How do idealized body images on Instagram affect young adult males’ self-control and how does this have an effect on their purchase behaviour?

(1)

How do idealized body images on Instagram affect young

adult males’ self-control and how does this have an effect on

their purchase behaviour?

Master’s Thesis

MSc in Business Administration University of Amsterdam

Meltem Türksoy - 11387246 Supervisor: Samuel Franssens

(2)

Statement of Originality

This document is written by the UvA student Meltem Türksoy who declares to take full responsibil-ity for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of the completion of the work, not for the contents.

(4)

Abstract

This study aimed to investigate whether exposure to idealized body images on Instagram could have an effect on adult males’ purchase behaviour, through the moderating effect of a person’s self-control. Since the growth of social network sites (SNS) popularity among this topic and literature has grown. Since most studies have focused on female SNS users and the relationship between SNS use and behavioural responses, it was a matter of time to investigate the relationship between SNS and male users. The purpose of this study was to answer the question, whether idealized body im-ages on Instagram’s #fitspo pim-ages affect male adults’ purchase behaviour of fitness products or not. This study was expected to add relevant insight for marketers in the new media, since it is in the interest of marketers to gain knowledge about SNS and how to incorporate them in increasing pur-chase intensions. With the means of an online survey male participants helped to create an overall idea about the relationship between idealized images and purchase behaviour. The hypothesis that toned models are more appealing than untoned models was supported, whereas H2 (After exposure to an idealized body image promoting a fitness product, intention to purchase the product will be higher when the model is toned rather than untoned) and H3 (The effect of toned vs untoned is stronger/ weaker for people with low/ high self-control) were not. In future research, more reliable measures should be implemented and the experiment better conducted and designed to overcome this study’s main limitations, which are addressed in the limitations section.

(5)

Acknowledgements

I would like to thank my supervisor Samuel Franssens for his guidance and expertise and his time throughout the process of writing my thesis.

(6)

1. Introduction

Have you ever thought about how you feel after browsing through Instagram’s newsfeed filled with beautiful and happy images?

The ever-growing number of social media users worldwide has reached 2.46 billion people in 2017 (Statista, 2018), which accounts for 1/3 of the world population. With the development and popular-ity of social media around the globe, information is easily shared and consumed among users (Akrimi & Khemakhem, 2012). This has changed the ways, in which individuals obtain and dis-tribute information about products, dating, health and more. Social network sites (SNS) are net-works of friends for social or professional interactions (Trusov, Bucklin & Pauwels, 2009). SNS have not only changed advertising and promotions of brands (Hannah, Rohm & Crittenden, 2011) but have also affected consumer behaviour throughout the pre-purchase, purchase and post-pur-chase phase (Mangold & Faulds, 2009). By connecting with individuals from all around the globe (Hagel & Armstrong, 1997), social media users are influenced by each other, which shapes and forms perceptions of attitudes towards products (Williams & Cothrell, 2000). Social media posts that are meant to inspire and motivate people to exercise and eat healthily are often referred to as Fitspiration posts. Fitspiration, or Fitspo, in short, is a term that describes inspirational, fitness and health-related posts on social media (Boepple & Thompson, 2015). A number of previous studies have investigated how such posts, portraying idealized body images, negatively affect women’s self-esteem, mood and their body view. However, only little is known about how social media use and exposure to idealized body images affects male consumers’ self-control and how that influences their purchase behaviour. When referring to purchase behaviour of men in this study, it is referred to as fitness-related products such as the pre-workout powder by the brand called Cellucor C4. The brand Cellucor heavily advertises on Instagram and makes use of endorsers that frequently post

(7)

pic-tures of themselves together with the products they use before a heavy workout. Typical hashtags used to capture those pictures are among others #fitspo.

Despite the popularity of Fitspo pages directed towards men, the research on that field is only limit-ed. Most studies have so far focused on Facebook as a social media platform, whereas Instagram is rapidly growing and is one of the most popular SNS in the world (Statista, 2018).

The main goal of this thesis is to investigate how and to what extent idealized male body images affect young adult males’ purchase behaviour of fitness-related products presented on Instagram’s Fitspo pages. In order to examine the effects of idealized body images, self-control is anticipated to have a moderating effect on the dependent variable (purchase behaviour), which is contributing to already existing studies. Previous researchers have found a relationship between self-control and behavioural responses such as low level of self-control and poor eating patterns or alcohol-abuse (Peluso, Ricciardelli & Williams, 1999). For the purpose of this study, the brief self-control scale by Tangney et al. (2004) is applied to measure participants’ level of self-control. The research method and measures will be discussed later on.

2. Literature Review

With the evolution of social media, the number of literature in that field is growing and covers a wide range of topics. However, only little is known about how social media use (SMU) and expo-sure to idealized body images affect male consumers’ purchase behaviour of fitness-related products advertised on Instagram’s Fitspo pages. Self-control is anticipated to have a moderating effect on this relationship, as detected by previous studies (Baumeister, 2002; Brandon, Oescher & Lottin, 1990; Fagen, Long & Stevens, 1975). This relationship and effects will be addressed in the follow-ing paragraphs.

(8)

This section provides a theoretical outline of existing literature in the field of social media use and effects on consumers’ behaviour. Firstly, an overview of social media engagement and idealized body images on mass media channels are introduced. This is followed by literature on Fitspiration content as well as on self-control. Lastly, the literature gap will be addressed and the research ques-tion formulated.

2.1 Idealized Body Images

In the following paragraph, effects of idealized body images (thin and toned females; muscular and strong males) will be addressed. Previous research has investigated how body dissatisfaction emerges from unrealistic societal beauty ideals, which are presented on mass media (Hargreaves & Tiggeman, 2004), such as TV advertisements. They have examined how exposure to commercials with models with ‘perfect’ bodies led to increased negative mood and higher levels of appearance comparisons among boys and girls. This effect results, among others, from the media, which por-trays a cultural ideal of beauty that is becoming more and more unattainable. Airbrushing and digi-tal alteration only raise the unrealistic standards for self-evaluation and body (dis)satisfaction (Thompson, Heinberg, Altar & Tantleff-Dunn, 1999). This consequently leads to frustration in some individuals (Clay, Vignoles & Dittmar, 2005). It is suggested that appearance comparison is auto-matically triggered when attractive models are used in advertisement (Cattarin, Thompson, Thomas & Willaims, 2000) and that comparison only increases with age (Martin & Kennedy, 1993). In an experiment in which participants are exposed to idealized body images, the overall effect of lower mood and dissatisfaction was significantly higher for female participants (Hargreaves et al., 2004). Due to the participants’ mean age of 14.3, it is assumed that males are less vulnerable to muscular-ideal images until reaching late adolescence or early adulthood, “…perhaps due to developmental changes in the salience of muscularity concerns” (Hargreaves et al., 2004, p. 358). Another possible

(9)

reason for the lower effects on males’ body dissatisfaction could be “because the media’s portrayal of the thin ideal is likely stronger and more prevalent than the muscular ideal” (Hargreaves & Tiggemann, 2004, p. 358) and thus, it was not possible to match perfectly idealized-body image conditions. Another assumption is that many boys do not feel comfortable expressing feelings of dissatisfaction about their bodies (Pope, Phillips & Olivardia, 2000).

