=Paper=
{{Paper
|id=Vol-2608/paper74
|storemode=property
|title=An informetric view on relations between global brands and research activity
|pdfUrl=https://ceur-ws.org/Vol-2608/paper74.pdf
|volume=Vol-2608
|authors=Serhiy Shtovba,Olena Shtovba
|dblpUrl=https://dblp.org/rec/conf/cmis/ShtovbaS20
}}
==An informetric view on relations between global brands and research activity==
https://ceur-ws.org/Vol-2608/paper74.pdf
An Informetric View on Relations between Global
Brands and Research Activity
Serhiy Shtovba 1, 2 [0000-0003-1302-4899], Olena Shtovba 1 [0000-0003-1418-4907]
1
Vinnytsia National Technical University, Khmelnytske Shose, 95, Vinnytsia, 21021, Ukraine
shtovba@vntu.edu.ua
olena.shtovba@yahoo.com
2
Vasyl’ Stus Donetsk National University, 600-richchia str., 21, Vinnytsia, 21021, Ukraine
s.shtovba@donnu.edu.ua
Abstract. The article identifies the relations between global brands and re-
search activity during 2009 – 2018. Two types of relations are studied: support
of the research by global brands and usage of the global brands in the research.
The research support by 27 global brands and the global brands usage in the re-
search are analysed over 2009 – 2018. The support is assessed by the number of
papers in which the global brand is mentioned in the funding section. The usage
is assessed by a number of papers in which the global brand is mentioned in the
presentation part. The most generous and simultaneously popular brands are
Google, Microsoft, IBM, Samsung and Facebook. In the last three years their
support dynamics and usage dynamics are positive or strong positive.
Keywords: informetrics, global brands, research support, brands in research,
Scopus, classification.
1 Introduction
Global brands rapidly penetrate into new spheres of influence, including research. The
level of the infiltration can be assessed by the public “reports” of the scholars – by the
scientific papers. Until now, such activity of the global brands has not been studied in
a systematic way, only some particular studies were carried. For example, a bibli-
ometric analysis of the scientific literature related to the use of Facebook in educa-
tional research is carried in [1]. The analysis is performed on Web of Science data.
Scopus-based bibliometric analysis of the papers related to the use of Facebook and of
YouTube in any research field is carried in [2] and in [3]. An evaluation of corre-
spondence between Coca-Cola’s “Transparency List” of funded researchers and a list
of papers with scientific research acknowledging funding from Coca-Cola is per-
formed in [4]. A list of the supported papers is taken from Web of Science Core Col-
lection database. Paper [5] describes the scope of partnerships between Coca-Cola
and 74 health organizations in Spain, examining marketing strategies contained in
scientific papers funded by Coca-Cola between 2010 and 2016. A list of the supported
health organizations was formed through PubMed. Papers [6], [7] and [8] study vari-
Copyright © 2020 for this paper by its authors. Use permitted under Creative
Commons License Attribution 4.0 International (CC BY 4.0).
ous conflicts of interest between donors (Coca-Cola Company, Mars Inc.) and results
of the supported research.
The impact of the global brands on the research activity is proposed to assess by
two indices. The first index is the number of papers, where the global brand is men-
tioned in the presentation part. This index roughly assesses how the global brand is
connected with the research activity. This connection can be in the form of some
brand tools, technologies, recipes, datasets etc. It is possible that the global brand
itself acts as an object of research. The second index is the number of papers, where
the global brand is mentioned in the funding research section. This index enables to
estimate roughly the financial support of the research by the global brands. Such in-
formetric approach to identifying the various brand relations is becoming more and more
popular. For example, paper [9] analyses B2B-branding papers during 1972 – 2015,
paper [10] analyses analysed publications related to researches in brand personality
during the 1995 – 2017, paper [11] identifies the leaders and the trends in branding re-
searches during 2000 – 2019.
Our paper studies not only current state of “brands – research” relations, not only
statistics on certain date. Informative figure is also the dynamics – change of the indi-
ces during some time interval. For data acquisition the Scopus base with the corre-
sponding search services is applied.
