ArticlePDF Available

Abstract and Figures

Purpose The purpose of the research was to measure the impact of post type (advertising, fan, events, information and, promotion) on two interaction metrics: likes and comments. The measuring involved two popular social media, Facebook and Instagram, and in business profiles of five different segments (food, hairdressing, ladies’ footwear, body design, fashion gym wear). Design/methodology/approach The method used was multiple regression analysis with an estimator of the Ordinary Least Squares (OLS) for 1,849 posts from five different companies posted on Facebook (680 posts) and Instagram (1,169 Instagram) over an eight-month posting period. Regression analysis was used to identify the relationship between the dependent variables (likes and comments) and, the independent variables (post typology, segments, week period, month, characters and tags). Findings It was seen that the post types events and promotion led to a greater involvement of followers in Instagram, in particular. In Facebook, the events post type was only significant in the like’s interaction. Another finding of the research is the relevance of the food and body design segment which was significant in both virtual social media. This indicates a user preference involving their day-to-day lives, in this case, having a tattoo done or seeing a photo of a dessert. Originality/value With the findings of this study, academics and social media managers can improve the return indicators of interactions in posts and broaden the discussion on the types of post and interaction in different virtual social media.
Content may be subject to copyright.
Online Information Review
Does social media matter for post typology? Impact of post content on Facebook
and Instagram metrics
Ricardo Limongi França Coelho Denise Santos de Oliveira Marcos Inácio Severo de Almeida
Article information:
To cite this document:
Ricardo Limongi França Coelho Denise Santos de Oliveira Marcos Inácio Severo de Almeida
, (2016),"Does social media matter for post typology? Impact of post content on Facebook and
Instagram metrics", Online Information Review, Vol. 40 Iss 4 pp. 458 - 471
Permanent link to this document:
http://dx.doi.org/10.1108/OIR-06-2015-0176
Downloaded on: 05 August 2016, At: 14:57 (PT)
References: this document contains references to 40 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 60 times since 2016*
Users who downloaded this article also downloaded:
(2016),"Research impact of general and funded papers: A citation analysis of two ACM international
conference proceeding series", Online Information Review, Vol. 40 Iss 4 pp. 472-480 http://
dx.doi.org/10.1108/OIR-08-2015-0249
(2016),"Understanding the determinants of human computation game acceptance: The effects of
aesthetic experience and output quality", Online Information Review, Vol. 40 Iss 4 pp. 481-496 http://
dx.doi.org/10.1108/OIR-06-2015-0203
(2016),"Online sustainability information in local governments in an austerity context: An
empirical analysis in Italy and Spain", Online Information Review, Vol. 40 Iss 4 pp. 497-514 http://
dx.doi.org/10.1108/OIR-05-2015-0161
Access to this document was granted through an Emerald subscription provided by emerald-
srm:333301 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald
for Authors service information about how to choose which publication to write for and submission
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as
well as providing an extensive range of online products and additional customer resources and
services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the
Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for
digital archive preservation.
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
*Related content and download information correct at time of download.
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
Does social media matter for
post typology? Impact of post
content on Facebook and
Instagram metrics
Ricardo Limongi França Coelho, Denise Santos de Oliveira and
Marcos Inácio Severo de Almeida
Faculty of Business, Accounting and Economy,
Federal University of Goiás, Goiânia, Brazil
Abstract
Purpose The purpose of this paper is to measure the impact of post type (advertising, fan, events,
information, and promotion) on two interaction metrics: likes and comments. The measuring involved
two popular social media, Facebook and Instagram, and in business profiles of five different segments
(food, hairdressing, ladiesfootwear, body design, fashion gym wear).
Design/methodology/approach The method used was multiple regression analysis with an
estimator of the ordinary least squares for 1,849 posts from five different companies posted on
Facebook (680 posts) and Instagram (1,169 Instagram) over an eight-month posting period. Regression
analysis was used to identify the relationship between the dependent variables (likes and comments),
and the independent variables (post typology, segments, week period, month, characters and hashtag).
Findings It was seen that the post types events and promotion led to a greater involvement of
followers in Instagram, in particular. In Facebook, the events post type was only significant in the likes
interaction. Another finding of the research is the relevance of the food and body design segment
which was significant in both virtual social media. This indicates a user preference involving their
day-to-day lives, in this case, having a tattoo done or seeing a photo of a dessert.
Originality/value With the findings of this study, academics and social media managers can
improve the return indicators of interactions in posts and broaden the discussion on the types of post
and interaction in different virtual social media.
Keywords Facebook, Comments, Social media, Instagram, Likes, Post typologies
Paper type Research paper
Introduction
Research focussing on social media has been producing considerable results about
interaction and dynamics among individuals and companies, but it is still in its early
stages, since not even the term social media has a universally agreed definition, and
there is no standard typology of social media platforms upon which everyone agrees
(Weller, 2015). Social media is being recognized as a tool and business managers
advocate for its inclusion to the overall strategy because customers often rely on these
platforms to interact with friends and brands (Rapp et al., 2013). Despite being
commonly used for research focussing on consumer metrics, preferences, and demand
prediction (Aral et al., 2013), challenges about the variety of user interactions in an
ever-changing environment like social media still persist (Weller, 2015) and empirical
efforts must address these issues.
Our research reconciles these challenges, as it considers brand content and
individual user interaction across social media. Extant research about brand content
and engagement metrics normally used single environments, reproducing the overall
Online Information Review
Vol. 40 No. 4, 2016
pp. 458-471
© Emerald Group PublishingLimited
1468-4527
DOI 10.1108/OIR-06-2015-0176
Received 1 June 2015
Revised 30 September 2015
Accepted 15 January 2016
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
458
OIR
40,4
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
publication trend, where Facebook and Twitter are most overemphasized (Weller,
2015). Over the last years, researchers have been trying to understand which brand
content results in more comments, likes and shares but have not considered the
potential mediating role of different social media. This is an important question to
regard, as recent research agendas point out to the heterogeneity effect of social media
design on relationship formation among individuals and companies (Aral et al., 2013;
Schultz and Peltier, 2013).
Companies use virtual social media to enhance interaction with their current and
potential customers by publishing posts. Posts use publications features such as text,
photos, and videos, which can facilitate interaction among users (De Vries et al., 2012).
Usually, empirical research focusses on post typology effects on Facebook, but
companies are also using Instagram. Launched in 2010 and designed only for
smartphones, the aim of the platform is to share photos and videos (up to 15 seconds)
freely. The resulting popularity disseminated to business and 65 percent of the world
leading brands have already an account on the platform (Statista, 2013). Not only big
brands have been using Instagram. Small business found a cheap way to promote and
sell their products, especially the fashion industry, such as clothing and accessories
(Forbes, 2013), and this must be taken into consideration by researchers.
We organize the paper as follows. First, we briefly present metrics for interaction on
the most popular virtual social media. Next, we outline the research background on of
post typology and provide the basis for hypotheses construction. Research method and
results are provided after the hypotheses, and we conclude our paper with research and
managerial contributions and limitations.
Virtual social media and metrics for interaction
Social media are important forms of virtual communications in which participants
share information, knowledge, and maintain social ties (Boyd and Ellison, 2007; Ellison
et al., 2007). They are defined as spaces where users create profiles, articulate
themselves, and interact in different levels with other people, brands, and companies
(Boyd and Ellison, 2007). Over the last years social media such as Facebook and
Instagram have put together millions of members. The last estimation indicate 1.49 billion
active users on Facebook and 300 million users on Instagram (Facebook, 2015).
This popularity seems to span different cultures. According to a social media
monitoring service SocialBakers, a posting referring to a release of a new Taco Bell
product onto the US market yielded 360,606 interactions (322,973 likes, 12,302
comments, 25,331 shares) (SocialBakers, 2015b) in just one month, while in Brazil a
similar posting about the release of a new footwear resulted in 353,321 interactions
(341,392 likes, 7,688 comments, 4,241 shares) (SocialBakers, 2015a). Given the
importance of virtual social media to interaction between companies or brands and
potential consumers, scientific researchers have been published about different issues.
