Not all hits in our literature search fulfil the inclusion criteria in our systematic vocabulary intervention review

In our literature review we want to find out about vocabulary intervention approaches, so we know what approaches have been used in earlier research and what works best with regards to improving vocabulary in children with Down syndrome.

The second article that we have selected in this series of our literature review posts is ‘Advancing imitation and requesting skills in toddlers with Down syndrome’ by Feeley, Jones, Blackburn, & Bauer (2011). This study was a hit in our literature search, however, the study will not be included in our review because it did not meet our inclusion criteria; this is since it is, a.) Not directly a vocabulary intervention, and b.) Not including a control group/condition. As the study was a hit in our search we will nonetheless discuss the study in this post, in relation to variability in intervention content and research quality across the various hits in our search.

The reason we are including only studies with control groups/conditions is a really important one; if children receive an intervention and their performance improves, but there is no control group or control condition, then it is not possible to determine whether they improved simply due to attention, time in general, or any given property of the specific intervention. Also, without comparing to another population, it is not possible to argue that the intervention is specifically effective for the population tested, e.g., Down syndrome. Without a control condition or control group researchers are unable to empirically test the theory underlying the given intervention.

The article by Feeley et al. (2011) aimed to investigate the result of an intervention program developed for children with Down syndrome that focuses on imitation and requesting. Four toddlers who have Down syndrome participated in their study. Due to the suggestion of relative strengths in social development in Down syndrome, based on a review by Fidler et al (2005), social interaction was used in the intervention by Feeley et al. to encourage and reinforce requesting and verbal imitation in toddlers with Down syndrome. The intervention involved providing specific prompts (telling the child to say a certain sound) to give the child opportunities to verbally imitate or encourage them to request. Corrective feedback was also given, along with social reinforcement such as hugging and social praise.

The sessions were run by the child’s speech therapist and their mother. Each intervention session lasted around 15 minutes, with 3 sessions per week. Sessions could involve up to 10 ‘opportunities’ (meaning opportunities to respond to prompts), but often consisted of between 5 and 7 opportunities. It was unclear why this variation was present rather than a set number of opportunities in each session for each child. Verbal imitation and requesting interventions were done at the same time so that the children could use any learnt verbal imitation skills when trying to request. For verbal imitation children were presented with a model of the target sound (spoken by the interventionist – mother or speech therapist), such as the ‘ahhhh’ sound. In earlier sessions the sound was exaggerated, the child was also given a physical prompt (by having gentle pressure on their chin). Over the course of the sessions the interventionist made less exaggeration of the target sound and put less pressure on the child’s chin. For the requesting training, the training followed a sequence, starting with training the use of a gaze shift to request social interaction with an object (the interventionist would prompt the gaze shift by, for instance, taking a deep breath or moving into the child’s line of sight). If the child looked at the object and then to the interventionist they were subsequently given the object. Training next moved on to the use of gaze shift and any vocalisation (whereby the interventionist additionally prompted the child by gently tapping on their mouth). Finally gaze shift was to be accompanied with an approximation of the appropriate requesting word (e.g., an ‘mmm’ sound for ‘more’). The interventionist’s prompt was to tell the child to ‘say ‘mmm’ for instance, and the object was placed in a jar such that the child was requesting for help (e.g., to get the object from out of the jar).

The children were assessed in terms of whether they reached a mastery criterion, e.g., 80% independent correct responding (independent correct responding refers to children’s ability to imitate the sound made by the interventionist without the need to hear an exaggerated verbal model and/or a physical prompt, and their ability to make a request without a prompt) in ‘two consecutive sessions on two days’ (pg. 2418), we assume this means a correct response one day and again the next day. The authors found that the children were able to acquire verbal imitation and requesting skills. Each child’s results are reported separately in terms of how many probe opportunities it took to reach the mastery criterion. It is difficult to see how many probe opportunities were needed on average as group level results are not provided.

Maintenance of skills was assessed via a follow up (3 to 4 weeks after intervention). The follow up was based on giving the child just one opportunity to respond to a probe, e.g., prompting the child to ‘say ‘mmm’, and seeing whether the child responds correctly in that one instance, or e.g., showing the child the object in the jar such that they are prompted to make a request for help getting the object out of the jar. The children were probed once for each of the sounds that they had learnt in the intervention and the most sophisticated form of requesting they had learnt in the intervention. For three of the four children follow up was assessed and showed that children continued to show the skills (i.e., same level of performance) that they showed at the end of the intervention.