2 Studied global brands
The studied global brands are selected from Top-100 list of the most expensive global
brands of 2018 according to “Best global brands rankings” by Interbrand. All the
homonymous brands that do not have the unique names are rejected. For instance, the
word “apple” in the scientific papers is used not only as the name of the most expen-
sive brand but also as an ordinary apple. Caterpillar is not only the producer of power-
ful road construction machinery but also ordinary caterpillar. Nike is not only sports-
wear but also one of the species of fish and the name of laser. Amazon and Honda
have the geographical homonyms. Some short names of the brands coincide with the
specific abbreviations, widely spread in certain research fields. For instance, NIKE is
used as the abbreviation for “Non-Interactive Key Exchange”, BMW – as “biomedi-
cal waste”. In case of homonymy, automatic processing the query outputs from Sco-
pus is complicated. For our study we have chosen the following 27 brands with
unique names: Google, Microsoft, Coca-Cola, Samsung, Toyota, Facebook, IBM,
Cisco, Louis Vuitton, Budweiser, Accenture, Hyundai, Ebay, Volkswagen, Goldman
Sachs, L'Oréal, Adidas, Hewlett Packard, Morgan Stanley, Harley-Davidson, Netflix,
Huawei, PayPal, John Deere, Spotify, Johnnie Walker, and Nintendo.
3 Brands classification based on statistics over 10 years
We have collected statistics about support of research by global brands as well as
using the global brands in research over 10 past years. For 2D-distribution of the
brands in the axes “support – usage” the data for the period of 10 years must be ag-
gregated in some way. The simplest way is summation of values for the 10 years. But
naturally, the fresh data have the greatest impact on the current perception of the
brand. The old papers produce the least impact. That is why, during the averaging we
will normalize the indices, taking into account the various importance of the data. For
the normalization of the support and usage, the following formulas are proposed:
1
UN wi U i ,
10 i 2009, 2018
(1)
1
SN wi Si ,
10 i 2009, 2018
(2)
where Si denotes the number of the papers of the ith year in which the corresponding
global brand is mentioned in the section Funding; U i denotes the number of the pa-
pers of the ith year, in which the corresponding global brand in mentioned in the pres-
entation part; wi denotes weight coefficient of the ith year.
We use arithmetic sequence of weight coefficients with decay 0.1, namely:
w2018 1 ; w2017 0.9 ; w2016 0.8 , ..., w2009 0.1 . Hence, the impact of the
events of the year 2009 is 10 times less than the events of the year 2018. 2D-
distribution of the brands according to the normalized indices is shown in Figure 1.
Fig. 1. Classification of the global brands in case of arithmetic sequence of weight coefficients
with decay 0.1 (log-log scale)
By the ordinal scale {Tiny, Low, Average, High} 9 clusters are allocated on Fig-
ure 1. For example, the leader cluster High-High, the cluster with High Normalized
Support and High Normalized Usage comprises the brands Google, Microsoft and
IBM.
In Figure 1 all the brands are located within the vicinity of the diagonal – each
brand has similar or neighboring linguistic values of the normalized support and us-
age. There is no brand with High Support level and Low Usage level or vice versa,
Low Support level and High Usage level. Greater part of the brands – 17 out of 27 are
in the clusters with the equivalent levels of the Normalized Support and Normalized
Usage. Twelve brands locate in the clusters High-High, High-Average, Average-High
or Average-Average. Eleven out of twelve of these brands are referred to technolo-
gies, business services, automotive, and only one brand – Coca-Cola belongs to the
food industry. The most generous and popular brands are located in the right upper
corner of Figure 1. These brands are Google, Microsoft, IBM, Samsung and Face-
book.
To checking the reliability of brands classification let us plot 2D-distributions in
case of other weight coefficients in formulas (1) and (2). Figure 2 shows 2D-
distribution for a geometric sequence of weight coefficients with ratio 0.9, namely:
w2018 1 ; w2017 0.9 ; w2016 0.9 2 0.81 , ..., w2016 0.99 0.39 .
Fig. 2. Classification of the global brands in case of geometric sequence of weight coefficients
with ratio 0.9 (log-log scale)
Figure 3 shows 2D-distribution for a geometric sequence of weight coefficients
with ratio 0.8, namely: w2018 1 ; w2017 0.8 ; w2016 0.82 0.64 , ...,
9
w2016 0.8 0.13 Figures 1–3 show slight fluctuations of the brands without any
change of the classifying results. Hence, the proposed classification is robust – it is
not sensitive to reasonable change of weight coefficients in (1) and (2).
