CONSUMER BOYCOTTS IN INDONESIA NATIONAL PRESS CONTEXT, 1982-MID 2010: A PRELIMINARY STUDY

The objective of this study is to describe how boycotts are released in Indonesia national press. The data were obtained from mass media using saturation sampling. Data obtained is analyzed using qualitative content analysis. The results showed that, as in other countries, in Indonesia, boycott has increased in quantity. In the boycott news, the country of origin target is released more than corporate target. Mostly the boycott in Indonesia is sponsored/triggered by groups. In this case, religious groups are the primary cause. In addition there are environmental and worker groups. Although the boycott in Indonesia is mostly intended to express dissatisfaction to particular corporate or country of origin actions, some boycotts are intended to establish social or political control to the boycott target.


Introduction
In Java, especially in Central Java, there is a proverb of "…, ajining salira seka busana" literally meaning "you are what you wear". In marketing, many researchers have confirmed this (i.e. Levy, 1959;Aaker, 1999;Hogg, Cox and Keeling, 2000). However, on the other hand there is only a few researchs about why people avoid particular products or brands. The avoidance to such particular products or brands is known as boycott, which is the contrast to what is known as brand loyalty. Sen, Gurhan-Canli, and Morwitz (2001) inform that the lack of researchs about boycott is not a good condition, because there are more and more consumers concerned in such behavior. Hoffmann and Muller (2009) write that based on the report of European Values Study Group and World Values Survey Association, boycott participation increased in USA and other industrial countries. For example, in a survey in 1982, 16% of respondents in USA stating that they have ever made boycott to a company's products and about 35% decided to boycott in the future time. This rate increased in 2002 to 26% stating that they have ever boycotted products and 51% stating to have boycott in future time. In Indonesia, for example, the KOMPAS daily of 28/12/2008 wrote that Tangerang Mayor, Bandung Mayor, Depok Mayor, Sukabumi Regent, Probolinggo Regent, Balikpapan Regent, and Indonesian Council of Religious Scholar of South Sumatera prohibited Dewi Persik from any performance. The dangdut singer famous for her "goyang gergaji" was prohibited to be performed in those regions with some reasons. Likewise is when Luna Maya released her first album. The Association of Indonesia National Private Broadcast Radio in Makassar (PRSSNI), South Sulawesi, had commitment not to play the songs of the album of Luna Maya. All members of PRSSNI promised not to broadcast any song of Luna Maya known as singer and presenter. Boycott was made after the police put Luna as the suspect of having committed in porn video. Previously, Indonesian Council of Religious Scholar of Bali Province has refused the arrival of Luna Maya, Nazriel Ilham alias "Ariel", and Cut Tari to Bali Island (BOG, 2010).
Some empirical researchs focusing on consumer boycott generally used quantitative approach (e.g. Klein, Smith, and John, 2002;Klein Smith, and John, 2004;Sen et al., 2001 Hoffmann andMuller, 2009;Farah and Newman, 2010), while qualitative technique is relatively rarely used (e.g. Kozinets and Handelman, 1998;Lee, Motion, and Conroy, 2009;Braunsberger and Buckler, 2010). From the aforementioned description, by using qualitative approach, this study contributes to the existing literature by describing boycott in Indonesia national press. So, the objective of this study is to describe how boycotts are released in Indonesia national press. To achieve such goal, the research questions of study are as follow: a) How mass media put boycott in news release, b) What is the target of boycott, and c) What triggers boycott in Indonesia.

