A Calorie for A “ Like ”

(1)

A Calorie for A “Like”

The influence of a social media post on the intention to engage in exercise and the

moderation effect of physical activity and the users’ involvement of social media.

By:

Liesel Zoeger Brambilla

University of Groningen

Faculty of Economics and Business

MSc. in Marketing Management

Master Thesis

June 2019

L.Zoeger@student.rug.nl

Student Number: S3563847

(2)

2

Abstract

Nowadays, we are facing an ever-growing obesity problem. This problem has consequences in health and well-being. Many barriers have been found in order to decrease this health problem. Nevertheless, one new tool that can help fight obesity is emerging and is part of our daily activities. Social media has been added to several field studies as a new and powerful tool aimed to influence users when struggling with obesity.

The purpose of this research is to understand how an intention to engage in physical activity can be boost through social media. An online survey was used to assess the proposed hypotheses and conceptual model. The survey had 150 respondents, each respondent had been randomly assigned to only one of the three types of posts. This helped better understand the different levels of intention users had across the three options. Using different studies and literature two different variables were established that can help increase the intention to exercise through social media: (1) past physical activity, (2) social media engagement. These two variables were used to test a moderation effect between seeing a post and the intention to engage in physical activity. As a second part of the research, a direct consequence of the intention to exercise was tested, that is the intention to post about doing physical activity.

The findings showed that user’s that saw a friend’s post about physical activity had a higher intention to exercise than the control group. Although there was no moderation effect with the stated variables, a new moderation variable was added later. The “age” variable showed that when using social media, the group called “millennials” had a higher response to a social media post about physical activity than “mature” group.

Key words: Social Network Sites, Social Media Engagement, Physical Activity, Intention to Exercise,

(3)

3

Chapter 1: Introduction

"Physical fitness is not only one of the most important keys to a healthy body, it is the basis of dynamic and creative intellectual activity."

(Kennedy, 1960)

Increasing well-being is not just important for oneself but also for society. Well-being can be described as a personal growth that should involve health, happiness, and prosperity (Mick, Pettigrew, Pechmann & Ozanne, 2012). But for the past years, the pressures of the modern lifestyle have made people not to focus on their own well-being. One of the consequences of this lack of focus is obesity. Which has been established as one of the most crucial health issues through different countries because of its high rates which are continually increasing (World Obesity Federation, 2019). According to the World Health Organization ([WHO], 2018), as of 2016, about 13% of adults worldwide were considered obese.

Obesity results from an imbalanced consumption of energy caused by genetics, low level of physical activity, sedentary behavior, or overconsumption of food (WHO, 2018). High obesity rates can have a negative psychological and physical effect and consequently can decrease health in a never-ending vicious cycle. According to the WHO (2018), overweight and obesity are largely preventable. Three possible ways to reduce obesity have been exposed: through education, the balance of food consumption, and physical activity (WHO, 2018).

(6)

6 education and food consumption are important variables for decreasing obesity, this study will focus on physical activity as the main driver to increase well-being by using social media.

Online interactions can help improve physical activity due to peer competition which can be intensified because of social comparison on social media (Centola, 2011; Zhang et al., 2016). Nowadays, social network sites work together with health and fitness apps. This can help provide accurate information and guidance to users and consequently, increase their physical activity by using personalized programs. By generating social media interaction through these apps, a larger quantity of users can benefit from the data provided by their peers. Overall, social network sites can be used as a channel aimed to promote health and well-being by increasing physical activity and consequently, decreasing obesity (Becker et al., 2014).

Even though social networks sites make people feel more connected and able to access rapidly to information, these can have a reverse effect on health. Social media can generate low involvement with real-life environment by making users highly motivated to live for posting online. Therefore, they can trigger consequences in users such as less connection to real-life peers, low self-esteem, depression, anxiety and loneliness (Oberst, Wegmann, Stodt, Brand & Chamarro, 2017).

All in all, many studies have shown that decreasing levels of obesity can improve well-being. However, many of these studies have missed to include social media as a new tool to influence physical activity and consequently decrease obesity. Although there have been no empirical results about the impacts of social media on changing and improving physical activity, many users are interested in interacting on social network sites about health and fitness topics. The number of social network sites related to physical activity has increased but the causal and the interactive effects of social network sites and physical activity have not been established (Cavallo et al., 2012). Thus, it is important to assess if social media interaction can improve the chances of engaging in physical activity.

1.1 Managerial relevance

(7)

7 Leading industries that can benefit from data collection are insurance companies and health care providers. According to Carroll et al. (2017), there is increasing usage of social network sites focused on health, and it can help with data collection and self-tracking. Social network sites’ users are actively providing and sharing their health information through health and fitness apps. Through these fitness apps, users share data that is being used in order to know about the individual and, at the same time, to formulate general conclusions (Crawford, Lingel & Karppi, 2015). The usage of big data provided by social network sites and health and fitness apps can help marketers and health care professionals to educate consumers and contribute to the user’s well-being by offering personalized products.

Social media focused on health and fitness is present in more than 100,000 health apps (Carroll et al., 2017). The usage of this kind of apps is continually growing and can have a positive effect for companies in the health and insurance industry because of the low-cost data when learning about the customer and the market (Chen, Chiang & Storey, 2012). According to a study regarding information sharing done by Houdek VonHolts et al. (2015), participants that specified they share information from health and fitness apps (59%), they prefer to share it through social networks (29%) and not so much with their health care providers (17%). Social media users have a higher intention to express personal health information through social media than with their own health providers. This indicates that there is an opportunity for companies to benefit from social media data that is now being neglected.

