A content analysis that investigates how shisha pipe smoking is portrayed on YouTube.

Elizabeth Johnston Year of submission: 2013 A dissertation presented in the University of Salford in partial fulfilment of the requirements for the degree of MA Public Health.

1

Abstract Shisha pipe smoking [SPS], although an ancient cultural tradition in the East, has in recent years experienced growing popularity in the Western world. It has become a growing concern for public health departments as emerging evidence sheds light on how harmful it is. The focus of this research is to find out how SPS is portrayed on YouTube in terms of how popular the videos are, what types of videos they are, identify the most common messages about SHS in these videos and look for any evidence that these videos are being used to market shisha tobacco brands. This study is important in order for public health departments to formulate responses to this growing concern. The research approach adopted in this dissertation was content analysis and a sample of one hundred and twenty six videos were analysed. The findings from this research provide evidence that the videos are popular (judged by the number of views they have), they are in the majority positive about SHS (portrayed as attractive, fun, powerful, pleasurable, relaxing or sexy) and that they not only advertise tobacco products, but numerous other shisha related products too. This dissertation recommends that public health departments use YouTube videos in health campaigns, that Governments around the World start to think about how websites like YouTube can be regulated for tobacco content and that YouTube itself reviews its own policies with regard to tobacco content allowed. keywords: Shisha, hookah, YouTube, smoking, internet.

2

Acknowledgements

I would first like to acknowledge my employers, as was Bolton Primary Care Trust and now Bolton Council, who funded my course and supported me to complete my dissertation. I would also like to thank my tutor Lindsey Dugdill for her support and enthusiasm throughout the project. I also have to thank my husband, Richard, and my children, Lydia and George, for their support and love throughout the process of writing this dissertation.

3

Statement of ownership

“I declare that no part of this dissertation has been taken from existing published or unpublished material without due acknowledgement and that all secondary material used herein has been fully referenced.”

Signed……………… Date………………….

4

Contents Abstract ...................................................................................................................... 2 Acknowledgements .................................................................................................... 3 Statement of ownership.............................................................................................. 4 Figures/tables ............................................................................................................. 7 Chapter One – Introduction ........................................................................................ 8 1.1 Background .......................................................................................................... 8 1.2 Research focus .................................................................................................... 9 1.3 Overall research Aim and individual Research objectives .................................. 11 1.4 Value of this research......................................................................................... 12 Chapter two – Literature review ............................................................................... 14 2.1 Introduction ..................................................................................................... 14 2.2 Background of shisha pipe smoking ............................................................... 14 2.3 Health risks of shisha pipe smoking ................................................................ 15 2.4 Young peoples’ health beliefs about shisha pipe smoking .............................. 17 2.5 Introduction to YouTube and public health ...................................................... 18 2.6 YouTube and tobacco content analysis .......................................................... 19 2.7 YouTube and tobacco marketing .................................................................... 20 2.8 The tobacco industry and young people ......................................................... 22 2.9 Youth exposure to smoking in the media ........................................................ 23 2.10 Summary and emerging issues ..................................................................... 24 Chapter 3 - Research methods ................................................................................ 25 3.1 Introduction ........................................................................................................ 25 3.2 Research strategy .............................................................................................. 26 3.3 Data Collection ................................................................................................... 27 3.4 Framework for data analysis .............................................................................. 35 3.5 Limitation and potential problems ....................................................................... 36 Chapter Four - Results and discussion ................................................................... 37 4.1 Introduction ..................................................................................................... 37 Chapter Five – Conclusion ....................................................................................... 49 5.1 Research objective 1 .......................................................................................... 49 5.2 Research objective 2 .......................................................................................... 50 5

5.3 Research objective 3 .......................................................................................... 51 5.4 Research objective 4 .......................................................................................... 51 5.5 Limitations .......................................................................................................... 52 5.6 Recommendations ............................................................................................. 53 References ............................................................................................................... 55 Appendices .................................................................Error! Bookmark not defined.

6

Figures/tables

3.1 Table : Definition of content analysis ......................................................................... 11

3.2 Table: Overall study design and methodology .......................................................... 13

4.1 Flowchart: Process of selecting sample ..................................................................... 15

4.2 Table : Video category and total number of view, likes and dislikes for each video .......................................................................................................................................... 19

4.3 Pie Chart: Class of the videos by their stance on shisha pipe smoking...................... 20

4.4 Bar Chart : The different messages about shisha pipe smoking and the number of videos they feature in ................................................................................................... 22

4.5 Table: The video quality and the presence of smoking and/or brand content in the videos ......................................................................................................................... 23

4. 6 Bar Chart: The age and gender of people appearing in the videos ........................... 28

4.7 Table: Brand names and flavours of shisha tobacco in the videos and the number of videos they featured in ..................................................................................... 29

7

Chapter One – Introduction 1.1 Background Shisha pipe smoking [SPS], also sometimes known as hookah, is traditionally a Middle Eastern custom dating back thousands of years (Maziak., et al, 2004). An ornate pipe is used to smoke flavoured tobacco in which the smoke is passed through a water basin (often glass based) before inhalation. Shisha pipes were traditionally smoked by adult men in the Middle East (Kandela, 2000). However, over the past twenty years, the usage of shisha pipes has risen across the Western world, including North America, Europe and the United Kingdom [UK] (Cobb, Khader, Nasim & Eissenberg, 2012 ; Combrink., et al, 2010 ; Jackson & Aveyard, 2008). Although prevalence studies for the usage of shisha pipes in the UK are limited, they have shown an distinct increase in prevalence (Jackson & Aveyard, 2008 ; Jawad et al., 2013 ; Jawad, Wilson, Lee, Jawad, Hamilton & Millett, 2013), and a 210% growth in the numbers of shisha cafes in the UK the past five years has been reported (British Heart Foundation, 2012). The figures show there were 179 known shisha bars in 2007 compared with 556 in 2012. Increasingly accessed by White British students as well as Asian students, the use of shisha pipes has crossed the cultural divide (Roskin & Aveyard, 2009) which has contributed to the rising numbers of cafes in the UK. Many research papers have been written on the health risks of SPS (Al-Belasy, 2004 ; Kiter, Ucan, Ceylan & Kilinc, 2000 ; Munckhof, Konstantinos, Wamsley, Mortlock & Gilpin, 2003 ; Safagoj & Mohammed, 2002 ; Yadav & Thakur, 2000) and found it to be at least, if not more, harmful than cigarettes smoking. A systematic review found that it is significantly associated with lung cancer, respiratory illness, low birth weight and periodontal disease (Akl., et al, 2010). These risks have not however been widely covered in the popular media. There is currently no advertising or literature produced by the Department of Health to warn users of the dangers of SPS. This void in clear health information has led to many popular myths about shisha smoking gaining widespread belief (Aveyard & Roskin, 2009 ; Jackson & Aveyard, 2008 ; NHS Harrow, 2011). Common misconceptions include that the tobacco is just ‘fruit’ because it is often flavoured and given names like ‘silver fox apple’ and ‘lemon and mint fusion’. Also, very commonly, it is thought that because the smoke goes through 8

water, it is filtered of any harmful chemicals which incorrect (Knishkowy & Amitai, 2005). The rise in the use of shisha pipes in the Western world over the past five years has coincided with the increased use of social media sites like Facebook, Twitter and YouTube. These are examples of so called Web 2.0 websites, where users not only view the website, but share and create their own content. The figures associated with social media use are quite extensive. Every minute of the day, 100,000 tweets sent, two million search queries are made on Google, 684,478 pieces of content are shared on Facebook and 48 hours of video are uploaded to YouTube (All Twitter, 2012). These mediums have become platforms for young people to exchange information and content for themselves about the advantages and disadvantages of using shisha pipes. The use of YouTube in particular to share views and information on shisha pipes has become very popular with tens of thousands of videos on the subject uploaded by users. Young people from around the world can upload videos of SPS and within seconds they are available for anyone else to view and then rate, either 'like' or 'dislike' (the viewer can press a button on under the video to give their opinion of it), or add a comment underneath the video. The use of social media, mainly YouTube, by tobacco companies, to market their products to young people has been noted in other research studies. Studies that have analysed YouTube videos for advertising of tobacco brands have found there to be significant amounts of brand advertising (Elkin, Thomson & Wilson, 2009 ; Freeman & Chapman, 2007, 2009). No studies, to the author’s knowledge, have specifically looked at the brand advertising of shisha tobacco products on You Tube, a gap that this research study aimed to fill.

1.2 Research focus

There has been much confusion about the health risks faced by the users of shisha pipes, with academic literature arguing one theory, that SPS is as dangerous as cigarettes and the general public generally believing it to be less harmful (Smith, Novotny & Edland, 2011; Smith-Simone, Maziak, Ward & Eissenberg,

2008).

Academic literature to date has tended to focus on a comparison of the health risks 9

of shisha use ‘compared’ to cigarettes use. This type of comparison adds to the evidence base but essentially researchers have been trying to compare two things which are impossible to make a proper comparison of. Cigarettes are of a similar shape, content and take the same time (on average) to smoke. Shisha pipes are extremely varied - they have different types of pipes, different types of tobacco in, can be smoked by multiple people and SPS sessions are of varying length. Trying to compare the two things has not always enhanced academic research. Another key issue in the misinterpretation of SPS evidence has been the misinterpretation of a statement that was included in the 2005 paper on SPS by the World Health Organisation [WHO] , 'Water pipe tobacco smoking: health effects, research needs and recommended actions by regulators'. A key line in the report states ' the waterpipe pipe smoker will therefore inhale as much smoke during one session as a cigarette smoker would inhale consuming 100 or more cigarettes’,p.3. This has however been misrepresented in some health information and in the news as meaning that the SPS is the equivalent of one hundred cigarettes in terms of the health dangers (Jawad, Bakir, Ali, Jawa & Akl, 2013). While academia has been publishing papers in health journals and academics have been disagreeing among themselves over the health dangers of SPS (see chapter two - Literature review) , shisha pipe users themselves have filled this vacuum in information using the free and widely available medium of social media. This research study focussed in particular on the medium of YouTube as a source of information sharing about SPS and the information that is being proliferated through these videos. While this study could have looked at any form of social media, YouTube in particular was chosen because of its unique way of sharing videos. Through Facebook, information sharing tends to be limited, with more focus on personal accounts and personal pictures. On Twitter, users are only able to use one hundred and forty characters per tweet and add single photos or links to videos. On YouTube, users can specifically search for videos about certain topics that they are interested in, add comments to other users videos and also upload their own content on a subject area.

10

As the usage of shisha pipes has grown and public health departments seek to educate the public about the dangers of using them, understanding the information that is on social media sites like YouTube will be of crucial importance. Once public health specialists have an understanding of what are the main beliefs about SPS being proliferated on the internet, public health departments can then use this information in health campaigns to negate these beliefs if they are incorrect.