A more recently conducted study focuses on idealized body images on SNSs and suggests a positive relationship between SNS, self-esteem and self-control (Wilcox & Stephen, 2012). The authors fo-cus on ties between close friends on social media networks. According to them, those individuals who focus on strong ties appear to have momentary higher self-esteem achieved by presenting a positive self-image on SNS, which is followed by lower self-control. They claim that elevated feel-ings of self-esteem can lead to impulsive behaviour, referred to as low self-control. Results of that particular study present that frequent Facebook use was associated with poor self-control in health, mental persistence and spending/ finances. The causality of this relationship has not been further investigated (Hargreaves & Tiggemann, 2004). The research question of this thesis is partially based on the prediction that SNS use has negative effects on individuals’ behaviour such as impulse purchases and that the extent of self-control enhances or lowers such behaviours.

2.2 Fitspiration

A relatively new trend, called Fitspiration, has emerged through individuals’ willingness to provide social media-based content (Carrotte, Prichard & Lim, 2017). Fitspiration, also called Fitspo, is meant to inspire individuals to have a healthier lifestyle and to motivate them into exercising. It is described as fitness inspirations and related content, such as exercise images and videos, as well as recipes and photos of food and people are shared on social networks, such as Instagram (Boepple et al., 2015). Social media users that are interested in such content, can follow Fitspiration pages, in

(10)

order to view related content on their newsfeed. The posts usually capture a hashtag, followed by short sentences or words. This way, it becomes easier for users to find #Fitspo posts when searching for popular hashtags (Carrotte et al., 2017). Despite the fact that Fitspo content was originally de-signed to positively engage individuals, there are recent studies, which detected negative effects of them on SNS users (Tiggemann et al., 2015; Boepple & Thompson, 2016). Since many of those posts feature thin and toned women or muscular men, it is argued that their main message is to em-phasize idealized and unattainable body appearances, rather than a body’s health and functionality. This in response leads to lower mood and appearance dissatisfaction on SNS users (Tiggermann et al., 2015), because those who are not conforming to that body ideal tend to have poor self-esteem and lower general body dissatisfaction.

Idealized images can be traced back to the ‘Objectification Theory’ (Frederickson & Roberts, 1997), which states that the human body in Western societies is perceived as an object and has sig-nificant negative effects on women. “Objectification theory posits that girls and women are typical-ly acculturated to internalize an observer’s perspective as a primary view of their physical selves. … which in turn, may help account for an array of mental health risks…” (Frederickson & Roberts, 1997, p. 173). Roughly defined, this means that a person is seen or treated as an object, which is only identified with his or her body and, thus, is reduced to the body only (Rae, 1993).

There is evidence that men also experience body dissatisfaction, such as muscle dissatisfaction (Olivardia, Pope, Borowiecki & Cohane, 2004). Previous research found that muscular images shown on mass media channels are linked to “increased drive for muscularity and depression” (McCray, 2005), lower satisfaction with their muscles (Hargreaves & Tiggemann, 2009) and decrease in satisfaction with their fitness levels and overall appearance (Mulgrew, John-son, Lane & Katsikitis, 2014). There is no evidence as of yet, that examines the effects of exposure to Fitspo content on male social media users.

(11)

2.3 Social Network Site: Instagram

It is interesting to investigate SNSs and the role of appearance-based pages, such as Fitspo for fol-lowing reasons: First, SNSs allow a two-way communication between brands/ endorsers and social media users and thus, makes it a far more interactive channel than the classic mass media channels. Users connect with other users from all around the world by “following” or “friending” them and share comment and like content posted online. The number of likes for a content shared on SNS gives an indication of what is popular or desired (Borzekowski, Schenk, Wilson & Peebles, 2010). Second, unlike traditional media, SNSs are available at all times and is continuously active, made possible by smartphones and internet connection. 91% of the 85% of individuals that own a smart-phone use SNSs and spend hours on their favourite social media channels (Pew Research Center, 2015). It allows SNS users to engage and participate in social-/ or appearance comparisons any time anywhere. It is indeed reported that women “…rate social comparison as a primary motive for viewing others’ profiles” (Haferkamp, Eimler, Papadakis & Kruck, 2012). The users’ profiles can be accessible to the public and can be viewed by anyone. This allows not only friends but also strangers to express their opinions on one’s personal pictures or on other content shared online. Lastly, as SNSs such as Instagram usually portray images of peers, they might be perceived as more realistic than images appearing on mass media channels (Kleemans, Daalmans, Carbaat & An-schütz, 2016). ‘Real’ people sharing pictures of their staged every-day lives, makes it more relatable and more desirable to achieve the same standards. Since selected photos are presented to the SNS users only, they appear to show more positive images of oneself and thus, social comparison might involve more upward comparison (Haferkamp & Krāmer, 2011; Lee, 2014). It is argued that ap-pearance comparison on social media is associated with more negative effects on body dissatisfac-tion and mood relative to comparisons with images presented on tradidissatisfac-tional media (Fardouly, Pinkus & Vartanian, 2017).

(12)

Instagram is well-suited to investigate effects of fitspo since its main purpose is to share pictures and videos with others. A study has found that greater body dissatisfaction results from appearance-oriented content on social media, such as on Instagram (McLean, Paxton & Wertheim, 2016). An-other reason why Instagram is a relevant channel for this study is that it allows an asymmetric rela-tionship between users since users do not have to be mutually connected in order to view each oth-ers’ contents. This means, unlike on Facebook, Instagram offers users a wider audience to reach with their photos and videos. Thus, not knowing other users in person, may create the biased per-ception, that these users live a more ‘attractive’ life (Blease, 2015).

With Instagram, users have great access to the content of their interest, anytime and anywhere. Fre-quent exposure to such enhanced photos of other Instagram users might lead to more pressure to achieve an idealized lifestyle or an ideal body image.

2.4 Self - Control

In the following paragraph, the term ‘self-control’ in general will be addressed and related literature will be utilized to do so. For examining the effects of idealized body images on social network sites on males’ purchase behaviour, self-control is anticipated to have a moderating effect on the pur-chase behaviour of fitness-related products. Self-control is described as the “capacity to change and adapt the self to produce better, more optimal fit between self and the world” (Tangney et al., 2004, p.275). The authors state that self-control can override and change individuals’ inner responses, as well as interrupting unwanted behavioural tendencies.