(8)

8

1.2 Research question

This thesis aims to examine and learn whether seeing a post on social media related to physical activity will generate an intention to exercise. Framework research about literature has been done together with an experimental study to answer the question presented below:

“Does seeing a post related to physical activity on social media, moderated by the level of physical activity together with users’ social media involvement, increase the intention to engage in exercise and

consequently to post the results on social media?”

(9)

9

Chapter 2: Theoretical framework

In this chapter, the background problem and the variables that will be considered for the research are provided together with information from previous studies and literature. At the same time, this chapter presents the hypotheses that will be used for the study together with the conceptual model.

2.1 Health, well-being and welfare

According to the WHO (1948), “health is a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity”. There are two types of perspectives on well-being: hedonic and eudaimonic. According to Kahneman, Diener & Schwarz (1999), hedonic well-being focuses on instant happiness by the reaching of pleasure and prevention of pain. On the other hand, there is eudaimonic being, which focuses on the actualization of human potential and defines consumer well-being in terms of the degree to which people are realizing their true nature (Ryan & Deci, 2001). In other words, being the best a person can be. Consumers are constantly deciding which perspective of well-being they are going to adopt. For example, drinking soda or green tea, one contains high amounts of sugar but gives instant pleasure and the other has antioxidant properties that have a long-term positive effect on the body. Looking at these examples, one can easily assume that consumers would choose the one that has better benefits to the body. According to Trope, Liberman & Wakslak (2007) consumers prefer smaller and sooner benefits over later and large ones. This helps us understand why consumers sometimes prefer vice food over virtue food.

Being healthy compromises the mind, body, and society. Hence, one might believe that consumers understand that helping society is, consequently, helping oneself. Nevertheless, there are some people that have a propensity to make decisions that outweigh individual benefits but discount social costs (Burroughs & Rindfleisch, 2011). For example, increasing the consumption of red meat is highly linked to heart disease. According to Burroughs & Rindfleisch (2011), thinking about health is also thinking about welfare, which is defined as the total sum of the well-being of all the citizens that comprise a society. Beyond any doubt, personal and collective welfare are highly related, and together they can increase well-being. Therefore, consumers need to focus on balancing hedonic and eudaimonic perspectives in order to increase their own welfare. At the same time, consumers need to think about the social consequences of their health decisions.

(10)

10 care system has become a burden for many countries. Health policies have become more expensive, and diseases like obesity, diabetes, osteoarthritis, hypertension, among others, have become common in society. Those diseases can be due to overconsumption of unhealthy food, living a sedentary lifestyle, low physical activity, among others.

2.2 Obesity, a world’s problem

Obesity is a disease in which large accumulation of fat in the body presents a risk for several chronic diseases which have negative consequences on the health (Tremmel, Gerdtham, Nilsson, & Saha, 2017). As a result, it can seriously decrease life expectancy. Obesity increases health care costs, but according to Tremmel et al. (2017), it is not the only economic expenditure it generates. It also generates costs like loss of productivity due to reduced labor days, permanent disability, and finally, leading to mortality (Tremmel et al., 2017). “In 2014, more than 2.1 billion people, nearly 30% of the global population, were overweight or obese and 5% of the deaths worldwide were attributable to obesity” (Tremmel et al., 2017, p. 435). These consequences of obesity pressure to settle this disease as a health priority, aiming to have a better diet and increment physical activity. Inadequate food consumption together with lack of exercise is attributed to the leading causes of obesity (Milani & Lavie, 2015). The recent modifications on the environment, as well as on the society have brought changes to a person’s diet and physical activities. Those changes are linked to the “development and lack of supportive policies in sectors such as health, agriculture, transport, urban planning, environment, food processing, distribution, marketing, and education” (WHO, 2018). Hence, many changes are needed in order to decrease obesity among society like increased education, lower calorie intake, and boost physical activity.

Although obesity is one of the greatest concerns, there is still no government that has been outstanding in decreasing obesity in their country (Kleinert & Horton, 2015). The absence of evidence has emphasized the urgency for improving research about obesity solutions (Kleinert & Horton, 2015). The WHO together with the World Health Assembly presented a “Global Action Plan for Prevention and Control of

Noncommunicable diseases 2013-2020” which included obesity as a principal disease (WHO, 2013). There

(11)

11 actions and too little political assessment of the effects of many programs and policies” (Roberto et al., 2015, p. 2401). All things considered, one possible way to decrease obesity is by increasing knowledge and lowering barriers and that can be obtained through the use of social network sites.

2.3 Social network sites as promoters of health

Internet has evolved from pure information retrieval to a social platform that has increased interaction and communication options around the world (Campbell, Pitt, Parent & Berthon, 2011). When webpages started providing users with the capability of interaction, social network sites were born. Social network sites can be described as an interactive service provided through internet (Ellison, 2007). This service allows internet users to create a profile, manage a directory of peers in order to connect, lastly, take a connect with new users that are within the system (Ellison, 2007).

Social media is not only used to communicate or share information, but customers are also able to learn about health topics by using these platforms (Edelman 2010). Social media users search the web for health topics for two main reasons: information and interaction with other users that have similar conditions (Magnezi et al., 2015). Cavallo et al. (2012) established that there is a high number of adults using social network site, more specifically, “more than one-third of adults” possess an account. This makes social media a potential tool to improve health education and even generate a potential behavioral change. More accordingly, social media can have a direct impact on a user’s activities regarding health (Consolvo, Everitt, Smith & Landay, 2006). According to Vandelanotte (2016), users’ interactive interventions can trigger greater usage and engagement with social network sites. Therefore, people’s behavior can be changed and improved by triggering the motivation with the incorporation of cultural dimensions and social context (Nicholson 2012).