1.3 Overall research Aim and individual Research objectives The overall aim of this research study is to greatly enhance what is currently known about the content of videos containing shisha pipe use or imagery on the video sharing website YouTube. While the imagery of cigarette use on You Tube has been covered in several research studies (Kim, Paek, & Lynn, 2010 ; Freeman & Chapman, 2007, 2009 ; Elkin, Thomson & Wilson, 2010), the use of shisha pipes and/or products has been given little attention in comparison. To facilitate this study, a literature review has been conducted of all up to date relevant literature and also a content analysis approach was used to gather the data necessary for the study to fulfil its aims. In order to give focus to this research, it was decided to focus on five specific areas of knowledge that would have the most benefit for other researchers, public health departments and the greatest impact for the general public. Specifically, within the context of public health, the objectives of this research are to: 1.Identify what are the most common public health messages about SPS on YouTube and explore if they are either positive (SPS portrayed as attractive, fun, powerful, pleasurable, relaxing or sexy) or negative (SPS portrayed as undesirable, unattractive or harmful) messages. 2. Categorize what types of videos about SPS are on YouTube (Instructional, Review, Trick, Comedy, Commercial, Health or fetish) and what the production quality is of the videos available (low, medium or high). 3. Explore how popular the shisha pipe videos are by their number of views. 4. Investigate the prominence of shisha tobacco products and/or packaging in these videos and which brands are mentioned most frequently. 11

5. Formulate recommendations for public health departments in Local Authorities and also for social networking sites themselves. These five objectives should not be seen in isolation however but together as part of a broader research study. Objectives one and two are related in that they are the two that will be answered through the content analysis of the YouTube videos. They will specifically look at the content and imagery of the YouTube videos and the prominence of SPS in them. This will be completed by using key search terms to see which videos are brought up by searches on YouTube. The qualitative data will then be examined to see exactly which public health messages are being proliferated through YouTube about SPS. The messages will then be coded into different categories to see if they are positive or negative about the usage of shisha pipes. Objective three will look specifically at the popularity of the videos that contain SPS on You Tube and then objective four will then look at the data in a different way, exploring images of shisha tobacco packaging on videos on YouTube to see if there usage might imply any marketing of shisha tobacco products on YouTube. Objective five will then complete the study by making recommendations through the findings of this study to Local Authority public health departments and also to public health bodies such as the Public Health England. As well as public health bodies, the study will then conclude with some recommendations for social media websites themselves.

1.4 Value of this research

As the use of social media in society increases in the UK and shisha cafes proliferate, it seems the ideal time to research how shisha pipes are portrayed on YouTube. There is currently only one study, which focused on obesity (Keelan, Pavri-Garcia, Tomlinson & Wilson, 2007), that has researched public health information on YouTube. Research has shown that young people think the internet is an important resource for health information (Borzekowski & Rickert, 2001; Gray, Klein, Noyce, Sesselberg & Cantrill, 2005 ; Skinner, Biscope, Poland & Goldberg, 2003). The importance of this research becomes even more paramount when it is considered that currently little is known about what messages are being shown on

12

YouTube about SPS and yet nearly every young person in the UK will have access to these videos on their computer or mobile device at their fingertips. This research adds value to the current studies that have been done in this area in a number of ways. Firstly, it complements the research that has shown that there is evidence to suggest tobacco companies are using YouTube to promote their products (Elkin, Thomson & Wilson, 2010), by for the first time examining YouTube videos for the presence of branded shisha tobacco products. By adding to this body of evidence, it could highlight the need for revision of YouTube's policies about what videos are allowed to be shown on YouTube. Currently there is no process for checking the videos before they are uploaded but users can press a 'red flag' button to report videos that they think are unsuitable or that contravene YouTube's policies. Currently tobacco use is not one of the things that is listed under YouTube's 'Harmful or dangerous content ' policy, meaning that anything listed in that policy cannot be encouraged via videos on YouTube. Secondly, this research will add value to the few studies that have so far looked at public health information on YouTube (Keelan et al, 2007). Content analysis of social media in all its wider forms is still in its primitive stages but as well as YouTube, other sites such as Twitter, Facebook and Storify are rich sources of information about the public’s perception of public health and health messages in general. Researchers need to define their methods effectively in order to gain the most information from them and this study can play a role in this. The next chapter will now go on to provide a critical literature review of all the available evidence on the subject matter. Starting with a review of the general background to the history of SPS, a review of the literature on the dangers of SPS and then a review of the evidence of young people’s health beliefs around SPS. The literature review will then turn to look at YouTube and the evidence of its use in public health before finally summarising the current research around YouTube and tobacco marketing. The chapter will then conclude with the emerging issues in the themes of SPS and YouTube and public health more generally.

13

Chapter two – Literature review 2.1 Introduction

Globally, 4.9 million deaths each year are attributed to tobacco use and this may increase to ten million within the next twenty to thirty years, with 70% of these deaths occurring in developing countries (Maziak et al, 2004). SPS is a widely used form of tobacco use in the Middle East among adolescents (with a prevalence as high as 37% in one study of 15-17 year old students in Israel, (Korn & Magnezi, 2008)) but recently has surged in popularity in the US and UK, particularly among students (Roskin & Ayeyard, 2009). The Chartered Institute of Environmental Health [CIEH] in the UK has issued guidance for local authorities regulatory officers on dealing with non-compliance (cafes which are breaching the 2007 smoke free legislation and allowing smoking inside their buildings) in shisha bars due to the increasing predominance of legal cafes and illegal shisha dens (CIEH, 2011 ; Blackburn Citizen, 2011). The World Health Organisation felt that there were sufficient ‘growing concerns’ about SPS to release an advisory note on the subject in 2005 (WHO, 2005).

This literature review shall firstly briefly examining health effects of SPS and young people’s beliefs and misconceptions about SPS. It will also critically analyse current research on YouTube and health topics including how tobacco is being promoted via this medium and why the tobacco industry may find young people a profitable sector of the market to target. Finally, it will summarize the research findings and identify gaps in the evidence and where this study can add to the knowledge base.

2.2 Background of Shisha pipe smoking

Shisha pipes have been used to smoke tobacco and other substances by the indigenous peoples of Africa and Asia for at least four centuries (WHO, 2005). The use of shisha pipes spread through the Middle East and Asia, and were widely used 14

in Turkey during the Ottoman Empire in the 15th Century. In the late 20th Century, sweeter additives and more flavour were developed in Egypt in an effort to attract more female consumers (American Lung Association, 2007). The shisha pipe (UK, US) is also known by a plethora of other names depending upon region; this includes such names as ‘goza’(Egypt, Saudi Arabia), ‘narghile’ , ‘nargile’, or ‘arghile’ (Israel, Jordan, Lebanon, Syria), ‘hookah’ (Africa and Indian subcontinent) and ‘hubble bubble’ (many regions) (WHO,2005). It has been claimed that more than one hundred million people worldwide smoke shisha pipes daily (Knishkowy & Amitai, 2005). Traditionally a Middle Eastern habit, SPS has managed to cross the cultural divide and in recent years become popular with White British students as well as British students from Asian Backgrounds (Jackson & Aveyard, 2008 ; National Health Service [NHS] Harrow, 2011; Roskin & Aveyard, 2009).

2.3 Health risks of shisha pipe smoking

A systematic review found that SPS is significantly associated with lung cancer, respiratory illness, low birth weight and periodontal disease and could not rule out associations with bladder cancer, nasopharyngeal cancer, oesophageal cancer or infertility (Akl., et al, 2010). SPS also appears to have a similar effect on the lungs as cigarette smoking (Raad., et al, 2011) and other studies have implicated SPS in the development of chronic obstructive pulmonary disease (COPD) (Salameh, Waked & Khayat, 2012). There is also a increasing body of research that is showing that SPS users may show signs of addiction to nicotine as cigarette smokers (Hammal, Mock, Ward, Eissenberg & Maziak, 2008 ; Jawaid., et al, 2008 ; Ward., et al, 2005). As well as the specific consequences of the tobacco itself, the sharing of the mouth pieces of the pipes has also created concerns about the possibility of transmission of infectious diseases, from serious diseases such as tuberculosis to minor ailments like oral herpes (Knishkowy & Amitai, 2005). Although a small number of studies have been conducted into the composition of the tobacco smoke that comes out of the shisha pipe, the tobacco used in the pipes and

15

the pipes themselves are variable and not standardised (Kiter et al, 2000 ; Sajid, Akhter & Malik, 1993 ; Shafagoj & Mohammed, 2002 ; Zahran, Ardawi & Al-Fayez, 1985). The studies that have been conducted are therefore not as robust as the evidence we have for cigarette use which come in a more standard format. As Jawad (2012, p67) comments, there is a : Wide variation in shisha smoking habits, even within a single session (such as, number of puffs taken, depth of each puff, sharing of the shisha pipe with peers, length of each smoking session), it is difficult to quantify accurately the duration and magnitude of an individual’s exposure. The studies do suggest however that shisha pipe smoke is at least as toxic as cigarette smoke (Knishkowy & Amitai, 2005), even if this is difficult to quantify. While it can be said with some certainty that a single cigarette would be smoked in between five to seven minutes, a SPS session would typically last from anywhere from twenty to eighty minutes during which the smoker may take between fifty to two hundred puffs (WHO, 2005). The WHO advisory note, (2005, p3) states that " the water pipe smoker may therefore inhale as much smoke during one session as a cigarette smoker would inhale consuming 100 or more cigarettes". This statement from the WHO has been interpreted as meaning that one water pipe session is as harmful as smoking 100 or more cigarettes. However, when read how it is meant, this is not the case. It is just making a statement about the amount of smoke that is inhaled, not that it causes the equivalent harm. It is this lack of clear, strongly evidenced based rigorous research, which has been built up for cigarette smoking that has contributed towards misconceptions and ill founded beliefs about SPS. One recent researcher has however tried to quantify the health risks caused by SPS, Jawad (2012) cautiously suggests an equivalence of ten cigarettes as one shisha session, which he recommends is what GP’s should be advising their patients. He has based this on the fact that an average session will last forty five minutes, ten times longer than a cigarette. There is no other reason given other than the duration of the smoking session and it seems that this figure is an estimation rather than based on medical research. He does however, call for more high quality trials so that evidence based public health programmes can be produced. It might be better to wait for more research rather than trying to guess the equivalence with cigarettes as 16

giving out misinformation to the public can lead them to lose trust in what health professionals are telling them.

2.4 Young people's health beliefs about shisha pipe smoking

Evidence about young people's health beliefs about SPS is limited and research is particularly scarce in the UK. The limited studies that there are however suggest that young people generally believe that SPS is less harmful than cigarettes. Roskin and Aveyard (2009), in their study of Canadian and English students’ beliefs about SPS, found that many felt that the smoke was smoother than cigarettes. This was however a very small study with six participants from the UK and six from Canada. Together with the fruit flavourings, many associated this more comfortable experience with less harmful health implications. When asked how healthy SPS was compared to cigarettes, nine respondents considered it to be less harmful, two considered the two methods of smoking to be equivalent, and one person considered SPS more harmful. A cross sectional survey conducted in Birmingham University, UK, found that out of twenty one regular shisha pipe smokers, nineteen (90.5%) thought SPS was bad for your health, but of these, thirteen (68.4%) thought shisha pipes were less damaging than cigarettes. Of the twenty one, only two (9.5%) thought shisha pipes were more damaging than smoking cigarettes (Jackson & Aveyard, 2008). Again however, the small sample size of this study means the results need to be treated with caution. Research recently conducted by NHS Harrow using a focus group of eleven shisha pipe smokers found that there were a number of misconceptions about the health risks of SPS. These included that because shisha tobacco does not have any health warning on the packets that there are not any dangers, that it is not that bad as the smoke goes through the water and that it does not have chemicals or tar in so it is not that harmful (NHS Harrow, 2011). The focus groups also found that participants were keen for clear evidence and facts around the risks and effects of SPS, specifically how it differs to smoking cigarettes and what the equivalent effects are. The shortcomings of these studies are however that the numbers of participants in

17

these three studies are small and larger quantitative and qualitative studies would help to understand British students’ beliefs further. The majority of evidence on young people's beliefs about SPS is that it is perceived as less harmful than cigarette smoking and common themes and misconceptions recur throughout the literature. This piece of research will help to add to this body of research and see which common misconceptions and messages are being shown to YouTube users about SPS. This literature review will look at the evidence of how YouTube is being used to market tobacco products in general, in the absence of any research on how shisha tobacco is being marketed on YouTube.

