Erasmus Mundus Masters
Journalism, Media and Globalisation
(Joint Degree)
GOOD OR BAD?
News Frames and Its Effect on Cognitive Evaluation of the
International Criminal Court in Kenya
By
Kenneth Owino
Student ID: 11300361
Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science
Supervisor/Examiner: dhr. Michael Hameleers, MSc
Abstract
News frames have effects on peoples’ evaluation of policies and entities but most of the studies that prove this have been done mostly in western contexts. There have been calls to situate framing studies in different cultural and socio-political contexts. In the Kenyan
setting, extant studies have tended to focus on content analysis to reveal dominant frames but seldom probed the effects of these frames. This study is among the first to disentangle the effects of frames on cognitive evaluations in an African and specifically a Kenyan context. In an online Experiment, the study randomly exposed respondents to positively and negatively framed news articles on the International Criminal Court (ICC) and later measured their evaluations of the ICC. Findings of the study note that irrespective of exposure to conditions, those who were government leaning were less positive in their assessments of the ICC while those who are opposition leaning tend to be more positive of the ICC.
Key Words: Valenced Frames, Ideology, ICC. Kenya, Cognitive Evaluations.
Introduction
Most of the 20th
century witnessed dramatic growth in various measures of international cooperation and institution-building including an overall increase in
intergovernmental exchange, treaties, and international governmental organizations. (Goldin & Reinert, 2012). At least in the near term, we can expect continued interest in developing and strengthening international institutions to respond to the problems of an increasingly interdependent world, even though building these institutions will not occur without
cement their position as individual nations or collectively as a continent in the international arena by not only joining international organizations but also seeking to reform or even withdrawing their membership these organizations in order secure their interests. The recent attempt by African countries to first amend the regulations of the International Criminal Court (ICC) to prevent prosecution of sitting heads of state (Momanyi, 2014) and their resolution to pull out en masse out of the Rome Statute that set up the ICC ( Igunza, 2017) attests to this desire to protect interests.
Given the important role played by international organizations in mitigating global challenges such as economic crises, underdevelopment, peace, justice among others which has seen international organizations take over the role of states to become most important actors on the international scene. It is necessary to explore how these international
organizations are framed in news media and how framing shapes the overall evaluation of the international organizations in the eyes of the public and support for policies emanating from the international organizations.
Ground-breaking generation of framing research compellingly showed that frames can affect citizens’ sentiments on broad range of issues (e.g. Iyengar, 1991; Nelson, Clawson, & Oxley, 1997; Zaller, 1992). More over with increased access to information, the media has today become the major source of political information (Brants & Praag, 2015; Strömbäck, 2008).This trend, conceptualized as the first phase of ‘mediatization’ by Strömbäck (2008, p. 236) has potential to intensify the effect of frames on people’s cognitive evaluations of a myriad of issues. In addition, approval and support of international organizations have been proved to be dependent on how the international organization is portrayed in the news and is integral in winning trust from members of the public who are the ones who feel the impact of policies, laws or binding resolutions made by the international organization.(e.g. de Vreese, van der Brug, & Hobolt, 2012 ; Schuck & de Vreese, 2006 ; Aarøe, 2011). Frames are also
contingent on specific settings and are “central to the production of hegemonic meanings” and are further shaped by competing stakeholders or claims-makers (Carragee & Roefs, 2004 p 222). Research of news frames, therefore needs to be located within a specific cultural, economic or political context (Anderson, 1997).
In view of the aforementioned, the current study seeks to add into the knowledge of framing effects in general and specifically in Kenya. The paper intends to explore the extent to which attitudes and support for the ICC in Kenya is influenced by valenced news frames and whether this relationship is moderated by support for political leaders, political
knowledge, ideology and ethnicity. The study seeks to contribute knowledge in framing studies in two ways; First by exploring the strength of valenced news frame and secondly doing this in an African context which not many studies have attempted. The study will also add into the discussion and give an indication on the contested ICC membership in Africa given that Kenya is among the countries that has recently called for the mass withdrawal from the ICC citing disproportionate targeting of African leaders by The Hague based court.
Literature Review
Framing in Communication
Research on framing has been popular within communication research and political science for a good four decades (Entman, 2007). Drawing on disciplines such as sociology, economics, psychology, cognitive linguistics and media studies, media framing specifically aims at linking news texts to the production of news and to recipient processes (see Entman, 1993; Scheufele, 2009). The theory of framing refers to a broad tradition of research that focuses on how we interpret reality. More specifically, “media framing is part of the study of
media content, its origin, form and meaning and effects on recipients” (McQuail, 2010, 340). Framing is an important component of political communication and news reporting (Aarøe, 2011; Chong & Druckman, 2007 p. 100). According to Entman (1993), framing
encompasses selecting certain aspects of a perceived reality and making them more prominent in a communication text, “in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described” (p. 52). In this regard, frames play four key roles, which are to; - define a problem, assign responsibility, moralize the issue at hand and suggest potential remedies (Entman, 2007).
Van Gorp (2007) contends that a frame reveals itself upon interaction with a text through a myriad of framing devices, these devices includes word choice, metaphors, exemplars, descriptions, arguments (especially in a largely verbal media), visual imagery, visual metaphors, stereotypes, catch phrases, dramatic characters, graphics (in a largely visual text like editorial cartoons) and lexical choices. Frames can be thematic or episodic.
Specifically, thematic frames focus on political issues and events in a broader context and present collective, abstract and general evidence (Van Gorp, 2007). Episodic frames on the other hand describes concrete events and specific cases that illuminate the issue (Gross, 2008).