(12)

12 occurs when we compare ourselves to someone we perceive as ‘better-off’ (Dibb, 2019). Therefore, a comparison can have positive consequences on the intention to exercise when seeing someone else doing it. Thus, the following hypothesis is established:

H1: Seeing a friend’s post related to physical activity on social media increases the intention to engage in physical exercise.

2.4 Physical activity as a driver of health

It is well known that physical activity can lead to good health and that also improves well-being by decreasing the levels of obesity around the world. Nevertheless, many people outweigh their daily activities over physical activity. Physical activity is defined by the WHO (2011b) as a “bodily movement that requires energy”. This means that it is not necessary to follow a planned schedule of fitness activities in order to improve physical activity. Physical activity can be increased by changing daily actions like walking or biking to work, using the stairs instead of the elevator, among others.

Physical activity has positive benefits for the health, for example, can improve the body functionality, increase the metabolism, help the cardiorespiratory fitness, and overall prevents premature death (Luzak, et al., 2017). There have been several policies and promotions by different entities that promote the adoption of physical activity as part of their daily routines. One entity that has advertised physical activity is the WHO. By making different studies, the WHO established a minimum of physical activity to increase well-being. According to the WHO (2011b), “adults aged 18–64 should do at least 150 minutes of moderate-intensity aerobic physical activity throughout the week or do at least 75 minutes of vigorous-intensity aerobic physical activity throughout the week or an equivalent combination of moderate- and vigorous-intensity activity”. By meeting this minimum, physical activity can enhance psychological and physical well-being and increase life quality regarding health issues.

(13)

13 generate athletic behavior and participate in physical activity while being adolescence to ensure future well-being.

All things considered, living a non-active lifestyle as an adolescent can help development obesity, making adults less prone to exercise, and having a low-calorie expense, consequently, increasing their fatness (Pietiläinen et al., 2008). Therefore, it is important to change consumers’ behavior since their adolescence, even though there is no fast or easy way to go about it. Also, it is a task that involves the community, environment, government, and overall, the consumer. Thus, the following hypothesis is established:

H2 (Moderator effect): The effect of seeing a friend’s post related to physical activity on social media on the intention to engage in physical exercise is moderated by physical activity. More specifically, it is

expected that the effect is stronger among those with higher levels of physical activity.

2.5 Social media involvement

Social network sites have triggered important changes regarding the approach users have in order to communicate and interact with others (Pantic, 2014). There has been an important change in how people socialize, shifting from offline communications to an ever-present online interaction by using different social platforms. Roughly 2,000 million internet users use using social network sites, it is expected that it will grow because of the increasing usage of mobile devices and mobile social networks (Statista, 2019b). In addition to an increase in users adoption of online behavior, the time users spend on social media platforms increased by 37% in 2012 (Lim, Al-Aali & Lim, 2013). As of 2017, the daily navigation on social media was more than two hours per day (Statista, 2019c).Due to the constant presence of social media in users’ lives, there is a constant misunderstanding of when life is offline and online, and this has a high social impact. The easy connection to social network sites together with the necessity of always staying online have made users highly involved with social network sites.

(14)

14 interaction with the site, will have a higher deposition to store important information and consequently use social media as a criterion for decision-making (Lim et al., 2013).

A users’ participation with social media is associated with social conformity, peer-recognition, and sense of belonging on the social media community (Lim et al., 2013). The presented experiences can boost users’ involvement with social network sites (Lim et al., 2013). Therefore, physically active users with high social media involvement will have a large interest in physical activity topics within social media. For example, they will be searching, interacting and posting about physical activity. Following the previous statement, the next is hypothesized:

H3 (Moderator effect): The effect of seeing a friend’s post related to physical activity on social media on the intention to engage in physical exercise is moderated by the level of social media involvement. More specifically, it is expected that the effect is stronger among those with high social media involvement.

2.6 Upward social comparison

Even though social network sites have the potential to improve societal well-being, there are also negative consequences. The constant access to the internet through smartphones has increased social problems which can be anxiety, high stress, low self-control, loneliness, among others (van Deursen, Bolle, Hegner & Kommers, 2015). Some negative consequences of the increasing social media activity are information overload, privacy issues, self-diagnosis, social comparison, among others. The last consequence is important to this study because social comparison made through online social networks can generate positive as well as negative consequences for one’s own well-being. This study focused on social comparison as a motivator to post physical activity goals and results on social media.

(15)

15 2009). Finally, for a feeling of being connected. Feeling connected can help users share experiences which they consider not easy to tell offline (Vaterlaus, Jones, Pattern & Cook, 2015).

All in all, interaction with social network sites peers can encourage physical activity by boosting the users’ desire to be more likable. At the same time, when the intention to participate in physical activity has been generated, posting on social media can generate a feeling of fulfillment and increase emotional empowerment by upwards social comparison. Therefore, users might feel the motivation to post their physical activity for other users to see. Following the previous statement, the next is hypothesized:

H4: The intention of engaging in physical activity has a positive effect on the likelihood of posting the results on social media.

2.7 Conceptual model

The previous researched literature together with the stated hypotheses helped develop the conceptual model, which is presented in Figure 1. The model shows that there is a positive effect on the intention to engage in exercise by seeing a friend posting about physical activity on social media. Also, this model presents the positive moderation effect of physical activity and social media involvement on the intention to engage in exercise. Finally, this study also hypothesized that once the intention to engage in exercise is established, there is a positive effect on the intention to post the results on social media.

(16)

16

Chapter 3: Research design

Chapter three focuses on describing the methodology applied to test the previously mentioned hypotheses together with the conceptual model. First, the research method is presented, next, the variables’ description and measurement are stated. Lastly, the sample and data collection are presented.