Self-control contributed to positive effects on individuals’ behavioural, affective and cognitive components (Tangney, Baumeister & Boone, 2004). It was found that higher scores in self-control were correlated with a higher grade point average (GPA), less binge eating and alcohol abuse, better relationships and interpersonal skills, as well as optimal emotional responses (Tangney et al., 2004).

(13)

The authors constructed a new scale to assess levels of self-control, which would include aspects around “breaking habits, resisting temptation and keeping a good self-discipline, which all reflected the ability of the self to control itself” (Tangney et al., 2004, p.275). The original scale consisted of 93 items, encompassing all aspect of self-control They were rated on a 5-point Likert scale, an-chored from 1 not at all to 5 very much (Tangney et al., 2004). Similar research conducted experi-ments among university students and came to the conclusion that there were positive relationships between high self-control and better grades (Feldman, Martinez-Pons & Shaham, 1995) and found that self-control contributed significantly to the prediction of GPAs (Wolfe & Johnson’s, 1995). These findings were supported by the findings of Peluso et al., (1999), who detected a link between poor self-control and alcohol abuse and poor eating patterns among university students.

Lastly, it was found that people who scored high in self-control, assessed by Rosenbaum’s scale (1998), were better at decision-making and at saving money (Romal & Kaplan, 1995). It was rather surprising that there were not many theoretically reliable and valid measures of self-control, how-ever, the scale assessing self-control by Tangney et al. (2004), found great support in reliability and validity.

2.5 Literature Gap

In this paragraph, the literature gap is addressed and the research question formulated.

There are numerous articles and studies about how idealized body images negatively affect individ-uals’ self-esteem and certain consumer behaviours, such as purchase behaviour. Most studies have investigated how this relationship affected women in particular and have widely neglected the ef-fects of idealized body images on male participants’ purchase behaviour. There is evidence that muscular images have an impact on males’ mood and body (dis)satisfaction, however further con-sequences such as consumer behaviour are not addressed. Additionally, most studies have taken into

(14)

account classic mass media channels, and only recently have added Facebook into the literature. Thus, Instagram as a novel yet rapidly growing platform with over 800 million active monthly users, occurs to be suitable for this study (Statista, 2018). Especially, because of its main purpose of sharing images and videos, it is relevant for studying idealized body images and their effects on consumers. Both, Instagram and idealized body images are highly appearance-oriented in nature. Additionally, other than traditional media channels or Facebook, Instagram offers almost unlimited virtual access to anyone’s profile or to a content of their interest by just searching for popular hash-tags, for example, #fitspo.

Another stream of research relevant to this study is that of self-control since it is anticipated to have a moderating effect on males’ purchase behaviour, for instance, impulse buying of fitness products after being exposed to idealized male body images promoting such fitness products. This prediction is based on previous research that has investigated the impact of low self-control on individuals (Feldman et al., 1995; Wolfe et al., 1995) and most importantly for this study, they have detected a link between low level of self-control and alcohol abuse and poor eating habits (Peluso et al., 1999). Alcohol abuse and poor eating habits can serve as a starting point for the assumption that from low self-control impulsive behaviours, such as poor eating habits or impulse buying are emerging be-haviours.

The main goal of this thesis is to expand on past literature on idealized body images and their ef-fects on consumers, by investigating how and to what extent idealized male body images on Insta-gram affect young adult males and whether self-control serves as a moderator for purchase be-haviour of promoted fitness products. It would be interesting to examine whether there are purchase patterns of consumers who are affected by such idealized body images and whether they tend to impulsively purchase fitness-related items that are promoted on Fitspo pages.

(15)

RQ: How do idealized body images on Instagram affect young adult males’ self-control and

(how) does this have an effect on their purchase behaviour?

3. Conceptual Framework

Based on the previously mentioned gap in the literature, regarding the effects of exposure of ideal-ized male body images on young adults’ purchase behaviour, the aim of this research will be to in-vestigate the relationship between these variables and the moderating variable self-control. This is the proposed conceptual model, based on the literature gap and the research questions (See Figure 1)

(16)

3.1 Explanation of the Model

Based on the past literature and the existing gap, this study’s aim is to analyze the relationship be-tween idealized body images posted on Instagram’ (independent variable) and young adult males’ purchase behaviour (dependent variable). In this research, we anticipate self-control to be the mod-erator between these two variables. It is hypothesized that exposure to idealized body images on Instagram has a direct effect on males’ purchase behaviour, of products, which are promoted on In-stagram. Specifically, it is expected that purchase intention will be higher after exposure to ideal-ized body images compared to exposure to non-idealideal-ized images. In this study, these images are re-ferred to as toned condition and the non-idealized image as the untoned condition.

3.2 Hypotheses

H1: The toned model is perceived as more appealing than the untoned model.

H2: After exposure to an idealized body image promoting a fitness product, intention to purchase the product will be higher when the model is toned rather than untoned.

H3: The effect of toned vs untoned is stronger/ weaker for people with low/ high self-control.

The mass media, traditional and new media are presenting beautiful, skinny and muscular models ever since advertising was born (Hargreaves et al., 2004). The media portrays a cultural ideal of beauty, which is becoming more and more unattainable, however, seems to attract the average con-sumer (Thompson et al., 1999). From the assumption that skinny and muscular models are more appealing to the average consumer, we hypothesize H1: The toned condition is perceived as more appealing than the untoned condition.

(17)

Following the first assumption, namely that the toned condition is, in general, more appealing to the average consumer, it is therefore presumed that idealized body images will increase purchase inten-tion. This reasoning leads to H2: After exposure to an idealized body image promoting a fitness product, intention to purchase the product will be higher when the model is toned rather than un-toned.

As self-control is anticipated to have a moderating effect in the direct relationship between idealized body images and purchase behaviour, it is hypothesized that levels of self-control moderate the strength of that relationship. From this assumption we come to our last hypothesis, H3: The effect of toned vs untoned is stronger/ weaker for people with low/ high self-control. So that when some-one with high self-control is exposed to the tsome-oned condition, is affected less and therefore has a weaker effect on purchase intention and vice versa.

4. Methodology

The following section consists of a detailed description of the method used to test the hypotheses and the developed instrument. Scales used for this study were selected from existing studies. This chapter also contains a description of procedures the questionnaire was distributed to participants, data collection and analysis of data.