Generalisation was not measured in a systematic way, e.g., a particular set of non-trained items tested across each child.  Rather, sounds that the interventionists said that they heard spontaneously during intervention, but that the child could not say at baseline, were tested again after the intervention. Generalization was assessed for just 2 of the 4 children and showed that they generalized to novel sounds. Given that this was only 2 children it does not tell us about what we should expect in the Down syndrome population more generally.

The authors note that they have demonstrated the effectiveness of the intervention, however we think this is a rather strong statement given the design and the limitations of the study. Improvement could be due to natural development over time rather than an effect of the intervention. The authors note the need for replication, and a replication was in fact subsequently carried out by Bauer and Jones (2014); we will discuss this study in another post in this series. However, we would argue that it is much more than replication that is needed here. The findings do not give us any idea of meaningful group level patterns for those with Down syndrome. A follow up intervention study would be needed to test the claims made by Feeley et al, but this should involve a larger sample and a control comparison in order to tell us something more meaningful about this intervention, whether it results in any significant effects, and test the validity of the theory underlying it.

Further, for the four children in the study it was difficult to compare individual differences in abilities. In the initial descriptions of the four children different measures were reported for the respective children. We were unclear about why the different measures for different children were reported. Other issues, such as one child missing a part of the intervention due to interventionist’s time constraints, and one child’s parents introducing a different intervention for one of the sounds, present a number of confounds in terms of interpreting the results in this very small scale study.


Fidler, D. J. (2005). The emerging Down syndrome behavioral phenotype in early childhood: Implications for practice. Infants & Young Children, 18(2), 86-103.

Feeley, K. M., Jones, E. A., Blackburn, C., & Bauer, S. (2011). Advancing imitation and requesting skills in toddlers with Down syndrome. Research in developmental disabilities, 32(6), 2415-2430.


Liz and Kari-Anne

The effect of a combined language and reading intervention on vocabulary growth in children with Down syndrome

We are currently in the process of undertaking a systematic literature review of existing vocabulary interventions carried out with children who have Down syndrome. Having a good insight into existing interventions in this area is very useful for us when planning our DSL+-intervention. As we work on this review we will write a post each week about a different vocabulary intervention article discovered through the review process.

The first article that we are going to present in this series of posts is by Burgoyne, Duff, Clarke, Buckley, Snowling, & Hulme (2012); they provided evidence that a combined reading and language intervention was beneficial for vocabulary development (as well as literacy development) in children with Down syndrome. This study, which is one of the very few Randomised Control Trial (RCT) studies within language intervention for children with Down syndrome, involved 57 children (primary school years 1–5) and was delivered by their teaching assistants who received training on how to implement the intervention sessions.

The experimental group received training every day in school and each session was 40 minutes long. A waiting control-group receive the intervention at a later point.

Various measures related to reading and language were completed by the children at four different time points: 1. At screening, 2. Immediately prior to intervention, 3. After 20 weeks of intervention, and 4. After 40 weeks of intervention. Each week children received 4 sessions on ‘new teaching’ and one session dedicated to revision. The reading strand of the intervention focused on teaching reading and phonics together, based on Hatcher, Hulme and Ellis (1994). The language strand of the intervention focused on teaching children new vocabulary items; aiming to result in children expressing these new words. The new words were chosen based on a set of parent checklists to determine which words many of the children did not already know or understand and therefore would be useful to the children in general. Visual support and various games e.g., matching, were used to teach the new words. 

After the first 20 weeks of individual daily intervention, the experimental group showed significantly greater improvement compared to the waiting control group on measures of taught expressive vocabulary, single word reading, letter-sound knowledge, and phoneme blending.  These measures were all proximal to the content of the intervention. No transfer effects to any other language related skills measured were found. After 40 weeks of intervention for the experimental group and 20 weeks of intervention for the waiting controls, the experimental group was still ahead of the control group on most outcome measures, but the differences between the groups were not statistically significant.
The study by Burgoyne and colleagues was well designed, using RCT, obtaining a larger sample size than other studies in this area, and providing a long training period. Nonetheless, some limitations in the study and results are also present. As the authors note, the effect size was very modest. Another point to note is the age spread of the sample; how a 5 year old with Down syndrome may respond to language and reading intervention may be very different of course to the way that a 10 year old with Down syndrome may respond to such intervention. Looking at correlations, Burgoyne et al. found that across group’s younger children experienced greater reading growth by the end of the 40 weeks. Having an age-spread sample makes it difficult to know what outcome and effect size to expect from a given intervention at a certain point in children’s development. Of course, this is a tricky issue, as achieving a large sample is important (e.g. due to power in the analysis), and due to various practicalities researchers may have to make a compromise between sample size vs age range.