Valence Frames
Beyond episodic and thematic frames, studies have also explored the valenced dimension of framing in Political communication. Schuck and de Vreese (2006) assert that Valence frames “evaluate political issues or situations in either positive or negative terms” (p. 6). In a study exploring the effect of valence frames on support for the European Union (EU)
enlargement, the scholars note that when EU enlargement is presented as a risk, people’s support is generally lower compared to when EU enlargement is presented as an opportunity. Entman (1991) analysed the different framing of an American and a Soviet plane accident as either ‘tragedy’ (in the US case) or ‘attack’ (in the Soviet case). de Vreese, van der Brug and Hobolt (2012) explored the differences in support for Turkish ascension into the EU between people exposed to – positive and negative – cultural and economic arguments, noting that the economic frames were persuasive across the board, with the positive frames being a little stronger whereas cultural (religious) frames are strongly conditioned by individual
predispositions on religious diversity. Nelson, Clawson & Oxley (1997) examined the effects of presenting actions of the Ku Klux Klan (a controversial U.S. hate organization that has employed terror in pursuit of their white supremacist agenda) either in a ‘free speech’ frame or in a ‘disrupting public order’ frame. Results showed a higher level of tolerance towards the Ku Klux Klan among participants in the positive frame (free speech) condition. Against the backdrop of the discussed notions of valenced news frames, the current study explores the effects of framing of the ICC in Kenya when the court is presented either as a good and desirable organization because of being a fair and impartial bulwark against impunity (positive condition) or a bad and undesirable entity because the court is partisan and an extension of colonialism (negative condition).
Framing studies in Kenya
Previous framing studies focusing on Kenya have largely applied content analysis to explore how prevalent issues are portrayed in the news media but seldom explored the effects of these portrayals on policy support nor cognitive evaluations by citizens. For instance, Lore, Imungi, & Mubuu (2013) note that on the coverage of genetically modified organisms
(GMO) newspaper articles were skewed in favour of either the positive agriculture frame in which GMOs are depicted as a worthy breakthrough in agriculture and a potential solution to mitigate food shortage or the negative safety frame which sees GMOs as dangerous for human consumption. Another study has explored the prevalence of media frames in op-ed articles published in Kenya's Sunday Nation newspaper tracing responsibility for the violence that rocked the country in 2007 after a hotly contested and disputed election. The findings indicate that the then president, Mwai Kibaki was blamed more, and thus framed more
negatively, than his protagonist, and later prime minister, Raila Odinga (Ireri, 2013). Omanga (2012) on the other hand probed the dominant frames in the first three months after the United States launched of the War on Terror as revealed through Kenya’s editorial cartoons. The study reveals that the War on Terror had already found a growing legitimacy that changed to varying degrees of illegitimacy as the promised war script increasingly veered from the actual war.
The current study takes these content features as a nascent point of departure in framing studies in the Kenyan context by disentangling the effects of valance frames to explore if the interactions between valenced news frames and cognitive evaluation which has been proved in largely western contexts would be similarly observed in an African context. The study explores the evaluation of the ICC as an international organization because it is a topic that has been debated widely in Kenya. In addition, in almost every issue area affected by international institutions, there are at least two groups with divergent preferences over compliance. This sets the ground for a political contest to access instruments of power which is then used to effect either compliance or defiance to the international organization.
(Chaudoin, 2016). Much of this political contest is mediated or takes place in the media as the media has transformed into modern day ‘public sphere’ (Sinekopova, 2006) where major political issues are articulated and discussed. However, unlike much of the western world, the
media in Kenya has emerged as the sole source for political information predominantly in the last two decades bellied on rapid media proliferation and expansion credited with the
opening-up of democratic space and deregulation of the media in the early 1990’s (Amutabi, 2013).
The ICC in Kenya
Before presenting the research questions and hypotheses, I will give an overview of the ICC as an international organization in Kenya and why it become a hot topic and a major political issue. The ICC came into force in 2002 through the Rome Statute which has since been ratified by 123 countries. (Mutua 2016). The Hague based court is mandated to bring to justice those responsible for the worst crimes such as genocide, crimes against humanity and war crimes (Igunza, 2017; Momanyi, 2014). Kenya ratified the Rome Statute in March 15, 2005 giving the ICC jurisdiction over serious crimes, but only in cases where the government proved unwilling or unable to investigate and prosecute those crimes. IJM (2014). The ICC involvement in Kenya came because of the violence that engulfed the East African nation following the 2007 presidential elections. After general voting, the Kenyan Electoral Commission declared the incumbent, President Mwai Kibaki of the PNU party, the winner. However, supporters of the challenging candidate, Raila Odinga of the ODM party,
contended that electoral results had been manipulated(Chaudoin, 2016; Mueller, 2014). The electoral crisis, combined with existing tensions, resulted in violence between supporters of each group. The post-election violence (PEV) caused over 1,000 deaths and internally displaced approximately 600,000 people. ( Ireri, 2013).
Violence subsided after Koffi Annan (former UN Secretary General) moderated a power-sharing agreement. The power sharing agreement obliged the Kenyan government to set up an international commission of inquiry into the post- election violence. The
deemed to have had the biggest responsibility of the violence. However, there was a caveat; if the tribunal was not set up within a fixed time frame of one year, the evidence collected by the commission would be handed over to the ICC including a sealed envelope with the names of those suspected to be most responsible for the violence. (International Justice Monitor, 2014). The Kenyan government failed to set up a local tribunal and the evidence was handed over to the then ICC chief prosecutor Louis Moreno Ocampo who after analysis of the evidence requested for authorization to commence investigations into possible crimes against humanity committed during the post-election violence in Kenya. (Lynch, 2014). This was the first time in the history of the ICC that the prosecutor applied the proprio motu powers granted to him to him under Article 15 (3) of the Rome Statute i.e. the right to initiate an investigation at his own instigation, without a referral from the State Party or the UN Security Council. (Mueller, 2014). This is where the rift between Kenya and the ICC seem to have begun, while the Kenyan government has continuously asserted that it did not invite ICC into its’internal affairs (Momanyi, 2014), others have argued that by failing to set up a local tribunal, Kenya had indirectly invited the ICC (Mutua, 2016).