2.5 Introduction to YouTube and public health

YouTube is a video sharing website which was founded in 2005 where users upload their own content and share it around the world. The website is the third most accessed website in the world after Google.com and Yahoo.com (Backinger., et al, 2010). As of February 2011, YouTube had 490 million unique users worldwide per month, who viewed an estimated 92 billion pages each month. This is the equivalent to 2.9 billion hours spent on YouTube each month (Mashable.com, 2011). With this rate of popularity, particularly among young people,YouTube is a medium that cannot afford to be ignored by public health professionals. A study by the Kaiser Family Foundation in the USA examined how young people use the internet as a health resource. It found that 75% of the young people (aged 15-24) had searched the internet for health information and information on smoking had been sought by 19% (n= 1209) (Kaiser Family Foundation, 2001). This suggests that young people could be receptive to health information found on social media website like YouTube. YouTube is a steadily growing source of research information, particularly around health topics, with several content analysis studies having been published using YouTube as their source of data. The topics include obesity(Yoo & Kim, 2011), kidney stone disease (Sood, Sarangi, Pandey & Murugiah, 2011), human papilloma virus [HPV] vaccine coverage (Briones, Nan, Madden & Waks 2012), tanning bed use (Hossler & Conroy, 2008), prostate cancer (Stienberg., et al, 2010) and diet and 18

acne (Qureshi & Lowenstein, 2011). Although all of the research papers use YouTube as their data source, because of the new and emerging nature of YouTube as a dataset, each study varies widely in terms of the numbers of videos that are viewed and how the study was set up. The only content analysis of YouTube videos of specific public health information examined the subject of immunization (Keelan., et al, 2007). It found that (para.4) "compared with positive videos, negative videos were more likely to receive a rating, and they had a higher mean star rating and more views." This void in evidence around what public health information is on YouTube ignores the huge potential of YouTube as an information medium, particularly to young people. This study can help to fill this information void for SPS. 2.6 YouTube and tobacco content analysis

YouTube and tobacco products have been one area of research that has been well evidenced through published literature. Since 2007, thirteen research papers have been published that have examined the tobacco content of videos on YouTube. Those papers which cover tobacco marketing on YouTube will be reviewed in the next section of this literature review. Excluding the four papers that examined marketing, that leaves nine studies of smoking on YouTube, covering such diverse topics as smoking fetish imagery (Kim, Paek & Lynn, 2010), quitting information (Backinger et al., 2011) and smokeless tobacco or ‘dip’ videos (Seidenberg, Rodgers, Rees & Connolly, 2011). This literature review will however concentrate on critically analysing the research studies which are most relevant to the topic that is being researched, that of SPS. There have been two research studies that have examined the prominence of these videos on YouTube previously. The first study was a comparison of cigarette and hookah/shisha related videos on YouTube (Carroll, Shensa, & Primack, 2012) in which the authors compared the characteristics of the two sets of videos. In their analysis of one hundred and twenty seven videos, the authors noted that shisha related videos were more likely (than cigarette videos) to portray shisha smoking in a positive light. There were fewer messages about the short term and long term health consequences of using a shisha pipe compared to cigarette use and the shisha pipe 19

videos more likely to be commented on. The authors only however used the word ‘hookah’ when performing a YouTube search and as such would have mostly found results from America and not Europe where it is more often referred to as ‘shisha’. The study also used only quite a small sample of hookah videos, coding only sixty one videos in total which as a sample size is quite small. A search for ‘hookah’ on YouTube currently (March 2013), retrieves 82,900 results. The other research study that is pertinent is the study conducted by Seitz et al. (2011) in which the authors conducted a content analysis of YouTube videos. The objectives were to investigate the types of videos about hookah on YouTube, their popularity and the demographics of those who post and view the videos. The study found that the most common types of videos related to hookah were from the categories of instructional, trick, and health. The results showed that hookah videos are popular, viewed frequently, rated positively and the videos were ‘liked’ six times more often than ‘disliked’. Viewers are able to press a button after watching each video that allows them to 'like' or 'dislike' it and the numbers of each is then displayed below the video. The demographic that most watched the videos were males aged 18-44 years old and the countries that most frequently watched hookah videos were in the United States, Canada, Australia, the United Kingdom, Germany, India, Norway, Finland and Sweden, showing the popularity of videos in both the east and west of the world. This research study used both search terms ‘shisha’ and ‘hookah’, enabling it to pick up more videos that will be viewed in Europe and it had a larger sample size of one hundred videos. In terms of enlarging the knowledge base about shisha videos on YouTube however, this study does not add much to the existing research. Also, as the authors admit as a limitation of their study, when trying to measure the demographics of those who watch the videos and the countries that they resided in, only those people who are registered users and log in to watch the videos will have their demographics listed in the viewer data statistics. That means that anyone watching the videos without logging in to YouTube is not included which could mean that the data is completely flawed. 2.7 YouTube and tobacco marketing

20

The medium of YouTube as marketing tool for tobacco companies has been the subject of several research projects. The first paper, which really first sparked an interest in the subject, was by Freeman and Chapman (2007), with their exploratory paper, 'Is ‘YouTube’ telling or selling you something?', the first real analysis of the content of YouTube videos and tobacco content. The authors did a YouTube search for the term ‘smoking’ and analysed the first fifty videos that were returned using two of the site's sort functions - relevance and number of views. They classified the videos into six different categories based on their tobacco content. The main finding of the study was that "smoking imagery is prolific and accessible on YouTube", p.210. They concluded that there was possible content produced by the tobacco industry on YouTube, masquerading as amateur footage. This was however just based on the fact that they found a YouTube video that was promoting Swedish snus (also known as snuff, a smokeless tobacco product used under the lip) that was professionally made and had a link to a company’s website. They were correct in pointing out that this contravenes the Swedish Tobacco act, however it seems then quite a leap to suggest that the tobacco industry would probably also be making videos too. It was admitted in the paper that "none of the pro-smoking videos were readily identifiable as tobacco industry creations" p.209, but the paper goes on to justify the conclusions by adding, " that given the industry’s historical use of stealth marketing techniques, we cannot definitively say that there is no tobacco industry presence on YouTube.",p.209. Although the findings of their study did not seem to back up their conclusion, it was this study that was the catalyst for other research into the topic. In 2010, a study by Elkin et al, looked in more detail at the branding of specific cigarette companies in videos on YouTube. The authors conducted a YouTube search using five leading non-Chinese cigarette brands worldwide. Of the one hundred and sixty three videos analysed, 71.2% had pro-tobacco content and 70.6% contained tobacco brand content. This study is the only study that has specifically looked at tobacco branding on YouTube to date. Of the one hundred and sixty three videos analyzed, in the authors' opinion, twenty of them appeared to be professionally made, possibly by the tobacco industry, adding more weight made to the initial claims made by Freeman and Chapman.

21

While these studies have examined cigarettes brands, no studies so far have examined if any brands of shisha tobacco are being marketed through YouTube. This gap in the research literature can be filled by this piece of research. This study can further add to the research on tobacco branding on YouTube and try to establish if the same patterns are observed for shisha tobacco brands.

2.8 The tobacco industry and young people

As part of this research study, it is imperative to understand why the tobacco industry might be so keen to promote their products via YouTube. In one Worldwide survey that measured the age of initiation of smoking (Global Health Survey [GHS], 2006), approximately two-thirds of respondents started smoking under the age of eighteen and almost two-fifths under sixteen years old, with virtually no respondents initiating smoking over the age of twenty. The twelve to eighteen age group is then the key target group of the tobacco industry which makes them the prime target for any industry marketing. A Cochrane review (2011, p2) on the impact tobacco advertising and promotion on adolescent smoking behaviours concluded that: Based on the strength and specificity of this association, evidence of a doseresponse relationship, the consistency of findings across numerous observational studies, temporality of exposure and smoking behaviours observed, as well as the theoretical plausibility regarding the impact of advertising, we conclude that tobacco advertising and promotion increases the likelihood that adolescents will start to smoke. It is now well documented that the tobacco industry has in the past sought to recruit and target young people through the analysis of internal records of the tobacco industry. The extract below was from a briefing put together for a tobacco company, showing how the tobacco industry were acutely aware of serving the emotional needs of the young people they were targeting. Background Gallaher are exploring the opportunity for the launch of a new brand, targeted at younger smokers, communicating an outgoing personality and offering a mellow flavoured product. Specifically, the research sought to [among other things]: Explore the opportunity for a mellow blend new brand from B&H, targeted at younger, image/fashion conscious smokers. 22

Main findings 1. The people - Many smoking as an ‘accoutrement to style’ statement, the image style statement of their preferred brand was important to them. Respondent characteristics Conservatives Surprisingly strongly conservative in values (given age) Cigarettes smoked must add to credibility, not challenge it Little desire to be ‘anarchic’ Strong desire to ‘conform’ Acceptance is all!.’ (Colquhoun Associates Debrief Presentation, 1996)

Given that their key marketing age groups are also the main users of social marketing websites like YouTube, it would be reasonable to surmise that the tobacco industry may well have decided to use YouTube, and other social networking sites as marketing tools. 2.9 Youth exposure to smoking in the media

As much as it is imperative to understand why the tobacco industry might be using YouTube to target young people, it is as crucial to understand what the significance of this might be to the young people watching the videos. Smoking in films decreased between 1950 and 1990, but then increased subsequently, despite an agreement by the tobacco industry to prohibit paid brand placement. By 2002, levels of smoking in films were back on par with those seen in the 1950’s (Glantz, Kacirk, & McCulloch, 2004). A large body of evidence now exists which shows the strong dose relationship between exposure to imagery of smoking in movies and increases in adolescent smoking (Charlesworth & Glantz, 2005). Movies have been the most researched genre of media for tobacco advertising, mostly because of their ease of recall by participants in data collection. High levels of smoking imagery have also been found on UK TV, including branding representation in popular TV shows including Coronation Street (Lyons, 2012).

