Digitalization of public sector organizations over time: The applicability of quantitative text analysis
| Published date | 01 June 2024 |
| DOI | http://doi.org/10.1177/00208523231183569 |
| Author | Karin Fossheim,Jonas Lund-Tønnesen |
| Date | 01 June 2024 |
| Subject Matter | Articles |
Digitalization of public
sector organizations over
time: The applicability of
quantitative text analysis
Karin Fossheim
Institute of Transport Economics and University of Oslo,
Norway
Jonas Lund-Tønnesen
University of Oslo, Norway
Abstract
In recent decades, public organizations have undergone significant changes related to
digitalization. These changes are the result of multiple, varying influences, such as exter-
nal institutional feedback. The issue of digitalization feedback development in public
organizations opens the discussion on how to study digitalization over time. In this art-
icle, we consider whether the method of quantitative text analysis can be used to
observe the same administrative trends in digitalization as found using other methods
and data sources in existing research. After conducting a co-occurrence analysis of
documents from the Norwegian Customs Agency (NCA) over a period of 20 years,
we found that the NCA’s digitalization-related language changed in this period and
that user feedback was the most frequently emphasized kind of external feedback.
These observations are consistent with the literature, which shows that the use of
co-occurrence methodology to study public administration has a positive role in future
research.
Points for practitioners
•Quantitative text analysis is an effective method to observe administrative trends.
•Findings of the study corroborate with previous research showing that current
administrative tr ends emphasize coll aboration between public organi zations and
Corresponding author:
Karin Fossheim, Institute of Transport Economics and University of Oslo, Gaustadalléen 21, 0349 Oslo,
Norway.
Email: karin.fossheim@psykologi.uio.no
Article
International
Review of
Administrative
Sciences
International Review of Administrative
Sciences
2024, Vol. 90(2) 318–335
© The Author(s) 2023
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/00208523231183569
journals.sagepub.com/home/ras
citizen participation as central development features of digitalization of public
organizations.
•The co-occurrence analysis has some limitations and needs to be further improved to
develop theory and identify new areas of public administration digitalization.
Keywords
digitalization, feedback, public organization, quantitative text analysis
Introduction
In recent decades, public organizations have undergone significant changes related to
digitalization (M argetts and Dunlea vy, 2013). Even histor ically rule-bound o rganiza-
tions, such as custo ms agencies, have beco me more user and custom er friendly and
have increased their use of digital technology, which are features not us ually associated
with these organizations (Grainger, 2008). Such changes in public organizations come
in many different forms and are the result of many different factors outside of organiza-
tions (Greve et al., 2020). One of these factors is external institutional influence
(Lember et al., 2018 ), which tends to chang e and develop over time. To observe
changes and external influences, the common appro ach has been to conduct int erviews
and surveys with public managers or to qualitatively analyze key documents. Although
these methods are effective, in practical terms, they can be highly time consuming,
expensive, and difficu lt when dealing with l arge volumes of docum ents (Grimmer
and Stewart, 2013).
Quantitative text analysis is a method that can help public administration researchers
overcome such practical problems. This method has been increasingly employed in the
social sciences (Brady, 2019; Krippendorff, 2018; Lazer and Radford, 2017; Molina
and Garip, 2019) because statistical software for conducting such analyses has become
more accessible and much less expensive (Grimmer and Stewart, 2013). In addition,
quantitative text analysis allows for systematizing the content and patterns of great quan-
tities of text and documents. In political science and public administration, written words
have great significance. Although close reading and careful consideration of text cannot
be substituted by automated computer analysis, an examination of the prevalence and
co-occurrence of words and concepts can help us gain insight into how digitalization
topics change over time, and how public organizations respond to external feedback
and stakeholders’input. In particular, it is worthwhile to test whether this kind of analysis
can identify the same administrative trends and external institutional influences related to
digitalization as previous research has in the same context using different methods and
data (e.g. Christensen and Lægreid, 2022; Greve et al., 2016, 2020; Lægreid and
Rykkja, 2016). This would help us understand the validity and applicability of the
method. Accordingly, we posed the following research questions:
Fossheim and Lund-Tønnesen 319
To continue reading
Request your trial