In March 2011, the ICC prosecution Chamber summoned six individuals to appear before the court. The six included then Deputy Prime Minister Uhuru Kenyatta, who was accused of facilitating violence against ODM supporters, and then Education Minister William Ruto, who was accused of supporting violence against PNU members. Around that time both Kenyatta and Ruto had declared themselves presidential candidates. With only months to the 2013 presidential elections, Kenyatta and Ruto formed the ‘Jubilee’ alliance. Pundits described the coalition as ‘alliance of the accused’, in reference to the fact the two politicians were both accused of organizing and financing violence between members of their respective ethnic communities who had been on opposing sides of the previous elections and subsequent violence. (Chaudoin, 2016; Lynch, 2014; Mutua, 2016). This alliance, later
catapulted Kenyatta and Ruto into power in 2013 with big winning margins in areas that are predominantly inhabited by the Kikuyu and Kalenjin communities while preforming dismally in areas mostly occupied by the Luo community where their closest challenger Raila Odinga, an ethnic Luo garnered a lot of votes. (Lynch, 2014; Chaudoin, 2016). It’s also worth noting that the support for the ICC declined as the popularity of the two candidates rose, Rothmyer (2012) posits that by 2009, support for the ICC in Kenya was at 65%, by 2012 it had dropped to 58% with the number projected to fall even further. Lynch (2014) contends that the waning support for the ICC was the result of Kenyatta and Ruto uniting their two communities
against a common foe in the name of the ICC. Mutua (2016) avers that the two leaders used their ethnic political base to win the elections like many other African leaders ostensibly to shield themselves from prosecution, an argument that is also advanced by Chaudoin (2016) who adds that the victory resulted in the two leaders embarking on a non- compliance with the ICC while the Kenyan opposition led by former Prime minister Raila Odinga maintained their support for the ICC. In 2015, the ICC dropped charges against president Kenyatta because of lack of evidence against a backdrop wide-spread witness interference and
intimidation while at the same time accusing the Kenyan government for lack of cooperation in the case (ICC, 2015). Charges against Ruto were dropped in April 2016 on similar grounds in what was perceived as a low moment for the ICC and victims of the 2007 violence
(Leithead 2016).
Although an international organizations such as the ICC become major election issue in Kenya, political mobilization is still largely done along ethnic basis where there “there is little to choose in ideological terms between political parties” (Nyamnjoh, 2005, p.34) even though others argue that the influence of ethnicity is declining due to an increasing middle class that is more informed and hinges political support on policies and track record of politicians as opposed to ethnicity (Cheeseman, 2015). In addition, journalists in Kenya list
supporting official policies as one of their major roles (Ireri, 2016). Against this backdrop, support for the ICC could either be as a result of ethnic polarisation with no regard to actual policies or the influence of the media which by design or default overly takes side with the government’s position with regards to the ICC albeit on the basis of playing their role of supporting official policies, more so, if we also take into account that unlike in the western world where media trust is waning, the Media in Kenya is one of the most trusted institutions ranking higher than the police and the judiciary (Ogola, 2011).
Framing the ICC
Nelson et al., (1997, p. 567) argue that “…A well-placed quote . . . will convey a construction of an issue that could ultimately benefit a particular interest.” In this regard, a quote that implies a favourable evaluation of the ICC would benefit and find congruence with those who support the ICC and a quote that conveys a negative assessment of the ICC would be beneficial and find favour with those who do not support the ICC. The study thus isolated prominent Kenyan political leader’s assessment of the ICC enmeshed in quotes in a
newspaper article to test if these quotes as part of framing device influence people’s evaluation of the ICC. People also have a greater propensity to readily accept information that matches their initial dispositions and beliefs while rejecting credible information that disputes their existing attitudes. This is trend described as confirmation/disconfirmation bias is more pronounced on people who feels mostly connected or passionate to the issue at hand. (Taber & Lodge, 2016; Kahan, 2013, p. 418). In this regard, it is expected that one’s prior political ideology or leaning would influence their cognitive evaluations of a given matter, policy or entity.
Research Questions and Hypotheses
The core aim of this study which has hitherto been unexplored, is to wade into this conundrum of locating what influences support for the ICC in Kenya weighed against news frames, political mobilization along ethnic lines for or against the ICC and other mitigating parameters. The study seeks to answer the principal question; to what extent does valenced
frames influence cognitive evaluation the ICC? In addition to the overarching research
question the study would also explore the following sub question;
RQ1: To what extent does positive framing of the ICC lead people to use positive explanations when commenting or describing the ICC?
RQ2: To what extent does negative framing of the ICC lead people to use negative explanations when commenting or describing the ICC?
These questions are modelled along the study by Valentino, Beckmann, & Buhr (2001)who apply similar criteria to test the effect of frames on voter turnout and trust in government. According to Aøre (2011) frames that make salient aspects of a broader
controversy that typically imply an evaluation in favour of a given policy leads to a positive description of the policy. In addition, people exposed to positive frames have been found to have a more positive evaluation of the entity at hand (de Vreese et al., 2012). In this regard, the following hypothesis is made;
H1 Exposure to a positive frame, would lead to a more positive assessment of the ICC compared to exposure to a negative frame.
Given that the ICC was a major issue in last Kenyan presidential elections in 2013 and continued to be an emotive issue which precipitated the government/ opposition divide in view of the ICC, I presuppose that based on the expected polarisation and disconfirmation bias on exposure to information (Taber & Lodge, 2016; Kahan, 2013), one’s political leaning on the government/opposition divide would moderate the effect of valenced frames one is exposed to and the eventual evaluation of the ICC. I therefore foreground the following hypotheses;
H2 Those who identify as pro government will have a more negative assessment of the ICC compared to those who identify as pro opposition.
As a court of last resort, cases adjudicated by the ICC generally point to nation-states that Brock, Holm, Sorensen, & Stohl (2012) describe as fragile. These states are marked by failed social, political and judicial system with citizens identifying more in ethnic, religious or other sectarian enclaves than in the nation-state. In the same line of thought, in most African countries, political support is often mobilized along ethnic lines and not on issues. Nyamnjoh (2005), I thus expected ethnicity to moderate news frames in influencing people’s evaluations of the ICC. I thus put forward the following hypotheses;
H3 Individuals from ethnic communities that overwhelmingly voted for the government will have a more negative assessment of the ICC compared to individuals from ethnic
Drawing from the findings of Schuck and de Vreese (2006), it also expected that Political knowledge moderates framing effects such that individuals with low levels of political knowledge are more affected by news frame than individuals with high political knowledge. In this regard, the following hypothesis is foregrounded;
H4 Evaluation of the ICC in Individuals with low knowledge will be more affected by news frame compared to Individuals with high knowledge of the ICC.