An online questionnaire (qualtrics.com) was used as the data collection method for this study. Basic demographic information was assessed, followed by questions regarding levels of education, time spent exercising and favourite SNS. These were followed by a self-control scale developed by Tangney et al. (2004) as well as a Body-Esteem scale developed by Mendelson et al. (2001). Final-ly, one Instagram image is shown, with either the toned or untoned condition, promoting the fitness

(18)

product Cellucor C4. Each of these images is followed by two questions, which were to be an-swered by participants.

4.1 Research Design

In order to arrive at an answer to the research question of this study, namely whether exposure to idealized body images on Instagram have an effect on male adults’ purchase behaviour, moderated by self-control, a survey was conducted and distributed among young adult males. This quantita-tive method of obtaining data is considered a good method to run correlations between the indepen-dent and depenindepen-dent variable (Bryman, 2015). The data was collected through convenient sampling among friends through social media platforms, such as Facebook and LinkedIn, as well as by utiliz-ing services of Prolific (www.prolific.ac), which is an application that helps researchers recruit par-ticipants. The survey is in a form of a self-report questionnaire, which contains a number of ques-tions assessing participants’ personal opinions on self-control and body-esteem.

4.2 Sample

Participants were recruited through two social media platforms, Facebook and LinkedIn. A link to the survey was shared on the personal accounts of the researcher, which directed volunteers directly to the questionnaire. Furthermore, utilizing the Prolific app, allowed to recruit additional partici-pants for this study. It is an online service, which researchers can make use of in order to recruit ad-ditional eligible participants for their study purposes. The conditions required for participation were to be male, between the ages of 18-36 and resident in the Netherlands. This was due to study’s tar-get group, young adults males, aged between 18-36 years old. To proceed with the survey,

(19)

partici-pants were to agree to the consent form, which was presented at the beginning of the survey, in or-der to continue. The collection of the data took place between 6th and 9th June 2018. A total of 193 people participated in the survey, however, only 154 responses were eligible for the purpose of the study. This is because one participant did not meet the condition of gender (one female) and was therefore left out of the analysis, as well as one participant, who did not fill in the survey complete-ly. These were deleted list wise, before proceeding with the analysis. The mean age of the partici-pants was 24.74 (SD = 4.69) ranging from 18-36 years. Most participartici-pants have a Bachelor’s degree (31.2%), followed by High school graduates (22.7%) and some college with no degree (22.1%).

Before participation, they were presented with information about the purpose and duration of the study and were asked to whether agree or disagree to participate (See Appendix B).

The average duration time for this survey was around five minutes and consisted of 14 questions.

4.3 Procedure

In order to answer the research question of this study, the survey was designed and run by means of Qualtrics, which is a software for research purposes, which allows collecting and analyzing data. A between-subjects (toned condition vs. untoned condition) 2 X 2 (+/- purchase behaviour) experi-ment was conducted in order to avoid carryover effects among participants (Charness, Gneezy & Kuhn, 2011). The withsubjects design may lead participants in the experiment to assume the in-tention of the study and might be more likely to change their behaviour when filling in the survey, which is also referred to as the demand-effect (Charness et al., 2011). The experiment contains two conditions, toned (n = 80) and untoned (n = 74), to which participants have been randomly as-signed.

(20)

The toned condition presents a male muscular and fit model in an Instagram picture, promoting the fitness product Cellucor C4, a pre-workout powder. The other condition shows an untoned male model in an Instagram picture, promoting the exact same product. Both images have the same amount of likes and same comments on Instagram, which is visible to each respondent. The results of the experiment will be further addressed in the following paragraphs.

Once participants have finished the survey, they were thanked for taking part in the study. Contact details of the researcher were provided in case they would be interested to learn more about the re-search as well as in the results afterwards.

4.4 Variables and Measures

In this section, the variables of the conceptual model will be addressed as well as the measures which were used to analyze the relationship between the independent, the dependent and the mod-erator.

4.4.1 Independent Variable  

The independent variable in this study is Exposure to Idealized Body Images on Instagram. In order to investigate whether idealized images on the SNS Instagram have an effect on male consumers’ purchase behaviour, participants are shown one image each, with either the toned or untoned condi-tion. The participants were randomly assigned (toned= 80, 51.8%; untoned= 74, 48.1%) to either one of the conditions, to allow a direct comparison later in the process of analysis. The independent

(21)

variable of idealized body images is hypothesized to have an effect on male participants’ purchase behaviour, as prior research argued that attractive models are most effective in promoting attrac-tiveness -related products, such as fitness products (Kahle & Homer, 1985; Kamins, 1990; Peterson & Kerin, 1977). Each image contained two statements: 1)Do you find this image appealing?; 2)Would you be willing to purchase the promoted product? Respondents were to select the most suitable answer, which was provided on a 5-point Likert scale. The answer options ranged from 1= strongly disagree, to 5= Strongly agree.

4.4.2 Dependent Variable

Purchase behaviour is the dependent variable for this study. The aim of this research is to investi-gate the relationship between idealized body images and their effect on purchase behaviour. Each participant is asked whether he would be interested in purchasing the promoted fitness product Cel-lucor C4, after viewing the Instagram image (toned vs. untoned). This way, the respondents’ answer to this questions is an indication of purchase behaviour. The scores were assessed on a 5-point Lik-ert scale, where 1= Strongly disagree and 5= Strongly agree. Consequently, a high score indicated a greater intention to purchase.

4.4.3 Moderator

Self- control is the moderator that is expected to affect the relationship between the independent and dependent variables. This is drawn from previous studies, in which self-control has been found to influence behavioural responses (Peluso, Ricciardelli & Williams, 1999). Researchers claim that elevated feelings of self-esteem can lead to impulsive behaviour, including purchase behaviour

(22)

(Peluso et al., 1999). For the purpose of this study, the brief self-control scale by Tangney et al. (2004) is applied to measure participants’ level of control. Instead of utilizing the initial self-control scale consisting of 36 items, only the reduced version comprising 13 items was included. The authors state that the items included in the brief self-control scale were highly reliable with a Cronbach’s Alpha above the acceptable level of 0.7 (α = .93) (Tangney et al., 2004). Hence, the moderating role of self-control was measured by the 13-item trait self-control scale, to which the answers ranged from 1= Never; 2= A little; 3= Sometimes; 4= A lot to 5= A great deal. Example questions are “I have a hard time breaking bad habits” and “I wish I had more self-discipline”. Some items of the scale have been reverted to ensure that the least self-control value to be the first (on the left side) and the highest self-control value to be the last (on the right side). In total, there were nine negative items, which were reverse-scored.