One specific topic highlights the impact of different types of content on social media
metrics. There are numerous social media metrics as pointed out by Peters et al. (2013) and
there is a controversy regarding the choice on the use of appropriate indicators. Apart
from the discussion on the usefulness of each metric to reflect business reality, a common
ground for the research on this field is to use data from social media to understand the
variation on non-economic variables, such likes, comments and shares (Luarn et al., 2015).
Usually, social media authorize the creation of individual and company/brand
profiles which are used as interaction tools. Users can incorporate personal and
professional information, upload photos and invite friends, while brands can connect to
459
Facebook and
Instagram
metrics
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
their consumers and publicize marketing related material (Boyd and Ellison, 2007;
Smith et al., 2012). After creating profiles to communicate to consumers, companies, and
brands incorporate contents such as news, photos, and videos, seeking to raise visitor
levels and the aforementioned metrics. The five most popular virtual social media and
their characteristics and metrics are shown in Table I.
Research background on post typology in single social media
Literature on the impact of companies posts typology is growing rapidly in the last five
years, but usually concentrate on semantic (text content) or richness (moving-pictures
that complement text, such as pictures and videos). Some studies analyze the number of
media elements or media type (text, photo, or video) and the presumable positive
impact in consumer responses, such as comments, likes, and shares (Kim et al., 2015;
Rauschnabel et al., 2012).
Another group classifies post types and media elements such as richness of content
and results indicate that images and videos are responsible for a positive influence on
likes and comments (De Vries et al., 2012; Sabate et al., 2014). One last group includes
the brand effect and text content as independent variables and results show that
avoiding hard sell types could increase the number of likes inasmuch as including
corporate brand names and emotional tones in Facebook posts (Swani et al., 2013).
Results on the impact of text content is controversial: some studies found
that contents such as entertainment and information raises, on average, the number
of likes, comments, and shares (Cvijikj and Michahelles, 2013) while others not
(De Vries et al., 2012). From all the recent empirical research conducted, one may
conclude that extant research classifies post typologies in different forms and do not
consider potential differences across social media, since empirical efforts concentrate
on Facebook.
Media characteristics can shape individual objectives and experiences in virtual
environments (Hoffman and Novak, 1996) and new studies are stressing the mediating
Virtual social
media Primary metrics Specific characteristics
Facebook Comments
Likes
Shares
Creation of groups, pages, events and advertisements
Use of applications
Add more than one user to a conversation
Instagram Comments
Likes
Postings originated exclusively from smartphones and tablets
Postings with images and short videos (up to 15 seconds)
Editing images and videos tool
LinkedIn Comments
Likes
Shares
Relationships based on professional contacts
Profiles containing professional information
Participation in groups
Twitter Favouriting a
tweet
Likes
Incorporate a
tweet (quoting)
Retweet (share)
Personal or professional messages with a maximum of 140
characters
YouTube Didnt like
Likes
Shares
Creation and interaction with personal or
third-parties videos
Enroll in content channels
Table I.
Characteristics
and metrics for
engagement in the
most known virtual
social media
460
OIR
40,4
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
role that different social media has on user interaction. Instagram, for example, poses
singular forms of engagement because alters temporal and vertical structures in favor
of spatial connectivitys. Based on photo sharing, this social media requires a specific
device (a smartphone) and exhibit a dynamic timestamp, as each image shows a
constantly changing representation of time(Hochman and Manovich, 2013).
Our empirical study incorporates these differences as it considers brand typologies
as a source of interaction, but contemplates the potential mediating role of social media.
Research design followed previous procedures of categorizing post typology of
business from different segments. However, the selection criteria included one
additional component, since the companies studied were active on the two most
popular social media sites, Facebook and Instagram. Table II illustrates the current
literature on post typology effects and the research gap addressed by our study.
Method
Our research is characterized by being causal look through quasi-experiment
procedures, establish relations of cause and effect between the independent and
dependent variables (Malhotra and Birks, 2007). The empirical study took place in a
quasi-experiment or econometric treatment since both the choice of the companies as of
the posts was not carried out by random assignment (Shadish et al., 2002). To perform
the data collection was selected five companies accessible to researchers to: ensure the
small business profile; and assess the free advertising of products and service offers by
small businesses on Instagram. The selection of companies obeyed the following
criteria: different segments; active accounts on Facebook and Instagram; at least three
weekly posts as each social network analyzed.
The segments of the chosen companies were the food, beauty, womens shoes,
fashion design, and body fitness. The operationalization of the research took place after
the collection and categorization of publications of 1849 posts (680 on Facebook and
1,169 on Instagram) of the chosen companies. The purpose of this choice was to
reproduce user interaction with different groups and social media. Posts refer to the
period between January and August of 2014. These choices differ from previous
research. De Vries et al. (2012) evaluated the food segments, accessories, leisure wear,
alcoholic beverages, cosmetics, mobile phones, and brought together 355 posts, while
Swani et al. (2013) collected 1,143 observations, both only on Facebook.
Study procedures and variable
Data collection was carried out on Facebook and Instagram of each company during
the months of August and September of 2014 through a browser feature that lets you
save all the page data. This procedure enables all contents to be saved for the extent
that the page has loaded, storing the data onto HTML file extension. The data
Description References
Single social media with multiple brand or
companies of one sector
Rauschnabel et al. (2012) and Sabate et al. (2014)
Single social media with multiple brand or
companies of various sectors
Cvijikj and Michahelles (2013), de Vries et al. (2012), Kim
et al. (2015) and Swani et al. (2013)
Multiple brand or companies of various
sectors across social media
Our empirical study
Table II.
Current literature on
post typology effects
461
Facebook and
Instagram
metrics
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
contained the variables of interest (type of posts), control (segment week period in
which the posting, month, and the text of the post) and dependent (like and
comments) empirical models. These data were then systematized in a
spreadsheet Excel software to be analyzed using Stata software (version 13.1).
The data were arranged in cross-section since they were not collected information on
the posts of time.
The categorization of types of threads extended previous research carried out in the
context of Facebook and was operationalized with a typology that considered the
content of messages posted on the social network. This categorization occurred after
analysis of investigations into similar objectives. Caseiro and Barbosa (2011) classified
the threads into three groups (advertising/services/campaign information and offers/
contests/hobbies). Rauschnabel et al. (2012) limited to classify posts according to
technical characteristics such as size, an amount of text, media elements (such as
pictures) and the presence of polls. De Vries et al. (2012) used six types, but with other
features in addition to the content, such as posting the location on the page. The work
of De Vries et al. (2012) appealed to four specific types: interactive, informative,
entertaining, and contrasting content. Smith et al. (2012) built six categories and
compared to platforms Facebook, Twitter, and YouTube, while Swani et al. (2013)
identified three types, based on psychological model choice Hansen (1976). The
categorization of Swani et al. (2013) classified the postings: those who use business
names, which refer to emotional content and that references to make instant purchases
of products or services.
The classification proposed to this paper is a quantitative improvement on the
mentioned studies and feature posts into five categories: advertising, events, fan,
information, and promotion. Table III presents and describes the variables of interest in
the study: the dependent variables, which are in a quantitative way, and the
independent variables in the qualitative form.
With the proposed categories of threads supported by Facebook and Instagram,
it was proposed five hypotheses to be evaluated in the study. As in the study of
Variable Description Nature Notation
Likes Quantity of likes received per post Dependent/
Quantitative
LIKE
Comments Quantity of comments received per post Dependent/
Quantitative
COM
Advertising Posts to promote brands in social media present
publicity items which cross the digital sphere and
posts with entertaining content, to attract the attention
of their followers and acquire larger numbers of
likes and comments
Independent/
Qualitative
ADV
Fan A fan is responsible for the main idea of post, or for
sending the photo. Their participation is always mentioned
in the post
Independent/
Qualitative
FAN
Events Posts, with photo and video media, directly connected to
brands or otherwise
Independent/
Qualitative
EVE
Information Content with data about events, places, opportunities, people,
or celebrities, directly connected to a brand or otherwise
Independent/
Qualitative
INFO
Promotion Posts with quizzes, which promote participation of followers
through rewards
Independent/
Qualitative
PROM
Table III.