Methods
Design
To answer the questions presented earlier, an experiment was conducted. This was preferred because in an ideal experiment set up, the researcher has full control over the independent variable (Harwell, 2011). Hence, he or she can say with certainty that it is this variable and no other things than is responsible for the observed change(s) in the dependent variable. However David and Sutton (2011), caution that it is often difficult in social sciences to randomly allocate subjects into two groups and one also has to grapple with the challenge of other external factors influencing the measured observations. To mitigate these challenges, the experiment was a 2x2 factorial design system with control, that is to mean 2 frames: (positive versus negative) X 2 (quoted political source: Uhuru Kenyatta versus Raila Odinga) between subjects with control. In this regard, there were 5 groups to compare.
Sample
Participants for the study were invited to participate in the study through a link shared on social media (Facebook and Twitter) and requested to invite others to participate in the
study. This snowball method of recruiting respondents was applied because it is affordable, cost-efficient and simple. This sampling technique requires little planning and fewer
workforce compared to other sampling techniques. In addition, snowball sampling enables a researcher to take advantage of the social networks of identified respondents to provide a researcher with a bigger and richer pool of diverse respondents (Emerson, 2015). In total 254 people took part in the study. Respondents for this survey were selected from among those who volunteered to participate. The data was not weighted to reflect the demographic composition of Kenya, the target population. The sample was based on those who initially self-selected for participation rather than a probability sample. In line with best practice benchmark as posited by AAPOR (2017) no estimates of sampling error could be calculated. The minimum age was 18 and the maximum age was 70 the average age was 28.3, the mode was 19, SD 8.55. Males accounted for 66.8%, females 33.2%. In terms of level of education 63% had reached undergraduate level, 30.3% post-graduate, 5.8% had high school Diploma and 1% had no formal education.
Procedure and Stimuli
The experiment was conducted online. Participants were first taken through a procedure of informed consent which filtered out those who did not want to take part. After this, the participants completed a pre-test which captured moderating variables and control variables. Next, the participants were randomly exposed to a newspaper article and later completed a post-test questionnaire measuring the dependent variable and demographics. At the end of the questionnaire, manipulation check for tone and source were evaluated. After this, the
respondents were debriefed and informed what the study entailed. The entire process lasted about 15 minutes.
Stimuli
The stimuli in the treatment conditions was made up of a neutral, two positively valenced and two negatively valenced online news articles on the ICC. The articles were framed in five different ways on differing on the independent variables; tone and source. Tone was
manipulated in terms of appraisal patterns and words that indicate a negative evaluation (e.g. Schuck & de Vreese, 2006). In the positive condition the favourable evaluation of the ICC was emphasised. The words/phrases ‘impartial and ‘withdrawal law rejected’ and ‘more Kenyans would face trial at The Hague’ were used to convey a positive assessment. In the negative condition, an unfavourable evaluation of the ICC was emphasised. The
words/phrases ‘disproportionate’ and ‘withdrawal law tabled’ and ‘No Kenyans would face trial at The Hague’ were used as indicators of a negative evaluation. The source was
manipulated to have one neutral source, a scholar and two Political leaders in Kenya, Leader A Raila Odinga, the Opposition leader (positive frame) leader and Leader B Uhuru Kenyatta the president (negative Frame). In all conditions the newspaper article include a picture of the entrance of the ICC in all conditions. This was done to control the framing effect of the image as it has been noted that frames are revealed not only through texts but also images. All the articles had three paragraphs were between 139 and 172 words long. All versions of the article are included in Appendix A.
Dependent Variable
The dependent variable (Evaluation of the ICC in Kenya) was measured by asking respondents to rate their evaluations of the ICC by responding to how much they agreed on
five of statements on the ICC measured on a 5 point Likert scale ranging from 1 Strongly disagree to 5 Strongly agree. See Appendix B for Item wordings. The constructed scale was found to be reliable when tested on the entire valid sample (N= 212 M= 3.48, SD = 1.11 Cronbach’s α 0.74).
Moderating Variables Knowledge
Knowledge on the ICC in Kenya, was measured by 5 factual open ended questions which were coded as right 1 or wrong 0. At the end of the study the mean score was used to demarcate the divide between high knowledge and low knowledge respondents. This measure has been successfully applied by (Schuck & de Vreese, 2006) to measure political
knowledge.
Ethnicity
The respondent’s actual ethnicity was also noted by asking which language they speak in addition to English and Kiswahili, the two official languages in Kenya.
Political ideology
In the absence of a left-right divide in Kenyan politics, Political ideology in the study was construed to in terms of government and opposition given that the main political divide on the issues of the ICC is between the government which is anti-ICC and the opposition which is pro ICC. To capture where participants lie on this divide, the survey asked them who are they are likely to vote for in the presidential elections slated for August 2017 between a government candidate, an opposition candidate, an independent candidate or a candidate from
another non-aligned party. This was later conflated into three groups i.e. Government leaning, opposition leaning and neutral leaning.
Manipulation Check
Manipulation was successful. The first check asked participants to write what the article was about. This was successful as descriptive data showed all participants could recall what the article was about and the sources quoted. Secondly as part of the post-test
questionnaire, participants were asked for their personal impression on how the news article presented the ICC as (1) ‘favourably or as (2) unfavourably. An analysis of this showed a statistically significant difference between groups as determined by one-way ANOVA (F (4,207) = 5.129, p = .001, η2 = 0.09. A Tukey post hoc test yielded a significant mean difference between participants in the positive and negative conditions in the expected direction.
