Safety culture and behavioural insights: Lessons from registered electrical contractors and registered gas installers in Ireland

How to cite

Shephard, Daniel; Cavassini, Filippo. 2019. "Safety culture and behavioural insights: Lessons from registered electrical contractors and registered gas installers in Ireland." AEA RCT Registry. AEARCTR-0003796. https://doi.org/10.1257/rct.3796-1.0

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

Daniel Shephard* , Filippo Cavassini

* Primary investigator

Registry ID
AEARCTR-0003796
DOI
https://doi.org/10.1257/rct.3796-1.0
First registered
Last updated
Status
On going
Trial dates
2018-12-03 – 2019-04-30
Countries
Private
JEL codes
D22, D29, J28, D91
Registry record
http://www.socialscienceregistry.org/trials/3796

Abstract

This study uses a randomized survey experiment to test to effects of using behavioural insights to frame safety processes and procedures among registered electrical contractors and registered gas installers in Ireland. The survey experiment examines (1) the salience of safety culture as associated with different entities in the regulatory regime, (2) the impact of personalisation, primacy, and implementation intentions on the predicted efficacy of (a) inspection processes and (b) notices; and (3) the effect of using different messengers to deliver safety messages.

Intervention

The survey experiment examines (1) the salience of safety culture as associated with different entities in the regulatory regime, (2) the impact of personalisation, primacy, and implementation intentions on the predicted efficacy of (a) inspection processes and (b) notices; and (3) the effect of using different messengers to deliver safety messages.

Experimental design

The study used stratified random sampling.

Design details

Randomization method
Randomization was done within each starta using the randomization algorithm of Qualtrics (computerized randomization).
Randomization unit
Individual respondents were randomly allocated.
Clusters
300
Observations
300
Arms
Arm 1: 110 Arm 2: 110 Arm 3: 80
Primary outcomes
The primary outcomes for each of the tests listed above are as follows: 1) The proportion of terms related to safety. 2a-2b) - The mean percentage of entities that the respondents expect would resolve (i) red non-conformances; (ii) amber non-conformances; (iii) yellow non-conformances; and (iv) [primary outcome] the mean of the the three. - The mean percentage of time that the respondents expect they themselves would resolve (i) red non-conformances; (ii) amber non-conformances; (iii) yellow non-conformances; and (iv) [primary outcome] the mean of the the three. 2b only) The mean percentage of entities that respondents think would read the notice if (i) emailed and if (ii) posted. 3) The mean percentage of entities that would (i) read and (ii) implement the notice respectively. And the mean percentage of time that the respondent themselves would (iii) read and (iv) implement the suggestion.
Secondary outcomes
Secondary outcomes will include any differences between the predicted responses of the respondents and their predictions about their peers. In addition, the stratification variable will be leveraged to investigate subgroup differences (for example are results different for electrical contractors and gas installers).

Cited by

No citation data available for this registration yet.

Keywords