Interest variables of
the empirical study
462
OIR
40,4
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
De Vries et al. (2012), we assume that the posting category has a positive relationship of
the dependent variables (interactions) choose:
H1. Posts of advertising category have a positive relationship with the interaction
of posting.
H2. Posts of the events category have a positive relationship with the interaction
of posting.
H3. Posts of fan category have a positive relationship with the interaction
of posting.
H4. Posts of information category have a positive relationship with the interaction
of posting.
H5. Posts of promotion category have a positive relationship with the interaction
of posting.
Additionally, our study found that there are five control variables against three of the
work of De Vries et al. (2012) and only two in the search Swani et al. (2013). First, the
business segment was controlled from the collection posts of five different segments
profiles, such as the research De Vries et al. (2012), who used as a control variable
product categories. The study also relied on De Vries et al. (2012) to assign as a control
variable for the week period in which the posting (weekday or weekend). The other
variables had to evaluate the seasonality of posting (month) and the characteristics of
the text used in the posting of the description (number of characters and the use of
tagging). The evaluation of the use of tagging on social network was previously used in
the study of Nam and Kannan (2014). In the survey the tagging was seen as new way
to share and online categorize content that enables user to their express thoughts,
perceptions, and feelings with respect to diverse concepts,lining up with the interest
of our research in evaluating text features that can influence threads. Table IV presents
and characterizes the study of the control variables.
We built two econometric models to evaluate each social network in which the
dependent variables are, respectively, like and comments. The models incorporate
quantitative and qualitative independent variables discussed in Tables III and IV.
The representations of the parameters and intercept of the models are reproduced
in Equations (1) and (2). The estimation was performed using the method of ordinary
Variable Description Nature Notation
Companies
segments
Food (restaurants), beauty (hairdressing), ladies
footwear, body design (tattoos), fashion gym wear
(womens fashion gym wear)
Control/
Qualitative
FOOD, HAIR, FOOT,
DES, FASH
Week
period
Posts that occurred in the middle of the week
(between Monday and Thursday) or the weekend
(Friday, Saturday or Sunday)
Control/
Qualitative
WDAY, WEND
Month of
the post
Posts of months January, February, March, April,
May, June, July, and August
Control/
Qualitative
JAN, FEB, MAR, APR,
MAY, JUN, JUL, AUG
Number of
characters
Number of characters used in the description of
the post
Control/
Quantitative
CHAR
Tagging Word or phrase after the sign in the description of
the post (#)
Control/
Qualitative
TAG
Table IV.
Control variables of
the empirical study
463
Facebook and
Instagram
metrics
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
least squares. This method estimates the parameters of the sample regression function so
that the sum of residuals is as low as possible and in which the estimated values are the
closest possible observed values (Greene, 2012). The dependent variables (like
and comments) and independent with quantitative trait (characters) were transformed
to logarithm notation, as well as in the study by De Vries et al. (2012) to facilitate
comparison of results.
Equations (1) and (2) follows:
like ¼b0þb1adv þb2fanþb3eveþb4inf oþb5ser þb6prom
þb7f ood þb8hair þb9f oot þb10 desþb11 fasþb12wday
þb13wend þb14 jan þb15 febþb16mar þb17apr þb18may
þb19 junþb20 jul þb21aug þb22 char þb23 has þm(1)
comments ¼b0þb1adv þb2fanþb3eveþb4inf o þb5ser
þb6prom þb7f ood þb8hai r þb9f oot þb10des
þb11 fasþb12wday þb13 wend þb14 jan þb15 feb
þb16mar þb17 apr þb18may þb19 jun þb20 jul
þb21aug þb22 char þb23hasþm(2)
Data analysis
Descriptive statistics
In Table V, is the summary of the dependent variables used in the econometric models
and the number of posts collected by segment. Like it is observed that has the highest
average among the variables. Instagram also has a higher average (like 4,210 and
69,686 comments) compared to Facebook (23,764 like and 1,091 comments). These
results suggest that the interaction between customer and company on Instagram is
more common, perhaps the unique interactions via smartphone and have the
characteristic of posting photos or videos. Another reason is the increased time spent
by users on the smartphone. Among the Americans, for example, the time from
40 minutes a day in 2010 to 134 minutes in 2014 (Statista, 2015). It is observed that like
Social media Variables Observations Average SD Minimum Maximum
Facebook LIKE 680 23.764 54.907 0 1037
COM 680 1.091 3.523 0 32
Instagram LIKE 1.169 69.686 60.288 6 362
COM 1.169 4.210 12.009 0 286
Business segments Facebook Instagram
M(like) M(comments) Posts (n)M(like) M(comments) Posts (n)
Food 75 3 78 57 5 53
Beauty 3 0 122 30 3 418
Ladies footwear 1 0 2 33 4 148
Body design 113 7 45 201 8 104
Fashion gym wear 11 0 433 90 5 446
Notes: M, mean of business segments by post; n, total number of posts
Table V.
Descriptive
statistics of the
dependent variables
and posts by
business segments
464
OIR
40,4
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
has the highest average among the segments. On the contrary, reviews have the lowest
average frequency between the two dependent variables. A likely reason for this result
is that individuals spend more effort in writing comments on posts on social networks.
Comments also appear as lower standard deviation variable, revealing lower dispersion
of data. It should be noted that: the maximum values of variables follow the same
layout: first, like (1,037 on Facebook and 362 on Instagram), followed by shares (32 on
Facebook and 286 in Instagram); and the body design segment has the highest average
of the dependent variables in the two social networks.
Specification tests, F-test and quality adjustment models
As a first step in the analysis, specification tests were performed and read from the
multiple linear constraints tests (test F) and the results of determination of
the coefficients of the models (R
2
and adjust R
2
). Table VI presents the estimates of the
Table VI.
