Employment, Social Affairs & Inclusion

Nuffield Early Language Intervention

Evidence level:
 
Evidence of effectiveness:
? - 0 + ++
Transferability:
? - 0 + ++
Enduring impact:
? - 0 +

Rationale  of practice

The Nuffield Early Language Intervention (NELI) is an intensive language support programme delivered by teaching assistants and designed to improve the spoken language ability of young children with relatively poor spoken language skills. The sessions focus on improving children’s vocabulary, listening, narrative skills and phonological awareness.

Practice Category

  • Helping Vulnerable Children

Recommendation pillar

  • Improve education systems’ impact on equal opportunities

Countries that have implemented practice

  • United Kingdom

Age groups

  • Young Children (age 0 to 5)

Target groups

  • Children

Years in operation

  • 2004 - Still operating

Scope of practice

  • Local level

Type of organisation implementing practice

  • Private Education Organization

The intervention was developed by researchers from the University of York, with funding from the Nuffield Foundation. The 20-week reception programme, which includes training provided by Elklan, is available for purchase from Oxford University Press.

Mode of delivery

  • Individual sessions,
  • Group sessions,
  • Face-to-face

Both the 30-week and the 20-week versions of the programme are delivered face-to-face by teaching assistants. Both versions include group sessions (two to four children); in reception, children also attend two individual sessions each week.

Delivery dosage

  • Frequency: Between 2-6 times a week
  • Duration: Less than one hour

Both the 30-week and the 20-week versions of the programme are delivered in 10-week blocks. Each week, teaching assistants hold three group sessions with two to four children, lasting 20 minutes each for children in nursery and 30 minutes each for children in reception. In reception, children also attend two 15-minute individual sessions each week.

Location of practice

  • Kindergarten-based

Two versions of the programme are included in this evaluation: a 30-week programme delivered in three ten-week blocks starting in the final term of nursery and continuing into the first two terms of reception in primary school, and a 20-week programme delivered in two ten-week blocks only in reception year.

Evidence of effectiveness

  • Practice has been evaluated

The evaluation, which was conducted in England, used a randomised controlled design with individual pupils within nurseries randomly assigned to treatment and control groups. The randomisation procedure minimised the difference between treatment and control groups in age, gender and language scores. There were 350 pupils in the trial: 114 pupils in the 30-week treatment group, 121 in the 20-week treatment group, and 115 in the control group. The overall attrition was 11%. There were significant and positive effects of both the 30- and 20-week intervention on the primary outcome, language (at least p < .10), though not on the secondary outcome, word-level literacy score. The effect on language increased six months after the intervention (at least p < .05).

Transferability

  • Practice has been evaluated in multiple populations

In addition to the evaluation reported here, a pilot study conducted by the programme developers took place in 19 preschool settings England in 2009-2010 (Fricke et al., 2013). The pilot study evaluated an earlier version of the 30-week programme, which was itself a modified version of a programme that had been tested, also in England, in 2004-2005 (Bowyer-Crane et al., 2008). However, the previously-evaluated versions of the programme differ from the version reported in this evaluation, and due to issues of comparability these earlier evaluations are not assessed as part of this review. The EEF is currently funding an evaluation of the 20-week programme. The programme has not been implemented outside of England.

While for this evaluation we have given a transferability score of “?”, for the overall programme we have taken into consideration that there have been previous evaluations of versions of the intervention that are similar, and that there is an ongoing evaluation of the 20-week programme, and thus given an overall programme score of “+”.

Cost information

  • Implementation cost information is available

The commercial cost of a resource pack and training for the 20-week programme is £1,040. At approximately £70 per pupil per year, the evaluation rated NELI’s cost as very low (on the EEF’s relative 5-point scale ranging from very low to very high). The evaluators considered the costs related to training, materials and the time needed for teaching assistants to deliver the programme, but did not include any further staff time or disruption costs.

Practice Materials

  • Practice materials available online

The 20-week reception programme is available for purchase from Oxford University Press (https://global.oup.com/education/content/primary/series/nuffield-intervention/?region=international). The programme cost includes training which is provided by Elklan. The commercial cost of a resource pack and training is £1,040. The 30-week nursery and reception programme is currently not available for purchase.