Agenda Setting Theory and Previous Research
Walter Lippman introduces agenda setting theory (Littlejohn and Foss, 2009). Rakhmat (2007) informs that McComb and Shaw do the first empirical research about this theory. McComb and Shaw (1972) examine the agenda setting in US presidential campaign in 1968 and made hypothesis that mass media determined the agenda for each political campaign, that affect the projected attitude to political issues. Furthermore, Severin and Tankard Jr (2001) report that agenda setting theory refers to media ability, with repeated news coverage, to raise the importance of an issue in public mind. Assumption in agenda setting theory is that mass media filter news, articles, or writings to be published. Selectively, the "gatekeepers" such as proofreaders, editors, and even journalists to determine which news deserve to be published, and which others have to be declined. Because the readers, audience, and listeners get information mostly from mass media, the agenda of media is certainly related with the agenda of society. For example, although only 33 miners saved, President Cile, Sebastian Pinera, is praised by many ordinary people and leaders around the world. His commitment and consistency in mine workers rescue have inspired many people and boosted his popularity. Not only is the President of Cile who became famous, but also the 33 workers who were rescued. Congratulations, a gift of goods from various people, and an invitation from two major football clubs, Manchester United and Real Madrid, to watch the game over to them for free are accepted. This is because of the media, electronic or print, which proclaim the news is constantly and repeatedly. If the media can make ordinary people become famous, then the media can also increase the attention of people about the boycott.
In line with agenda setting, McQuail (1983) stated that: First, mass media can serve as a mirror of public life. Second, mass media can also be viewed as a social agent or power. As a mirror, mass media is assumed to be an institution that process social facts in the society. In this context, the mass media creates agenda of social facts. In other words, what is released by the mass media is important social document that portrays the real condition of the society. On the contrary, when the events are concealed, the events are considered unimportant. Meanwhile as an agent, the mass media is assumed to be a social institution taking part in the creation of public opinion and public attitude on particular issues or events. The mass media can create the agenda of mind and perception determinant on particular issues and events. What is not released by the mass media is considered unimportant for the society.
Hoffmann and Muller (2009) inform that the literature about consumer boycott consists of three main categories. They are a) the literature discussing the frequency, causes, and goal of boycott. In this category, most data of researchs are historical data that describes the frequency and nature of consumer boycott, b) the literature discussing about the consequences of boycott. In this category, researchs focus on the impact of boycott on financial performance and reputation of the corporates, and c) the literature discussing about individual motivation that becomes the basis to participate in boycott. In this category, most researchs rely on cost-benefit approach and socio-psychological theories.
Friedman in Hoffman and Muller (2009) distinguished boycott based on the purposes. Those are: instrumental boycott and expressive boycott. Klein et al. (2002) added another purpose of boycott. Thus there are instrumental boycott, expressive boycott, and clean hands boycott. Instrumental boycott is the boycott intended to force the target of boycott to change the policy or behavior. Expressive boycott is a general form of boycott to protest the target of boycott. The difference between the instrumental boycott and expressive boycott can be seen from the purpose. The objective of instrumental boycott is clearly stated and specifically measured while expressive boycott is not clearly stated and not specifically measured because it is only a form of anger expression or frustration to the target of boycott. Meanwhile clean hands boycott is the boycott in which consumers participate to have comfort and avoid guilt. Shaw in Braunsberger and Buckler (2010) inform that the purpose of consumers' participation in boycott is to express disappointment to particular corporate or country of origin action. Sen et al., (2001) inform that presently demonstrated boycotts are intended to be social and political control to the target of boycott and try to change or cancel unethical and irresponsible behavior. Further Klein et al., (2004) inform that there are some costs related with the one's availability to participate in boycott. The lower the perceived cost the easier the person participated in boycott. Meanwhile, Sen et al., (2001) inform that consumers are more likely to participate in boycott when there are equally good substitute products. From the aforementioned description, there are 5 propositions in this study: P 1 : Mass media put product boycott in balance or cover both side P 2 : There is a different frequency of boycott news release in both inside and outside Indonesia in mass media P 3 : For news release of boycott in Indonesia, there is different frequency of news release about the targeted corporate product boycott and targeted product origin in mass media P 4 : There is different frequency of news release about the reasons (political, economic, laws, environmental) of boycott in mass media P 5 : Mass media report that most product boycotts are sponsored/triggered by groups rather than individuals

Research Methods
The method employed is content analysis. Wright (1985) proposes that content analysis is a technique to obtain a description of objective, systematic, and relevant mass communication content. Data collected from this content analysis is used particularly to describe a phenomena. Berelson (1952) notes that content analysis in the 1930s was used by journalistic researchers. However, later, sociological and political researchers and other researchers of different fields including economic researchers also use this method.