Results
H1 posited that exposure to a positive frame, would lead to a more positive
assessment of the ICC compared to exposure to a negative frame. To test this hypothesis, I conducted an Analysis of variance (ANOVA) to test the interaction between frames in all the conditions and ICC evaluation. A repeated measures ANOVA with correction determined that mean ICC Evaluation did not differ statistically significantly between the different conditions. (F(4, 207) = 1.44, p = 0.22, partial η2 = 0.27). Post hoc tests using Bonferroni revealed that ICC evaluation had slight differences in the hypothesised directions, these differences were found to be not significant (p= 0.22). Therefore, it can be concluded that
Exposure to positive/negative frames had no effect on evaluation of the ICC, as such H1 is
hereby rejected. Table 1 summarises these findings.
Table 1: ICC Evaluation by Experimental Conditions
Characteristic Condition p-value Neutral Positive Tone with Neutral Source Positive Tone with source B Negative Tone with Neutral Source Negative Tone with source A Number of Respondents 28 46 47 46 45 0.22 ICC Evaluation Mean (SD) 3.57 (0.98) 3.78 (1.00) 3.48 (1.05) 3.24 (1.20) 3.37 (1.24)
Note: Cell entries are mean scores of ICC Evaluation on a five-point scale; high score denotes positive evaluation, standard deviations in parentheses, (F(4, 207) = 1.44, p = 0.22, partial η2 = 0.27.
H2 projected that those who identify as pro government will have a more negative assessment of the ICC compared to those who identify as pro opposition. I conducted an Analysis of variance (ANOVA) to test the effect of Political leaning acting independently or in concert with the effect of treatment conditions on ICC evaluation. The test of between subjects effects showed that acting in concert (Political leaning*Treatment Condition) had no effect on ICC evaluation (F (8,194) = 1.358, p= 0.217, partial η2
=0.157. Acting solely, Political leaning had effect on ICC evaluation (F (4, 194) = 7.837, p= 0.01, partial η2
A Bonferroni post hoc test revealed that the mean ICC evaluation was statistically significantly lower in those who are government leaning compared to those who are
opposition leaning. as shown in table 2. It can thus be concluded that irrespective of exposure to valenced frames, all participants have a positive evaluation of the ICC with those who are opposition leaning being more positive in their evaluation of the ICC compared to those who are government leaning. Consequently, hypothesis H2 is partly confirmed.
Table 2: ICC Evaluation by Political Leaning
Note: Different subscripts indicates discernible difference, * indicates significant difference with p=
0.01, higher values indicate a positive evaluation of the ICC, (F (4, 194) = 7.84, p= 0.01, partial η2 =0.16.
H3 projected that Individuals from ethnic communities that overwhelmingly voted for the government will have a more negative assessment of the ICC compared to individuals from ethnic Communities that overwhelmingly voted for the opposition. A two-way ANOVA was conducted that examined the effect of ethnicity and exposure to treatment condition on evaluation of the ICC. There was a statistically significant interaction between the effects of ethnicity on evaluation of the ICC., (F (3, 177) = 3.16, p = .0003, η2
= 0.15 Characteristic Neutral Government Leaning opposition Leaning Number of Respondents 34 50 125 ICC Evaluation Mean (SD) 3.40b (0.92) 2.90*b (1.16) 3.72*a (1.05)
A combined condition and ethnicity interaction yielded no statistically significant interaction between the effects of ethnicity and treatment condition on evaluation of the ICC. (F (11, 177) = 1.03, p =0.42, η2
= 0.15). Post hoc tests using the Bonferroni further revealed statistically significant mean differences in ICC evaluation in the hypothesised directions. All participants had a positive evaluation of the ICC with ethnic Luos (who mostly voted for the opposition) having a more positive evaluation of the ICC M = 3.71, SD 1.03 compared to ethnic Kikuyus (who largely voted for the government) M = 3. 04, SD 1.21 These differences were found to be significant (p= 0.14) in homogenous subsets but not in the experimental conditions. In this regard, H3 is partly confirmed.
According to H4, evaluation of the ICC in Individuals with low knowledge were supposed to be more affected by news frame compared to Individuals with high knowledge of the ICC. An ANOVA test was done to compare mean evaluations of the ICC between high knowledge and low knowledge individuals across all treatment conditions. Knowledge in concert with exposure to the treatment conditions had no effect on ICC evaluation (F (4,202) = 2.13, p=0.78, η2 =0.91. As noted in the table 3, the evaluations of the ICC did not
differ significantly between high and low knowledge participants. In addition, a post hoc test applying the Bonferroni correction further revealed that there was a distinct difference in evaluations by participants with high knowledge from those with low knowledge that went against the hypothesis in the Positive Tone and Source B Condition, however this difference was not significant (p>0.05). It can thus be concluded that knowledge does not mitigate the interaction between exposure to valenced news frames and evaluation of the ICC. Therefore,
Table 3: ICC Evaluation by knowledge in all Experimental Conditions Condition
Control Group Positive Tone with Neutral Source Positive Tone with Source B Negative Tone with Neutral Source Negative Tone with Source A High N=13 3.44 (0.98) Low N=15 3.68 0.90) High N=24 3.76 (1.05) Low N=22 3.77 (0.97) High N=26 3.92 (0.81) Low N=21 2.93 (1.08) High N=23 3.34 (1.25) Low N=23 3.15 (1.18) High N=17 3.78 (1.10) Low N=28 3.11 (1.28)
Note: Cell entries are mean scores of ICC Evaluation; high score denotes positive evaluation,
standard deviations in parentheses, (F (4,202) = 2.131, p=0.78, η2
=0.91.
Discussion
News frames have been proved to have effect on peoples’ evaluation of policies and entities ( e.g. Gross, 2008; de Vreese et al., 2012). Most of these studies have been done in western contexts, even though there have been calls to situate framing studies in different cultural and socio-political contexts (Anderson 1997). This study is among the first to explore the effects of frames on cognitive evaluations in an African and specifically a Kenyan context where previous studies have tended to focus on content analysis to reveal dominant frames but seldom the effects of these frames. The results of the study reveal that overall participants had a positive evaluation of the ICC although there was a marginal difference in ICC
evaluations in the different conditions such that those exposed to a positive frame had a marginally though statistically insignificant, more positive evaluation. This rejected the hypothesis that those exposure to a positive frame would lead a positive evaluation and those
exposed to a negative frame will have a more negative assessment of the ICC. Considering the debate about Kenya pulling out of the ICC, the current results indicate that the ICC is presently positively evaluated. Accordingly, Kenya’s withdrawal from the Rome Statute would not be favoured.