23

2.10 Summary and emerging issues

In summary, shisha pipe use is growing the UK and Europe with young people believing it to be mainly harmless, fruit tobacco. A growing evidence base however is showing that it is not harmless (Akl et al., 2010 ; Raad et al., 2011 ; Salameh, Waked & Khayat, 2012) and is at least as harmful as smoking cigarettes. Although only one study so far has researched public health information on YouTube (Keelan et al, 2007), with YouTube growing quickly as a medium for information, this will surely only rise in the future. There have been lots of studies however that have explored other forms of tobacco content on YouTube. The studies have found differing degrees of evidence of how tobacco companies are using YouTube as a marketing tool with the overall inference that to some extent they are using it, but it cannot currently be quantified because they are disguising their techniques. However, what is clear is that the main users of YouTube are in the exact same age groups that the tobacco industry would want to target, making it a logical conclusion, when seeing their past marketing techniques, that they would want to use such mediums. The strong dose response relationship between exposure to smoking in the movies and smoking initiation rates by young people, gives strong cause for concern that any smoking imagery is freely available, with no age restrictions, for young people to access. Although there is now a small but growing body of evidence of what videos on YouTube contain, to the authors knowledge there are no studies that have explored the relationship between watching information on YouTube and consequential behaviour changes, particularly around smoking imagery. Although this is outside the scope of this research study, it is a large gap in the research knowledge base at this current time that needs urgent attention.

24

Chapter 3 - Research methods

3.1 Introduction This research study focused on the research objectives as set out below: 1. To identify what were the most common messages about SPS on YouTube and explore if the videos are either positive ( shisha smoking portrayed as attractive, fun, powerful, pleasurable, relaxing or sexy) or negative (shisha smoking portrayed as undesirable, unattractive or harmful). 2. To categorize what types of videos about SPS were on YouTube (Instructional, Review, Trick, Comedy, Commercial, fetish or Health) and what the production quality was of the videos available (low, medium or high). 3. Explore how popular the shisha pipe videos were by their number of views, likes and dislikes. 4. Investigate the prominence of shisha tobacco products and/or packaging and which brands were mentioned most frequently. 5. Formulate recommendations for public health departments in Local Authorities and also for social networking sites themselves. A valuable part of this research study was related to objectives one, two and three; the thorough investigation of the videos on YouTube through the research strategy that is explained in this chapter. As identified in Chapter two – literature review – there is a clear gap in the evidence base about what messages are being portrayed about SPS on YouTube and objective one addressed this evidence gap. Objective two categorised the videos that are on YouTube while objective three was answered by the number of ‘likes’ and ‘dislikes’ each video received. An important part of this study was the focus on objective four, the prominence of shisha tobacco branding and products in these videos, a part of this research that has never been analysed before, to the author’s knowledge.

25

This section – research methods – will outline the research strategy that was adopted and the justification for its use in this research study. It will then go on to discuss how the data was collected and what methods were used for data analysis. This section will conclude with a discussion of the strengths and limitations of the particular research strategy adopted.

3.2 Research strategy The research strategy that was used to implement the empirical research was content analysis. What is content analysis and why was it suitable for this research? Content analysis has many definitions as shown in table 3.1 below: Table 3.1 Definitions of content analysis (Bauer, 2000) The statistics semantics of political discourse. (Kaplan, 1943) A research technique for the objective, systematic and quantitative description of the manifest content of communication (Berelseon, 1952) Any technique for making inferences by objectively and systematically identifying specified characteristics of messages (Holsti, 1969) Information processing in which communication content is transformed, through objective and systematic application of categorisation rules. (Paisley, 1969) A research technique for making replicable and valid inferences from data to their context (Krippendorff, 1980) A research methodology that utilizes a set of procedures to make valid inferences from text. These inferences are about senders, the message itself, or the audience of the message. (Weber, 1985).

As shown above, different researchers have given their own interpretation of what content analysis is. Silverman (2011, p.64), however, simplified it well when he wrote "[it] involves establishing categories and then counting the number of instances when 26

those categories are used in a particular item of text". This is how this piece of research was carried out; categories were predetermined before the research started and then the number of instances were recorded. As such, this research study used both qualitative research and quantitative research to answer the research objectives. The YouTube videos were the source of qualitative data, however once the data had been collected, quantitative methods were used to analyse that data and to find the answers to the research objectives. It was felt that this was the only research design that fitted exactly what the research objectives were asking. For example, to answer objective one, ‘identify what are the most common messages about SPS on YouTube’, it was necessary to have predefined (produced after watching a sample of videos) categories and then to code each video as it was watched in order to answer the question. This study followed the steps set out below by Marvasti (2004), for how content analysis should be conducted: 1. Define the research problem. (See research objectives). 2. Decide where the source of the visual material will be. (YouTube) 3. Identify the categories or features that will be the focus of your research. (Set out in data collection). 4. Sample documents from the sources previously defined. (The empirical research). 5. Measure or count the occurrence of the pre-established categories. (See framework for data analysis).

3.3 Data Collection

The data for this study was obtained from the video sharing website YouTube (wwww.youtube.com). The sample was limited to videos from the website YouTube because they hold a large (43%) share of the online video market, far ahead of their nearest rival (Elkin et al., 2010). YouTube is the third most viewed website on the internet (Alexa, 2013).

27

Originally, the author had planned to use five different search terms. These were: hookah, shisha, narghile, sheesha and hubble bubble. The terms were chosen as they were the most common names used for shisha around the world. However, after a search of each of these terms on YouTube, it soon became apparent that the terms narghile and sheesha did not return any videos that were in English, which is one of the criteria for selection. The search term ‘hubble bubble’ returned videos of a pop group called ‘Hubble Bubble’ in the large majority and was felt not to be appropriate. It was therefore decided not to include the terms ‘narghile’, ‘sheesha’ and ‘hubble bubble’ as originally planned but keep ‘hookah’ and ‘shisha’. The author decided to replace these search terms with different ones. It was decided that the best way to do this was to try different search terms in YouTube to see which ones brought up the most relevant results. This was confirmed firstly by the titles of the videos that appeared on the first page of ‘hits’ for each search terms and also the number of videos listed for each search term. After this process, the author decided to also search using the terms ‘shisha smoking’ and ‘hookah smoking’, both returning over sixty thousand videos on YouTube. Two shisha tobacco brand names were also decided to be used as search terms as it was felt that these were important to be used in order to assess the prominence of shisha tobacco branding on YouTube. Further detail on how these were chosen is discussed below, however these search terms were: ‘Al Fakher tobacco’ and ‘Starbuzz tobacco'.

In total, for clarity, after pilot testing the original search terms for the actual study, six search terms were finally chosen for use. The next decision was to decide how to prioritise the results that would be used as part of the analysis. There are four methods of prioritising searches on YouTube, these are ‘relevance’ (which is used as a default setting), ‘upload date’, ‘view count’ and ‘rating’. It was decided to use the method used by Carroll et al. (2012) as this sampling frame has been used previously and shown to be successful. That method was to search for the terms selected by both ‘relevance’ (because most users probably don’t change the default settings) and also by ‘view count’ (to get an understanding of which videos are the most popular and being watched with the most frequency). The first two pages of 28

results (twenty videos) are collected; a number which is supported by previous public health and internet research (Carroll et al, 2012) and in line with research which suggests that viewers will generally only view the first page of results returned (Jansen & Spink, 2006). Videos were excluded from the study if they were not in English language, were irrelevant (defined as having no reference to SPS either through audio or video) or if they were duplicates (defined as having more than half of the same material of a video already coded). Videos that focussed on ‘E-shisha’ (electronic shisha pipe) were discounted as although interesting, it was a completely different topic to this research study.

The exact methodology used for each research objective is detailed below, for ease of understanding: 1.Identify what are the most common public health messages about SPS on YouTube and explore if they are either positive (SPS portrayed as attractive, fun, powerful, pleasurable, relaxing or sexy) or negative (SPS portrayed as undesirable, unattractive or harmful) messages All the videos will also be categorised using categories previously used in two studies that have looked at tobacco content on YouTube but will be adapted slightly so they fit shisha use, not just tobacco use (Elkin et al., 2010 ; Jenssen et al., 2009). The categories are: a. Pro shisha (containing a positive description of shisha use or shisha company promotion) b. Anti – shisha (containing a negative description of shisha use) c. Complex (containing positive and negative descriptions) d. Unclear (containing no clear shisha message) e. Neutral ( shisha content which was statistical or historical)

By using the above categorisation, the research will show what percentage of videos that are positive or negative about SPS. 29

In order to answer the objective about what are the most common messages about SPS on YouTube, any messages that the author has heard anecdotally and referenced in section two - ‘literature review’ - were used and then a sample of ten videos were watched and any further messages added to the spreadsheet for coding. Any new categories were added as they emerged.

2. Categorize what types of videos about SPS are on YouTube (Instructional, Review, Trick, Comedy, Commercial Health or fetish) and what the production quality is of the videos available (low, medium or high). For this category, the author decided to follow the categories already used by Sietz et al. (2011) for coding the category of each video. The category 'fetish' was also added to the study because during the analysis, there were some videos that fitted into this category and could not be put into any of the others. The description for each category was as follows: Instructional: Viewers are taught how to make/assemble/clean the pipe, also the etiquette of smoking the pipe and how to perform tricks like smoke rings etc. Review: shisha pipe users that are reviewing either the pipe itself or the tobacco product used. Trick: shisha users perform smoking tricks for the camera. Comedy: Involves the user participating in funny challenges involving a shisha pipe. Commercial: Advertisements for shisha cafes/pipes Fetish: Women smoking shisha pipe for sexual purposes Health: videos that focus on pro shisha and anti shisha health messages both come into this category This objective also included the production quality of each video and was coded for in one of the previous research papers on SPS videos on YouTube (Carroll, Shensa & Primack, 2012). This was included because production factors may influence a viewer's interpretation of and response to a video. There is some evidence that 30

viewers may be more likely to believe the messages coming from high production values (Austin, Pinkleton & Fujioka,1999).

3. Explore how popular the shisha pipe videos are by their number of views and ‘likes’ and ‘dislikes’. This was relatively straight forward, for each video that was watched, the number of views that it had received were recorded next to the video name in an excel spreadsheet and the number of likes and dislikes.

4. Investigate the prominence of shisha tobacco branding/ advertisements on YouTube. In order to assess the prominence of shisha tobacco marketing on YouTube, it was decided to search for the most famous brands of shisha tobacco in the UK. Originally, it had been determined to code whether there were pictures or references to particular brands in the other videos that were going to be coded for the search terms ‘hookah’ and ‘shisha’. On reflection however, it was decided that it would be preferable to follow the method that Elkin et al., (2010), used in their study on the prominence of cigarette tobacco branding on YouTube, in which the authors conducted a search using five leading cigarette brands. They chose the five leading non-Chinese tobacco brands worldwide as search terms, these were Marlboro, L & M, Benson and Hedges, Winston and Mild Seven. It was the original intension to replicate the findings of the Elkin study for shisha tobacco so it was decided that this would give the research study the most valid and useful results to answer research objective four. It proved to be quite difficult to ascertain exactly what were the most popular or best selling shisha tobacco brands as there is, to the author’s knowledge, no literature that details this information. To gain this information, firstly, the author did a Google search ‘buy shisha tobacco’ and the first webpage that that search returned was for the shisha tobacco company ‘Al Fakher’. The next website that was brought up was www.hookah-shisha.com and a quick search under the ‘tobacco’ section brought up 31

a list of different brands and based on how they are listed on the website, the author chose the next brand on that website as another search terms. That was ‘Starbuzz’. To see how prominent shisha tobacco branding is widely shown on YouTube, following the methodology of Elkin et al. (2010), all videos were coded for: A. The presence of any smoking in the video (defined as anyone touching or holding a shisha pipe or hose) B. The presence of any tobacco brand content (defined as the presence of the brand (in audio, text or as logo). The brand name, if featured, was also recorded.