Population and Sample
Media used in the study is the print media. The population at this phase is all news in KOMPAS daily. This study selects KOMPAS daily for several reasons: a) KOMPAS daily is a national newspaper, b) KOMPAS daily is the largest circulation, c) KOMPAS daily is independent (neither it is left-oriented nor right-oriented) and impartial (neither to the government nor to non-government agency), and d) KOMPAS daily has all-Indonesia coverage, both the news content and the distribution. This can be seen from the fact that KOMPAS daily has remote printing system, KOMPAS daily update, as well as the website www.KOMPAS.com. Meanwhile, the scope of the study is the news that meet the following criteria: a) the news is about boycott product, and b) the news is put in KOMPAS daily since the beginning to the end of the observation, that is from mid 1965 to 17 June 2010 or during 45 years.
Because of little news found in KOMPAS daily, the sample used in this study is saturation sampling. Black and Champion (1976) inform that saturation sampling is defined as the gather of all sample elements in a particular population having the characteristics needed by the researcher.

Operational Definition
Operational definition is quantification of a nominal/ dictionary definition (Black and Champion, 1976). Kerlinger (2006) states that the operational definition gives meaning to a variable by specifying the operations or activities needed to measure these variables. While Kuncoro (2003) writes that operational definition is the procedure to be followed by researchers in the measurement of a variable. In this study, operational definitions are as follows: a. Cover both side is the balance of reporting made by a journalist to cover both parties in conflict (Kusumaningrat and Kusumaningrat, 2007). b. Target is an object, person, or place that is deliberately chosen to be attacked (Longman, 2003). In this study, target is defined as a country of origin or a corporate, choosen purposively for boycott. Country of origin is the country of manufacture, production, or growth where a product comes from. Corporate is a legally-based business activity intended to generate profit while conforming the prevailing regulation. c. Purpose or aim is something hoped to achieve by a plan, action, or activity (Longman, 2003). In this study, boycott purposes are instrumental boycott, expressive boycott, and clean hands boycott (Klein et al., 2002). d. Reason is the cause or fact that explains why something has happened (Longman, 2003). In this study, reasons are defined as the cause/ trigger (politic, lawsl, economics, environment) why boycott has done. e. Sponsor is group or someone who officially introduces or supports a proposal for a new law (Longman, 2003). In this study, a sponsor is the individual or group that provides the support to boycott action. Berelson (1952) reports that in most cases validity does not seem to be a major problem in the content analysis. With a careful operational definition and good and correct chosing of indicators, the coding sheet is assumed to measure what it should be measured. Thus, to test the validity this study use face validity. Face validity here is a judgment by scientific community that the indicator in coding sheet really measures the construct. Neuman (2000) informs that inter-coder reliability is a common type of reliability reported in content analysis. Inter-coder reliability arises when there are several observers, raters, or coders of information. This study measures reliability by calculating the value of inter-coder realibility as Holsti (1963) suggests:

Data Analysis
In which: C 1,2 : The number of category assignments on which all coder agree C 1 , C 2 : The sum of all category assignments by all coder After face validity and inter-coder reliability testing, the data was descriptively analyzed. Descriptive analysis was performed by contextualizing the news. Contextualization was performed by coding the consensus and difference among the text and presented some citations from the news to strengthen arguments.