In view of previous framing effect studies, the results differ from other studies in political communication that have found a robust relationship between news frames and cognitive evaluations (e.g., Schuck & de Vreese 2006). This finding could be explained in different ways. To start with, frames have been found to have a greater effect on people when the frame elicits strong emotions (Gross, 2008; Aøre, 2011). It therefore potentially implies that the frames tested in the study in themselves were not strong enough to elicit a reaction in the respondents that is strong enough to precipitate a change or reinforce opinion. Secondly, frames are by nature simplification of complex issues which as Hilgartner and Bosk (1988) assert, have to compete with other issues of the day in an ‘arena’ such as newspapers, television, radio and even new media (Maratea, 2008) which are constrained in terms of the capacity to carry these issues or social problems. Given that the ICC has dominated the News in Kenya for a very long time, a period of almost 7 years from when the cases started up to the time when the last case was dismissed, frames on the ICC could have been over taken by newer ‘social problems.’
The study did not also find a correlation between political knowledge and cognitive evaluations, contrary to other previous studies. This finding underscores the contextual differences in the media and political environment in Kenya and the Western world. One major highlight of this contextual difference is the media and democratic systems. Strömbäck (2005) posits that in a procedural democracy, citizens are not obliged to search for
information or be informed about political activities, their responsibility mainly is to vote. In such a setting, information or knowledge ceases to be important as opposed to a deliberate
democracy where issues are exhaustively discussed thus necessitating the need for citizens to be informed. With insights from this study bellied on the argument by Strömbäck (2005) frames strengths could be influenced by the type of democracy and the expectations of citizens. This is an area future studies should explore to empirically explain why knowledge in political affairs plays a pivotal role in one context but hardly does the same in another context.
The study found that regardless of the exposure to different frames, people from the community of a leader who positively evaluates the ICC had a more positive assessment of the ICC compared to those from the ethnic community of a leader who has had negative appraisal of the ICC. Whereas the study did not find evidence for the effect of ethnicity on ICC evaluations, it did find a correlation between the two. This correlation could possibly point to the argument by Nyamnjoh (2005) that political mobilization in Kenya is still done largely along ethnic lines. Moreover, other studies (e.g. Chaudoin, 2016; Lynch, 2014; Mueller, 2014) have found that the Kenyan presidential elections in 2013 was marked by the interpretation of the ICC as a performance of injustice which was successfully used in contingent with other narratives to mobilize political support along ethnic lines in which two major ethnic communities (Kikuyu and Kalenjin) united against the ICC but also behind the respective leaders. As such, the correlation between ethnicity and ICC evaluation could also indicate that those communities’ perceptions on the ICC have presently not significantly changed.
The study also noted that those who are government leaning were less positive in their assessments of the ICC while those who are opposition leaning tended to be more positive of the ICC irrespective of exposure to the different conditions. This implies that perceptions of the ICC are largely stable along government/opposition lines in Kenya. This finding could point to other studies that have found a relationship between political ideology and cognitive
evaluations in a conservative/liberal divide (e.g. (Kahan, 2013; Taber & Lodge, 2016). The point of departure with those studies is that in Kenya like in most African democracies there is no clear left or right. In fact, “there is little to choose in ideological terms between political parties” (Nyamnjoh, 2005, p 34). Future studies should explore what constitutes ideological difference in an African context such as the government/opposition divide in Kenya as noted by this study and how it affects people’s cognitive evaluations and political decisions.
Limitation and Conclusion
This study is one of the first in exploring the effects of valance framing in an African and more specifically Kenyan context. The findings would be vital in creating a stepping stone into more studies of framing effects in contexts hitherto unexplored.
Nonetheless, the study had some limitations. To start with, the study relied on a convenient snowball sample which may not be representative of the Kenyan population. As such the findings may not be generalizable to a wider Kenyan population. The study also used a sample of 254 respondents. In addition, the study suffered major drop out in responses especially in the control condition. This resulted in having very small groups whose
comparisons are subject to errors and do not give a rich and diverse basis for identifying trends. The study only tested the effects of frames in an online and textual medium, as such these findings cannot be generalised to other medium like Radio and Television which equally have an impact on audiences. That notwithstanding, the study makes worthwhile contribution in political communication studies and adds knowledge and vigour to the study of framing effects.
AAPOR (American Association for Public Opinion. (2017). Opt-In Surveys and Margin of Error. Retrieved May 25, 2017, from https://www.aapor.org/Education-Resources/For-Researchers/Poll-Survey-FAQ/Opt-In-Surveys-and-Margin-of-Error.aspx
Aarøe, L. (2011). Investigating Frame Strength: The Case of Episodic and Thematic Frames.
Political Communication, 28, 207–226. https://doi.org/10.1080/10584609.2011.568041
Amutabi, M. N. (2013). Media boom in Kenya and celebrity galore. Journal of African
Cultural Studies, 25(1), 14–29. https://doi.org/10.1080/13696815.2013.765357
Anderson, A. (1997). Media, Culture and the Environment. New Brunswick, NJ: Rutgers University Press.
Brants, K., & Praag, P. Van. (2015). Beyond media logic. Journalism Studies, 9699 1–14. https://doi.org/10.1080/1461670X.2015.1065200
Brock, L., Holm, H.-H., Sorensen, G., & Stohl, M. (2012). Fragile States - Violence and
Failure of Intervention. War and Conflict in the Modern World.
Carragee, K. M., & Roefs, W. (2004). The neglect of power in recent framing research.
Journal of Communication, 54(2), 214–233. https://doi.org/10.1093/joc/54.2.214
Chaudoin, S. (2016). How Contestation Moderates the Effect of International Institutions : the International Criminal Court and Kenya, Journal of Politics 78(2), 1–37.