Enduring impact

  • Follow-up has been conducted

Follow-up tests were conducted about 6 months after the intervention phase, which found that the impact of both the 20-week and 30-week programmes may have increased over time. However, the evaluators were cautious about this finding as not enough is known about what other activities the children were involved in during the six-month follow-up period. Still, please note that to be considered for the enduring impact evaluation criteria of EPIC the practice needs to have at a minimum a two-year follow-up which identifies significant positive results that are maintained through time. 

At follow-up, there were still no statistically significant differences between treatment and control groups in terms of the secondary outcome, word-level literacy scores.

Evaluation 

Reference:  

Sibieta, L., Kotecha, M., and Skipp, A. (2016). Nuffield Early Language Intervention Evaluation report and executive summary, Education Endowment Foundation.

The evaluation examined two versions of the NELI programme: a 30-week programme delivered in three ten-week blocks starting in the final term of nursery and continuing into the first two terms of reception in primary school, and a 20-week programme delivered in two ten-week blocks only in reception year.

The study used a randomised controlled trial design, with participants randomised at the pupil level. The evaluators used a minimisation procedure in randomisation, or order to minimise the difference between treatment and control groups in age, gender and language scores. Although not explicitly planned for, groups were also broadly similar in terms of other characteristics, including speaking English as an additional language and speech and language difficulties, and ethnicity. The only exceptions were that compared to the control group, the 20-week intervention group appeared to be more deprived, in that a higher proportion of pupils were eligible for free school meals, and that fewer pupils had special educational needs or were of White-British ethnicity; however, none of these differences were statistically significant.

The primary outcome for the evaluation was a language skills score, which comprised a composite score of four validated measures of language skill, namely the Renfrew Action Picture Test (two components; information and grammar), the CELF-Preschool 2 UK: Expressive Vocabulary and Listening Comprehension (based on the York Assessment of Reading Comprehension test, YARC).The secondary outcome was word-level literacy, which comprised a composite of YARC: Letter Sound Knowledge, YARC: Early Word Reading and Spelling. All tests were carried out by researchers who were blind to the allocation of children to groups.

The tests were carried out at three time points. Children received baseline tests in April 2013, before the start of the intervention phase. The 30-week intervention began in April 2013 and the 20-week intervention began in September 2013. Children’s outcomes were tested between May and July 2014 after the end of the intervention phase. Follow-up tests were conducted about 6 months after the intervention phase, between October and December 2014.

At the end of the intervention phase, the evaluators estimated the impact of the programme using Fully Interactive Linear Matching (FILM) to account for differences at baseline in pupil characteristics and pre-test scores. They found that the 30-week intervention had an effect size of 0.27 (p < .001) while the 20-week intervention had an effect size of 0.16 (p < .10) on the primary outcome, language skills. The impact on the secondary outcome, word-level literacy skills, was not significant.

Six months after the intervention, the post-tests found that the differences between the treatment and control groups on the language composite score had actually increased, compared to the immediate post-intervention scores. The effect size for the 30-week treatment group increased from 0.27 to 0.37 standard deviations (p < .01), while for the 20-week treatment group the difference increased from 0.16 to 0.21 standard deviations (p < .05).

Please note that there are two other evaluations which were conducted in European countries (England) but which evaluated earlier, slightly different, versions of the intervention. Due to issues of comparability these earlier evaluations are not assessed as part of this review. We list these in the Bibliography below.

Summary of Results for Evaluation 1

Table 1: Impact of Nuffield Early Learning Intervention on language and literacy skills

 

Raw means

Effect size

Outcome

Intervention group

Control group

 

 

n

(missing)

Mean

(95% CI)

n

(missing)

Mean

(95% CI)

n in model (treatment or control)el

Estimated Impact (95% CI)

p-value

Language skills (primary outcome)

30-week treatment

114 (3)

0.11

(-0.07; 0.29)

115 (4)

-0.17

(-0.35; 0.01)

229

(114, 115)

0.267***

( 0.073; 0.461)

0.007

20-week treatment

121 (3)

0.05

(-0.13; 0.24)

115 (4)

-0.17

(-0.35; 0.01)

236

(121, 115)

0.161*

(-0.016; 0.339)

0.075

Word-level literacy skills (secondary outcome)

30-week treatment

114 (3)

-0.01

(-0.17; 0.14)

