Children’s early childhood experiences are crucial for their positive development and educational success. These experiences vary largly across families and thus a child’s social background is strongly linked to its educational pathway (Burger & Walk, 2016) – especially for children from psychosocially disadvantaged families. Their risk is increased to experience high familiar stress and without balancing ressources this is highly likely to impact their development and education (Heilig, 2014). Therefore, programs which aim at strengthing children’s early family environment – like home-based early childhood interventions – might be one path to increase the educational opportunities of children at risk. The bio-psychosocial approach describes children’s development as a reciprocal process of their preconditions, parental behaviour and learning opportunities (Sameroff, 2010). Hence, changes in parental behaviour are considered to induce change within the child as well as in the reverse direction (Bronfenbrenner & Morris, 2006). Home-based programms are based on this reasoning. Results from several studies indicate that such parenting interventions have a positive effect for several developmental child characteristics (e.g. early child cognitive, language, motor, socioemotional development) as well as for parenting skills (e.g. parenting knowledge, parenting practices, parent–child interactions; Jeong et al., 2021). However, very few studies to date have examined the effectiveness of early intervention over longer term. Studying long-term effects is a meaningful step in shedding light on the role of early intervention in improving children's educational opportunities. 

The longitudinal RCT study ZEPPELIN investigates the effectiveness of the «Parents as Teachers (PAT)» program in the urban area surrounding Zurich, Switzerland. PAT was developed in Missouri (USA) in 1981 and follows a systemic approach designed to strengthen the family as a learning environment. The program was adapted in Germany in 2004 (Sindbert, 2010) and implemented in the German-speaking part of Switzerland in 2011 (Lanfranchi & Neuhauser, 2011). Previous results from ZEPPELIN 0 – 3 showed that PAT at the age of three years improved children’s adaptive behavior and language skills. Problem behavior was reduced in families with the highest psychosocial risk (Schaub et al., 2019). Furthermore, results from ZEPPELIN 5 – 8 showed that the early intervention was two years later still effective: Children from PAT scored higher in language skills, self-regulation, and showed less behaviour problems (Schaub et al., 2021). The aim of this study is to investigate whether these early effects continue to last into early school age and thus provide a first insight into the mid-term effectiveness of the program in psychosocially disadvantaged families. To examine this research question, we used data from the ZEPPELIN Follow-up Study 5-8.

The ZEPPELIN sample was recruited in 2011 and 2012 at three project sites in the canton Zürich, Switzerland. To meet the inclusion criteria, at least two risk factors had to be present on the personal (e.g., mental disorder), the family (e.g., single parent), the social (e.g., no social network), or the material level (e.g., restricted living conditions). The final sample were 248 families (261 children including twins). These families were allocated using a stratified block randomization to the intervention group (IG, n = 132 families, 139 children) and the control group (CG, n = 116 families, 122 children). During the first three years of their child’s life the IG received support by a trained PAT-Trainer: monthly home visits and group connections. The families from the control group had access to regular community services but were not supported by PAT. During the program period and starting with the child’s first year in kindergarten, child and family characteristics were collected annually up to the current time point. In the 1st year of primary school (t7) 165 children (IG, n = 83; CG, n = 82) participated. Following competencies were assessed: Writing (SLRT, Moll & Landerl, 2010), reading (WLLP-R, Schneider, Blanke, Faust & Küspert, 2011), mathematic (MBK-1+, Ennemoser, Krajewski & Sinner, 2010) and nonverbal intelligence (CFT 1-R; Weiss & Osterland, 2013). Additionally, teachers were asked to report on children’s problem behaviour (SDQ, Koglin, Barquero, Mayer, Scheithauer, & Petermann, 2007), motivation (Renzulli et al., 2010), participation (PIQ; Venetz, Zurbriggen, Eckhart, Schwab, & Hessels, 2015) and child-teacher relationship (STRS; Pianta, 2001). Furthermore, we collected data from the parents regarding children’s problem behaviour (SDQ, Koglin, Barquero, Mayer, Scheithauer, & Petermann, 2007). All multiple regression analyses were conducted in R. For standardized psychometric test the standard values were used (MBK-1+, WLLP-R, CFT 1-R), for writing (SLRT) and for all non-standardized instruments, the reached score was used. In all analyses the following variables were considered: project location, single parent, sex, proportion of years in Switzerland, no post-compulsory education, german skills of the mother, ISEI, maternal sensitivity as well as psychosocial stress on the family. The children’s age and months since school enrollment were included for examining the effect on writing. To reduce bias due to sample fade-out we used multiple imputed data additionally to the manifest data (Little, 1988). Predicted Mean Matching was used to estimate with 40 Iterations 100 data sets.