Cheeseman, N. (2015). “No bourgeoisie, no democracy”? The political attitudes of the Kenyan middle class. Journal of International Development, 27(5), 647–664. https://doi.org/10.1002/jid.3057
Chong, D., & Druckman, J. N. (2007). Framing Theory. Annual Review of Political Science,
10(1), 103–126. https://doi.org/10.1146/annurev.polisci.10.072805.103054
David, M., & Sutton, C. D. (2011). Social research: An introduction. New York: Sage. de Vreese, C. H., van der Brug, W., & Hobolt, S. (2012). Turkey in the EU?: How cultural
Politics, 10(2), 218–235. https://doi.org/10.1057/cep.2011.27
Emerson, R. W. (2015). Convenience Sampling, Random Sampling, and Snowball Sampling: How Does Sampling Affect the Validity of Research? Journal of Visual Impairment &
Blindness, 109(2), 164–168. Retrieved from
http://library.gcu.edu:2048/login?url=http://search.ebscohost.com/login.aspx?direct=true &db=ccm&AN=2012938256&site=eds-live&scope=site
Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of
Communication, 43(4), 51–58. https://doi.org/10.1111/j.1460-2466.1993.tb01304.x
Entman, R. M. (2007). Framing bias: Media in the distribution of power. Journal of
Communication, 57(1), 163–173. https://doi.org/10.1111/j.1460-2466.2006.00336.x
Goldin, I., & Reinert, K. (2012). Background and Context. In Globalization for Development:
Meeting New Challenges (New, pp. 1–17). New York: Oxford University Press.
Gross, K. (2008). Framing persuasive appeals: Episodic and thematic framing, emotional response, and policy opinion. Political Psychology, 29(2), 169–192.
https://doi.org/10.1111/j.1467-9221.2008.00622.x
Harwell, M. R. (2011). Research Design in Qualitative/Quantitative/ Mixed Methods. In The
SAGE Handbook for Research in Education: Pursuing Ideas as the Keystone of Exemplary Inquiry (pp. 147–182). https://doi.org/10.4135/9781412961288.n380
Hilgartner, S., & Bosk, C. L. (1988). The Rise and Fall of Social Problems: A Public Arenas Model. American Journal of Sociology, 94(1), 53. https://doi.org/10.1086/228951 Igunza, E. (BBC). (2017). African Union backs mass withdrawal from ICC. Retrieved April
17, 2017, from http://www.bbc.com/news/world-africa-38826073
International Justice Monitor, (IJM). (2014). Kenyan Cases at the International Criminal Court.
election violence. Ecquid Novi: African Journalism Studies 34(2), 109-127.
Ireri, K. (2016). A National Survey on the Professional Role Conceptions of Journalists in Kenya. Journalism Practice, 1–20. https://doi.org/10.1080/17512786.2016.1213137 Iyengar, S. (1991). Is anyone responsible? How television frames political issues.
Kahan, D. M. (2013). Ideology, motivated reasoning, and cognitive reflection. Judgment and
Decision Making, 8(4), 407–424. https://doi.org/10.2139/ssrn.2182588
Keohane, R. O., & Nye, J. S. J. (2000). Introduction. In J. S. Nye & J. D. Donahue (Eds.),
Governance in a Globalizing World (pp. 1–44). Cambridge, Massachusetts: Visions of
Governance for the 21st Century.
Lore, T. A., Imungi, J. K., & Mubuu, K. (2013). A Framing Analysis of Newspaper Coverage of Genetically Modified Crops in Kenya. Journal of Agricultural & Food Information,
14(2), 132–150. https://doi.org/10.1080/10496505.2013.774277
Lynch, G. (2014). Electing the “alliance of the accused”: the success of the Jubilee Alliance in Kenya’s Rift Valley. Journal of Eastern African Studies, 8(1), 93–114.
https://doi.org/10.1080/17531055.2013.844438
Maratea, R. (2008). The e-Rise and Fall of Social Problems: The Blogosphere as a Public Arena. Social Problems, 55(1), 139–160. https://doi.org/10.1525/sp.2008.55.1.139 McQuail, D. (2010). McQuail’s Mass Communication Theory. Book.Sage publications. Momanyi, B. (2014). Kenya urges Africans to support ICC changes. Retrieved April 17,
2017, from http://www.capitalfm.co.ke/news/2013/11/kenya-urges-africans-to-support-icc-changes/
Mueller, S. D. (2014). Kenya and the International Criminal Court (ICC): politics, the election and the law. Journal of Eastern African Studies, 8(1), 25–42.
https://doi.org/10.1080/17531055.2013.874142
Knottnerus, & E. de Volder (Eds.), Africans and the ICC Hypocrisy, Impunity, and
Perversion (pp. 47–61). New York: Cambridge University Press.
Nelson, T. E., Clawson, R. a., & Oxley, Z. M. (1997). Media Framing of a Civil Liberties Conflict and Its Effect on Tolerance. American Political Science Review, 91(3), 567– 583. https://doi.org/10.2307/2952075
Nyamnjoh, F. B. (2005). Africa’s Media, Democracy and Politics of Belonging (pp. 34–35). Zed Books. London.
Ogola, G. (2011). The Political Economy of the Media in Kenya: From Kenyatta’s Nation- Building Press to Kibaki’s Local- Language FM Radio. Africa Today, 57, 77–95. Omanga, D. (2012). Editorial cartoons and the War on Terror in Kenya’s print media.
Journal of African Media Studies, 4(1), 75–96. https://doi.org/10.1386/jams.4.1.75_1
Rothmyer, K. (2012). Kenyans Put Criminal Court on Trial. Nation, 294 (22), 20–25. Scheufele, D. A. (2009). Journal of Communication, 59(2), 205–406.
Schuck, A. R. T., & de Vreese, C. H. (2006). Between Risk and Opportunity. European
Journal of Communication, 21(1), 5–32. https://doi.org/10.1177/0267323106060987
Sinekopova, G. V. (2006). Building the public sphere: Bases and biases. Journal of
Communication, 56(3), 505–522. https://doi.org/10.1111/j.1460-2466.2006.00298.x
Strömbäck, J. (2005). In Search of a Standard: four models of democracy and their normative implications for journalism. Journalism Studies, 6(3), 331–345.
https://doi.org/10.1080/14616700500131950
Strömbäck, J. (2008). Four Phases of Mediatization: An Analysis of the Mediatization of Politics. The International Journal of Press/Politics, 13(3), 228–246.
https://doi.org/10.1177/1940161208319097
Taber, C. S., & Lodge, M. (2016). Motivated Skepticism in the Evaluation of Political Beliefs
755–769.