3.1 Research method

The purpose of this research is to investigate whether seeing a post on social media about physical activity has a positive effect on the willingness to engage in physical activity, and also investigates the moderation effect of physical activity and the users’ involvement with social media. Additionally, this study focuses on the intention to post about physical activity on social. This study is based on quantitative research focused on the previously mentioned variables. In order to test this relation, a cross-sectional survey has been used as a primary tool.

Online surveys can be an important source for researchers who want to study people in their own social and cultural conditions without the usage of laboratory settings (Eysenbach & Wyatt, 2002). The usage of online surveys, in the form of a questionnaire designed for self-completion, is a good way to collect information from a big sample. The online survey has been divided into four blocks and has been modeled with multiple-choice questions. One of the three different conditions within the online survey were randomly assigned to the participants. By assigning people to one of the different random conditions, the causal relationship between seeing a post on social media and the intention to exercise can be determined (Aronson, Wilson & Brewer, 1998). This permits the accurate manipulation of the independent variable and to exclude possible intrusiveness of other independent variables (Aronson, et al., 1998).

Although the social media population is unrepresentative of the global population, it can be a good source for assessing online social media research. Data obtained through an online survey can be related, in terms of the validity and reliability, to the data obtained in different classical methods (Eysenbach & Wyatt, 2002). The questionnaire will be self-completed in an anonymous way to enhance the willingness to share information and therefore increase reliability.

3.2 Variables and measurements

(17)

17

3.2.1 Intention to exercise after seeing a friend’s post

The effect of seeing a friend’s post about physical activity was measured through three randomized blocks, see Table 1 for the visual representation. Each participant could only see one condition. Two out of the three blocks were used to measured two different activities: non-physical activity and physical activity. Two different pictures and each with its respective passage was used for the non-control groups. The first passage asked to see a friend’s post about reading a book. The second passage asked to see a friend’s post about physical activity. Additionally, the third randomized block was the control group which did not include any passage. The dependent variable “intention to exercise” was measured by the 7-point Likert scale questions, ranging from (1) Strongly disagree to (7) Strongly agree. Each of the three conditions included the same two questions about intentions to read a book and engage in physical activity.

(18)

18

3.2.2 Moderation effect of physical activity

Physical activity and health perception were measured by using a 7-point Likert scale provided by a study done by Hoj et al. (2017), where physical and health perception questions were asked. The second question “What do you do to stay healthy?” was taken from 8Fit-app webpage regarding things users do to stay healthy.

For the next two questions, the number of days spent doing physical activity was retrieved from the WHO’s (2011b) webpage and was distributed between a 4-scale question using days as a measure. Additionally, a study done by Luzak et al. (2017) was considered, which comprised the duration patterns of physical activity. The questions were adapted to fit the purposed of the study.

3.2.3 Moderation effect of social media involvement

On the first block of the questionnaire, social media involvement was measured. For the question regarding “Which social media is used?” the top social media platforms were retrieved from Statista’s webpage (2019b).

The scale and measurement of social media involvement were taken from a study done by Amaro & Duarte (2015) and adapted to fit the purposes of this study. A good indicator of high social media involvement is social media time consumption and content creation behavior (Amaro & Duarte, 2015). Therefore, different questions regarding the likelihood of content creation and time spent on social media were asked on the 7-point Likert scale.

3.2.3 Posting results on social media

Together with the dependent variable “intention to exercise”, the dependent variable “intention to post on social media” was measured. The three randomized conditions included two additional questions regarding the intention to post on social media about reading a book and the intention to post on social media about doing physical activity. Both questions had a 7-point Likert scale ranging from (1) Strongly disagree to (7) Strongly agree.

3.2 Sample and data collection

(19)

19 “Do not use social media” and did not finish the questionnaire, therefore, the final sample was composed of 150 respondents.

(20)

20

Chapter 4: Analysis and discussion

This chapter focuses on the results and findings of the online survey. In the first part, a pre-analysis is presented. Which consists of the reliability analysis and the descriptive statistics of the variables considered. On the second part of this chapter, each of the hypotheses is analyzed with the support of the SPSS program. Lastly, a discussion section is provided with a summary of the results.

4.1 Pre-Analysis

Before doing any analysis for the hypotheses, the different variables were measured for a better understanding of the data.

4.1.1 Factor analysis and reliability analysis

A Factor Analysis and a Reliability Analysis of the moderation variables “Social media involvement” and “Physical activity” was carried out. This paper evaluates the Cronbach’s alpha for internal consistency. Social media involvement was comprised of six items based on question Q3 (1-6) (see Appendix #1). A Factor Analysis was done which showed a KMO of 0,668 and p = 0,000. Although the KMO was sufficient, the Communalities for the question “I use social media for information searching purposes” was not sufficient and had an extraction of 0,115, which was not accepted. A second Factor Analysis showed a KMO of 0,665 and p = 0,000. Like the last time, the Communalities for the variable “I feel like I spend too many hours in social media”, showed an extraction of 0,395, which is not sufficient. A third Factor Analysis was done without the previous item, which showed a KMO of 0,671 and p = 0,000, this time all Communalities were accepted. Next, a Reliability Analysis was done for the four items and showed a Cronbach’s alpha of α = 0,757 (N=150). The final social media involvement variable was computed as a mean of four items with a mean of 3,223.

Physical activity factor was comprised by two-items, “Past physical activity” in question Q9 and “Present physical activity” in question Q10. A Factor Analysis was done and showed a KMO of 0,500 and p = 0,000. Next, Reliability Analysis was done for the two items and presented a Cronbach’s alpha of α = 0,628 (N=150). The final physical activity variable was computed as a mean of two items with a mean of 2,697.

4.1.2 Descriptive statistics

(21)

21 older. To further analyze the data, age was grouped into two classes “Millennials” (18-34) and “Mature” (35 to older). “Millennials” group account for the majority of the sample, with a 51% of participation. The occupation variable was comprised of 72% of respondents who work. Table 2 presents the results for the demographic variables.