4.4.4 Body-Esteem Scale

For exploratory reasons, the Body-Esteem scale by Mendelson et al. (2001), was included in the survey. In line with previous studies investigating the role of body-esteem (Mendelson et al., 2001; Mendelson, McLaren, Gauvin & Steiger, 2002; Mendelson & White, 1982), the 14-item scale was used to measure participants’ Body-esteem. Example questions are “I like what I look like in pic-tures” and “I am proud of my body”. The answers ranged from 1= Never; 2= A little; 3= Sometimes; 4= A lot and 5= A great deal. Four negative items were reverse-scored, before computing the total scores of participants. This ensures that all items, negatively-keyed as well as positively-keyed, are consistent with each other (Barnette, 2000).

(23)

4.4.5 Control Variable: Time spent Exercising

The control variable time spent exercising was included in the analysis to investigate whether pur-chase intention for fitness products was different between participants who work out not at all or less than participants that work out more frequently. Further analysis with this variable will be dis-cussed in the analysis and result section.

4.4.6 Other Variables

Other variables such as age, level of education and social media use were assessed in the survey. This was to obtain a general overview regarding the characteristics of the samples participating in the experiment. Since it was relevant for this particular study, to only recruit young male adults, only male participants were asked to take part. The age range for young adults was anchored from 18- 35 years (Petry, 2002). Putting a threshold of 35 years is also in line with recent statistics stating that the majority of SNS, in particular, Instagram users’, are represented by individuals aged be-tween 18-35 (Statista, 2018).

5. Data Analysis and Results

A total of 181 participants took part in the study. The subjects were between 18 and 35 years old, which was one of the conditions which needed to be satisfied in order to participate in the experi-ment. Another condition that had to be met was the gender, male. Females that took the survey were removed list wise from the data set. Furthermore, participants that did not finish the survey, were removed list wise as well. A remaining of 154 relevant responses was taken into account for the analysis and the purpose of this study. All respondents were randomly assigned to either the toned

(24)

or untoned condition, which was presented as an Instagram picture, showing each a male model promoting the fitness product Cellucor C4, with one model being toned and the other being un-toned. As aforementioned, respondents were recruited by means of Facebook and LinkedIn as well the online service Prolific. All respondents were from the Netherlands, in order to limit cultural dif-ferences that might have an effect on the results and outcomes of this study.

The analysis will be processed throughout several steps. First, the data will be revised, in order to detect missing items or errors. Thereafter, the scales utilized for the study will be tested for reliabili-ty by means of the Cronbach’s Alpha. Subsequently, correlation and regression analyses will be run among the variables. In the following section, the analytical strategy will be addressed more de-tailed.

5.1 Analytical Strategy

As aforementioned, the first step of the analysis was to check whether either was any errors or miss-ing data in the collected data set. One missmiss-ing item was found and was deleted listwise from the dataset since the participant did not completely finish the survey until the end. Another item was deleted list wise, because the participant did not meet the requirements of gender and thus, was deleted from the data. The self-control items are represented as sc1 to sc13. In total nine counter-indicative items of the self-control were re-coded. The Body-Esteem items are labelled as be1-be14. In total, four items of the Body-Esteem scale were counter-indicative and thus had to be re-coded. Afterwards, it will be proceeded with testing the reliability of the set of used items.

(25)

5.2 Reliability Check

The self-control items consist of 13 items, indicate a high reliability (α = .83) with Cronbach’s Al-pha significantly higher than the accepted point of 0.7 (Bland & Altman, 1997). Additionally, the self-control scale items and its total score was analyzed, which exhibited overall satisfactory corre-lation (Pearson’s r > 0.2). The corrected item-total correcorre-lations indicate that most items have a good correlation with the total score of the scale (above .30). However, some items correlated less and thus, the scores ranged between -0.049 and 0.738. Sc7 had highly satisfactory positive correlation with total self-control (r = 0.738), whereas items sc7 and sc4 had a low negative correlation coeffi-cient (r = -.002). Then the reliability of Body-Esteem was analyzed. The Body-Esteem scale by Mendelson et al. (2004) has extremely high reliability (α = .92). The body-esteem scale items and its total score were analyzed. The scale showed a good correlation with the total score of the scale, with the items’ Corrected Item-Total correlation well above the 0.2 point. Most items within the body-esteem scale had relatively high item-total correlation and ranged from 0.238 (b12 and b6). To 0.866 (b6 b4). It is therefore concluded that both scales overall, are good and reliable for the purpose of this study.

5.3 Independent Samples t-Test: Interested & Appealing

In order to compare the means of two groups (exposure to toned vs. untoned image), an indepen-dent samples t-test for both depenindepen-dent variables (interested; appealing) is carried out. The reason for conducting two separate t-tests, is due to an only moderate correlation between interested and appealing, which did not allow to average both variables in order to create one single dependent variable. Therefore, two separate t-tests were carried out. This study’s aim was to understand a) whether idealized body images are considered as more appealing and, b) whether idealized body

(26)

images are associated with greater interest to purchase. One assumption to proceed with a t-test is the assumption of normality, which requires the dependent variable to be approximately normally distributed.

Before carrying out an independent samples t-test for the first dependent variable interested, the as-sumption of homogeneity of variances needs to be checked. The null hypothesis of equal variances, that the toned and the untoned condition have approximately an equal effect on interest to purchase, is not rejected (p = .056) and therefore, a t-test that assumes equal variances can be conducted. The output (See Appendix 1). presents that the toned condition (N = 80) had a numerically higher mean, not statistically significant, for interest in purchasing than the untoned condition (N = 74) (Mtoned = 2.05, SD = 1.25 vs. Muntoned = 1.85, SD = .87, t(152) = 1.13, p = .258). This result shows that the groups are similar between each other, and is an indication for the results to have occurred by chance.

In the following paragraph, the second t-test, namely for the second dependent variable appealing will be addressed. As aforementioned, the assumption of homogeneity needs to be examined by means of Levene’s test. The null hypothesis, that the toned and the untoned condition have approx-imately equal variances, is is rejected (p = .000). Thus, a t-test that does not assume equal variances will be conducted. The output (see Appendix 2) indicates that the toned condition (N = 80) had again a numerically higher mean, and statistically significant, for finding the image appealing than for the untoned condition (N = 74), (Mtoned = 3.25, SD = 1.38 vs. Muntoned = 2.05, SD = .74, t(122.6) = 6.76, p = .000). The large t-score and low p-value represents that the two groups are 6.76 as different from each other and thus, is an indication for the results to have happened by manipula-tion rather than by chance.

(27)

5.4 Independent Samples t-Test: Self - Control & Condition

In this section, the comparison between the means of self-control as well as the means of toned vs. untoned will be compared. This will be conducted by means of an independent samples t-test. First, the assumption of homogeneity needs to be examined by means of Levene’s test. Following Lev-ene’s Test for Equality of Variance and according to the p-value one can conclude that equal vari-ances are assumed. Thus, we accept the null hypothesis (p = 532). Again, in the case of toned v.s untoned, the mean is greater than the mean of untoned, however had no statistically significant ef-fect (Mtoned = 3.18, SD = .588 vs. Muntoned = 3.07, SD = .574, t(152)= -1.15, p = 252). This re-sults is indicating that the groups are similar between each other, and are likely to have occurred by chance (see Appendix 3).