Potential limitations may also relate to teaching assistants delivering the intervention; specifically, their opportunities for preparation time and training, such as in pedagogical knowledge and how this impacts intervention implementation. Webster et al. (2011) notes that there is often an ‘instinctive, but mistaken assumption that less pedagogical skill is required when teaching pupils with SEN; if anything, a higher level of skill is needed’ (pg. 15).  Those teaching assistants who have a pedagogical background appear to ensure better implementation of training, resulting in greater results (Alborz et al., 2009).

Finally, because the program included both vocabulary and reading components, it is not possible to conclude whether it is the vocabulary training or the reading that is critical or if it is the combined approach that results in the observed effect. The intervention resulted in greater reading gains relative to language gains. It would be interesting to explore the effects of only the vocabulary training or only the reading training components in a comparison group, to see how this impacts the same outcome measures.

To summarize, this RCT study by Burgoyne and colleagues had many strengths and did improve reading and language outcomes (including vocabulary) in children with Down syndrome. As is so often the case in training studies however, the effects of the intervention did not generalize to other aspects of reading and language more distal from the intervention content. As with any study, there are some issues to consider regarding these outcomes.

Look out for our second post next week discussing another vocabulary intervention article carried out with children who have Down syndrome.

Alborz, A., Pearson, D., Farrell, P., & Howes, A. (2009). The impact of adult support staff on pupils and mainstream schools. London: Department for Children, Schools and Families and Institute of Education.
Burgoyne, K., Duff, F. J., Clarke, P. J., Buckley, S., Snowling, M. J., & Hulme, C. (2012). Efficacy of a reading and language intervention for children with Down syndrome: a randomized controlled trial. Journal of Child Psychology and Psychiatry, 53(10), 1044-1053.
Hatcher, P. J., Hulme, C., & Ellis, A. W. (1994). Ameliorating early reading failure by integrating the teaching of reading and phonological skills: The phonological linkage hypothesis. Child development, 65(1), 41-57.
Webster, R., Blatchford, P., Bassett, P., Brown, P., Martin, C., & Russell, A. (2011). The wider pedagogical role of teaching assistants. School Leadership and Management, 31(1), 3-20.

-Liz Smith and Kari-Anne B. Næss-

Ungen i mørket

Kronikk i Klassekampen, torsdag 17. september 2015, av Ingrid Holmboe Høibo

Foto: Ingrid Holmboe Høibo

Barn bruker meir og meir tid med nettbrettet. Kvifor bryr ikkje vaksne seg om kva dei gjer der inne?

Guten sit krumbøygd og med si fulle merksemd vendt mot skjermen. Nokre gonger kjem latter andre gonger song og prat. Ivrig tapping med finger mot skjerm. Merksemda er høg og avleiing er vanskeleg. Han er nett fylt fire og sit aleine med eit av sine fyrste kulturmøte.

I 82 %  av dei norske heimane er iPaden eit dagleg innslag. Barn i alderen 1-12 år bruker i snitt 43 minuttar på å spele data-/TV-/mobilspell kvar dag (tal frå medietilsynet, 2014). På få år har barns møte med kulturuttrykk endra seg totalt. Barne-TV til fastsett tid, sensurert og presentert av TV-vert, er bytta med nett-TV. Dataspel, som ein tidlegare måtte vera ganske stor for å kunne logge seg inn på, er no tilgjengeleg ved eit fingertapp. Bøker presentert i eit fang ligg no tilgjengeleg med opplesarstemme og integrerte moglegheiter for interaksjon med mediet. Ein kan teikne, filme, fotografere og animere med iPaden, men kven sørgjer for at det som blir servert barnet er av god kvalitet?

Dei siste fem åra har salet og konsumet av nettbrett eksplodert. Særleg barn har fått stor tilgang til dette mediet. Barn finn fram til dei applikasjonane dei ønskjer å bruke sjølv. Dei lever i ein kultur der dei blir visuelt bombardert kvar dag, og dei har lært seg å manøvrerer i dette mylderet av uttrykk. Mykje av manøvreringa går på intuisjon. Nettbrett spelar nettopp på denne intuitive måten å manøvrere på. Ofte er det kjennelege figurar, stereotypar og spin-off-produkt frå TV og film som blir valt til fordel for meir utfordrande og mangfaldige visuelle og tekstlege uttrykk.