Table 2: Descriptive Statistic for demographic variables

(22)

22 descriptive statistic is presented in Table 3. The overall mean of the variable “Intention to exercise” is 3.74, which regarding the coding, would be placed between “Somewhat disagree” and “Neither agree or disagree”. The overall mean of the variable “Intention to post” about physical activity is 4.6, which regarding the coding, would be placed between “Neither agree or disagree” and “Somewhat agree”.

Table 3: Descriptive Statistic for overall intention to exercise and intention to post

As this was an experiment condition, 3 different randomized conditions were elaborated and only one per each respondent was presented. The descriptive statistics are shown in Table 4. The overall mean for “Intention to exercise” after seeing an exercise-related post is 3.69, which regarding the coding, would be placed between “Somewhat disagree” and “Neither agree or disagree”.

(23)

23 The variable “Intention to post on social media” about doing physical activity after seeing a physical-related post was measured with a 7-point Likert scale, ranging from (1) Strongly disagree to (7) Strongly agree. As well as “Intention to exercise”, this variable had 3 different randomized conditions and only one was presented to each respondent. The descriptive statistics are shown in Table 5. The overall mean for “Intention to post on social media” after seeing a physical-related post is 4.33, which regarding the coding, would be placed between “Neither agree or disagree” and “Somewhat agree”.

Table 5: Descriptive Statistic for intention to post

(24)

24 Table 6: Descriptive Statistics for social media involvement and physical activity

4.2 Assessing the hypotheses

The hypotheses were tested with Correlation Analysis, One-Way ANOVA and Regression Analysis by Haynes. This allows assessing the effects of the dependent variable from the independent variable (Field, 2009), as well as the moderation effect. The proposed model is as follows:

Intention to exercise = βo + β1PostPhysicalActivity + β2PhysicalActivity + β3SocialMedia +

β2PhysicalActivity*M + β3SocialMedia*W + e

4.2.1 Intention to exercise

H1: Seeing a friend’s post related to physical activity on social media increases the intention to engage in physical exercise.

In order to test the intention to exercise across the different types of posts, a new dummy variable was constructed by the name “Type_of_Post”. This variable indicated which type of post the respondents have seen. This variable was derived from a combination of the three conditions measured on the questionnaire: reading a book = 1, physical activity = 2, and the control variable = 0. In order to identify the differences in the variance of intention to exercise across these different types of posts, a one-way ANOVA test was conducted. The ANOVA test is significant (p = .000) which indicates that at least one group condition is different (see Table 7).

(25)

25 In order to know which groups are not the same a post-hoc analysis was performed, which consists of Homogeneity of Variance. The Homogeneity of Variance test is not statistically significant (p = .547) which indicates that equal variance is assumed (see Table 8). Consequently, Tukey and Bonferroni tests were used (see Appendix #2). This indicates that when seeing a post about physical activity compared when seeing a post about reading a book there is a statistically significant difference in the intention to exercise. Nevertheless, the post-hoc test showed no statistically significant difference when seeing a post about physical activity and the control group on the intention to exercise. Figure 2 shows the mean of three types of post regarding the intention to exercise.

Table 8: Homogeneity of variances

*The main difference is significant at the 0.05 level

Figure 2: Intention to exercise across different types of post 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Control Variable Book Physical Activity

(26)

26 On the next step, three dummy variables were created: Control, Post_Book, and Post_PA. They were recoded into different variables in order to gain two levels. Then a Regression Analysis was performed where the Control group was not included (see Table #9). The results were significant (p = ,000) and each variable was also positive and significant. Post_PA has a b = 0,814 and p = ,028. This means the intention to exercise will increase in 0,814 when seeing a post about physical activity in comparison to the control group. Post_Book as compare to Control will have a 1,767 unit increase on the intention to exercise (see Table #10). Based on the results, hypothesis #1 is supported.

Table #9 – One-way ANOVA

Table #10 – individual significance

(27)

27 relationship between the type of post and intention to exercise is moderated by age (see Table 11 for details). “Millennials” were classified as 0, and “mature” as 1. When age is low (millennials) there is a significant positive relationship between type of post and intention of exercise, b = ,818, CI [0,288 , 1,348], t = 3,049, p = 0,002. When age is high (mature) no significant relationship was found (p = 0,940). These results tell us that the relationship between the type of post and the intention to exercise is different for the two levels of age. Specifically, for millennials, as the type of post shows physical activity the level of intention to exercise also increases (see Figure #3 for intention to exercise moderate by age).

Table 11 - moderation results from age

(28)

28

4.2.2 Intention to exercise moderated by physical activity

H2 (Moderator effect): The effect of seeing a friend’s post related to physical activity on social media on the intention to engage in physical exercise is moderated by physical activity. More specifically, it is

expected that the effect is stronger among those with higher levels of physical activity.

On the first step, a Correlation Analysis was performed. Table 12 shows that the relationships between intention to exercise and type of post is statistically significant (p = ,036) with a Pearson Correlation of ,171. At the same time, intention to exercise and exercise are statistically significant (p = ,003) with a Pearson correlation of ,240. On the other hand, the relationship between the type of post and exercise is not significant (p = ,442) with a Pearson correlation of -,063.