5.5 Moderation Analysis

In this section, it is described whether self-control effectively moderates the relationship between idealized body images and purchase behaviour. Self-control is assumed to have a moderating effect, based on previous literature and studies, as aforementioned in the literature review. In order to ex-amine this effect, separate linear regression analyses were conducted. Each time, the self-control variable was combined with both dependent variables appealing and interested. Appealing refers to the extent the participants felt the image was appealing and interest refers to purchase intention. The first step was to test whether self-control accounted for a significant proportion of variance for the dependent variables, interest and appealing. The overall model does not statistically significantly predict the dependent variable, appealing (F (1,152) = 1.239, p = .267). Afterwards, it was tested whether self-control predicted the second deponent variable, namely, whether participants were

(28)

in-terested in purchasing the product. Again, the model did not show any statistically significant effect on predicting the dependent variable (F (1,152) = .433, p = .512).

Following these outcomes, it can be concluded that there was no moderating effect of self-control on purchase behaviour (interested; appealing) found.

This analysis will be followed by analyzing the interaction between the continuous and categorical variables. It will be examined whether or if there is a relationship between self-control and pur-chase behaviour for participants after exposure to the toned vs. untoned Instagram picture.

5.6 Interaction between a continuous and a categorical independent

variable

In order to test whether the relationship between levels of self-control and purchase behaviour (in-terested; appealing) is different for SNS users after exposure to idealized body images, the interac-tion between these variables is tested. For this, two individual scatterplots were created: one for the interaction between self-control and appealing, which suggests positive linear relationship, (see Figure 2) for toned vs. untoned images and another scatterplot for the interaction between self-con-trol and interested (see Figure 3) for toned vs. untoned images. The scatterplot in Figure 3 suggests a negative linear relationship between interest to purchase and self-control for both groups. With an increase of unit of self-control, interest to purchase the promoted fitness product is reduced.

The regression lines in both figures seem parallel, so the effect of toned vs. untoned, indicated by the distance between the lines seem equal for self-control. The slope of the lines indicate that the effect of self-control is most likely equal for toned. v.s untoned images. From this, one can conclude that there is no interaction between idealized images and self-control. Then it was examined whether the interactions were significant predictors by means of the ANOVA linear model. Howev-er, again both models did not show to have a statistically significant effect

(29)

(30)

(Fselfcontrol*appealing(33, 120) = 1.163, p = .274 vs. Fselfcontrol*interested(33, 120) = 1.034, p = .431).

5.7 Control Variable: Time spent Exercising

The control variable time spent exercising was incorporated in the analysis, after no significant ef-fects were found. Therefore, it was an additional variable, to investigate whether the results were due to differences among participants. By means of this control variable, it was to detect whether there was a difference in interest for people who work out more and for those who work out less. In order to see whether the results were due to low or no initial interest in fitness-related products, a Pearson correlation was run. This was to investigate whether participants who spend a lot of time in the gym (4-6 hours and 8 hours or more) were more inclined to purchase the product. For partici-pants that spend 4-6 hours exercising, there was a positive statistically not significant correlation between hours spent in the gym and interest to purchase the fitness product (r(152) = .007, p = . 928). In order to see whether people that spend even more time in the gym (8 hours or more) are interested more in purchasing Cellucor C4 another correlation was run. The results show that it is moderately more correlated than people spending 4-6 hours (r(152) = .213, p = .008) (see Appendix 5).

6. Conclusion and General Discussion

This study aimed to investigate whether exposure to idealized body images on Instagram can pre-dict purchase behaviour of adult men. Additionally, it was examined whether and if so, to what

(31)

ex-tent self-control moderated the relationship between idealized images and purchase behaviour. It was hypothesized that in general the idealized body image (toned) would be perceived as more ap-pealing than the untoned model. Subsequently, after exposure to an idealized body image promoting a fitness product, which in this case was Cellucor C4, a pre-workout powder, the intention to pur-chase would be higher when the model was toned rather than untoned. Lastly, the hypothesized re-lationship between idealized body image and purchase behaviour was presumed to be moderated by the variable self-control. Therefore, it was hypothesized that the effect, after exposure to an ideal-ized body image, would be stronger for people with low self-control, and weaker for participants with high self-control. Meaning that someone who has low self-control, would be more interested in purchasing the fitness product after being exposed to an idealized body image.

To begin with, the first hypothesis, that idealized body images would be perceived as more appeal-ing, was accepted. This outcome does not seems very surprisappeal-ing, ever since beautiful images of at-tractive models have been used in traditional and new media to promote any kind of product or ser-vice (Hargreaves et al., 2004). The media portrays a cultural ideal of beauty, which is becoming more and more unattainable, however, seems to attract the average consumer (Thompson et al., 1999). The consumers look up to those models, compare themselves with them, hoping to look like them eventually after using the products that are being promoted (Richins, 1991). The present study showed that the mean for finding the toned image appealing was greater than the mean of the un-toned body. Since a between-subject experiment was conducted, the result is expected not to be bi-ased due to the demand-effect which would have affected the results if the experiment was conduct-ed as a within-subject design (Charness et al., 2011). Following this result, the assumption that skinny and muscular models are more appealing to the average consumer , H1 was supported. In regards to H2: after exposure to an idealized body image promoting a fitness product the inten-tion to purchase would be higher when the model was toned rather than untoned. There was no

(32)

re-sults to support this hypothesis. The rere-sults indicated a positive trend towards finding the toned im-age more appealing however, it had no effect on interest to purchase the product.

A possible explanation might be following: since, the selected images with both, the toned and un-toned conditions, were promoting a fitness-related product and as the overall results indicated low interest in purchasing the product, it can be assumed that participants might simply not be interested in the product category. In order to predict a stronger outcome on purchase behaviour, participants could have been asked for preferences in product categories, thus participating in an experiment that is more personalized to their preferences. The Pearson Correlation indicated that there was only a weak correlation between hours spent in the gym and interest to purchase Cellucor C4 (see Ap-pendix 4). Then after comparing whether the results would change between 4-6 hours and 8 hours or more in the gym, it showed that there was indeed a moderately higher correlation between hours spent in the gym and interest to purchase the fitness product promoted on Instagram. From this, one can conclude that in future research, participants ants could be divided into groups according to their categories of interest.