Results of the
estimates of the
coefficients and
statistics of the two
models constructed
for each social media
Facebook Instagram
Model 1 like Model 2 comments Model 1 Like Model 2 comments
Variables of interest
Post typology
adv 0.2948 (1.08) **** 0.1884 (1.24)*** 0.1459 (0.44)
fan 0.2287 (0.83) 0.3276 (0.59) 0.0280 (0.18) 0.0672 (0.20)
eve 0.2742 (0.88)** 0.0066 (0.01) 0.3437 (2.07)* 0.5745 (1.57)***
info 0.2483 (0.86) 0.1572 (0.25) 0.2085 (1.30) 0.2780 (0.80)
prom 0.1393 (0.47) **** 0.1715 (0.93)* 0.5236 (1.33)***
Control variables
Segments
food 2.0031 (3.63)*** 0.6657 (1.50) 0.3456 (4.30)*** 0.4482 (2.48)**
hair 0.7182 (1.30) **** 0.9983 (16.54)*** 0.0961 (0.68)
foot 1.1274 (1.67) **** 0.8026 (11.74)*** 0.4781 (3.10)**
des 2.4400 (4.42)*** 1.3220 (2.83) 0.8510 (16.57)*** 0.8482 (7.70)***
Week period
wday 0.079 (0.15) 0.0360 (0.21) 0.034 (1.22) 0.09888 (1.51)
Month post
jan 0.7330 (5.02)*** 1.0651 (2.47) 0.0061 (0.10) 0.05736 (0.42)
feb 0.5959 (3.89)*** 1.3132 (3.02) 0.0025 (0.05) 0.2434 (2.06)*
mar 0.5628 (3.71)*** 1.5776 (3.48) 0.1323 (2.46)* 0.5247 (4.23)***
apr 0.2237 (1.60) 0.8824 (2.15) 0.1694 (3.42)** 0.2941 (2.52)**
may 0.2144 (1.49) 0.8902 (2.18) 0.1746 (3.49)** 0.1164 (0.98)
jun 0.3468 (2.41)** 1.02555 (2.41) 0.2344 (4.92)*** 0.1967 (1.75)
jul 0.3363 (2.25)* 0.8379 (1.95) 0.0657 (1.36) 0.2543 (2.20)*
Text post
carac 0.8063 (3.69)*** 0.1724 (2.87) 0.0061 (0.63) 0.1177 (4.83)***
tag 0.2634 (0.48) 0.0249 (0.05) 0.1551 (2.91)* 0.2141 (1.74)
Constant 1.74 (2.78) 0.67 (0.77) 4.00 (22.54) 0.29 (0.75)
F108.61*** 4.80*** 166.43*** 13.49***
R
2
0.7599 0.3371 0.7337 0.2240
Adjust R
2
0.7529 0.2669 0.7292 0.2074
Number of obs 680 680 1.169 1.169
Notes: The t-statistics are in parentheses, located below the estimates of the variables of interest
and control. *Significant differences ( po0.05); **Very significant signal values ( po0.01);
***Highly significant values ( po0.001); ****Omitted because present collinearity
465
Facebook and
Instagram
metrics
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
models for social network operationalized with robust procedures about
heteroscedasticity. Wooldridge (2013) indicates that estimators variances are biased
without the possibility of homoscedasticity; as a result, inadequate inferences are
conducted in the presence of heteroscedasticity (White, 1980). The identification of this
feature happened after performing the Breusch and Pagan (1979) test, which returned
high values of the χ
2
statistic and allowed reject the null hypothesis of constant variance
(Chi ¼3138.87 in Model 1; qui ¼1616.39 in Model 2 on Facebook, qui ¼621.68 in Model 1,
qui ¼27138.47 in Model 2 on Instagram). Another test identified values of the variance
inflation factor (VIF) models. Gujarati and Porter (2011, p. 337) argue that this factor
shows how the variance of an estimator is inflated by the presence of multicollinearity.
The values returned by the tests indicate VIF average closer to the value 1 and away from
10, which indicates the existence of an acceptable collinearity between the independent
variables (average VIF equal to 1.32 to Facebook and average equal to 1.73 for Instagram).
It is necessary to highlight the organization procedure of data and allocation of
reference categories in qualitative variables so that they properly interpret Table VI. As a
general principle to include dummy variables that indicate different groups in the case of
regression model shows ggroups or categories, there is the need to include the g one
variables in the models (Wooldridge, 2013). The reference variables defined in both models
for social media Facebook and Instagram were: promfor posting typology group, fash
for the enterprise segment, wendfor the week period and augfor month post. This
grouping was selected, after conducting several tests between variables, for the following
reason: this combination results in intercepts that are not statistically significant, allowing
appropriate comparisons between the intercepts groups of dummy variables to intercept
the group base (Wooldridge, 2013). The no statistical significance guarantees the constant
comparisons between the posting types of coefficients when they are significant.
Model 1 returned in the two largest social media values of R
2
, which measures the
proportion of total variation in the dependent variable explained by the regression model
(Gujarati and Porter, 2011; Wooldridge, 2013). When comparing the determination
coefficients of the models presented in this work with marketing researches conducted in
the context of social media, it is observed that are superior: the study by De Vries et al.
(2012) in the context of Facebook, the results were 15 percent in the like model and
30 percent in the comment model against 75.18 and 27 percent, respectively on Facebook
and 73 and 20 percent in Instagram. An argument should be put on the agenda on this
issue, this paper advances in understanding the impact of the type of posting by
considering additional shares as the dependent variable.
The models for number of like and comments are jointly statistically significant.
The test F-values allow reject the null hypothesis that the slope coefficients are
simultaneously equal to 0 (Gujarati and Porter, 2011). The R
2
values and adjust R
2
explain the variance of the dependent variables in a reasonable way for Models 1
(F¼108.61 value, p-value o0.01, R
2
¼75.99 percent and adjust R
2
¼75.29 percent on
Facebook and value F¼166.43, p-value o0.01, R
2
¼73.37 percent and adjust
R
2
¼72.92 percent in Instagram) and 2 (F-value ¼4.80 po0:01, R
2
¼33.71 percent
and adjust R
2
¼26.69 percent on Facebook and value F¼13:49, p-value o0.01,
R
2
¼22.40 percent and adjust R
2
¼20, 74 percent in Instagram).
Analysis of the impact of the variables of interest
The second stage of analysis involved the results of hypothesis tests on the individual
estimates of the regression coefficients. Two important considerations about the
variables of interest should be highlighted: the types of posts that impact like and
466
OIR
40,4
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
comments are different in the analyzed social media, indicating a theoretical and
empirical way for future work aimed at classifying and identify the influence of these
types in different virtual social networks; and the event category is the one that has an
impact on Facebook (like) and Instagram (like and comments), which indicates a
typology for digital content managers consider during the planning of posts that will
be posted on social networks like Instagram and Facebook.
Category advertising posts are statistically significant at a confidence level of
99.9 percent in the like on an Instagram model and is characterized by linear, positive
impact on the dependent variable like. A post published with the kind of advertising in a
business segment receives 18.84 percent more like than posts of fan categories, event,
information, and promotion, the latter category of reference. Unlike threads promotional
type, characterized as advertising are not linked to contests, sweepstakes or rewards and
are designed to promote the brand, sometimes referring to festive dates like Christmas.
The statistical significance of this type appears to be related to brand management in the
post-internet reality process: it is a more dynamic context that seeks to involve consumers
at key stages of the process of brand building (Christodoulides, 2009).
The second type of posts that promotes linear impact on like on Facebook and like and
comments on Instagram is the event. However, the results of this category are slightly less
consistent in econometric terms: they are significant at a level of 95 percent to like on
Facebook and 90 to 99.9 percent and like for comments on Instagram. The event posting
receives 27 percent of like on Facebook. For Winer (2009), virtual social networks like
Facebook are equipped with features that enable companies to communicate with specific
segments and engage individuals through interactivity. One of those ways would be to
invite the followers for events organized by companies. However, as the create feature
events can be used between users and between company and followers, the feature does
not arouse positive intention of followers enjoy posting on Facebook. Instagram on the
event type has two different behaviors. Increases by 34 percent the number of like, but
reduces the number of comments on 57 percent. This behavior can occur by the usersease
interacting with the post only with the like, not producing a text or informing another user
inthecommentsasawaytopublicizetheevent.
The promotion typology promotes linear and positive impact on the dependent
variables only on Instagram. At a 90 percent confidence level posts receive promotional
17.15 percent more like than threads categories of services. About comments and a
99.9 percent confidence level, the typology receives 52.36 percent more comments.
Promotional activities may go beyond the immediate effect on sales and influence
consumer learning and their behavior in the long run (Van Heerde and Neslin, 2008).
It seems to be the goal of companies when they encourage some activity follower in
their profiles in exchange for participation in sweepstakes and contests.
These mechanisms also part of the scope of promotional activities, although the goal
is no effect on sales or profitability, the behaviors and attitudes of individuals
(Chandon, 1995). In this sense, an important managerial implication refers to the use by
marketers, postings characterized as promotional.
The other two types of posts analyzed (fan and information) does not promote
statistically different impacts of the reference variable (promotion). These results
together early indications of no significance of these typologies and additional studies
are needed to identify the recurrence of this pattern with companies in other industries.
Specifically dealing on informative posts, literature had already given evidence of no
statistical significance in that category. De Vries et al. (2012) did not fail to reject the
hypothesis that informative posts and comments get more like than those characterized
467
Facebook and
Instagram
metrics
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
as non-informative. The decision to include it in the group of five variables of interest
was due to the need to test a broader group of typologies that De Vries et al. (2012) and
Swani et al. (2013) (Table VII).