Table #12 – Correlation for exercise

(29)

29 *Confidence intervals significant at the 0.05 level

**Confidence intervals not significant at the 0.05 level

Figure 4: Moderation result for physical activity

4.2.3 Intention to exercise moderated by social media involvement

H3 (Moderator effect): The effect of seeing a friend’s post related to physical activity on social media on the intention to engage in physical exercise is moderated by the level of social media involvement. More specifically, it is expected that the effect is stronger among those with high social media involvement To test if there is evidence about the effect of social media involvement as a moderation between seeing a post about physical activity and the intention to exercise, a correlation analysis was performed. The correlation coefficients were studied, that is, the links between each of the variables. Table 13 shows that the relationship between the intention to exercise and the type of post are statistically significant (p =

,036) with a Pearson Correlation of ,171. At the same time, intention to exercise and social media

(30)

30 Table #13: Correlation for social media involvement

Next, to test the moderation effect of social media involvement a Regression Analysis via Process by Hayes was done. The overall test was significant (p = ,015). Social media involvement is significant (p = ,015) and explains that for every 1 unit increased in social media involvement the intention to engage in exercise increases with 0,334. The interaction effect of type of post with social media (X * W) is not significant (p

= ,698). Therefore, social media has no moderation effect and hypothesis #3 is rejected (see appendix #5

for Hayes Matrix). Figure #5 best describes the results.

*Confidence intervals significant at the 0.05 level **Confidence intervals not significant at the 0.05 level

(31)

31

4.2.4 Intention to post in social media about the results of physical activity

The second part of the hypothesis included the direct and positive effect of intention to exercise on the intention to post the results on social media. The proposed model is as follows:

Intention to post = βo + β1IntentionToExercise + ℮

H4: The intention of engaging in physical activity has a positive effect on the likelihood of posting the results on social media.

To test the effect of engaging in physical activity into the likelihood of posting the results on social media, a Regression Analysis was done. To assess the quality of the model both R square (0,185) and adjusted R squared (0,179) were considered, and both are adequate. The overall regression is significant (see Table #14). The variable intention to exercise has a positive effect on the intention of posting on social media. The beta of the intention to exercise is 0,395 and significant (p = ,000). Which can be translated as when the intention to exercise increase in 1 unit, the user’s intention to post on social media increases in 0,395. (see Table #15). Based on the results, hypothesis #4 is supported.

Table #14 – ANOVA

(32)

32

4.3 Additional analysis

To further explore the possibilities of different variable affecting the model a mediation test was done. The variables used were Type_of_Post as the independent variables, Intention_to_Exercise as a mediator, and Intention_to_Post as the dependent variable. To test the effect a Regression Analysis via Process by Haves was done (see appendix #6 for Hayes Matrix). The effect of Type_of_Post on the Intention_to_Exercise showed up by a significant moderation effect, b = ,4116, CI [,0275 , ,7957], t = 2,118 , p = ,359. The effect of Intention_to_Exercise on the Intention_to_Post showed up by a significant moderation effect, b = ,4102, CI [,2739 , ,5466], t = 5,994 , p = ,000. The effect of Type_of_Post on the Intention_to_Post is not significant (p = ,8048). This means that there is no direct effect of the Type_of_Post on the Intention_to_Post, but there is a mediation effect between the three variables. Figure #6 best describes the results of the mediation effect.

*Confidence intervals significant at the 0.05 level **Confidence intervals not significant at the 0.05 level

(33)

33

4.4 Discussion

This study focused on testing the variables which predict the intention to engage in exercise and consequently post on social media. Table #16 shows a summary of the hypotheses tested.

Table #16: Summary of hypotheses

Although the two proposed moderators did not have a significant effect, they had a direct effect on the intention to exercise. Additionally, it was found that the age group “Millennials” did have a moderator effect between seeing a post on social media and the intention to exercise. Based in the results presented on this chapter, the final conceptual model based on the supported and rejected the hypothesis is presented in Figure 7.

(34)

34

Chapter 5: Conclusions

This chapter focuses on the conclusion of the presented research. First, key findings are specified. In the second part of this chapter, an indication of the research implications is presented. Lastly, the research limitations, along with future research are mentioned.

5.1 Key findings

The purpose of this paper is to contribute to existing literature by adding outcomes based on the study done about social networks sites. Specifically, how social media can increase physical activity and consequently decrease obesity. Previous researches already mentioned physical activity as an influential tool to decrease obesity, but they missed to include social media as a new tool to improve the intention to exercise. The results show that indeed when people see a post about physical activity on social media there is a positive and direct relationship with the intention to exercise. The intention of this study is to help to build a conceptual framework with variables that can influence the type of post’s effect on the intention to exercise. Two out of four hypotheses were supported at 95% confidence level.

Although the proposed moderation variables were not statistically significant, another controlled variable was used in order to seek moderation effect. Age was used as the moderation effect between seeing a post about physical activity and the intention to exercise. These results showed that the millennials group are much more influential by a post about physical activity than the mature group. Meaning that across the three different types of posts when millennials see physical activity they engage in a higher intention to exercise than when seeing a book or in the control group. As for the mature group, there is no significant difference between the three different types of posts.

Lastly, when the intention to exercise is already achieved there is a direct and positive effect on the intention to post the results. This could mean that there is a possibility that social media could generate an incremental effect. This could be interpreted as when a user sees a post about physical activity and consequently post about an intention to engage in exercise that in turn can generate, on other users, the intention to exercise.

5.2 Implications

(35)

35 other hand, managerial implications emerged from the study regarding the variable “age”. Moreover, the study showed that there is a different result between the two age groups into the intention to exercise when seeing a post about physical activity. This is an important distinction because the Millennials are a group that is starting or has a few years into the working force. Therefore, health care and insurance companies can start persuading this group into engaging in more physical activity via social media. Consequently, it is important for marketers to start focusing on the Millennials group, and according to the result, social media could be a helpful tool. Although, as mentioned in the first part of this study, many users prefer to share information on social media than with health care providers. This is an implication for the marketing department, they must close this gap in order to benefit from user’s data. At the same time, this study could be a good background for health care and insurance companies to consider developing appropriate apps that work with different social media platforms.