Table 3.2 : Overall study design and methodology General study design This study used the methodology content analysis and used the video sharing website ‘YouTube’ to answer the research objectives. The six search terms used were: ‘hookah’, ‘shisha’, ‘hookah smoking’, ‘shisha smoking’, ‘Al Fakher tobacco’, ‘Starbuzz tobacco’. For each search term, the first twenty videos were watched when sorted by both ‘relevance’ and ‘view count’. All of the videos were then coded for the following categories or numbers recorded: Research Objective

Method of retrieving data

1.Positve/Negative about shisha

Pro shisha (SPS portrayed as attractive, fun, powerful, pleasurable, relaxing or sexy) Anti shisha (SPS portrayed as undesirable, unattractive or harmful) messages. Complex (containing positive and negative descriptions) Unclear (containing no clear shisha message) Neutral ( shisha content which was statistical or historical)

This data was analysed and put into a pie chart which shows how many of the videos came under each category.

32

Common messages about shisha on YouTube

A list complied of commonly known messages about SPS : Less harmful than cigarettes Natural Effective water filter/cool smoke Tobacco is ‘baked’ Fun Cool Just fruit As bad as smoking cigarettes Can cause cancer/lung diseases Equivalent of 100 cigarettes Addictive Can spread disease via the mouthpiece Can go rusty inside and cause coughing up blood Refuting health warnings Stomach bugs can be passed through the water Harmful additives in shisha tobacco Shisha smoke is cooler so more tar goes in your lungs The smoke is lighter than cigarettes No controlled levels of CO, nicotine and tar Fruit and nuts contain pesticides in shisha Herbal shisha is just as bad as tobacco shisha Shisha will not make new users sick Cleaner than cigarettes Fresh Strong/potent smell/flavour

33

Sweet taste Full thick smoke/cloud Delicious Sociable Nice Juicy Relaxing Buzz Spiritual/helps you to study This data is displayed in a bar chart, showing which messages about SPS occur most frequently in these videos. The gender and the age of the people appearing in the videos was also recorded 2. What categories Categorize what types of videos about SPS are on YouTube: are the videos

Instructional Review Trick Comedy Commercial Health Fetish What the production quality is of the videos available : Low (hand held camera/mobile phone) Medium (More professional equipment/audio) High (Television quality production/audio/visual)

This data was put into a bar chart which will show which categories are the most popular. 34

3.How popular are Number of : the videos

A. Views B. Likes C. Dislikes This data has been put in a table so a comparison between the percentage of videos that are pro or anti shisha and use the number of ‘views’,’ likes’ and ‘dislikes’ to see which videos are more popular and most watched.

4.

Prominence

shisha branding

of The prominence of shisha tobacco branding, all videos were

tobacco coded for : A. The presence of any smoking in the video B. The presence of any tobacco brand content

This data looked for what percentage of the videos had smoking in them and also the percentage of videos that show a specific branded product. The most popular brand names mentioned or shown in the videos have been recorded and displayed in a pie chart.

3.4 Framework for data analysis The data was coded using an Excel spreadsheet, which had headings at the top and then the codes were added as the videos were being watched. For example, the heading ‘category’ – and then in the box below it was entered as either:

A,

Instructional, B, Review, C, Trick, D, Comedy, E, Commercial, F, Health or G, Fetish. The use of an excel spreadsheet allowed for easy data storage and also enabled pivot tables to be used to extract data for analysis. 35

3.5 Limitation and potential problems Using content analysis as a research method can have its disadvantages. It has been argued that the coding methods may not be very reliable because it is the end product of the personal interpretations of the researcher (McNeil & Chapman, 2005). Ideally this research project would have had at least two coders watching each video but unfortunately, without any funding to pay anybody else, there was only one researcher. This would have avoided potential bias in the study because of the researchers own personal views and interpretations. This possible source of bias was avoided as much as possible by having clear and concise coding rules and categories at the beginning of the study. Gilbert (2003) also argued that content analysis ignores context and the potential for multiple meanings. In this study however, only the overt messages from the video clips were studied, as the idea was that the researcher saw exactly what a person would see – it did not need to analyse any deeper meanings. Content analysis does have some advantages in that it provides a very concise and clear method of analysing data and vast amounts of information can be coded relatively quickly. It is also a very cost effective way of extracting data and can be easily replicated by other researchers if necessary (Denscombe, 2007). This method of researching involves using information and data sources that are already in the public domain so the ethical issues of this research are also minimal. It is a recognised weakness of this study that only English language videos were included and that also only six search terms were used. Another limitation of this study was that the data was captured over a small period of days and it is unknown how quickly the view count and the content of videos posted will change.

36

Chapter Four - Results and discussion 4.1 Introduction This chapter will now display the results of the content analysis approach used to code videos on YouTube of their content which was described in Chapter 3 research methods. For each research objective, the exact data that was coded for is set out in table 3.2. Each set of results has been displayed so that they can easily be interpreted by the reader, be this a table, pie chart or bar chart. At the end of this section, there will be a more general discussion about the overall results of this empirical research and how the study could have been improved in a more ideal setting. Each search term was searched for twice by the two different criteria (relevance and most viewed) and the first twenty results brought back were included in the sample. Videos were then excluded from the study if they were duplicates, no mention of SPS, were not in the English language or if they featured an electronic shisha pipe. The numbers of videos excluded and the sample size left is set out below (Figure 4.1).

37

240 Relevant videos identified: shisha - 40 shisha smoking - 40 hookah - 40 hookah smoking - 40 114 videos were excluded: Al fakher tobacco - 40 Duplicates - 67 Starbuzz tobacco - 40 No shisha pipe mention - 16 None English language - 26 126 videos included:

E -Shisha - 5

shisha - 23 shisha smoking - 25 hookah - 15 hookah smoking – 14 al faker tobacco - 25

Figure 4. tobacco - 24 Starbuzz

Figure 4.1 Flowchart of how many videos were included in the study

Out of the 240 videos that were included in the sample, only one hundred and twenty six met the exact criteria set out by study. The majority (67) were excluded because they were duplicates as the search terms 'shisha' and 'shisha smoking' brought up a lot of the same videos as did the searches for 'hookah' and 'hookah smoking'. Twenty six videos were not in the English language and therefore excluded. The videos were all placed into one of seven categories as displayed in the table below (4.2). These categories had been used before in a similar study by Sietz et al. (2011), however the 'fetish' category was added during the research because of several videos coded that only seemed to fit in this category and none of the others already specified.

38

Table 4.2 Video categories and the number of view, likes and dislikes for each video Video Category (n=126) Instructional (n =24)

Views

Likes

Dislikes

7,211,091

22,542

2,835

Review ( n=56)

6,572,096

13,295

1,492

Trick (n =29)

30,269,261

52,793

6,796

Comedy (n =8)

878,194

8,457

280

Commercial (n=1)

406,362

444

140

Pro shisha (n=3)

111,977

763

76

Anti Shisha (n=2)

103,374

400

646

Fetish (n=3)

67,671

259

21

Total

45,620,026

98,953

12,286

Health (n=5)

As can be seen in the table above, the majority of videos were either instructional, review or trick category, with 87% of the videos falling into these three categories which answers objective two of this study. The trick videos got by far the most views, having just over thirty million views for just twenty nine videos, averaging a million per video. The video with the highest number of views was a trick video called 'Smoke table hookah trick' and had nearly six million views. It is a twenty three second video of a young male blowing smoke over a table that appears to then stick to the table and move very slowly across it. The trick videos also got the most 'likes' with over fifty thousand. The instructional videos were also very popular - these videos were mostly explanations of how to set up shisha pipes correctly with just over seven million views and twenty two thousand likes. There were five videos that were coded as being in the health category, three of which were pro-shisha in their message and two which had been made to warn of the dangers to health of SPS. The pro-shisha videos got an average of two hundred and fifty four likes and twenty five dislikes and the anti-shisha videos got two 39

hundred likes and three hundred and twenty three dislikes each. While the videos got a similar number of the likes, the anti-shisha videos got thirteen times as many dislikes and got the most dislikes out of any of the categories. This perhaps suggests that people react negatively to hearing the health effects of shisha or do not believe the health information that they are presented with. Objective three of this study set out to find out how popular the shisha pipe videos are and from the results in the table above, it would appear that they are very popular to watch. The one hundred and twenty six videos had an overall view count of over forty five million views and nearly one hundred thousand likes. Although this does not take into account the length of time each video has been on the site, just the sheer number of views still suggests they are very popular to watch. To find part of the answer to objective one (explore if they are either positive or negative about SPS) the videos were coded for their stance on SPS and the results and percentages for the different classifications are shown in the pie chart below (figure 4.3).

Classification of videos 0%

Positive (79)

35%

Negative ( 2) Complex (2) Unclear (44) 62%

Neutral (0)

2% 1%

Figure 4.3 Classification of the videos by their stance on shisha pipe smoking As demonstrated above, sixty two percent (79) of the videos were positive in their message. The term 'positive' was determined as any video where SPS was portrayed as attractive, fun, powerful, pleasurable, relaxing or sexy. Any video that 40

talked about shisha in a positive light - including reviews of shisha products that were talked about positively - were included in this category. There were only two videos coded that were actively 'negative’, determined as SPS portrayed as undesirable, unattractive or harmful. One of these videos was produced by Brent Council and NHS Brent as a promotional video to raise awareness of the dangers of using a shisha pipe and had received 22587 views at the time of coding. The other video was produced by a man who did mock instructional video, but advising viewers how to do the most damage to themselves along the way. That only two negative videos were found is consistent with previous research (Seitz et al., 2011). It is however concerning, if not surprising, that out of this sample, only 2% of the videos contained any messages about the health effects of shisha and 79% of the videos portrayed SPS in a positive light.

Another large category was videos that were 'unclear', determined as having no clear shisha message. The vast majority of videos in this group were 'trick' videos where the person was usually blowing smoke rings and these videos couldn't be described as having any real clear message about SPS. The different messages that each video gave in the audio about SPS were coded to answer part of objective one and the results are shown below in a bar chart (figure 4.4).

41

35 30 25 20 15 10 5 0

less harmful natural water filter baked tobacco fun cool fruit as bad as cigs cause cancer 100 cigarettes Is addictive infectious mouthpiece rusty pipe/ diseases refuting warnings stomach bugs contains additives more tar than cigs lighter smoke than… no control over CO… pesticides herbal shisha no sickness cleaner fresh strong smell sweet taste thick smoke delicious sociable nice juicy relaxing buzz spirtual

Number of videos

Messages about shisha pipe smoking

Messages about shisha Figure 4.4 The different messages about shisha pipe smoking and the number of videos they featured in. As evident above, there were a myriad of different messages about SPS in the videos, the majority conveying positive messages about shisha. There were a few negative health messages but they were confined to two videos. The most common messages were those that were frequently repeated in review videos, where the presenter, reviews a particular brand or flavour of shisha tobacco. By far the most frequent messages related to 'thick smoke' or 'clouds' as they are usually referred as, appearing in thirty one videos, followed by 'strong smell or flavour' , twenty three videos, and 'sweet taste' , in nineteen videos. Although the only messages that were recorded for this study were the direct words that were used, it is interesting to surmise what were the covert messages were that were given out by the videos. Of course those messages would be different to each viewer, but it seems obvious that the videos that may not directly say in the audio that SPS is 'cool' , but if the video features hip hop music, a group of young men dressed in fashionable cloths, sat around laughing and relaxing while smoking a shisha pipe, might say to some people that SPS is 'cool' without actually directly mentioning the word 'cool' itself. This type of analysis was outside the breadth of this study, however it would certainly make for a further research project.

