Estimating under-reporting of teachers’ use of physical violence in schools in Nyarugusu Refugee Camp

How to cite

Fabbri, Camilla; Devries, Karen. 2020. "Estimating under-reporting of teachers’ use of physical violence in schools in Nyarugusu Refugee Camp." AEA RCT Registry. AEARCTR-0005312. https://doi.org/10.1257/rct.5312-1.0

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

Camilla Fabbri* , Karen Devries

* Primary investigator

Registry ID
AEARCTR-0005312
DOI
https://doi.org/10.1257/rct.5312-1.0
First registered
Last updated
Status
In development
Trial dates
2020-01-23 – 2021-06-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/5312

Abstract

Schools in refugee settings do not only provide children with opportunities to learn and play, but also represent a safe environment to recover from trauma and to re-built a sense of normality routine that allow children and adolescents to find purpose and motivation. However, schools are also a place where children are likely to experience violence; this may be even more likely in emergency settings. In Nyarugusu refugee camp in Tanzania reports of violence in schools perpetrated by teachers and peers are common, however official estimates do not exist. The present study will employ a double list experiment to produce unbiased estimates of teacher's use of phyisical violence against students in primary and secondary schools in Nyarugusu refugee camp. The estimates from the list experiment will be compared to answers provided by teachers to direct questions about use of physical violence to assess the extent of of underreporting and therefore the extent ot social desirability bias. Further sub-group analyses will be conducted to assess whether use of physical violence and underreporting is higher among teachers with specific characteristics.

Intervention

The sample includes around 700 teachers from 27 schools in Nyarugusu refugee camp. We will use a double list experiment which represent an extension of the traditional list experiment (or item list technique) which uses two sets of lists such that all respondents receive one control list and one treatment list. All participating teachers will received 2 lists (list A and list B), each containing a set of non-sensitive statements. Individuals will not be asked to to say whether they agree with each of the statements but only with how many they agree. A sensitive item will elicit use of physical violence against children: the same sensitive item will be used in the two different lists of non-sensitive items.Teachers will be individually randomised into two groups (group 1 and group 2): group 1 will receive the sensitive item in list A (and will act as control group for list B) and group 2 will received the sensitive item in list B (and will act as control group for list A). In the double list experiment each group of teachers serves sequentially as treated and control group or vice versa.
All teachers will be also asked a direct question about their use of physical violence in a subsequent section of the survey therefore use of violence will be elicited directly and indirectly in the same survey.

Experimental design

see above Intervention detail

Design details

Randomization method
Randomization by computer (Stata).
Randomization unit
Individual teachers.
Clusters
n/a
Observations
700 teachers
Arms
350 teachers group 1, 350 teachers group 2
Primary outcomes
The outcome is a measured as a count of the number of items with which each respondent agrees in each list (it will not be known which items make up for this count for any individual).

Cited by

No citation data available for this registration yet.

Keywords