5.3 Limitations and future research

The first two limitations are regarding the demographic variables. The size of the sample was comprised of 150 respondents. The limited size of the sample can restrict this study as it could be considered low in order to make a general conclusion from the results provided. The nationality of the respondents was in their majority (80.7%) from the country Peru. This factor can influence the general conclusions regarding intentions due to cultural and regional traits.

Additionally, this study focusses in both the intention to exercise and the intention to post, which is a limitation of this paper. Although there are multiple studies that indicate that intention predicts behavior there is still some studies that show that changing an intention does not necessarily change the behavior (Sheeran & Webb, 2016).

Another limitation is the type of post that has been used on the three-randomized groups. For example, the first photo about reading a book showed part of the body of a person but the second photo which was about physical activity only showed numerical results. This might also affect the results of the intention to exercise because one picture can be more influential than the other one.

(36)

36 actual behavior of physical activity could be included in the study. A laboratory experiment could work better to understand the behavior of each participant when seeing a social media post.

One important future study is to find different moderations or mediations effects of other variables. This would help understand what factors can help enhance the intention to exercise or better yet the actual behavior of exercise. For example, environmental factor, different psychosocial factors, among others, can be used as moderation or mediation variables.

An interesting approach for future studies would be understanding different types of social media posts that better boost the intention to exercise. For example, if seeing the numerical results or a person doing the physical activity is more effective than the other, at the same time if this person is a friend or a blogger/influencer. Therefore, marketers can have a better approach to influence the intention and hopefully the behavior of those users. Additionally, further studies that include social media as a tool should be done by studying only “Millennials”, because according to this study they are more influential by social media.

(37)

37

References

Adams, P., Baumer, E. P., & Gay, G. (2014). Staccato social support in mobile health applications. In

Proceedings of the SIGCHI conference on human factors in computing systems. pp. 653-662.

Amaro, S., & Duarte, P. (2015). Social Media Involvement: A Proposed Measure. Information and

Communication Technologies in Tourism. pp. 213-226.

Appel, H., Gerlach, A. L., Crusius, J. (2016). The interplay between Facebook use, social comparison, envy, and depression. The current opinion in psychology. Volume 9, pp. 44-49

Aronson, E., Wilson, T., & Brewer, M. (1998). Experimentation in Social Psychology. The Handbook of

Social Psychology. 1:99-142

Becker, S., Miron-Shatz, T., Schumacher, N., Krocza, J., Diamantidis, C., & Albrecht, U. V. (2014). mHealth 2.0: experiences, possibilities, and perspectives. JMIR mHealth and uHealth, 2(2), e24.

Burroughs, J.E., & Rindfleisch, A. (2011). What welfare? On the definition and domain of transformative consumer research and the fundamental role of materialism.

Campbell, C., Pitt, L., Parent, M., & Berthon, P. (2011). Understanding consumer conversations around ads in a web 2.0 world. Journal of Advertising, 40(1), pp. 87-102.

Carroll, J. K., Moorhead, A., Bond, R., LeBlanc, W. G., Petrella, R. J., & Fiscella, K. (2017). Who Uses Mobile Phone Health Apps and Does Use Matter? A Secondary Data Analytics Approach. Journal of medical Internet research, 19(4), e125.

Cavallo, D. N., Tate, D. F., Ries, A. V., Brown, J. D., DeVellis, R. F., & Ammerman, A. S. (2012). A social media-based physical activity intervention: a randomized controlled trial. American journal of preventive medicine, 43(5), 527-32.

Centola, D. (2011). An experimental study of homophily in the adoption of health behavior. Science. 334:1269-1272.

Chang, T., Chopra, V., Zhang, C., & Woolford, S. J. (2013). The role of social media in online weight management: systematic review. Journal of medical Internet research, 15(11).

Chen, H., Chiang, R.H.L. & Storey, V.C. (2012). Business Intelligence and Analytics: From Big Data to Big Impact. MIS Quarterly Vol. 36; pp.1165-1188.

(38)

38 Crawford K., Lingel J., & Karppi T. (2015). Our metrics, ourselves: A hundred years of self-tracking from the weight scale to the wrist wearable device. European Journal of Cultural Studies, v18. pp. 479-496. Dibb, B. (2019). Social media use and perception of physical health. Science direct. Volume 5, issue 1. Edelman, D. C. (2010). Branding in the Digital Age: You’re Spending Your Money in All the Wrong Places.

Harvard Business Review, 88(12), pp. 62-69.

Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer‐

Mediated Communication, 13(1), 210-230.

Eysenbach, G., & Wyatt, J. (2002). Using the Internet for surveys and health research. Journal of medical

Internet research, 4(2), E13.

Field, A. P. (2009). Discovering statistics using SPSS: and sex and drugs and rock 'n' roll. SAGE Publications. Houdek VonHoltz, L. A., Hypolite, K. A, Carr, B. G, Shofer, F., Winston, F. K., Hanson, C. W., & Merchant, R.M. (2015). Use of mobile apps: A patient-center approach. Official Journal of the Society for Academic Emergency Medicine.

Hoj, T. H., Covey, E. L., Jones, A. C., Haines, A. C., Hall, P. C., Crookston, B. T., & West, J. H. (2017). How do apps work? An analysis of physical activity app users' perceptions of behavior changes mechanisms. JMIR

mHealth and uHealth, 5(8),

Kahneman D., Diener E., & Schwarz N. (eds). (1999). Well-Being: The Foundations of hedonic psychology. New York, NY: Russell Sage Foundation.

Kennedy, J.F. (1960). The soft American. Sports illustrated.