Analyzing the impact of control variables
The first set of control variables refers to enterprise segments. The variable qualitative
reference was chosen gym fashion segment wear and according to the results, it is apparent
that only the feed body segments and drawing are statistically significant in the two
studied social networks. Posts on meal segment increase like on Facebook at 200.31 percent,
representing a decrease of 34.56 percent on Instagram while comments have an increase of
44.82 percent over the gym fashion wear segment. Instagram on the hair segment has a fall
of 99.83 percent in the number of like to a confidence level of 99.99 percent. The foot
segment has a linear impact on Instagram with a decrease of 80.26 percent in the number of
like and an increase of 47.81 shares. As well as the power segment, the body design
influences Facebook and Instagram. There is an increase in the number of like Facebook
(244 percent) and like (85 percent) and comments (84.82 percent) on Instagram. Especially
theincreaseofinteractioninthepowersegment when they are posted photos of desserts,
and photos in the body design firm present the new tattoo will be made on a consumer.
Both the photo of dessert as the body design that will be done, characteristics were present
in the posts with the most like and comments.
Posts published during the week are statistically the same as the weekend. This
means that do not cause impacts on reviews and like. These results corroborate what was
found in the study by De Vries et al. (2012), which also included days of the week as a
control variable and did not find statistically significant results. That way, you can
ensure that users like, share and comment threads with equal frequency on weekdays
and weekends and make use of Facebook and Instagram in different ways that e-mail
services, for example. Another variable about post, month, showed different behaviors.
Facebook, the month of January and March, had a positive impact while the months of
February, March and June had a negative impact, and Instagram returned different
results on the models evaluated, but all positive. In the case of like, the months of March,
April, May, and June showed an average increase of 17.76 percent and comments the
increase occurred in February, March, April, and July. In particular, the increase in the
number of like from February and March is explained to represent a period of national
holidays in the country of the companies surveyed, and July a vacation, which tends to
increase the release of stocks of companies with their followers.
The third control group of variables analyzed characteristics the type of text of
posts. Regarding the like Facebook, the number of characters representing an increase
of 80.63 percent. As for the Instagram, only comments have a positive linear impact on
the number of characters of the post. This behavior can be caused by the preference of
Likes Comments
Post typology Expectation Instagram Facebook Instagram Facebook
H
1
: advertising +Supported Unsupported Unsupported Unsupported
H
2
: event +Supported Supported Supported Unsupported
H
3
: fan +Unsupported Unsupported Unsupported Unsupported
H
4
: information +Unsupported Unsupported Unsupported Unsupported
H
5
: promotion +Supported Unsupported Supported Unsupported
Table VII.
Summary of results
468
OIR
40,4
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
consumers interact with posts that have not only photos and informative text. The
Instagram like model returned to a significant 90 percent, an increase of 15:51 percent
in posts that have tags, a common feature in threads companies in Instagram, by
posting describing the characteristics using different tags.
Conclusions and further research
This study aimed to measure the impact of the posting on important metrics in two social
medias. Event postings exert a linear impact on the dependent variable of the analyzed
social medias, suggesting a standard, and indicates a theoretical and empirical way for
future studies directed to classify and identify the influence of this type. We conclude that
virtual social medias like Facebook and Instagram are more efficiently utilized when used
as a means of promotion that provides hedonic benefits to users, rather than commercial
benefits through direct promotion of products, services, and prices (Chandon, 1995;
Chandon et al., 2000). The types that were statistically significant possibly promote it
Subramani and Rajagopalan (2003) classified asemotional engagement and connection of
individuals with the transmitted message. This connection allows the diffusion of posting
to social media friends in the social circle.
Future investigations should proceed from the model presented in this work and
analyze, for example, the dependent variables as members of simultaneous equations
or a system in which the terms of disorder are highly correlated (Zellner, 1962).
Exploratory analyzes with the database of this research show that the coefficient of
determination increases significantly when including shares and like as independent
variables Models 1 and 2. Understanding this dynamic is important since there
seems to be a relationship between the dependent variables: individuals who enjoy a
particular post, also seem to comment on it and share it.
References
Aral, S., Dellarocas, C. and Godes, D. (2013), Introduction to the special issue social media and
business transformation: a framework for research,Information Systems Research,
Vol. 24 No. 1, pp. 3-13.
Boyd, D.M. and Ellison, N.B. (2007), Social network sites: definition, history, and scholarship,
Journal of Computer-Mediated Communication, Vol. 13 No. 1, pp. 210-230.
Breusch, T.S. and Pagan, A.R. (1979), A simple test for heteroscedasticity and random coefficient
variation,Econometrica: Journal of the Econometric Society, Vol. 47 No. 5, pp. 1287-1294.
Caseiro, B. and Barbosa, R. (2011), Empresas no Facebook: O Caso da TMN e da Optimus,
Internet Latent Corpus Journal, Vol. 2 No. 1, pp. 6-15.
Chandon, P. (1995), Consumer research on sales promotions: a state-of-the-art literature review,
Journal of Marketing Management, Vol. 11 No. 5, pp. 419-441.
Chandon, P., Wansink, B. and Laurent, G. (2000), A benefit congruency framework of sales
promotion effectiveness,Journal of Marketing, Vol. 64 No. 4, pp. 65-81.
Christodoulides, G. (2009), Branding in the post-internet era,Marketing Theory, Vol. 9 No. 1,
pp. 141-144.
Cvijikj, I.P. and Michahelles, F. (2013), Online engagement factors on Facebook brand pages,
Social Network Analysis and Mining, Vol. 3 No. 4, pp. 843-861.
De Vries, L., Gensler, S. and Leeflang, P.S.H. (2012), Popularity of brand posts on brand fan
pages: an investigation of the effects of social media marketing,Journal of Interactive
Marketing, Vol. 26 No. 2, pp. 83-91.
469
Facebook and
Instagram
metrics
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
Ellison, N.B., Steinfield, C. and LAMPE, C. (2007), The benefits of Facebook friends: social
capital and college studentsuse of online social network sites,Journal of Computer-
Mediated Communication, Vol. 12 No. 4, pp. 1143-1168.
Facebook (2015), Investor relations, available at: http://investor.fb.com/results.cfm (accessed
August 10, 2015).
Forbes (2013), The future of social media? Forget about the US, look to Brazil, available at:
www.forbes.com/sites/ciocentral/2013/09/12/the-future-of-social-media-forget-about-the-u-
s-look-to-brazil/ (accessed May 28, 2015).
Greene, W.H. (2012), Econometric Analysis, 5th ed., Prentice Hall, NJ and New York, NY.
Gujarati, D.N. and Porter, D.C. (2011), Essentials of Econometrics, 5th ed., McGraw-Hill,
New York, NY.
Hansen, F. (1976), Psychological theories of consumer choice,Journal of Consumer Research,
Vol. 3 No. 3, pp. 117-142.
Hochman, N. and Manovich, L. (2013), Zooming into an Instagram city: reading the local through
social media,First Monday, Vol. 18 No. 7, available at: www.firstmonday.dk/ojs/index.
php/fm/article/view/4711/3698
Hoffman, D.L. and Novak, T.P. (1996), Marketing in hypermedia computer-mediated
environments: conceptual foundations,Journal of Marketing, Vol. 60 No. 3, pp. 50-68.
Kim, D.-H., Spiller, L. and Hettche, M. (2015), Analyzing media types and content orientations in
Facebook for global brands,Journal of Research in Interactive Marketing,Vol.9No.1,pp.4-30.
Luarn, P., Lin, Y.-F. and Chiu, Y.-P. (2015), Influence of Facebook brand-page posts on online
engagement,Online Information Review, Vol. 39 No. 4, pp. 505-519.
Malhotra, N.K. and Birks, D.F. (2007), Marketing Research: An Applied Approach, 3rd ed., Pearson
Education, Upper Saddle River, NJ.
Nam, H. and Kannan, P.K. (2014), The informational value of social tagging networks,Journal
of Marketing, Vol. 78 No. 4, pp. 21-40.