Results and Discussions
The preparation of the coding sheet is consulted to researchers at the Center for Cultural and Popular Media Research. The objective of the consultation was to have face validity test in the coding sheet. In order to assess the face validity of the coding sheet, I submitted the coding sheet and a letter introducing my research objectives. These respondents provided feedback and qualitative comments about coding sheet in general. As a result of this procedure, I reworded some operational definition and indicator to integrate terminologies currently used by professionals. After that, with the assistance from a reasercher from the Institute of Research and Community Service, Gadjah Mada University, reliability is tested. There is a difference in determining the threshold score of acceptance of reliability coefficient. Krippendorf (1980) reports that the lower limit of acceptance of reliability coefficient is 0.80; Scott in Hasrullah (2001) put it over 0.75; while Berelson (1952) put the coefficient score between 0.79 and 0.96. the calculation of reliability at this study is 0.88 or above the score proposed by Krippendorf and Scott, and between the range suggested by Berelson.

How Mass Media in Indonesia Put Boycott in News Release
In democratic society, information is very important. Journalists are expected to provide information needed by the society for decision making (Mencer in Rianto, 2009). In providing such information, journalism has to meet the prevailing principles (Kovach and Rosenstiel, 2001): a) journalism is obliged to hold truth, b) journalism is loyal to the public, c) the essence of journalism is discipline in verification, d) journalists have to be independent of information source, e) journalism has to monitor governmental power, f) journalism has to provide public forum to accommodate criticism and support from society, g) journalism has to provide interesting and relevant matters, h) journalism has to keep the news comprehensive and proportional, and i) journalists have to use intuition.
In collecting the data, 50 news releases containing the phrase "boycott product' was collected. The first news about boycott was put on KOMPAS on July 9, 1982. Since then the news about product boycott is sometimes either present or absent in a year. After the reform era of 1998, there was higher rate of news about boycott. This is partly caused by better democracy in mass media than before. There were various kinds of mass media and they started to have courage to put the ongoing real life on mass media at the risk of possibly offending public officials. The distribution of news about boycott in KOMPAS is presented in Figure 1.

Figure 1. The Distribution of News about Boycott in KOMPAS Daily
Of the news collected, it is seen that KOMPAS daily has covered both sides when putting boycott in news. However, not all news covered both side, especially foreign source news and the news quoted from other news sources. The data of the news that cannot cover both sides is presented in Table 1. Most journalists merely take function as official news dealers (Kusumaningrat and Kusumaningrat, 2007). They attend press conferences held by governmental officials, make notes, and write news. In other words, journalists only follow the scheduled agenda from the source persons. In the news release boycott, KOMPAS daily performs investigative journalism. This is reflected in the exploration of news and the crosscheck to the reality.  As the most reputable newspaper in Indonesia, not only does KOMPAS daily provide domestic news. Especially about boycott, KOMPAS daily also provide news about boycott in foreign countries. The news originates from foreign countries' news sources. Table 3 illustrates the release of boycott news in foreign countries including the targets and products.  In contrast with the news of boycott in foreign countries, news of boycott in Indonesia is more complete. The news includes interview with competent sources. For detail, the following Table 4 describes about boycott in Indonesia, targets, products, and reasons of boycott.

4.2.What is the Target of Boycott in Indonesia?
Of 32 news releases about boycott in Indonesia, there is a difference about target of boycott (See Table 4). Country of origin was dominantly put in news (20 times), while corporate as target of boycott ranked second (9 times). For detail about target of boycott in Indonesia, please see Table 5.