42

Each video was also coded for the production quality (low, medium or high) and the presence of any smoking in the videos or any brand content (either in audio or visual) was recorded. The table below (4.5) shows the results. Table 4.5 The video quality and the presence of smoking and/or brand content in the videos Video Quality

Presence of smoking

No smoking present

Presence of any brand content

No brand content

Low (n =102)

N=90 (88%)

N=12 (12%)

N=61 (60%)

N=41 (40%)

Medium (n=24)

N= 21 (88%)

N=3 (12%)

N=15 (63%)

N=9 (37%)

High (n=1)

N=0 (0%)

N=1 (100%)

N=1 (100%)

N=0 (0%)

This tables answer objective two of this study and demonstrates that the videos were in the large majority low quality, one hundred and two (81%) of the videos fell into this category. These were all recorded on hand held devices such as camcorders or even on mobile phones. The medium category was much less frequent, twenty four videos (17%). These were determined as videos that had special effects such as added music or video titles. There was only one video which was classed as 'high' quality; determined as being of television style production quality in terms of sound, audio and special effects. Interestingly however, this video was for a new brand of shisha pipe called 'Desvall', billed as the 'world's most exclusive shisha pipe'. This TV style advert for this new shisha pipe make is advertising shisha to a different audience, turning the shisha pipe into an exclusive, luxury item. Although only posted online on YouTube in January 2013, this video has already had just over two hundred thousand views when coded in on the 6th of June and by the 24th of July had received another two hundred thousand hits. Receiving this volume of hits shows how popular shisha has become; however with a cost of up to $60,000, it can’t be 43

imagined that they are selling that many products. Also coded were the basic demographic information about the people who appeared in the videos, the results can be seen below in figure 4.6. Figure 4.6 The age and gender of people appearing in the videos

Age and gender of people in the videos 140 120 100 80

Age and gender of people in the videos

60 40 20 0 Male Female 18-24 24-30 30-40 40-50 50-60

As expected, it is clear that the vast majority of videos feature young males between the ages of 18-24 with one hundred and nine videos fitting into this category. This finding is consistent with the prior research (Seitz et al., 2011) and supports research that young males are more likely to use shisha pipes than young females (Eissenberg, Ward, Smith-Simone & Maziak, 2008 ; Maziak et al., 2004). The three videos that only featured females were all classed as being in the ‘fetish’ category. These videos all featured a young female inhaling and exhaling from the shisha pipe in a sexual manor. In a study in 2010 (Kim, Paek & Lynn), a content analysis of two hundred videos on YouTube only identified one fetish video that featured a shisha pipe. The presence of three shisha pipe fetish videos on YouTube perhaps reflects the growing popularity of shisha pipes around the world generally. To answer objective four, as well as recording smoking and brand content presence, particular brand names that feature, either in audio or visual , were recorded and the results are listed in the table (4.7) below. Any particular flavours that were mentioned or shown are listed and then the number of times that particular brand appeared in a video are shown in brackets.

44

Table 4.7 Brand names and flavours of shisha tobacco in the videos and the number of videos they featured in. Shisha tobacco brands and flavours

Types of coals

Types of hookahs

Other

AF peach and vanilla (1)

Lemonara coals (1)

khalil mamoon (3)

Al Fakher travel briefcase (1)

Al Fakher :peach, double apple, cappuccino, Eskandaranl Apple, grape, grenadine, gum cinnamon, kiwi, mint, orange, plum mint, strawberry, sweet orange, vanilla, watermelon, After nine, pina colada(28)

Japanese coals (1)

Desvall shisha pipe (1)

Namore hose (1)

Dazed and confused (1)

Chronic hookah natural coals(3)

Fumo hookah (1)

Blueberry ice drops hookah gel (1)

Tangiers tobacco (2)

Zamookah coals (1)

Leila portable hookah (1)

Harmony hookah bowl (1)

Haze : cantaloupe, peach, twice the apple, trash can punch (5)

Holland coals (1)

Mya hookah (1)

Meduse hookah pipe(1)

Mellow tobacco (1)

Three kings coals (1)

Exhale bowl (1)

Natural American spirit (2) Noir Kashmir apple (1)

Social smoke potion 9 (1)

Starbuzz : Xmas mix, acid red, blue mist, chocolate mint, citrus mist, coco jumbo, double apple, exotic sex on the beach, 45

fantasia ice mint, Lebanese bombshell, simply mint, tangerine dream, UFO, Watermelon, white mint (25) Triple XXX Fantasia hookah tobacco (1)

Eleven different brands of shisha tobacco were mentioned, with no surprises that both Al Fakher and Starbuzz came up the most frequently as they were two of the search terms used, mentioned twenty five and twenty eight times respectively. The presence of so many other brands however was slightly surprising but what was interesting and unexpected was the amount of other products, other than shisha tobacco, that were mentioned in these videos. Particularly in videos produced by regular reviewers of shisha flavours and tobacco, they would not only be talking about and reviewing the tobacco, but also numerous other products. A typical video could mention myriad products including the type and flavour of tobacco being used, the brand of the hose, the type of bowl and even the brand of the coals that were being used. Although this research originally only set out to the record the brands of shisha tobacco that were mentioned, as seen in the table above (table 4.7), it soon became apparent that multiple products could be talked about and in some cases promoted in a single video. Twenty seven of the videos also promoted specific websites that they appeared to be advertising - either through mentioning the name of the website where they bought their products from or by having a graphic at the beginning of the video that advertised a website name. For example, one of the most prominent authors of the videos was called 'Hookah Frank'; thirteen of his videos were included in the sample. He is a professional shisha reviewer, with hundreds of review videos on YouTube and he has over eighteen thousand people who subscribe to his YouTube channel. That means that if somebody chooses to subscribe, they will have every new video that he posts on YouTube sent to them via an email. Hookah Frank starts each video with the line 'Hi this is Frank from the Royal Hookah Forum.com'. Many other websites that sell shisha products were also heavily promoted via the videos in the sample. These included: 46

www.hookahpro.com www.nazarhookah.com www.5starhookah.com www.hookah-shisha.com www.hookahforum.com www.shishaforum.com www.hookahdomain.com www.hookahjohn.com www.shishashop.com www.hookahwisperer.com www.hookahdaily.com www.smoking-hookah.com

As demonstrated above, the videos were not only featuring numerous different products, but also acting as advertisements for various different websites. The power of advertising via YouTube has obviously not been lost on the makers of the products either as in the videos there were several mentions of suppliers sending reviewers like Hookah Frank free samples so that they would review them and put them on YouTube. Three of the videos in the sample were also competition entries for a competition run by Starbuzz tobacco in December 2012. The competition was promoted via the companies blog and was promoted as follows: Starbuzz® Giveaway Make a video commercial about why Starbuzz® is the best tobacco in the market! Tell the world what separates your favorite brand from the rest. Passion and enthusiasm is the key here! Rules: -

Film a video of your commercial idea. The theme is why Starbuzz® Tobacco is the

best. You can be as simple or elaborate as you want but the best commercial will be chosen as winners.

47

-

Upload your video to YouTube on to your own channel, and e-mail the link to

[email protected]. Please also include a paragraph written about why you personally chose Starbuzz® Tobacco over the rest. -

You must also go to www.starbuzzblog.com and under this contest, write a

comment with your paragraph on why Starbuzz® Tobacco is the best as well as post the link to your video you uploaded. Prizes: -

1st Place winner will get:



1 x 1000g of Starbuzz® Tobacco unreleased flavor



1 x Starbuzz Egyptian Hookah



Cash prize of $1,000!

That Starbuzz ran such a competition suggests that they see YouTube as a very powerful tool in the use of the marketing of their product and as a way of communicating with the audience that they want to attract. This is particularly pertinent when comparing the banning of advertising of all tobacco products in the UK in 2002 through the tobacco advertising and promotion act. No advertisement for any tobacco product is allowed on television or a website, however Starbuzz tobacco can run a competition for its customers to make advertisements for them and they can be put on YouTube for eternity and viewed by thousands of people in the UK without even any age restrictions. It suggests that companies like Starbuzz are utilising a gap in the tobacco advertising regulations through the use of YouTube as a way of promoting its products in a way no longer possible through normal media channels in the UK and now in many part of the world.

48

Chapter Five – Conclusion Research objectives: summary of findings and conclusions 5.1 Research objective 1 Summary of findings

Objective one of this study really has two parts, the first part was the find out what were the most common messages about SPS on YouTube. There was no evidence in the literature review which answered this question. The results of this study showed the three most common messages to be very literal messages, usually mentioned when reviewing a shisha tobacco. The three most common messages by far were related to 'thick smoke' or 'clouds' as they are usually referred as, appearing in thirty one videos, followed by 'strong smell or flavour' , twenty three videos, and 'sweet taste', in nineteen videos. Part two of objective one was also to assess whether the videos showed explore if they are either positive (portrayed as attractive, fun, powerful, pleasurable, relaxing or sexy) or negative (portrayed as undesirable, unattractive or harmful) messages. This objective was quite clearly answered, as 62% of the videos were classified as being positive and only 1% being negative. Conclusion

In conclusion, it would seem that both parts of this research objective were answered by the empirical research. The positive messages about SPS were much more practical than had been anticipated, the three most popular messages were all messages that were repeated in shisha review videos. Review videos on YouTube have become a popular means for shisha pipe smokers all over the world to review new flavours and share their opinions on brands. The main concern with this finding however is that only two of the one hundred and twenty six videos contained any warning about the health consequences of using a shisha pipe. With the emerging evidence identified in the literature review about the health effects of using these pipes, it is of great concern that any young person searching YouTube for videos about shisha pipes, would, in the vast majority of videos, only hear positive 49

messages. With tobacco advertising banned in many countries now, it is of great consternation that young people, of any age, can log on and see these videos that are, in the large majority, pro-shisha in their stance.

