /sl, st, sn/ Initial: /k, v/Final: /S/ Final: /k, g, f, S/

B14 3SR, -ed, ptirr, copula be Initial: /g, ‰/ & /sl, st, sn/ Initial: /k, S, ¸/Final: /S/ Final: /k, g, ‰, ¸/

B19 3SR, -ed, ’s, copula be Initial: /f/ & /sn, sk, sp/ Initial: /g, v/Final: /k, s/ Final: /g, z, f, v, S/

C30 3SR, -ed, ptirr, auxiliary be Initial: /k, s, ¸, l/ Initial: /g, z, ‰/Final: /s, z, ¸/

C31 3SR, -ed, ’s, copula be Initial: /T, ¸, r/ & /st, sn, sk/ Initial: /l, D/Final: /T, ̧ /

C32 3SR, -ed, ptirr, copula be Initial: /k, ‰, r/ & /st, sn, sk, sl/ Initial: /g, ¸/Final: /¸/

C33 3SR, -ed, ’s, copula be Initial: /T, ¸, r/ & /st, sn, sk/ Initial: /D, ‰/Final: /T, ‰, ¸/

C34 3SR, -ed, ptirr, copula be Initial: /‰, k, r/ & /sn, st, sk, sl/Final: /¸/

C35 3SR, -ed, ’s, copula be Initial: /T, ¸, r/ & /st, sn, sk/ Initial: /l, ‰, D/Final: /T, ‰, ¸/

C36 3SR, -ed, ptirr, copula be Initial: /k, ‰, r, la/ & /sn, st, sk, sl/ Initial: /g, ¸/Final: /‰, ¸/

C37 3SR, -ed, ptirr, auxiliary be Initial: /k, l/ & /st, sk, sp, sn/ Initial: /‰, ¸/Final: /‰/ Final: /p, t, k, ¸/




C38 3SR, -ed, ptirr, auxiliary be Initial: /f/ & /st, sp, sk, sn/ Initial: /v, s, z, ‰, ¸/Final: /z, ‰/ Final: /f, v, s, ¸/

C39 3SR, -ed, ’s, auxiliary be Initial: /f/ & /sn, sp, st, sk/ Initial: /v, s, z, ‰, ¸/Final: /z, ‰/ Final: /f, v, ¸/

C40 3SR, ’s, -ed, auxiliary be Initial: /f/ & /sn, sp, st, sk/ Initial: /v, z/Final: /z, ‰/ Final: /s, ¸/

D21 3SR, -ed, ptirr, auxiliary be Initial: /g, f, l/ & /sp, sn, st/ Initial: /k, v/Final: /v/

D22 3SR, -ed, ptirr, copula be Initial: /r/ & /sn, st, sk, sl/ Initial /s, z, ‰/Final: /z, ‰/ Final: /¸/

D23 3SR, -ed, ptirrr, copula be Initial: /r/ & /sn, st, sk, sl/ Initial: /z, ¸/Final: /z, ‰/ Final: /s/

D24 3SR, -ed, ptirr, copula be Initial: /g, f, l/ & /sp, sn, st/ Initial: /k, v/Final: /k, g, f, v/

D25 3SR, -ed, ptirr, copula be Initial: /g, l/ & /sn, st, sk/ Initial: /k, f, v/Final: /f/ Final: /k, g, v/

D26 3SR, -ed, ’s, copula be Initial: /p, f, l/ & /sn, st, sk, sl/ Initial: /t, k, v, s/Final: /f, v/

D27 3SR, -ed, ptirr, auxiliary be Initial: /p, f, l/ & /sn, st, sk, sl/ Initial: /b, m, v, s/Final: /b, v, p/

D28 3SR, ’s, ptirr, auxiliary be Initial: /g, f, r/ Initial: /k, s, z, v/Final: /z/ Final: /k, g, f, v, s/

D29 3SR, ’s, ptirr, auxiliary be Initial: /g, f, r/ Initial: /k, v, z/Final: /z/

Note. 3SR = third person singular regular; -ed = regular past tense; ptirr = irregular past tense; ’s = possessive;-s = regular plural; A = morphology first; B = phonology first; C = alternating; D = simultaneous.aIntervention target during individual sessions.

Documents

Outcomes of Different Speech and Language Goal Attack ... · PDF fileTyler et al.: Outcomes of Different Interventions 1079 that was similar to that achieved by the phonological intervention