Fig. 3. Classification of the global brands in case of geometric sequence of weight coefficients
with ratio 0.8 (log-log scale)
4 Support Dynamics
According to Figures 1–3 all the brands are divided into 4 groups by normalized sup-
port index. Let’s analyse support dynamics of the brands from the groups with high,
average and low support level. The dynamics of tiny support group is out of interest
due to statistically insignificant number of cases.
Five brands Google, Microsoft, Samsung, IBM and Volkswagen are in the group
with high support level. Time series of research support by those global brands are
shown in Figure 6. There is no dominant leader among these brands. Just one brand –
Google increased the number of supported papers every year. Google’s support dy-
namics is very close to a cube function. Figure 4 shows that all the brands considera-
bly increased their support during last 3 years. In 2018 Volkswagen supported 2.7
times more papers than in 2015, Samsung – 4.4 times, other brands – approximately
three times. Average dynamics shows that 2015 looks like as a breaking point. An
annual pace of support is equal to 22 during 2009 –2015 and has increased drastically
up to 297 during 2015–2018.
Fig. 4. Support dynamics for High support group
Eight brands are in the group with the average level of support. Annually they sup-
port 2–4 times less papers than the brands from the High support group. All the
brands from Average group increased also their support in 2016-2018 but did it very
unevenly (Figure 5). The slowest three-year pace demonstrates Coca-Cola – 2.3
times. The most rapid are Facebook – 4.7 times, and, especially, Huawei – 7.8 times.
Huawei supported 727 papers in 2018, and only 93 papers in 2015. Huawei with such
pace has a good change to move into the High support group during next year. Toyota
is another brand in the candidate-list for the High support group.
Fig. 5. Support dynamics for Average support group
On average, the brands of Average support group increased considerably the sup-
port during the last three years. We draw the attention to Average support group that
includes 2 brands from food industry and FMCG-industry. However, the level of their
support is far less than IT, telecommunication and automotive brands.
Eight brands are in the Low support group (Figure 6). The greater part of the
brands has zero-dynamic, i.e. the number of supported papers remains approximately
the same. Exception is 2018, when all the brands besides Nintendo increased the
number of supported papers significantly.
Fig. 6. Support dynamics for Low support group
5 Usage Dynamics
Among the analysed brands the researchers most often use Google. Since 2016 the
number of papers, in the presentation part of which Google is mentioned, exceeded
5000 annually. Facebook, Microsoft and IBM are also among the leaders (Figure 7).
All four brands stable growth of their usage in research during the last several years.
However, Google has highest pace with almost quadratic dynamics during all the
years.
Fig. 7. Usage dynamics for High usage group
Average usage group comprises nine brands (Figure 8). Samsung and Huawei have
good positive dynamics during almost all the years. The usage of remaining brands of
this group is either stable of slightly decreases. As a result, the average usage is stable
for this group.
Fig. 8. Usage dynamics for Average usage group
Nine brands has formed Low usage group (Figure 9). Ebay demonstrates
strong negative dynamics. Probably, Ebay-phenomenon as a research object loses the
attractiveness. Spotify has reliably positive dynamics. Usage of the rest of the brands
of this group during 10 years is more or less stable.
Resuming the data from all the groups, we conclude that strong positive dynamics
of the usage is demonstrated only brands related with information services, informa-
tion resources, media, and electronics.
Fig. 9. Usage dynamics for Low usage group
6 Conclusion
The research support by 27 global brands as well as their global brands usage in the
research during the last 10 years has been studied. The analysis is based on the infor-
metric approach with Scopus data. The support was assessed by the number of papers
where the global brand is mentioned in the funding section. The usage was assessed
by the number of papers, where the global brand is mentioned in the presentation part.
The statistics about support of research by global brands as well as using the global
brands in research over 10 past years were aggregated. The aggregation in form of
weighted sum with the highest impact for data of 2018 and the lowest impact for data
of 2009 is proposed. Nine clusters of brands are inducted by 2D-distribution accord-
ing to the normalized indices of support and of usage in log-log scale. The proposed
classification is robust – it is insensitive to reasonable change of weight coefficients.