Peters, K., Chen, Y., Kaplan, A.M., Ognibeni, B. and Pauwels, K. (2013), Social media metrics a
framework and guidelines for managing social media,Journal of Interactive Marketing,
Vol. 27, No. 4, pp. 281-298.
Rapp, A., Beitelspacher, L.S., Grewal, D. and Hughes, D.E. (2013), Understanding social media
effects across seller, retailer, and consumer interactions,Journal of the Academy of
Marketing Science, Vol. 41 No. 5, pp. 547-566.
Rauschnabel, P.A., Praxmarer, S. and Ivens, B.S. (2012), Social media marketing: how design
features influence interactions with brand postings on Facebook, in Eisend, M.,
Langner, T. and Okazaki, S. (Eds), Advances in Advertising Research, Vol. III, Springer,
The Netherlands, pp. 153-161.
Sabate, F., Berbegal-Mirabent, J., Cañabate, A. and Lebherz, P.R. (2014), Factors influencing
popularity of branded content in Facebook fan pages,European Management Journal,
Vol. 32 No. 6, pp. 1001-1011.
Schultz, D.E. and Peltier, J. (2013), Social medias slippery slope: challenges, opportunities and
future research directions,Journal of Research in Interactive Marketing, Vol. 7 No. 2,
pp. 86-99.
Shadish, W.R., Cook, T.D. and Campbell, D.T. (2002), Experimental and Quasi-Experimental
Designs for Generalized Causal Inference, Cengage Learning, Boston, MA.
Smith, A.N., Fischer, E. and Yongjian, C. (2012), How does brand-related user-generated content
differ across Youtube, Facebook, and Twitter?,Journal of Interactive Marketing, Vol. 26
No. 2, pp. 102-113.
470
OIR
40,4
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
SocialBakers (2015a), April 2015 social marketing report Brazil, available at: www.socialbakers.
com/resources/reports/regional/brazil/2015/april/ (accessed May 28, 2015).
SocialBakers (2015b), April 2015 social marketing report United States, available at:
www.socialbakers.com/resources/reports/regional/united-states/2015/april/ (accessed
May 28, 2015).
Statista (2013), Statistics and facts about brands on social media, available at: www.statista.
com/topics/2057/brands-on-social-media/ (accessed August 10, 2015).
Statista (2015), Average daily media use in the United States from 2010 to 2014 (in minutes),
available at: www.statista.com/statistics/270781/average-daily-media-use-in-the-us/
(accessed August 10, 2015).
Subramani, M.R. and Rajagopalan, B. (2003), Knowledge-sharing and influence in online social
networks via viral marketing,Communications of the ACM, Vol. 46 No. 12, pp. 300-307.
Swani, K., Milne, G. and Brown, B.P. (2013), Spreading the word through likes on Facebook,
Journal of Research in Interactive Marketing, Vol. 7 No. 4, pp. 269-294.
Van Heerde, H.J. and Neslin, S.A. (2008), Sales promotion models, in Wierenga, B. (Ed.),
Handbook of Marketing Decision Models, Springer, The Netherlands, pp. 107-162.
Weller, K. (2015), Accepting the challenges of social media research,Online Information Review,
Vol. 39 No. 3, pp. 281-289.
White, H. (1980), A heteroskedasticity-consistent covariance matrix estimator and a direct test
for heteroskedasticity,Econometrica: Journal of the Econometric Society, Vol. 48 No. 4,
pp. 817-838.
Winer, R.S. (2009), New communications approaches in marketing: issues and research
directions,Journal of Interactive Marketing, Vol. 23 No. 2, pp. 108-117.
Wooldridge, J. (2013), Introductory Econometrics: A Modern Approach, 5th ed., Cengage
Learning, Boston, MA.
Zellner, A. (1962), An efficient method of estimating seemingly unrelated regressions and tests
for aggregation bias,Journal of the American statistical Association, Vol. 57 No. 298,
pp. 348-368.
About the authors
Ricardo Limongi França Coelho is an Assistant Professor of Marketing at the Federal University
of Goiás. He coordinates the ADMKT Marketing Group at the university. His research interests
include social media, consumer experience and quantitative methods/models in marketing
research. Ricardo Limongi França Coelho is the corresponding author and can be contacted at:
ricardolimongi@gmail.com
Denise Santos de Oliveira has Masters Degree in Business Administration from the Federal
University of Goiás. His research interests include innovation and marketing.
Marcos Inácio Severo de Almeida is an Assistant Professor of Marketing at the Federal
University of Goiás. He coordinates the ADMKT Marketing Group at the university. His research
interests include quantitative methods/models in marketing research.
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
471
Facebook and
Instagram
metrics
Downloaded by Cornell University Library At 14:57 05 August 2016 (PT)
... In this regard, firmgenerated online posts have been classified into product (informational and remuneration posts) and non-product categories (entertainment and social posts) (De Vries et al., 2012;Cvijikj and Michahelles, 2013;Shen and Bissell, 2013;Tafesse, 2015). Some researchers have concluded that more efficient use of social media involves posting non-product content that provides emotional benefits to users, rather than commercial benefits through direct promotion of products (Luarn et al., 2015;Tafesse, 2015;Coelho et al., 2016;Lee et al. 2018;Ferrer-Rosell et al., 2020). ...
... In the context of social media, the aim is to boost customer engagement, an important concept in determining the effectiveness of social media marketing efforts (So et al., 2014). As pointed out by Peters et al. (2013), there are numerous social media metrics, but a common ground for the research is to use data from social media to understand the variation of noneconomic variables, such as likes, comments, replies, and shares and retweets (De Vries et al., 2012;Bonsón and Ratkai, 2013;Cvijikj and Michahelles, 2013;Oviedo-García et al., 2014;Luarn et al., 2015;Bonsón et al., 2016;Coelho et al., 2016;Muñoz-Expósito et al., 2017;Mariani et al., 2018). ...
Article
Full-text available
The aim of this paper is to analyse the effects of the COVID-19 crisis on the accommodation sector by looking at the measures in place, prospects and future strategies. Using a multiple-case-study methodology, an electronically self-administered questionnaire was sent to the managers of 10 establishments located in the province of Huelva (Spain), between March and June 2020. Additionally, a thematic analysis was performed to compare the messages posted on social media over the same four-month period in both 2019 and 2020. The results showed that those establishments belonging to large chains diversified their strategies to obtain better chances of success. Also, reactions to the COVID-19 crisis appeared to differ by type of accommodation. An important change in the online communication strategy during the lockdown in relation to the same period in 2019 was observed. It was not just the different themes of the messages that were in different proportions, but so too were the levels of engagement expressed in each one. This study provides a snapshot of the effect that the economic shutdown had on Spain’s tourist accommodation and contributes to the thematic social media strategies deployed by the accommodation sector in situations of natural disasters and global pandemics.
... Importantly, user engagement is defined as behavioral manifestations toward a brand which are expressions of underlying psychological states resulting from a consumer's interactive relationship with a brand (Brodie et al., 2011). Consumer engagement is usually measured through likes, comments, and shares (Coelho et al., 2016). Likes are a way for consumers to react to brand-initiated posts by not only showing consumers endorsement of brandinitiated posts, but also giving them credibility. ...
Article
Full-text available
With the development of social media, advertising has migrated from traditional media to social media. Marketers are increasingly using social media’s brand pages to actively create humorous dialogue interactions with other brands for brand communication to achieve positive business outcomes. Especially brand-to-brand’s aggressive humor dialogue can also be an effective brand communication strategy. Based on benign violation theory, we have studied the influence mechanism and boundary condition of the brand-to-brand’s aggressive humor styles (low-aggressive and high-aggressive) on consumer engagement behavioral intention in social media context. Through experiments, it is indicated that low-aggressive humor could promote consumer engagement behavioral intention more than high-aggressive humor. Benign appraisal mediates the relationship between low-aggressive humor and consumer engagement behavioral intention. Furthermore, brand personality not only moderates the effect of low-aggressive humor on consumer engagement behavioral intention, but also moderates the mediating role of benign appraisal between low-aggressive humor and consumer engagement behavior intention.