What Reasons and Triggers Boycott in Indonesia
Most boycotts in Indonesia' mass media are caused by political matters including human rights and religions. Most boycotts in Indonesia reported in KOMPAS daily are sponsored or triggered by groups. Friedman (1985) informs that in US, boycotts are initiated by consumer groups, labor unions, minority groups, religious groups, and women groups.  The results of the analysis of this study suggested different findings. In Indonesia, the groups that triggered boycott consisted of religious groups (27.7%), environmental reservation groups (16.7%), labor groups and consumers groups (11.1%). Different from the groups, individuals suggested boycott in KOMPAS daily in the form of special feature of opinion. For detail please see Table 7.
It is interesting to note that not only do mass media reported about boycott, but also about the loss if the boycott was realized. In addition, it is also stated that boycott is a lawful behavior because it is one of the consumers' rights and it is based on market logics. It is even illustrated how to have appropriate and effective boycott. As quoted, Priyono said: "... If a performance of a corporate endangers the environment and local society, we can take a piece of paper and make some columns. The first column contains the names of corporates. The second column contains trademarks of goods / services along with the visual display. The fourth column contains detail of market, buyers, local, national, and global targets. What for? For pubic movement of product boycott until the corporate makes betterment..." (KOMPAS, 4/10/2006).
It was also stated that every individual/ group in Indonesia holds the rights to boycott any product related with Indonesia, because the product is produced in such a way against the prevailing law. As quoted, Transtoto said: "The government requested European and Asian countries to have selective purchase, and boycott wooden products from China, Malaysia, and Russia the raw materials of which are stolen from Indonesia. This is easy because the woods, such as merbau, are only found in Indonesia, not in other countries" (KOMPAS, 19/2/2005) Meanwhile, for boycott beyond Indonesia, it is interesting to quote when Hajime Nakatsuji, a Japanese student said. "What I have done may not make any significant difference, but I have to do something. People will suffer". Boycott to American products was went on after the phone talk between the Prime Minister of Japan and President Bush to show support the disarmament of Saddam Husain (American attack to Iraq) (KOMPAS, 22/3/2003).

Conclusion, Limitation, and Suggestion
The objective of this study is to describe how boycotts are released in Indonesia national press. In reporting boycott, not only do journalisms attend, note, and write what they see, but they also have investigative journalism. In Indonesia, the reported boycott is primarily aimed at the countries of origin rather than the companies. Mostly, boycott in Indonesia is triggered by political reason. Meanwhile, when compared to other groups, religious groups are the groups that most frequently triggered boycott. Most reported boycotts are intended to express consumers' anger or dislike to companies or countries of origin rather than for the purpose of social or political control.
The paper uses data from mass media. And then, data obtained is analized using content analysis. Stokes (2006) mentioned that the limitations of the content analysis are: a) insensitivity; sometimes content analysis becomes a rather ineffective instrument. For example, in the content analysis of violence in television, researchers sometimes only calculate violence without differentiating between the revenge, justice, or action performed wild animals to domesticated animals, b) content analysis sometimes is criticized to be too descriptive, c) content analysis is sometimes less credible, and d) using content analysis is a tiring and time-consuming activity. The limitations, in this paper, are minimized by: a) the insensitivity can be minimized by making categories (operational definition) of the unit of analysis, b) the criticized model of being too descriptive can be minimized in the nearly same way as the first disadvantage, that is by making suitable operational definitions and using simple statistical analysis, c) the third limitation can be minimized by inter-coder reliability test, in order to reduce inter-coder bias. Krippendorf (1980) inform that the method usually used is asking other two researchers to coding similar data, and then comparing the results, and d) the last limitation can be minimized by using computer, and by analyzing constantly "what is to be analyzed" and excluding the unnecessary analysis. However, although this paper tried to overcome the weaknesses, other analysis method (such as discourse analysis and meta-analysis) may be used to measure the consistency of this study. Despite the limitations, the most important is that articles used in the study were merely taken from KOMPAS daily. Further it will be more beneficial when incorporating other mass media such as Bisnis Indonesia, Kontan, and others.
Further researchs should identify whether boycott is a proactive behavior (internally driven) or reactive behavior (externally driven). In other words, it is important to identify what makes people participate in boycott; the difference between politically-driven boycott and boycott as a lifestyle; the effective method to deal with boycott. In addition, interviews with observers of the media can be added in the future research.