Moreover, this study was unable to detect a moderating effect of the self-control variable, as it was anticipated prior to the research. Contrary to the hypothesis H3, that the effect, after exposure to an idealized body image, would be stronger for people with low self-control, and weaker for pants with high self-control, no effect supporting this was found. As it was expected that partici-pants with lower levels of self-control would be affected more by idealized body images in regards to purchase behaviour, this effect showed no significant relationship. When comparing this result with those of participants with high self-control, approximately the same effect, namely none, could be found. The scatterplots above (Section 5.6, Figure 2 and Figure 3), show that both lines run par-allel, indicating equal relationship for people with low and high self-control after exposure to both conditions, toned vs. untoned. These results may imply that there simply is no effect between ideal-ized images, self-control and purchase behaviour. Previous literature incorporated self-control for

(33)

several studies regarding impulsive behaviours, such as binge eating, alcohol abuse and in general weakly controlled behavioural responses. Based on previous studies, purchase behaviour was treat-ed as a category of impulsive behaviour throughout the process of writing this paper. Research sug-gested that elevated feelings of low self-control were associated with poor self-control in health, mental persistence and spending/ finances (Wilcox et al., 2011). However, in this particular study, self-control did not show any effect on participants interest to purchase.

Another reason accounting for this result, could be that the conducted experiment of this study was not sufficient to detect a relationship, either due to the survey’s set up or missing measures.

More about limitations and future research will be addressed in the following section.

7. Limitations and Future Research

One of the main limitations of this study is the ability to find existence of the moderating effect of self-control on purchase behaviour. Originally, self-control was associated with predicting behav-ioural responses, including impulsive behaviours. More research in this field of studies might have provided more insights and might have suggested to perform different measures or scales to incor-porate self-control as a moderator.

Moreover, multiple indictors for measuring purchase behaviour could have been developed and im-plemented in this study, in order to increase the chance and reliability of the relationship between idealized body images and purchase behaviour. Two questions (Do you find this picture appealing” and “Would you be interested in purchasing this product”) used to assess purchase behaviour were not sufficient and thus, contributed partially to the limitations of the study. The addition of existing and reliable measures for purchase behaviour cold be essential as it could provide more insights into consumers’ behavioural responses to idealized images. In order to replicate this study, additional

(34)

measures or scales might improve the quality of measurement and therefore increase reliability for the effect between idealized body images and purchase behaviour.

Another limitations is the chosen product category, a pre-workout drink called Cellucor C4. The limitations to only one specific product, might have excluded valuable results from participants which were not particularly interested in fitness or products related to that category. For this reason, for future research it is suggested to create number of different groups for different product cate-gories and subsequently assign participants respectively, according to their personal interests. In order to do this, a pre-test could be conducted, which evaluates participants product category inter-ests first, before assigning them into groups.

Additionally, the inability to obtain or create two images of the same person with the same facial attributes, presented in two conditions, toned vs. untoned, might have affected the results to some extent. Naturally, individuals have personal preferences towards attributes in a different person which they find attractive or not. The fact that two different models were chosen for this experiment could have limited the reliability of comparison between these two conditions.

Lastly, another limitation is the lack of previous studies on idealised body images and its effects on purchase behaviour. Most studies approach the relationship between idealized body images on SNS and social comparisons by focusing on consumers’ levels of self-esteem or elevated negative feel-ings about oneself. Thus, there is only limited research investigating how such elevated negative feelings about oneself affect purchase behaviour in general. Further research could investigate this relationship in order to understand whether exposure to idealized images, which results in negative feelings, still positively affect intention to purchase.

Overall, the main limitations of this paper were the inability to provide images portraying the same model, the chosen product category, inadequate measures for purchase behaviour and missing pre-vious research on self-control.

(35)

References

Akrimi, Y., & Khemakhem, R. (2012). What Drive Consumers to Spread the Word in Social Media? Journal of Marketing Research & Case Studies, 2012, 1-14.

Appel, H., Gerlach, A. L., & Crusius, J. (2016). The interplay between Facebook use, social com parison, envy, and depression. Current Opinion in Psychology, 9, 44–49.

Barnette, J. J. (2000). Effects of stem and Likert response option reversals on survey internal con sistency: If you feel the need, there is a better alternative to using those negatively worded stems. Educational and Psychological Measurement, 60(3), 361-370.

Baumeister, R. F. (2002). Yielding to temptation: Self-control failure, impulsive purchasing, and consumer behaviour. Journal of Consumer Research, 28(4), 670-676.

Bland, J. M., & Altman, D. G. (1997). Statistics notes Cronbach's alpha. BMJ, 314(7080), 572. Blease, C. R. (2015). Too many “friends,” too few “likes”? Evolutionary psychology and “Face

book depression.” Review of General Psychology, 19(1), 1–13.

Boepple L, Thompson JK. A content analytic comparison of fitspiration and thinspiration websites. Int J Eat Disord. 2015 Mar, 16, 98–101.

Borzekowski, D. L., Schenk, S., Wilson, J. L., & Peebles, R. (2010). e-Ana and e-Mia: A content analysis of pro-eating disorder websites. American journal of public health, 100(8) 1526-1534.

Brandon, J. E., Oescher, J., & Loftin, J. M. (1990). The self-control questionnaire: An assessment. Health Values, 14, 3–9.

Bryman, A. (2015). Social research methods. Oxford university press.

Carrotte, E. R., Prichard, I., & Lim, M. S. C. (2017). “Fitspiration” on Social Media: A Content Analysis of Gendered Images. Journal of Medical Internet Research, 19(3), 95. Cattarin, J. A., Thompson, J. K., Thomas, C., & Williams, R. (2000). Body image, mood, and tele

vised images of attractiveness: The role of social comparison. Journal of Social and Clinical Psychology, 19, 220–239.

Charness, G., Gneezy, U., & Kuhn, M. A. (2012). Experimental methods: Between-subject and within-subject design. Journal of Economic Behavior & Organization, 81(1), 1-8.

Clay, D., Vignoles, V. L., Dittmar, H. (2005). Body Image and Self-Esteem Among Adolescent Girls: Testing the Influence of Sociocultural Factors. Journal of research on adolescence, 15(4), 451-477

Fagen, S., Long, N., & Stevens, D. (1975). Teaching children self-control. Columbus, Ohio: Charles E. Merrill.

Fardouly, J., Pinkus, R. T., & Vartanian, L. R. (2017). The impact of appearance comparisons made through social media, traditional media, and in person in women’s everyday lives. Body Image, 20, 31–39.

Feldmann S. C., Martinez-Pons M., Dan Shaham. Psychological Reports, 77(3), 971-978

Fredrickson BL, Roberts T. Objectification theory: Toward understanding women's lived experi ences and mental health risks. Psychol of Women Q. 1997; 21(2), 173–206.