5.2 Research objective 2 Summary of findings

Again objective two had two parts, the first being to categorise what types of videos are on YouTube that include shisha pipes(instructional, review, trick, comedy, commercial, health or fetish). The majority of videos were either instructional, review or trick category, with 87% of the videos falling into these three categories. The other part of this objective was to find out what the production quality was of the videos available (low, medium or high). This was something that had not been covered by any other study of shisha pipe videos. The videos were in the large majority low quality, one hundred and two (81%) of the videos fell into this category. These were all recorded on hand held devices such as camcorders or even on mobile phones. This style of very low quality, hand held videos taken on mobile phones was typical of the trick or instructional videos. There were however a minority of medium quality videos that had things like animated titles, usually advertising a website name, which showed that some of these videos were becoming a bit more sophisticated. Conclusion

The findings confirm that there are numerous types of videos on YouTube featuring shisha pipes, with the review videos being the most frequent category in this study. YouTube is fulfilling a need for shisha pipe users worldwide to swap information about different flavours and brands that no other traditional form of media is currently doing . Most of these videos take place in the home or garage and have a very 'home made' feel, linking with the finding that the majority of videos are of low production quality. The 'medium' quality videos with titles, music and a more professional feeling may however suggest a graduation to a more sophisticated and professional use of YouTube by companies who want to use videos to advertise their websites that sell shisha related products or tobacco. 50

5.3 Research objective 3 Summary of findings

Objective three was to find out how popular the videos of shisha pipes were using the number of views and 'likes' and 'dislikes' as measures. The results clearly suggest that the videos are popular, the one hundred and twenty six videos have over forty five million views, one hundred thousand 'likes' and just twelve thousand 'dislikes'. Although the views span varying lengths because of the different upload dates, it seems clear that the videos are popular based on the sheer number of views. The videos also had a huge breadth in the number of views, ranging from five and half million for a video called 'smoke table hookah trick' to just twenty nine for a review of Al Fakher mint tobacco, uploaded by 'TheNickshift'.

Conclusion

In conclusion, the SPS videos are popular, with lots of views and also attracted five times as many likes as dislikes. The trick videos did however have the majority of the views, with just over thirty million views for twenty nine videos. There were only five videos on the health theme, perhaps suggesting that people still see YouTube as a somewhere to watch fun videos rather than seek out health information in the main. However, with the videos being so popular, perhaps YouTube would be exactly the right place to target health messages about SPS. Currently the two videos that promote health messages are far outnumbered by other types of videos.

5.4 Research objective 4 Summary of findings

Objective four set out to investigate the prominence of shisha tobacco products and find out which brands were mentioned most frequently. The majority of the videos (61%) featured a tobacco brand, with eleven different brands featured. Obviously Al Faker and Starbuzz which going to be mentioned frequently as they were two of the search terms so this part of trying to answer the objective was slightly flawed. 51

However, even excluding those brands, nine other tobacco brands were mentioned (see Table 4.7) It wasn't just tobacco brands that were featured either, many videos featured several different products including the brand of hose, bowl, coals or shisha pipe itself. Twenty seven of the videos also included references to specific websites that they were either advertising for money possibly or in return for freebies.

Conclusion

In conclusion, shisha tobacco branding was widespread in the videos that were watched, usually through being talked about in the audio and then the packaging held up to the camera for the viewer to see. Smoking in these videos was also prolific with 88% of the videos featuring a person or persons smoking a shisha pipe. Slightly unexpected however were the prominent featuring of other products other than tobacco products, like coals, for example, and also the preponderance of advertising of shisha related websites. That shisha tobacco and other products are being advertised via YouTube videos is of great concern as it is managing to circumvent tobacco advertising laws.

5.5 Limitations

This study was not of course done in the most ideal setting and its limitations need to be taken into account when reading the results. Firstly, this study could only capture the results that were available on YouTube on one specific date and as the inherent nature of the internet means that it is changing on a daily basis, different results could have been produced on a subsequent date. The impact of this was lessened by both searching by the 'relevance' and also by the 'most viewed' function, and also by watching a large enough sample of videos. However, the intrinsic nature of studying the internet means that this problem can never fully be eradicated completely. This study was also limited by time; if time had permitted another part of the research project that would have added extra information to this study would have been to measure the views of the videos at two different points in time. This would 52

have enabled the study to get a better idea of how popular the videos were. However, the actual research part of this study was done after the introduction, literature review and research methods part of the dissertation, meaning that there was then no time to leave a three or six month gap to take a second measurement of views.

5.6 Recommendations

Objective five of this study was to formulate recommendations for public health departments in Local Authorities and also for social networking sites themselves. Starting with recommendations for public health departments, it is clear from the results of this study that there are very few videos on YouTube that demonstrate the health dangers of using shisha pipes. Only two such videos were found in our sample. Making a video no longer has to be expensive or time consuming and uploading videos to YouTube is free. Therefore it would seem a logical recommendation of this study that public health departments think about using a YouTube video when formulating public health campaigns about SPS. Leaflets and posters are the most common components of usual health campaigns but by using YouTube, public health departments can get their messages to young people both cheaply and effectively. Another recommendation of this study is based on conclusion's two and four of this study, that the majority of videos on YouTube are positive in their stance and also advertise or show tobacco products with no age restrictions applicable. This suggests that YouTube is being used as a way of circumventing tobacco advertising rules around the world. You would not be able to advertise shisha tobacco on a website, like pay for an advert on the Daily Mail website for example, however YouTube seems to operate in a vacuum of regulation. Another recommendation of this research is that Governments start to think about how they can apply regulation to website like YouTube so that shisha tobacco companies cannot bypass regulation and still advertise to underage consumers.

53

Another part of the research objective was to provide recommendations for YouTube itself. Based on the conclusions of the study, it seems to suggest that YouTube itself needs to tighten up its own procedures for what it allows to be viewed with no age restrictions. Only two of the videos out of the sample were age restricted, meaning that the viewer had to sign in with a profile where they had stated, by entering their date of birth, that they were over eighteen. It is especially worrying when considering that, in this sample, 88% of the videos contained an image of a person smoking and 62% featured a branded product, usually shisha tobacco. YouTube has lots of polices on its website about what is banned and what videos are age restricted, for example violence, impersonation, hate speech and harassment. Even under their dangerous and harmful content policy, that states 'we draw the line at content that intends to incite violence or encourage dangerous or illegal activities that have an inherent risk of serious physical harm or death.' There is then however no mention of tobacco under the list of activities it considers to cause serious physical harm or death, which is strange considering tobacco kills half of the people who use it. YouTube has a moral responsibility to protect the users of its website which it is currently not fulfilling by allowing positive shisha footage and not even giving it an age restricted status. This research project has provided a comprehensive study of videos on YouTube that contain SPS, but there are still further areas of research that could complement this study. For example, as referred to in the 'results' section, this study only coded words that were actually spoken. It is interesting to surmise what were the surreptitious messages were that were given out by the videos. This type of analysis was outside the breadth of this study, however it would certainly make for a further research project. It would also be an interesting project to ascertain young people's views on the videos themselves, and what messages they took away from them about SPS. Of course there are other websites such as Facebook, Twitter, Instagram and a whole host of other social media websites that will also contain messages and images of SPS. It is important to further the conclusions of this research study that other websites are researched in a similar manner. While this may be already happening for cigarettes, so far SPS has not received the same attention.

54

References Akl, E. A., Gaddam, S., Gunukula, S. K., Honeine, R., Jaoude, P. A., & Irani, J. (2010). The effects of waterpipe tobacco smoking on health outcomes: a systematic review. International Journal of Epidemiology, 39(3), 834-857. Al-Belasy, F. A. (2004). The relationship of “shisha”(water pipe) smoking to post extraction dry socket. Journal of Oral and Maxillofacial Surgery, 62(1), 10-14.

Alexa,

2013.

Alexa

top

500

global

websites.

(online)

Available

at

http://www.alexa.com/topsites accessed 21/01/2103

All Twitter, 2012. Twitter, Facebook, Google, YouTube – What Happens On The Internet

Every

60

Seconds?

(online)

http://www.mediabistro.com/alltwitter/data-never-sleeps_b24551

Available

at

accessed

31/01/2013 American Lung Association. (2007). Tobacco Policy Trend Alert. An emerging deadly trend: Waterpipe tobacco use. Austin, E. W., Pinkleton, B., & Fujioka, Y. (1999). Assessing prosocial message effectiveness: Effects of message quality, production quality, and persuasiveness. Journal of health communication, 4(3), 195-210. Aveyard, P., & Roskin, J. (2009). Canadian and English students’ beliefs about waterpipe smoking: a qualitative study. BMC Public Health.9 (10). Backinger, C, Pilsner, A, Auguston, E, Frydl, A, Philips, T & Rowden J. (2010). YouTube as a source of quitting smoking information. Tobacco Control. 20. pp.119 – 122. Bauer, M, W. (2000) Classical content analysis : A review. In : Bauer, M. W., & Gaskell, G. (2000). Qualitative researching with text, image and sound: A practical handbook for social research. Sage publications.

55

Berelson, B. (1952). Content analysis in communication research. Glencoe, IL: Free Press.

Borzekowski, D. L., & Rickert, V. I. (2001). Adolescent cybersurfing for health information: a new resource that crosses barriers. Archives of Pediatrics & Adolescent Medicine, 155(7), 813. Bowling, A. (2002). Research methods in health – investigating health and health services. 2nd edition. Berkshire: Open University Press. Briones, R., Nan, X., Madden, K., & Waks, L. (2012). When vaccines go viral: An analysis of HPV vaccine coverage on YouTube. Health Communication, 27(5), 478485. British Heart Foundation.(2012). Rise in ‘shisha bars’ prompts warning over health dangers. (online). Available

at

http://www.bhf.org.uk/default.aspx?page=14417

accessed 31/01/2013. Carroll, M. V., Shensa, A. S., & Primack, B. A. (2012). 50. A Comparison of YouTube Videos Related to Cigarette and Hookah Tobacco Smoking. Journal of Adolescent Health, 50(2), S39-S39.

Charlesworth, A., & Glantz, S. A.(2005). Smoking in the movies increases adolescent smoking: a review. Pediatrics, 116(6), 1516-1528.

Cobb, C. O., Khader, Y., Nasim, A., & Eissenberg, T. (2012). A multiyear survey of waterpipe and cigarette smoking on a US university campus. Journal of American College Health, 60(7), 521-527.

Combrink, A., Irwin, N., Laudin, G., Naidoo, K., Plagerson, S., & Mathee, A. (2010). High prevalence of hookah smoking among secondary school students in a disadvantaged community in Johannesburg. SAMJ: South African Medical Journal, 100(5), 297-299.

56

Dar-Oheh, N, Bakri, F, Al-Omiri, M, Al-Masni, H, Eimar, H, Khraisat, ... AbuHammad, O. (2010) Narghile (Waterpipe) smoking among university students in Jordan: prevalence , pattern and beliefs. Harm Reduction Journal. 7 (10). Denscombe, M. (2007). The good research guide for small-scale social research projects. 3rd edition. Berkshire : Open University Press. Devlin, E., Eadie, D., & Angus, K. (2003). Tobacco marketing and young people. NHS Scotland. Eissenberg, T., Ward, K. D., Smith-Simone, S., & Maziak, W. (2008). Waterpipe tobacco smoking on a US College campus: prevalence and correlates. Journal of Adolescent Health, 42(5), 526-529. Elkin, L., Thomson, G., & Wilson, N. (2010). Connecting world youth with tobacco brands: YouTube and internet policy vacuum on Web 2.0. Tobacco Control. 19 (5), 361-366. Freeman, B. (2012). New media and tobacco control. Tobacco Control, 21(2),139144. Freeman, B., & Chapman, S. (2007). Is “YouTube” telling or selling you something? Tobacco content on the YouTube video-sharing website. Tobacco Control. 16, 207210. Freeman, B., & Chapman, S. 2009. Open source marketing: Camel cigarette brand marketing in the “Web 2.0” world. Tobacco Control. 18, 212-217. Gilbet, N. (2003). Researching social life. 2nd edition. London: Sage publications. Glantz, S. A., Kacirk, K. W., & McCulloch, C. (2004). Back to the future: smoking in movies in 2002 compared with 1950 levels. Journal Information, 94(2).