Preliminary results from regression analysis with manifest and imputed data showed marginally significant positive effects on children’s mathematical competencies (on imputed data) and significantly less behavior problems regarding hyperactivity in the first grade. No significant group differences were found regarding language skills (reading, writing), nonverbal intelligence, other problem behaviour, motivation, participation, or child-teacher relationship. However, children in the IG showed slightly higher scores in most investigated achievement characteristics. In addition, we are planning to investigate which subgroups benefit the most from PAT. It may be that the effects are more pronounced among families with high levels of stress. Nevertheless, this first examination of PAT effects in children’s school age revealed fewer effects as expected. One reason for the lack of significance could be the low statistical power of the study in the first grade. However, it is noticeable that the positive effects on language skills cannot be replicated for early school age. This slightly fading out effect is in line with previous studies (e.g., Neuhauser, 2014). One possible explanation is that the child’s learning environment at home and at school after the intervention can enhance, limit, or neutralize the effects of programs in the early childhood (Reynolds, Magnuson, & Ou, 2010). Investigations in the following school years must show if this is just a temporary effect or a continuous fading out. Furthermore, contextual effects (school and classroom environment, student-teacher relationship) might be promising to investigate. Especially this transition phase from kindergarten to primary school might be sensible to various familiar and context characteristics (Dockett & Perry, 2004). Complementary programs through school age may be needed to achieve long-term impact. However, the question arises how to reach and effectively support families up to school age.

Burger, K., & Walk, M. (2016). Can children break the cycle of disadvantage? Structure and agency in the transmission of education across generations. Social Psychology of Education, 19(4), 695–713. Bronfenbrenner, U., & Morris, P. A. (2006). The bioecological model of humandevelopment. In R. M. Lerner, W. Damon, R. M. Lerner, & W. Damon (Eds.), Handbook of child psychology: Theoretical models of human development (Vol. 1) (6th ed., Vol. 1, pp. 793–828). Hoboken, NJ: John Wiley & Sons Inc. Dockett, S., & Perry, B. (2004). Starting school: Perspectives of Australian children, parents and educators. Journal of Early Childhood Research, 2(2), 171-189. Heilig, L. (2014). Risikokonstellationen in der frühen Kindheit: Auswirkungen biologischer und psychologischer Vulnerabilitäten sowie psychosozialer Stressoren auf kindliche Entwicklungsverläufe. Zeitschrift für Erziehungswissenschaft, 17(S2), 263–280. Lanfranchi, A., & Neuhauser, A. (2011). ZEPPELIN 0–3 – Förderung ab Geburt mit „PAT – Mit Eltern lernen“. Sonderpädagogische Förderung heute, 56(4), 437–442. Little, R. J. A. (1988). Missing-data adjustments in large surveys. Journal of Business & Economic Statistics, 6(3), 287–296. Neuhauser, A. (2014). A closer look at the effectiveness of early childhood education in at-risk families. Mental Health and Prevention, 2(3–4), 43–57. Reynolds, A. J., Magnuson, K. A., & Ou, S.-R. (2010). Preschool-to-third grade programs and practices: A review of research. Children & Youth Services Review, 32(8), 1121–1131. Sameroff, A. J. (2010). A unified theory of development: A dialectic integration of nature and nurture. Child Development, 81(1), 6–22. Schaub, S., Ramseier, E., Neuhauser, A., Burkhardt, S. C. A., & Lanfranchi, A. (2019). Effects of home-based early intervention on child outcomes: A randomized controlled trial of Parents as Teachers in Switzerland. Early Childhood Research Quarterly, 48, 173–185. Schaub, S., Eberli, R., Ramseier, E., Neuhauser, A. & Lanfranchi, A. (2021). Förderung ab Geburt mit dem Programm «PAT – Mit Eltern Lernen»: Effekte im ersten Kindergartenjahr. Schweizerische Zeitschrift für Bildungswissenschaft, 43(2), 285–296. Sindbert, R. (2010). PAT – Mit Eltern lernen. Bessere Bildungschancen für Kinder aus sozial benachteiligten Familien durch frühe Förderung und Elternempowerment. In C. Leyendecker (Eds.), Gefährdete Kindheit. Risiken früh erkennen, Ressourcen früh fördern (p. 322–349). Kohlhammer. Jeong, J., Franchett, E. E., Ramos de Oliveira, C. V., Rehmani, K., & Yousafzai, A. K. (2021). Parenting interventions to promote early child development in the first three years of life: A global systematic review and meta-analysis. PLoS medicine, 18(5).