Haferkamp, N., Eimler, S. C., Papadakis, A.-M., & Kruck, J. V. (2012). Men are from Mars, women are from Venus? Examining gender differences in self-presentation on social networking sites. Cyberpsychology, Behavior, and Social Networking, 15(2), 91–98.

Haferkamp, N., & Krämer, N. C. (2011). Social comparison 2.0: Examining the effects of online profiles on social-networking sites. Cyberpsychology, Behavior and Social Networking, 14(5), 309–14.

(36)

Hagel, J., & Armstrong, A. G. (1997). Net gain: Expanding markets through virtual communities. Boston, MA: Harvard Business School Press.

Hanna, R., Rohm, A., & Crittenden, V. L. (2011). We’re all connected: The power of the social me dia ecosystem. Business Horizons, 54(3), 265-273.

Hargreaves, D.A., & Tiggemann, T. (2004). Idealised media images and adolescent body image: ‘’Comparing” boys and girls, Body Image, 1, 351-361

Hargreaves DA, Tiggemann M. Muscular ideal media images and men's body image: social com parison processing and individual vulnerability. Journal of Psychol Men

Masc. 2009;10(2), 109–119.

Instagram Business, (2017). Retrieved from: https://business.instagram.com

Jelenchick, L.A., Moreno, M.A. (2013). “Facebook Depression?” Social Networking Site Use and Depression in Older Adolescents. Journal of Adolescent Health, 52(1), 128-130

Kahle, L. R. & Homer, P. M. (1985). “Physical attractiveness of the celebrity endorser. A social adaptation perspective”, Journal of consumer research, 11, 954-961

Kamins, M. A. (1990). An investigation into the ‘Match-up’ hypothesis in celebrity advertising: when beauty may be only skin deep”, Journal of advertising, 19 (1), 4-13

Kleemans, M., Daalmans, S., Carbaat, I., & Anschütz, D. (2016). Picture perfect: The direct effect of manipulated Instagram photos on body image in adolescent girls. Media Psychology, 0(0), 1–18.

Langton, R., (1993). “Beyond a Pragmatic Critique of Reason”, Australasian Journal of Philosophy, 71(4): 364–384.

Lee, S. Y. (2014). How do people compare themselves with others on social network sites? The case of Facebook. Computers in Human Behavior, 32, 253–260.

Mangold, W. G., & Faulds, D. J. (2009). Social media: The new hybrid element of the promotion mix. Business Horizons, 52, 357-365.

Martin, M. C., & Kennedy, P. F. (1993). Advertising and social comparison: Consequences for female preadolescents and adolescents. Psychology and Marketing, 10, 513–530.

McLean, S. A., Paxton, S. J., & Wertheim, E. H. (2016). Does Media Literacy Mitigate Risk for Exposure to Thin-Ideal Media? Journal of Youth and Adolescence.

Mendelson, B. K., McLaren, L., Gauvin, L., & Steiger, H. (2002). The relationship of self-esteem and body esteem in women with and without eating disorders. International Journal of Eat ing Disorders, 31(3), 318-323.

Mendelson, B. K., Mendelson, M. J., & White, D. R. (2001). Body-esteem scale for adolescents and adults. Journal of personality assessment, 76(1), 90-106

Mendelson, B. K., & White, D. R. (1982). Relation between body-esteem and self-esteem of obese and normal children. Perceptual and motor skills, 54(3), 899-905

Mulgrew KE, Johnson LM, Lane BR, Katsikitis M. The effect of aesthetic versus process images on men’s body satisfaction. Journal of Psychol Men Masc. 2014, 15(4), 452–459.

Olivardia R, Pope Jr HG, Borowiecki III JJ, Cohane G. Biceps and body image: the relationship between muscularity and self-esteem, depression, and eating disorder symptoms. Journal of Psychol Men Masc. 2004, 5(2), 112–120.

Peluso, T., Ricciardelli, L. A., & Williams, R. J. (1999). Self-control in relation to problem drinking and symptoms of disordered eating. Journal of Addictive Behaviors, 24, 439–442.

Peterson, R. A. & Kerin, R. A. (1977). The female role in advertisements: some experimental eve dence”, Journal of marketing, 41, 59-63.

Petry, N. M. (2002). A comparison of young, middle-aged, and older adult treatment-seeking patho logical gamblers. The Gerontologist, 42(1), 92-99.

(37)

Pope, H. G., Phillips, K. A., & Olivardia, R. (2000). The Adonis complex: The secret crisis of male body obsession. New York: Free Press.

Richins, M. L. (1991). Social comparison and the idealized images of advertising. Journal of con sumer research, 18(1), 71-83.

Romal, J. B., & Kaplan, B. J. (1995). Difference in self-control among spenders and savers. Psy chology—A Quarterly Journal of Human Behavior, 32, 8–17.

Rosenbaum, M. (1980). A schedule for assessing self-control behaviours: Preliminary findings. Be haviour Therapy, 11, 109–121.

Statista (2018). Number of Daily Active Instagram Users from October 2016 to September 2017 (in Millions). Retrieved from: https://www.statista.com/statistics/657823/number-of-daily-ac tive-Instagram-users/

Statista (2018). Number of Social Media Users Worldwide from 2010 to 2021 (in Billions). Retrieved from: https://www.statista.com/statistics/278414/number-of-worldwide-social- network-users/

Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good adjust ment, less pathology, better grades, and interpersonal success. Journal of personality, 72(2), 271-324.

Thompson, J. K., Heinberg, L. J., Altabe, M., & Tantleff-Dunn, S. (1999). Exacting beauty: Theory, assessment and treatment of body image disturbance. Washington, DC: American Psychological Association.

Tiggemann M, Zaccardo M. 'Strong is the new skinny': a content analysis of #fitspiration on Instagram. Journal of Health Psychol. 2016(1), 9.

Trusov, M., Bucklin, R. E., & Pauwels, K. (2009). Effects of -of-Mouth Versus Traditional Marketing: Findings from an Internet Social Networking Site. Journal of Marketing, 73, 90-102.

Wilcox, K., Stephen, A. T., (2012). Are close friends the enemy? Journal of Consumer Research,12-57

Williams, L., & Cothrell, J. (2000). Four smart ways to run online communities. Sloan Management Review, 41, 81-91.

Wolfe, R. N., & Johnson, S. D. (1995). Personality as a predictor of college performance. Journal of Educational and psychological measurement, 55(2), 177-185

(38)

(39)

(40)

(41)

(42)

(43)

(44)

(45)

(46)

(47)

(48)

(49)

How do idealized body images on Instagram affect young adult males’ self-control and how does this have an effect on their purchase behaviour?