Valentino, N. A., Beckmann, M. N., & Buhr, T. A. (2001). A Spiral of Cynicism for Some: The Contingent Effects of Campaign News Frames on Participation and Confidence in Government. Political Communication, 18(4), 347–367.
https://doi.org/10.1080/10584600152647083
Van Gorp, B. (2007). The constructionist approach to framing: Bringing culture back in.
Journal of Communication, 57(1), 60–78.
https://doi.org/10.1111/j.1460-2466.2006.00329.x
Zaller, J. (1992). The nature and origins of mass opinion. New York: Cambridge University Press.
Appendix A
Stimulus Materials for Experimental Conditions Image used in all conditions
ICC Moves to New Office
October 25, 2016 | By Francis Juma
Next month, the International Criminal Court (ICC) will move into its permanent premises, still within The Hague, Netherlands. The permanent premises were handed over last
Thursday.
ICC President Silvia Fernández de Gurmendi received a symbolic key from Courtys project manager Bart van Eijk.
Courtys was the consortium in charge of the construction works. The handover was attended by the ICC Judges, the ICC Prosecutor and the ICC Registrar.
The new building is close to the ICC’s detention centre and to major roads. It is part of the International Zone of The Hague, which also contains the Peace Palace, Europol, ICTY, OPCW and The Hague World Forum. The building complex consists of six towers connected on the ground and first floors and offers more than 1,200 workplaces.
Version 2 (Positive Tone/Neutral Source) ICC withdrawal law rejected
October 25, 2016 | By Francis Juma
A bill seeking to have Kenya pull out of the Rome Statute has been rejected in the National Assembly.
This comes after cases against six Kenyans — among them President Uhuru Kenyatta and his deputy William Ruto — were dismissed by the International Criminal Court.
The tabling of the Bill by Bumula MP Boniface Otsiula was not expected.
It now widely anticipated that future crimes against humanity cases would be handled in The Hague.
Prof Michelle Majabo, an expert in International law said the ICC has been impartial in Africa. Prof Majabo added that the recent establishment of an International Crimes Division by the Judicial Service Commission would ensure all future investigations and prosecutions of crimes such as those committed in the early months of 2008 are handled collectively.
Version 3 (Positive Tone/ Source B)
ICC withdrawal law rejected
October 25, 2016 | By Francis Juma
A bill seeking to have Kenya pull out of the Rome Statute has been rejected in the National Assembly.
This comes after cases against six Kenyans — among them President Uhuru Kenyatta and his deputy William Ruto — were dismissed by the International Criminal Court.
The tabling of the Bill by Bumula MP Boniface Otsiula was not expected.
It is now widely anticipated that future crimes against humanity cases would be handled in The Hague.
Prof Michelle Majabo, an expert in International law said the ICC has been impartial in Africa. Prof Majabo added that the recent establishment of an International Crimes Division by the Judicial Service Commission would ensure all future investigations and prosecutions of crimes such as those committed in the early months of 2008 are handled collectively. Speaking in Nakuru, Opposition leader Raila Odinga, a staunch supporter of The Hague based court said more Kenyans would be allowed to face trial at the ICC in future.
Version 4 (Negative Tone/Neutral Source)
ICC Withdrawal Law tabled
October 25, 2016 | By Francis Juma
A bill seeking to have Kenya pull out of the Rome Statute has been tabled in the National Assembly.
This comes after cases against six Kenyans — among them President Uhuru Kenyatta and his deputy William Ruto — were dismissed by the International Criminal Court.
It is now widely anticipated that future crimes against humanity cases would be handled locally.
Prof Michelle Majabo, an expert in International law said the ICC has
disproportionally targeted Africa. Prof Majabo added that the recent establishment of an International Crimes Division by the Judicial Service Commission would ensure all future investigations and prosecutions of crimes such as those committed in the early months of 2008 happen in Kenya.
Version 5 Negative Tone/Source A
ICC Withdrawal Law tabled
October 25, 2016 | By Francis Juma
A bill seeking to have Kenya pull out of the Rome Statute has been tabled in the National Assembly.
This comes after cases against six Kenyans — among them President Uhuru Kenyatta and his deputy William Ruto — were dismissed by the International Criminal Court.
The tabling of the Bill by Bumula MP Boniface Otsiula was expected.
It is now widely anticipated that future crimes against humanity cases would be handled locally.
Prof Michelle Majabo, an expert in International law said the ICC has
disproportionally targeted Africa. Prof Majabo added that the recently establishment of an International Crimes Division by the Judicial Service Commission would ensure all future investigations and prosecutions of crimes such as those committed in the early months of 2008 happen in Kenya. Speaking in Nakuru President Kenyatta, a harsh critic of The Hague based court said no Kenyan would be allowed to face trial at the ICC in future.
Appendix B
Moderators
Political Knowledge
1. Who is the Current Chief Prosecutor of the ICC?
2. Can you name one of the judges who presided over any of the Kenyan cases at the ICC?
3. Which International law was signed by member states to set up the ICC? 4. Which of the following countries is not a member of the ICC
A. Uganda B. South Africa C. Rwanda D. Nigeria
Ethnicity
In Kenya, people speak other languages in addition to English and Kiswahili which other local languages do you speak?
Political Leaning
Who are you likely to vote for in the upcoming Presidential Elections? A. Government/Government Coalition
B. Opposition/Opposition Coalition C. Independent Candidate
D. Other party/Coalition Candidate
Dependent Variable (ICC Evaluation)
To what extent do you agree with the following statement. On a scale of 1-5 (1= Strongly disagree, 5 Strongly agree
1. I support the ICC’s actions in Kenya
3. The ICC should prosecute more cases in Kenya