According to the normalized level of support, all the brands are divided into four
groups with high, average, low and tiny levels. All the brands with high and average
levels are characterized by the considerable growth of the number of the supported
papers during 3 last years. This can be explained by the increase of funding and (or)
growth of the requirements to the grant recipients, concerning the obligation to men-
tion the source of funding in the corresponding section of the paper. Regarding to
group with low level, the greater part of the brands has zero-dynamics. Exception is
2018, when all the brands besides Nintendo increased the number of supported papers
significantly.
According to the normalized level of usage all the brands are divided into 4 groups
with high, average low and tiny levels. For the group with high level stable consider-
able growth of the brand usage in the papers during all 10 year is characteristic. High
usage group consists of Google, Microsoft, Facebook and IBM. Also Samsung and
Huawei from Average usage group and Spotify from Low usage group have good
positive dynamics during almost all the years. Hence, the strong positive dynamics of
the usage is demonstrated only information-centered brands.
It is revealed that the most generous and simultaneously popular brands are
Google, Microsoft, IBM, Samsung and Facebook. All these five top brands are con-
nected with the information systems and services. In the last three years their support
dynamics and usage dynamics are positive or strong positive. This enables to put
forward the hypothesis that the strong connection of the brand with research stimu-
lates further strengthening of these relations. That is, the effect that “strong becomes
stronger” is observed also in the relations between global brands and research activity.
For the verification of this hypothesis the additional statistics for several future years
is required.
References
1. Lopes, R. M., Fidalgo-Neto, A.A., Mota, F.B.: Facebook in educational research:
A bibliometric analysis. Scientometrics. 111, 1591–1621 (2017). doi:
10.1007/s11192-017-2294-1.
2. Baran, K.S., Ghaffari, H.: The manifold research fields of Facebook: A bibli-
ometric analysis. Journal of Information Science Theory and Practice. 5, 33–47
(2017). doi: 10.1633/JISTAP.2017.5.2.3.
3. Noruzi, A.: YouTube in scientific research: A bibliometric analysis. Webology.
14, 1–7 (2017).
4. Serôdio, P. M., McKee, M., Stuckler, D.: Coca-Cola – a model of transparency in
research partnerships? A network analysis of Coca-Cola’s research funding
(2008–2016). Public health nutrition. 21, 1594–1607 (2018). doi:
10.1017/S136898001700307X.
5. Rey-López, J. P., Gonzalez, C.A.: Research partnerships between Coca-Cola and
health organizations in Spain. European journal of public health. 29, 810–815
(2018). doi: 10.1093/eurpub/cky175.
6. Steele, S., Ruskin, G., McKee, M., Stuckler, D.: Always read the small print: A
case study of commercial research funding, disclosure and agreements
with Coca‑Cola. Journal of public health policy. 40, 273–285 (2019). doi:
10.1057/s41271-019-00170-9.
7. Greenhalgh, S.: Science and serendipity finding Coca-Cola in China. Perspectives
in Biology and Medicine. 62, 131–152 (2019). doi: 10.1353/pbm.2019.0007.
8. Nestle, M.: Corporate funding of food and nutrition research: science or market-
ing?. JAMA Intern Med. 176, 13–14 (2015). doi:
10.1001/jamainternmed.2015.6667.
9. Seyedghorban, Z., Matanda, M.J., LaPlaca, P.: Advancing theory and knowledge
in the business-to-business branding literature. Journal of Business Research. 69,
2664-2677 (2016). doi: 10.1016/j.jbusres.2015.11.002
10. Llanos-Herrera, G.R., Merigo, J.M.: Overview of brand personality research with
bibliometric indicators. Kybernetes. 48, 546-569 (2019). doi: 10.1108/K-02-
2018-0051.
11. Shtovba, S., Shtovba, O., Filatova, L.: The current state of brand management re-
search: An overview of leaders and trends in branding research over the past 20
years. The Bottom Line. 33, 1-11 (2020). doi: 10.1108/BL-08-2019-0106