... Future studies can also be carried out using qualitative and quantitative approaches to determine the role of government policies, economic factors, human capital, and the social issues that might explain the mounting number of Instagram-related studies at those academic institutions. There is a need to better understand the role of Instagram in influencing users' behavior as argued by Coelho et al. (2016), who state that Instagram imposes singular engagement forms due to its ability to alter vertical and temporal structures in favor of spatial connectivity. Future scholars may also synthesize and contrast qualitative findings with the quantitative ones presented in this study in order to grasp the motivations and impact of Instagram use. ...
Article
Full-text available
Instagram is a leading social network for information sharing and communication. Rigorous studies are missing that leverage bibliometric techniques to comprehensively portray the field. To fill this knowledge gap, this study carries out a holistic bibliometric and network analysis of Instagram research, illustrating the dynamic evolution from 2013 to 2021. On the basis of 2,242 publications from the Web of Science database, which were authored by 6,206 researchers, this study identifies the most prominent scholars and articles in the literature. Furthermore, it analyzes diverse bibliometric networks, such as citation, co-citation, collaboration and keyword co-occurrence networks, and presents two intellectual structure maps (i.e., conceptual structure map, thematic map). The results indicate that the number of academic studies about Instagram has been growing significantly over time and that the dominant topics are the psychological motivation of Instagram use, the COVID-19 pandemic, Instagram marketing, social media platforms, and healthcare.
... Devido a seus baixos custos de operacionalização, essas plataformas proporcionam que qualquer indivíduo tenha meios de divulgar suas ideias e por isso são consideradas plataformas democráticas (Coelho, Oliveira & Almeida, 2016 Ressalta-se que esses discursos violentos são usados, algumas vezes, de forma estratégica, pois apresentam uma co-construção textual, que reflete as contribuições dos indivíduos, dando uma identidade de grupo, que se denominam web ativistas e que se protegem, em alguns casos, com perfis fictícios (Amossy, 2012;Araújo, Penteado & Santos, 2015;Kerbrat-Orecchioni, 2017;Cabral, 2019). ...
Article
Full-text available
O objetivo geral do estudo consistiu em compreender a dinâmica dos discursos deódio na militância política dos indivíduos nas redes sociais digitais. A partir de uma perspectiva qualitativa, desenvolveu-se uma pesquisa multimétodo, com análise temática de conteúdo para avaliar as motivações da militância e as consequências dos conflitos que ocorrem nas redes e com análise de conteúdo automatizada para identificara incidência de discursos de ódio e os tipos de postagens que causam essa reação. As redes sociais digitais são ambientes democráticos tornando-se um terreno fértil para militância política ideológica, que em seu debate pode gerar discussões violentas e discursos de ódio, o que torna esses ambientes, por vezes, um local violento e tóxico. Dentre os principais resultados teóricos e práticos tem-se a autoconstrução de um refúgio online que corrobora para o aumento do efeito da bolha social e a proposta de duas tipologias, a do discurso de ódio e a das suas consequências na vida do indivíduo.
... In this regard, a number of authors have measured the effectiveness of communication strategies on social networks focusing on the impact of the type of content generated in social networks and have performed comparative analyses of textual content and complementary elements, such as image and video, and their impact on consumer response to this type of content in terms of likes, shares and comments (França-Coelho, Santos de Oliveira and Severo de Almeida, 2016; Kim and Spill-erHettche, 2015;Rauschnabel et al., 2019). Others have focused on how more emotional, informative or promotional discourse affects such metrics in social networks (Coelho et al., 2016;Cvijikj and Michahelles, 2013;De Vries et al., 2012;Swani, et al., 2013). Li and Xie (2020) empirically examine the influence of image content on social media engagement using data sets of social media posts about major airlines on Twitter and Instagram. ...
Article
This comparative study analyses the effectiveness of the communications delivered via Facebook by two Spanish airlines, Iberia and Air Europa. Using various indicators, the publications posted by the two Spanish airline companies (Iberia and AirEuropa) on this social media site during the COVID-19 pandemic are examined using two time-frames, namely the “strict lockdown” (between March 14th and June 21st, 2020) and the “new normal” (between June 22nd and September 30th). In this study we examined a total of 39 Iberia's and 49 AirEuropa's posts for the former period and 49 Iberia's and 89 AirEuropa's posts for the latter one. By analysing the followers' reactions to these posts, the work seeks to identify the relationship between use of different contents and three variables: brand popularity, customer brand engagement, and virality. Two time periods are selected, one spanning national lockdown as a result of the state of emergency declared in Spain and the other relating to the subsequent ‘new-normal’ as emergency restrictions began to be lifted. The results show that the Facebook posts created by Iberia that included informative messages, references to COVID-19, with hashtags, and allusions to corporate social responsibility—were more popular than Air Europa posts. Iberia posts generated greater customer brand engagement and virality, and received more positive reactions in terms of likes and loves.
... It refers to the virtual engagement with an influencer in any social network site (Bond, 2018). According to Coelho et al. (2016), Instagram is considered the best platform for interacting with the audience. Interaction on social media can be analyzed by the number of likes, comments, and shares and the communication between the influencer and its followers. ...
Article
Purpose Instagram health and wellbeing influencers (HWIs) have been increasingly considered as important sources of information and advice for their followers. This study aims to investigate the key antecedents of followers' attitude towards HWIs as well as their influence on their followers' intent to purchase organic products. The moderating effect of gender is also taken into account. Design/methodology/approach Based on data collected from 251 Instagram HWIs followers, the authors empirically tested the conceptual model using structural equation modeling. Findings First, the authors demonstrate that attitude towards HWIs positively impacts followers' attitude towards the promoted brands as well as their intention to purchase organic food brands. Second, followers' attitude towards HWIs is mainly influenced by perceived congruence, influencer credibility, and physical attractiveness. Finally, gender acts as a moderator, e.g. attitude towards HWIs is more likely to be influenced by perceived congruence and physical attractiveness among female followers. Practical implications The findings allow organic brands' managers to understand the key antecedents of followers' attitudes toward HWIs, and therefore, better select talented influencers who are able to create purchase intentions among both existing and potential customers. Originality/value This original research bridges a gap pertaining to the potential use of HWIs to shape consumer intention to purchase organic products. To the authors' knowledge, this study is the first of its kind to investigate the impact of attitudes toward influencers on both brand attitude and purchase intention in the organic food industry.
... Researchers classified Instagram posts into different types based on their perspectives, for instance, Coelho et al. (2016) used five categories for Instagram and Facebook posts; advertising, fan, events, information, and promotion, their findings indicate an increased involvement using events and promotional posts in Instagram. Anagnostopoulos et al. (2018) classified Instagram posts into product-related and non-product-related, their observations favoured product-related posts for a higher number of likes and comments. ...
Article
This study investigates the utilization impact of Instagram on restaurants' brand management in emerging economies, particularly in the capital of Lebanon, Beirut. Instagram utilization will include brand personality, brand awareness, brand retention, brand image, and brand advocates as the dimensions of brand management. A quantitative approach "survey" is adopted to collect primary data. The restaurant sector is operating in the capital of Lebanon, Beirut, using convenience sampling to target 549 consumers of these restaurants. It was seen that using Instagram posts in the restaurant's sector will lead to a positive impact on their brand management overall. Instagram posts have the most significant impact on brand awareness and brand image, whereas brand retention and brand advocates are less affected by using them. An insignificant relationship between Instagram and brand personality was found. This is the first study conducted in Lebanon's restaurants' context, and this is crucial since Lebanon is highly dependent on tourism and travel-related services.