Gray, N. J., Klein, J. D., Noyce, P. R., Sesselberg, T. S., & Cantrill, J. A. (2005). Health information-seeking behaviour in adolescence: the place of the internet. Social science & medicine, 60(7), 1467-1478.

57

Hammal, F., Mock, J., Ward, K. D., Eissenberg, T., & Maziak, W. (2008). A pleasure among friends: how narghile (waterpipe) smoking differs from cigarette smoking in Syria. Tobacco Control, 17(2), e3-e3.

Heck, G., & Moule, P. (2006). Making sense of research: an introduction for health and social care practitioners. London: Sage publications. Holsti, O.R. (1968) ‘Content analysis’ in G. Lindzey and E. Aronson (eds), Handbook of social Psychology. Reading, MA: Addison-Wesley, 596-692. http://mashable.com/2011/02/19/youtube-facts/ -‘Ten fascinating YouTube facts that may surprise you’, accessed 19th of April 2011. http://www.blackburncitizen.co.uk/news/8867513.Children_paying_to_smoke_in_ille gal _Blackburn_shisha_dens/# Jackson, D and Aveyard, P. (2008). Waterpipe smoking n students: prevalence, risk factors, symptoms of addiction, and smoke intake. Evidence from one British University. BMC Public Health.8:174 Jansen, B. J., & Spink, A. (2006). How are we searching the World Wide Web? A comparison of nine search engine transaction logs. Information Processing & Management, 42(1), 248-263 . Jawad, M., Abass, J., Hariri, A., Rajasooriar, K. G., Salmasi, H., Millett, C., & Hamilton, F. L. (2013). Waterpipe smoking: prevalence and attitudes among medical students in London [Short communication]. The International Journal of Tuberculosis and Lung Disease, 17(1), 137-140.

Jawad, M., Bakir, A. M., Ali, M., Jawad, S., & Akl, E. A. (2013). Key health themes and reporting of numerical cigarette–waterpipe equivalence in online news articles reporting on waterpipe tobacco smoking: a content analysis. Tobacco control. Online First edition.

Jawad, M., Khaki, H., & Hamilton, F. (2012). Shisha guidance for GPs: eliciting the hidden history. The British Journal of General Practice, 62(595), 66. 58

Jawad, M., Wilson, A., Lee, J. T., Jawad, S., Hamilton, F. L., & Millett, C. (2013). Prevalence and Predictors of Water Pipe and Cigarette smoking among secondary school students in London. Nicotine & Tobacco Research.Online First edition.

Jawaid, A., Zafar, A. M., Rehman, T. U., Nazir, M. R., Ghafoor, Z. A., Afzal, O., & Khan, J. A. (2008). Knowledge, attitudes and practice of university students regarding waterpipe smoking in Pakistan. The International Journal of Tuberculosis and Lung Disease, 12(9), 1077-1084.

Jenssen, B. P., Klien, J., Salazr, L., Daluga, N & DiClemente, R. (2009). Exposure to tobacco on the internet: Content analysis of adolescent’s internet use. The American Journal of Paediatrics. 124 (2), 180- 186. Kaiser Family Foundation. (2001). Generation Rx.com: How young people use the internet for health information. Menlo Park, California, Kaiser family foundation. Kandela, P.(2000). Narghile smoking keeps Arabs in wonderland. The Lancet. 356:1175. Kaplan, A. (1943). Content analysis and the theory of signs. Philosophy of Science, 230-247.

Keelan, J., Pavri-garcia, V., Tomlinson, G & Wilson, K. (2007). YouTube as a source of information on immunization: A content analysis. Journal of the American Medical Association. 298 (21). Kim, K., Paek, H. J., & Lynn, J. (2010). A content analysis of smoking fetish videos on YouTube: regulatory implications for tobacco control. Health Communication, 25(2), 97-106.

Kiter, G., Ucan E., Ceylan, E & Kilinc, O. (2000). Waterpipe smoking and pulmonary functions. Respiratory Medicine. 94. 891-894. Knishkowy, B & Amitai, Y. (2005). Water-pipe smoking : An emerging health risk behaviour. The American Academy of Paediatrics. 116. 113-119. 59

Korn, L., & Magnezi, R. (2008). Cigarette and nargila (water pipe) use among Israeli Arab high school students: prevalence and determinants of tobacco smoking. The Scientific World Journal, 8, 517-525.

Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Sage Publications. London.

Lovato, C., Watts, A., & Stead, L. F. (2011). Impact of tobacco advertising and promotion on increasing adolescent smoking behaviours. Cochrane Database Systematic Review, 10.

Lyons, A. (2012) Tobacco and alcohol in films and on television. PhD thesis, University of Nottingham.

Lyons, A., McNeill, A., Chen, Y., & Britton, J. (2010). Tobacco and tobacco branding in films most popular in the UK from 1989 to 2008. Thorax, 65(5), 417-422.

Marvasti, A. (2004). Qualitative research in sociology. London. Sage. Mason, J., 2002. Qualitative researching. 2nd edition. London: Sage publications. Maziak, W, Eissenberg, T, Rastam, S, Hammal, F, Asfar, T, Bachir, M & Ward, K. (2004). Beliefs and attitudes related to narghile (waterpipe) smoking among university students in Syria. Annals of Epidemiology. 14 (9). pp646-654 Maziak, W., Ward, K.D., Soweid, R., Afifi, A & Eissenberg, T. (2004). Tobacco smoking using a waterpipe : a re-emerging strain in a global epidemic. Tobacco Control. 13. 327-333. McNeill, P & Chapman, S. (2005). Research methods. 3rd edition. New York: Routledge. Munckhof, W. J., Konstantinos, A., Wamsley, M., Mortlock, M., & Gilpin, C. (2003). A cluster of tuberculosis associated with use of a marijuana water pipe. The International Journal of Tuberculosis and Lung Disease, 7(9), 860-865.

60

NHS Harrow, (2011).Shisha, smoking and young people research. Forster. London. Office for national statistics (2008). General household survey, 2006. London : The Stationary Office. Paisley, W.J. (1969) ‘Studying style as deviation from encoding norms,’, in Gerbner, G., Holisti, O.R., Krippendorff, K., Paisley, W.J., and Stone, P.J. (eds), The Analysis of Communication contents : Developments in scientific theories an computer techniques. New York: Wiley. Primack, Brian, A. Walsh, M, Bryce, C and Eissenberg, T. (2009). Water-pipe smoking among Middle and High schools students in Arizona. The American Journal of paediatrics. 123. 282-288. Qureshi, N., & Lowenstein, E. (2011). The role of nutrition in acne pathogenesis: YouTube as a reflection of current popular thought. Skinmed, 9(5), 279-280.

Raad, D., Gaddam, S., Schunemann, H. J., Irani, J., Jaoude, P. A., Honeine, R., & Akl, E. A. (2011). Effects of Water-Pipe Smoking on Lung FunctionA Systematic Review and Meta-analysis. CHEST Journal, 139(4), 764-774.

Sajid, K., Akhter, M & Malik ,G. (1993). Carbon monoxide fractions in cigarette and hookah (hubble bubble) smoke. Journal of the Pakistani medical Association. 43. 179-182. Salameh, P., Waked, M., & Khayat, G. (2012) Waterpipe smoking and dependence are associated with chronic bronchitis: a case control study. Eastern Mediterranean Health Journal. 18: 996-1004.

Seidenberg, A. B., Rodgers, E. J., Rees, V. W., & Connolly, G. N. (2011). Youth access, creation, and content of smokeless tobacco (“dip”) videos in social media. Journal of Adolescent Health.

Seitz, C. M., Milroy, J. J., Orsini, M. M., Wyrick, D. L., Herring, J., & Koech, M. (2011). You Tube: An International Medium for Sharing Videos about Hookah Smoking. International Electronic Journal of Health Education, 14, 142-149. 61

Shafagoj, Y & Mohammed F. (2002). Levels of maximum end-expiratory carbon monoxide and certain cardiovascular parameters following hubble-bubble smoking. Saudi Medical Journal. 23. 953-958. Shafagoj, Y., Mohammed, F., & Hadidi, K. (2002). Hubble-bubble (waterpipe) smoking: levels of nicotine and cotinine in plasma, saliva and urine. International Journal of clinical pharmacology and therapeutics. 40. 249-255. Silverman, D., (2000). Doing qualitative research – a practical handbook. London: Sage publications. Silverman, D., (2011) Interpreting qualitative data. London : Sage publications. Skinner, H., Biscope, S., Poland, B., & Goldberg, E. (2003). How adolescents use technology for health information: implications for health professionals from focus group studies. Journal of Medical Internet Research, 5(4).

Smith, J. R., Novotny, T. E., Edland, S. D., Hofstetter, C. R., Lindsay, S. P., & AlDelaimy, W. K. (2011). Determinants of hookah use among high school students. Nicotine & Tobacco Research, 13(7), 565-572.

Smith-Simone, S., Maziak, W., Ward, K. D., & Eissenberg, T. (2008). Waterpipe tobacco smoking: knowledge, attitudes, beliefs, and behaviour in two US samples. Nicotine & Tobacco Research, 10(2), 393-398.

Sood, A., Sarangi, S., Pandey, A., & Murugiah, K. (2011). YouTube as a source of information on kidney stone disease. Urology, 77(3), 558-562.

Steinberg, P. L., Wason, S., Stern, J. M., Deters, L., Kowal, B., & Seigne, J. (2010). YouTube as source of prostate cancer information. Urology, 75(3), 619-622.

The Chartered Institute of Environmental Health. (2011). Implementation of smokefree legislation in England: supplementary guidance for local authority regulatory officers on dealing with non-compliance in shisha bars. London.

62

Varsani, S., Ganz, I., Eldor, N. and Garenkin, M. (2003). Waterpipe smoking among school children in Israel: frequencies, habits and attitudes. Harefuah. 142 (11), 73641. Ward, K. D., Hammal, F., VanderWeg, M. W., Eissenberg, T., Asfar, T., Rastam, S., & Maziak, W. (2005). Are waterpipe users interested in quitting?. Nicotine & Tobacco Research, 7(1), 149-156.

Weber, R.P. (1985). Basic content analysis. Beverley Hills, CA: Sage. World Health Organisation. (2005). Advisory note: Waterpipe tobacco smoking – Health effects, research needs and recommended actions by regulators. WHO document production services: Geneva. Yadav, J. S., & Thakur, S. (2000). Genetic risk assessment in hookah smokers. Cytobios, 101(397), 101.

Yoo, J. H., & Kim, J. (2012). Obesity in the new media: A content analysis of obesity videos on YouTube. Health Communication, 27(1), 86-97.

Zahran, F., Ardawi, M and Al-Fayez, S. (1985). Carboxyhaemoglobin concentrations in smokers of sheesha and cigarettes in Saudi Arabia. British Medical Journal. 291, 1768-1770.

63