Kleinert, S., & Horton, R. (2015). Rethinking and reframing obesity. The Lancet, 385(9985), 2326-2328. Lim, J-S., Al-Aali, A., Heinrichs, J.H., Lim, K-S. (2013). Testing alternative models of individuals’ social media involvement. Computers in human behavior. Vol 29, pp 2816-2828.

Luzak, A., Heier, M., Thorand, B., Laxy, B., Nowak, D., Peters, A., & Schulz, H. (2017). Physical activity levels, duration pattern and adherence to WHO recommendation in German adults.

Magnezi, R., Grosberg, D., Novikov, I., Ziv, A., Shani, M., & Freedman, L. S. (2015). Characteristics of patients seeking health information online via social health networks versus general Internet sites: a comparative study. Informatics for Health and Social Care, 40(2), 125-138.

Mick, D.G., Pettigrew, S., Pechmann, C., & Ozanne, J.L. (2012). Transformative Consumer Research for Personal and Collective WellBeing. Routledge, New York, NY, pp. 3–24.

(39)

39 Moss, P., Ascari, A., Bakshi, A., & Grijpink, F. (2010). mHealth: a new vision for healthcare. McKensey &

Company.

Nicholson, S. (2012). A user-centered theoretical framework for meaningful gamification. In Games Learning Society 8.0.

Oberst, U., Wegmann, E., Stodt, B., Brand, M., Chamarro, A. (2017). Negative consequences from heavy social networking in adolescents: The mediating role of fear of missing out. Journal of adolescences. Vol 55, pp 51-60.

Pantic I. (2014). Online social networking and mental health. Cyberpsychology, behavior and social

networking, 17(10), 652-7.

Roberto, C.A., Swinburn, B., Hawkes, C., Huang, T.T-K., Costa, S.A., Ashe, M., Zwicker, L., Cawley, J.H., Brownell, K.D. (2015). Patchy progress on obesity prevention: emerging examples, entrenched barriers, and new thinking. The Lancet,385(9985), 2400-2409.

Pietiläinen, K. H., Kaprio, J., Borg, P., Plasqui, G., Yki-Järvinen, H., Kujala, U. M., Rose, R. J., Westerterp, K. R., Rissanen, A. (2008). Physical inactivity and obesity: a vicious circle. Obesity (Silver Spring, Md.), 16(2), 409-14.

Ryan R. M., & Deci E. L. (2001). To be happy or to be self-fulfilled: a review of research on hedonic and eudaimonic well-being. Annual Review of Psychology. Vol. 52. Palo Alto, CA. pp. 141–166.

Sheeran, P. & Webb, T. (2016). The Intention–Behavior Gap. Social and Personality Psychology Compass. 10. 503-518.

Statista (2019). Number of smartphone users worldwide from 2014 to 2020 (in billions). Retrieved from: https://www.statista.com/statistics/330695/number-of-smartphone-users-worldwide/.

Statista (2019b). Most popular social networks worldwide as of January 2019, ranked by number of active users (in millions). Retrieved from: https://www.statista.com/statistics/272014/global-social-networks-ranked-by-number-of-users/.

Statista (2019c). Daily time spent on social networking by internet users worldwide from 2012 to 2017 (in minutes). Retrieved from: https://www.statista.com/statistics/433871/daily-social-media-usage-worldwide/

Tremmel, M., Gerdtham, U., Nilsson, P. M., & Saha, S. (2017). Economic Burden of Obesity: A Systematic Literature review. Environmental research and Public Health.

(40)

40 van Uden-kraan C.F., Drossaert, C.H., Taal E., Seydel E.R., van de Laar, M.A. (2009). Participation in online patient support groups endorses patient’s empowerment. National Institute of Health.

van Deursen, A. J., Bolle, C. L., Hegner, S. M., & Kommers, P. A. (2015). Modeling habitual and addictive smartphone behavior: The role of smartphone usage types, emotional intelligence, social stress, self-regulation, age, and gender. Computers in human behavior, 45, 411-420.

van Ittersum, K. (2015). Inaugural speech for the acceptance of the chair Marketing and Consumer Well-Being.

Vandelanotte, C., Müller, A. M., Short, C. E., Hingle, M., Nathan, N., Williams, S. L. & Maher, C. A. (2016). Past, Present, and Future of eHealth and mHealth Research to Improve Physical Activity and Dietary Behaviors. Journal of Nutrition Education and Behavior, 48(3), 219-228.

Vaterlaus, J. M., Jones, R. M., Patten, E. V., & Cook, J. L. (2015). An exploratory study of time spent with interactive technology and body mass among young adults. Computers in Human Behavior. Vol 52. Pp. 107-114.

World Health Organization. (1948). WHO Constitution. Official Records of the World Health Organization,

2, 100.

World Health Organization (2011a). New horizons for health through mobile technologies. Global

Observatory for eHealth series, Vol. 3; pp. 1-104.

World Health Organization (2011b). Information sheet: global recommendations on physical activity for health 18 - 64 years old.

World Health Organization (2013). Global Action Plan for the Prevention and Control of NCDs 2013-2020. Retrieved from: https://www.who.int/nmh/events/ndc_action_plan/en/

World Health Organization (2018). Fact Sheet. Retrieve from: https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight.

World Obesity Federation. (2019). Retrieved from: www.worldobesity.org/.

Zhang, J., Brackbill, D., Sijia, Y., Becker, J., Herbert, N., Centola, D. (2016). Support or competition? How online social networks increase physical activity: A randomized controlled trial. Preventive medicine

(41)

41

Appendix

(42)

(43)

(44)

(45)

(46)

(47)

(48)

(49)

(50)

(51)

(52)

(53)

(54)

(55)

(56)

(57)

(58)

A Calorie for A “ Like ”