Article
Full-text available
Businesses should implement TikTok as a social media marketing tool. In Indonesia particularly, there are more than 22 million active users. To enhance the social media marketing efficacy on TikTok, there is a need for an empirical study discussing factors driving social media engagement on TikTok. This research aims to investigate the effects of visual complexity and content types on social media engagement (i.e., likes, comments, and shares) on TikTok. A total of 647 posts were collected from 7 business accounts marketing their products to Indonesian consumers. This study employed content analysis, and data were computed using negative binomial regression on SPSS. Results prove that TikTok content with high visual complexity has negative effects on shares. Respond to Comment variable has positive impacts on likes; Product, Respond to Comment, and Review posts have positive effects on comments; Product and Review posts have positive effects on shares. Further, this study also evidences that Respond to Comment and Humor parameters negatively influence shares. Theoretically, this research expands social media marketing literature. Practically, insights from this study can guide brands in creating viral TikTok content that eventually can enhance business performance.
Article
Purpose With the growing trend of omni-channel marketing, brands are increasingly looking to offer a seamless experience to their online fan base by connecting with them across multiple social media platforms. This paper explores the relationship between brand posts' characteristics and popularity for start-ups across four different social media platforms: Facebook, Twitter, Instagram and LinkedIn. Design/methodology/approach A total of 1,200 social media posts from 10 start-ups were subjected to content analysis. Regression analysis was employed with brand posts' popularity (likes, comments and shares/retweets) as the dependent variable. Findings The results reveal several nuances in brand post popularity for start-ups across Facebook, Twitter, Instagram and LinkedIn. Antecedents of the popularity measures of likes, comments and shares/retweets also fared differently. Originality/value The paper reports one of the earliest empirical studies to better understand how the qualities of brand posts are related to their appeal across multiple social media platforms. It advances the literature on social media marketing and offers insights to social media managers of brands, particularly start-ups, on how to offer smoother customer journeys across numerous digital touchpoints.
Article
Full-text available
Purpose –This study examined how various characteristics of brand posts influence online engagement on Facebook brand pages. Design/methodology/approach –The data used for this study was obtained from the posts of 10 popular official brand pages. For the selected period between March 1 and May 1, 2014, a total of 1,030 posts were obtained and manually processed. To assign post categories to the posts created by page administrators, we performed manual coding, following the coding development strategy. Findings –The results demonstrated that the media and content type of posts exert a significant effect on user online engagement. This study used liking, commenting, and sharing behavior as a measure of users’ online engagement to specify the new phenomena. Originality/value –The findings are relevant for the theory of information dissemination and provide valuable and directly applicable implications for the social media marketing of companies.
Article
Full-text available
Social tagging is a new way to share and categorize online content that enables users to express their thoughts, perceptions, and feelings with respect to diverse concepts. In social tagging, content is connected through user-generated keywords-"tags"-and is readily searchable through these tags. The rich associative information that social tagging provides marketers new opportunities to infer brand associative networks. This article investigates how the information contained in social tags can act as a proxy measure for brand performance and can predict the financial valuation of a firm. Using data collected from a social tagging and bookmarking website, Delicious, the authors examine social tagging data for 44 firms across 14 markets. After controlling for accounting metrics, media citations, and other user-generated content, they find that social tag-based brand management metrics capturing brand familiarity, favorability of associations, and competitive overlaps of brand associations can explain unanticipated stock returns. In addition, they find that in managing brand equity, it is more important for strong brands to enhance category dominance, whereas it is more critical for weak brands to enhance connectedness. These findings suggest a new way for practitioners to track, measure, and manage intangible brand equity; proactively improve brand performance; and influence a firm's financial performance.
Article
The authors address the role of marketing in hypermedia computer-mediated environments (CMEs). Their approach considers hypermedia CMEs to be large-scale (i.e., national or global) networked environments, of which the World Wide Web on the Internet is the first and current global implementation. They introduce marketers to this revolutionary new medium, propose a structural model of consumer navigation behavior in a CME that incorporates the notion of flow, and examine a series of research issues and marketing implications that follow from the model.
Article
In this paper a method of estimating the parameters of a set of regression equations is reported which involves application of Aitken's generalized least-squares [1] to the whole system of equations. Under conditions generally encountered in practice, it is found that the regression coefficient estimators so obtained are at least asymptotically more efficient than those obtained by an equation-by-equation application of least squares. This gain in efficiency can be quite large if “independent” variables in different equations are not highly correlated and if disturbance terms in different equations are highly correlated. Further, tests of the hypothesis that all regression equation coefficient vectors are equal, based on “micro” and “macro” data, are described. If this hypothesis is accepted, there will be no aggregation bias. Finally, the estimation procedure and the “micro-test” for aggregation bias are applied in the analysis of annual investment data, 1935–1954, for two firms.
Article
Purpose: Academics and the business community are interested in learning how social media can benefit (or harm) consumer-brand engagement. As more branding activity goes social, marketers are not always welcome in all social media spaces. In this invited commentary, the authors aim to lay out the challenges that social media faces for enhancing consumer-brand engagement. In doing so, they seek to turn social media challenges into future research directions. Design/methodology/approach: The paper reviews prior literature on social media and brand engagement. Findings: The majority of social media marketing initiatives take the form of communicating sales promotions to already engaged consumers. Practical implications: Marketers need to find ways to use social media to create lasting brand engagement rather than to merely utilize this communication technology to enhance short-term revenue. Originality/value: This critical review provides marketing academics and practitioners avenues for future research and applied considerations. It is an adaptation and extension of Schultz's 2013 paper.
Article
Purpose – The purpose of this paper is to introduce a new viewpoint series, Monitoring the Media: Spotlight on Social Media Research, by providing an overview of the key challenges in social media research and some current initiatives in addressing them. Design/methodology/approach – The paper considers publication output from disciplines dealing with social media studies and summarises the key challenges as discussed in the broader research community. Findings – The paper suggests that challenges originate both from the interdisciplinary nature of social media research and from the ever-changing research landscape. It concludes that, whilst the community is addressing some challenges, others require more attention. Originality/value – The paper summarises key challenges of social media and will be of interest to researchers in different disciplines, as well as a general audience, wanting to learn about how social media data are used for research.
Article
Purpose – This study aims to examine current practices of social media marketing among major global brands across five product categories (namely, convenience, shopping, specialty, industrial and service). Assessing the frequency, media type and content orientations of corporate Facebook pages, this study aims to isolate the qualitative factors of a brand’s social media message that are most likely to facilitate a consumer response. Design/methodology/approach – A content analysis of 1,086 social media posts was conducted from the corporate Facebook pages of 92 global brands during a one-month (snapshot) time horizon in July 2013. The data collected from each individual post include its media type (i.e. text, photo or video), its content orientation (i.e. task, interaction and self-oriented) and the number and type of consumer response it generated (i.e. likes, comments and shares). Findings – Research findings reveal that global brands actively utilize social media, posting on average three messages per week and generally use photos (as a media type) and interaction-focused content (as a content orientation) to secure consumer responses. However, differences in consumer responses exist along various product categories, message media type and message content orientation. Practical implications – Findings imply that marketers should not only carefully consider the media type they use to message consumers on social media but should also try to consider the individual consumer’s motive for interaction. Originality/value – This article suggests a new way to study social media content by applying pre-existing communication frameworks from salesmanship literature as a way to define message content orientation.
Article
How are users' experiences of production, sharing, and interaction with the media they create mediated by the interfaces of particular social media platforms? How can we use computational analysis and visualizations of the content of visual social media (e.g., user photos, as opposed to upload dates, locations, tags and other metadata) to study social and cultural patterns? How can we visualize this media on multiple spatial and temporal scales? In this paper, we examine these questions through the analysis of the popular mobile photo-sharing application Instagram. First, we analyze the affordances provided by the Instagram interface and the ways this interface and the application's tools structure users' understanding and use of the "Instagram medium." Next, we compare the visual signatures of 13 different global cities using 2.3 million Instagram photos from these cities. Finally, we use spatio-temporal visualizations of over 200,000 Instagram photos uploaded in Tel Aviv, Israel over three months to show how they can offer social, cultural and political insights about people's activities in particular locations and time periods.