115 (4)

-0.09

(-0.29; 0.10)

229

(114,115)

0.062

(-0.159; 0.283)

0.582

20-week treatment

121 (3)

0.10

(-0.10; 0.30)

115 (4)

-0.09

(-0.29; 0.10)

236

(121, 115)

0.127

(-0.082; 0.337)

0.234

Note: * indicates that the treatment effect is significant at the 10% level ** at the 5% level *** at the 1% level. Impact estimates are calculated using Fully Interacted Linear Matching (FIM). Covariates included are: age; gender; EAL status; known speech and language difficulty status; standardised language composite baseline score; standardised baseline scores for components of language composite; standardised baseline scores for CELF sentence structure, vocabulary naming, vocabulary definitions and the British picture vocabulary scale.

 

 

Table 14: Follow-up test impact estimates

30-week treatment

20-week treatment

 

Follow-up language composite score (primary outcome)

Sample

Treatment effect

95% Confidence Interval

n

Treatment effect

95% Confidence Interval

n

All pupils

0.367***

( 0.170; 0.563)

229

0.211**

( 0.015; 0.407)

236

Pupils with observed NPD records including NPD covariates

0.464***

( 0.189; 0.739)

156

0.265*

(-0.008; 0.538)

159

Follow-up word-level literacy composite score (secondary outcome)

Sample

Treatment effect

95% Confidence Interval

n

Treatment effect

95% Confidence Interval

n

All pupils

0.054

(-0.259; 0.366)

229

0.066

(-0.151; 0.283)

236

Pupils with observed NPD records including NPD covariates

-0.05

(-0.332; 0.233)

156

-0.012

(-0.425; 0.401)

159

Note: * indicates that the treatment effect is significant at the 10% level ** at the 5% level *** at the 1% level.

Covariates included are: age; gender; EAL status; known speech and language difficulty status; standardised language composite baseline score; standardised baseline scores for components of language composite; standardised baseline scores for CELF sentence structure, vocabulary naming, vocabulary definitions and the British picture vocabulary scale.

Additional NPD covariates are: eligibility for FSM when in reception; ethnicity; whether the pupil has ever received SEN support; and IDACI rank percentile.

NOTE: Both tables above were taken directly from Sibieta, Kotecha and Skipp (2016) page 29 and 33.

Issues to consider

The overall attrition was 11%, due to three schools that withdrew from the intervention and 27 pupils in participating schools that dropped out because they changed schools between nursery and reception. Partial outcome data was recorded from the schools that dropped out from the trial, though the impact estimates are largely unchanged whether these schools are included or excluded.

The evaluators also highlighted some common factors that threatened the fidelity of the programme, including a lack of a discrete space where group and individual sessions could be delivered; failure to protect a teaching assistant’s time to deliver the programme, and a lack of support for the programme amongst wider school teaching staff.

Available resources

The Education Endowment Foundation’s webpage on NELI:

https://educationendowmentfoundation.org.uk/projects-and-evaluation/projects/nuffield-early-language-intervention

The Nuffield Foundation’s webpage on NELI:

http://www.nuffieldfoundation.org/nuffield-early-language-intervention

The Oxford University Press webpage on NELI, which provides more information on how to purchase the resources:

https://global.oup.com/education/content/primary/series/nuffield-intervention/?region=international

Bibliography

Sibieta, L., Kotecha, M., and Skipp, A. (2016). Nuffield Early Language Intervention Evaluation report and executive summary, Education Endowment Foundation.

Fricke, S., Bowyer-Crane, C., Haley, A. J., Hulme, C. and Snowling, M. J. (2013) ‘Efficacy of language intervention in the early years’. The Journal of Child Psychology and Psychiatry, 54: 3, 280–290

Bowyer-Crane, C., Snowling, M.J., Duff, F.J., Fieldsend, E., Carroll, J.M., Miles, J., & Hulme, C. (2008). Improving early language and literacy skills: Differential effects of an oral language versus a phonology with reading intervention. Journal of Child Psychology and Psychiatry, 49, 422–432.

Contact information

Email  epic@rand.org

Last updated

May 2018

Find a practice (within the 'Evidence based practices'):

The practices can be also searched along the three policy pillars of the Recommendation for Investing in Children:


Share this page