Managing Digital Technologies: Evidence from a Field Experiment 3

How to cite

Kneller, Richard; Li, Cher; Adem, Anwar. 2022. "Managing Digital Technologies: Evidence from a Field Experiment 3." AEA RCT Registry. AEARCTR-0009702. https://doi.org/10.1257/rct.9702-1.0

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

Richard Kneller* , Cher Li , Anwar Adem

* Primary investigator

Registry ID
AEARCTR-0009702
DOI
https://doi.org/10.1257/rct.9702-1.0
First registered
Last updated
Status
In development
Trial dates
2022-08-01 – 2024-03-31
Countries
United Kingdom of Great Britain and Northern Ireland ()
JEL codes
D22, D24, D83, L25, L66
Registry record
http://www.socialscienceregistry.org/trials/9702

Abstract

The question how firms improve their productivity is of paramount policy im-
portance for every nation. To harness the benefits from the recent acceleration
of investment in digital technologies, firms must put in place appropriate man-
agement practices. In this paper, we aim to use a field experiment among
UK firms to causally identify the effect of information provision on the perfor-
mance management of a key digital asset - business websites - and the software
technologies used to build them. Specifically, we investigate the importance of
information gap in hindering technology adoption, performance and manage-
ment of digital technologies. For this end, we provide a randomly selected firm
with an information about own website performance, industry common and
best practice together with inputs of digital management and monitoring tools.
After that we assess if these change the adoption of web-based technologies,
and website performance. The information provision or encouragement design
will be carried out to randomly selected firms in the retail industry.

Intervention

The intervention is an encouragement design, which will provide firms with infor-
mation on their website performance, common and best practice within the indus-
try. The information will be framed to be consistent with best practice performance
management processes for this technology. The information will be sent using mail
and/or email addresses of firms.

Experimental design

This is a four-arm block or Stratified randomization design. Within each block,
firms are randomly assigned to one of two arms:
1. Control group [C]
2. Benchmark information treated group[T].

Arm-blocks will based on previous adoption of analytics software in the baseline period, and non-adoption, plus the above/below average performance of website performance measures in the baseline.

Design details

Randomization method
We used Stata’s randtreat command to carry out the randomization
Randomization unit
Random assignment occurs at the firm-level
Clusters
7089 firms
Observations
7089 firms
Arms
4492 firms using analytics software in baseline; 2597 firms not using analytics in baseline. Of the 4492 analytics users in baseline; 2246 above average performance in baseline and 2246 below average. Treatment arms 1123 treated and 1123 control firms for above/below average analytics users in baseline. Of the 2597 analytics users in baseline; 1298.5 above average performance in baseline and 1298.5 below average. Treatment arms 649 treated and 649 control firms for above/below average analytics non-users in baseline.
Primary outcomes
Our primary outcome will be indicators for website performance
Secondary outcomes
Secondary outcomes include the adoption and retention of website softwares

Cited by

No citation data available for this registration yet.

Keywords