The Effects of Mobile-Learning on Language and Integration outcomes: Evidence From a Randomized Experiment in Integration Courses in Germany

How to cite

Kassaballi, Zouhier. 2018. "The Effects of Mobile-Learning on Language and Integration outcomes: Evidence From a Randomized Experiment in Integration Courses in Germany." AEA RCT Registry. AEARCTR-0003524. https://doi.org/10.1257/rct.3524-5.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Zouhier Kassaballi*

* Primary investigator

Registry ID
AEARCTR-0003524
DOI
https://doi.org/10.1257/rct.3524-5.0
First registered
Last updated
Status
In development
Trial dates
2019-03-01 – 2021-02-01
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/3524

Abstract

It is generally agreed upon that the most vital step towards integrating into a new culture is learning the spoken language. The language barrier issue could delay social integration and productivity outcomes. The use of mobile-assisted language learning applications along with integration course attendance may improve, through several channels, the language-learning and social integration outcomes among immigrants in Germany. We will conduct an RCT among migrants and refugees, in order to estimate the effect of encouraging the use of the German language-learning and similar applications on the study population´s German skills and social integration outcomes. We tend to focus on the application "Ankommen" since it includes several social integration topics as well as the fact that it has been developed by the German federal office for migration and refugees. We hypothesis that using the help of mobile learning to get to know important cultural aspects would increase social integration efforts and improve language level outcomes.

Intervention

In order for the study to be policy relevant, the intervention will be encouraging the use of existing and free of charge German language learning applications.

The sample will consist of 2 equally distributed groups: 1 treatment and 1 control group.

The intervention will include:
1) A video promoting highly-rated and free German learning apps, we will tend to focus on the application “Ankommen” in specific. The video will explain how the application "Ankommen" works, why it is useful and how to download it. The video also includes the experiences of migrants with MALL and it's advantages in general.
2) Flyers will be handed out in order to induce constant usage and provide traceable download links for language learning apps.

Experimental design

The aim is to estimate the effect of encouraging the use of Mobile-assisted language learning on immigrants' and refugees' German language skills. A randomized post-experimental research design will be used, with a treatment group and a control group. The treatment being in its nature non-excludable, we will have to use randomized promotion experiments, which will take place in integration Institutions in Germany. There will be no pre-test since we are in a suitable position to assume that on average the language level of students starting an A1-level or A2-level integration course is the same, moreover, students that want to start an integration course are subjected to an obligatory placement test which identifies the student’s language level.

A typical integration school offers one A1-level/A2-level course every 2 months, clustering will occur on a class level within schools. 4 classes from each participating school will be randomly assigned to treatment or control. Our sample will include about 60 schools offering integration courses.

We will employ a matched-pair cluster-randomized design, assigning treatment to half of the study population and clustering on a class level. We will use optimal Mahalanobis matching. Covariates to be matched on are the size of the school, the maximum German level offered by the school, size of the city and socioeconomic status of the neighborhood the school is located. Within each pair, treatment is randomly assigned for every treatment phase. All A1 and A2-level students of the courses are invited to participate.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
Cluster on a class level.
Clusters
The study is anticipating around 30 clusters (Integration Classes).
Observations
Around 400 students are planned to participate in the study.
Arms
15 classes in the control group 15 classes in the treatment group
Primary outcomes
1) Self-reported German language skills 2) Self-reported social integration levels

Cited by

No citation data available for this registration yet.

Keywords