For mange handlar det "å gjer iPad" i stor grad om tidtrøyte og barnevakt. Ein skil ikkje på kva denne bruke går ut på. Eg har gjort undersøkingar i barnehagar der iPad samvitsfult er kjøpt inn, som eit ledd i målet om å utvikle såkalla digital kompetanse. Men bruken av verktøyet har gått ut på å spele Angry birds, Lego, Kittykat eller liknande, som ei løn for andre utførte aktivitetar eller oppgåver.

Denne typen aktivitet kan vera kjekk som underhalding, men den utviklar verken evna til å lese bilde eller tekst, heller ikkje barnets kreativitet eller nysgjerrigheit. Målet med slike apper er å nå ein masse førehandsdefinerte mål. Fascinasjonen for spela kjem frå gjenkjenninga av figurar og den hyppige bruken av påskjønning og repetisjon i applikasjonen.

Det seier seg sjølv at denne typen applikasjonsaktivitet ikkje legg til rette for nokon litterær, biletleg eller filmatisk kunstoppleving. Barn treng å utfordre, bygge og øve på sitt visuelle og tekstlege vokabular, til dette treng dei gode kultur- og kunstmøte som gir dei den breidda og kvaliteten dei fortener.

Mange av applikasjonane på marknaden er så spekka av interaktive oppgåver, intens bruk av visuelle og auditive effektar at dei ikkje gir plass til barnet sin eigen refleksjon. Førskulelærar og pedagog Cathrine Darre i Myrertoppen barnehage, skriv at pedagogane ønskjer at digitale verktøy skal brukast til noko meir enn å fortelle barna kva dei trenger å vite.  Ho ønskjer seg bort frå applikasjonar der ein skal nå ein masse førehandsdefinerte læringsmål. Gjennom faktaprega læring automatiseras lett kunnskap, utan at vi veit om ungane lærer noko. Ein bør i staden bruke teknologien til å skape engasjement hjå ungar ved at dei får bidrar sjølv, då vil dei lære meir.

Den digitale bildeboka har på få år blitt ein stor suksess, og sjølv om den ofte tek utganspunkt i gode forteljingar er kvaliteten på det ein finn i Appstore og Googleplay svært varierande. Søker ein vilkårleg etter barnebok, er det fyrst og fremst dei kommersielle bildebøkene som dukkar opp. Særleg er det mange bildebokapplikasjonar som baserer seg på kjente eventyr og forteljingar der opphavsretten er gått ut.

Ein får også inntrykk av at bildebokapplikasjons-utviklarane berre har eit minimum av visuell kompetanse. Dei spelar difor på klisjéaktig uttrykk, sjablongar, fattige bilde og flate karakterar. Ofte brukar dei det som er av effektar, og resultatet blir heftig sansestimmulering som endar opp i bråk og kaos - utan innhald. Det er med andre ord mykje ræl. skriv Atle Berg at kritikken av norske barnebokappar nesten er fråverande. Sjølv om dei fleste forlaga no har komme i gong med å lage og tilby bildebokapplikasjonar, og salet av desse har steget i været, manglar me kriterium, og omgrep for å kjenne att kvalitet. Sjølv om omfanget av bokapplikasjonar aukar, salet og bruken aukar, blir det ikkje møtt med tilsvarande engasjement frå litteraturkritikken. Ofte blir dei berre omtala med ein kort presentasjon, ofte med fokus på funksjonslyte og eit terningkast. Kritikken tar ikkje opp den litterære eller visuelle kvalliteten i bildebokapplikasjonen.  

Slik det er no, er det barnet sjølv som sit med oppgåva, ofte åleine utan nokon å drøfte dei ulike kulturuttrykka og kunstopplevingane med. Me treng diskusjon og me treng engasjement frå vaksne, frå vitarar innanfor dei ulike fagfelta. Me treng pedagogar som har ambisjonar for barns møte med applikasjonar utover å opp øve digital kompetanse.

Me treng vaksne som tek mediet på alvor.

Barnet sin nærmaste vaksne må få ein forbrukarguid, vokabular og kriterium å støtte seg på for å finne fram til gode kunstmøte og kulturuttrykk på nettbrett. 

Mest lest