Media Framing Effects
Puerto Rican Attitudes towards the Status of the Island
within the United States
________________________________________
Sandra Rullo Pichel
11300477
Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science
Supervisor: Thomas E. Powell
02/06/2017
2 Abstract
Framing of news articles can affect how people perceive a political issue and their attitudes towards it. However, little is known about how this theory plays out in
countries like Puerto Rico, where the definition of their political status is under constant watch and the main newspapers are biased based on whether they support independence of the island from the United States (U.S.), association with America, or to become the 51st state of the union. Since a status referendum is to take place on June 11th 2017, it is interesting to study how media framing affects voters. In order to see these effects in such a political environment, a survey-experiment was carried out to see how media coverage of the new U.S. administration regarding Puerto Rico has influenced these citizens’ attitudes towards the coming referendum, the status of the island, and the solution of the issue. Results show that news framing does not influence Puerto Ricans in their status preferences or in their voting behavior. However, it does play a role when it comes to attitudes towards holding the referendum and definitely solving the issue: compared to those exposed to the negative article, people exposed to positive one were significantly more likely to support the consultation, as well as to believe in the solution of the status issue.
3 The Oldest Colony in the World
Ever since the United States took control of Puerto Rico in 1898 the island has struggled with the definition of its status within -or outside- the United Stated. The so-called neo-colony is torn among the current territorial status –also known as Free Associated State or E.L.A. in Spanish-, statehood, and different forms of independence (Thornburgh, 2007, p. 74); a dilemma that has driven local politics throughout the 20th century and continues to do so nowadays (Meléndez & Meléndez, 1993). The most recent example of this is the referendum scheduled for June 11th 2017, in which the Puerto Rican government aims to determine the status of the island based on its citizens’ preferred option. Although this is the latest attempt to solve the problem, it is not the first. This will actually be the fifth consultation to be held in the Caribbean island since it became a U.S. territory over a century ago1.
The main issue lies in the fact that Puerto Ricans are considered second-class citizens (Malavet, 2004, p.10): while they hold an American passport and are allowed to freely travel, live and work in the U.S., they cannot participate in presidential elections and their representative in Congress has no right to vote (Library of Congress, 2017; Regis, 2017; Trías Monge, 1997). Furthermore, even though they can vote in the U.S. primaries and elect a new governor every four years (Puerto Rican Constitution, 1952), the American Constitution is above the Puerto Rican one, which allows Congress to modify and overrule political and economic decisions made by the Caribbean
government (Constitution of the United States, 1787; Subervi Vélez, Hernández López, & Frambes-Buxeda, 1995, p. 49).
When on November 8th, 2016 Puerto Ricans casted their ballots to elect their new governor -pro-statehood party leader Ricardo Roselló Nevares- most polls in the
1
4 U.S. gave Hilary Clinton a win over Donald Trump by an average of four points
(RealClear, 2016). A few hours later, the Republican candidate was elected president against all odds. Not only had some of the country’s biggest media outlets like The New York Times heavily criticized the Republican candidate (Craig, 2016), they had also believed that he would never make it to the White House (Bradner, 2016). On
November 9th Business Insider reported that “almost every major forecasting aggregator (…) heavily favored a Clinton victory in the lead-up to Tuesday’s race” (Tani, 2016). Media’s conviction that Trump would not win played such a big role that even markets plunged when these unexpected news were made public (Tani, 2016). In addition to this, the president’s attacks against “the dishonest media” and his use of Twitter as a communication tool have helped enlarge the role of journalism both in his campaign and in his first term in the White House (Jacobs, 2017). With this change in the U.S. administration and the impact it is having in American policies, it would be interesting to see how media coverage of this presidency has influenced Puerto Rican perceptions regarding the status of the island. After all, the U.S. has political power over this Caribbean territory and can ultimately determine the legal validity of the status referendum of June.
Newspapers in the Caribbean
When looking at Puerto Rican media, and particularly at newspapers, it is interesting to see that the main outlets in the island do not align with the traditional left-right editorial line (Subervi Vélez et al., 1995). As explained above, determining the status of Puerto Rico has been an issue for decades, and this has also had an impact on the development of national newspapers.
Although it is common for media everywhere to explicit- or implicitly support one side of the political spectrum or the other, newspapers in Puerto Rico create their
5 content based on their stand on the island’s status (Subervi Vélez et al., 1995, p. 49). As a consequence, actions and reactions of the U.S. to Puerto Rican affairs will be reported differently depending on the outlet. For instance, about a month after Trump was elected, two Puerto Rican newspapers published a story on the future of the island and the new U.S. administration: while El Vocero (Méndez Núñez, 2016) was positive about the elected presidency and acknowledged its commitment with the results of the status referendum, Claridad (2016) was negative about it and stressed the new U.S. cabinet’s interventionism in the status issue.
In academia, using these different points of view when reporting a specific event is linked to media framing and its effects on people’s attitudes. In line with this, I will examine how different media frames influence Puerto Ricans’ opinions on the island’s status options: E.L.A., statehood, independence and free association.
Concepts and Theoretical Framework Of Attitudes and Framing
As explained by Entman (1993), to frame is to “select some aspects of a
perceived reality and make them more salient [noticeable, memorable for the readers] in a communicating text” in order to stress a particular problem, its causes, its moral consequences and a way to solve it (p. 52). Edelman (1993) explains that the world is “a kaleidoscope of potential realities” and that these can be portrayed differently
depending on how they are framed (p. 232). In line with this, Gamson and Modigliani (1987) state that media frames are a central “story line that provides meaning to an unfolding strip of events” (p. 143), while Tuchman (1978) says it is a way to organize everyday reality. In addition, he states that it is an “essential feature of news” (p. 193), which relates to the explicit Puerto Rican newspapers’ bias mentioned in the
6 so from a pro-independence one. When thinking about media framing in the context of this study, it could be defined as the textual representation of an event in which certain aspects of the reality being portrayed are emphasized, therefore leaving others
undermined. In this case, a positively-framed article towards the referendum would consist on highlighting the chance to finally solve the status issue over the costs of such consultation, for instance. However, if the costs are the ones highlighted instead, the frame would be negative towards the referendum.
Previous research has shown that media framing can have noticeable effects on readers and audiences. Studies focused “on information processing and on how citizens interpret and ‘understand’ a news frame” (Lecheler, Keer, Schuck, & Hänggli, 2015, p. 341) have demonstrated that public opinion is malleable and vulnerable to framing effects (Entman, 1993). Frames -as defined above- establish a problem, diagnose its causes, make moral claims by evaluating the causer(s) and its/their effects, and suggest solutions to the issue (Entman, 1993); as a result, Scheufele and Tewksbury (2007) argue that salience of certain aspects of a problem affects readers’ perceptions of it. Furthermore, research has shown that media framing does not only have an effect on people’s interpretation of an issue, but also on attitudes, opinions and behaviors towards it (Nelson, Oxley, & Clawson, 1997; Schuck & de Vreese, 2006). Chong and Druckman (2007) even note that opinion leaders -newspapers, for instance- use media to create opposing frames in order to “influence public preferences” (p. 100). Therefore, for the purposes of this paper media framing effects refers to how media framing affects public opinion by influencing people’s interpretation of an issue, as well as their attitudes and behaviors towards it. When applied to this study, it means how the framing of news articles affects Puerto Ricans regarding their attitudes and behaviors towards the status
7 issue, be it the preferred status option, the referendum itself, the definite solution of this problem or whether to vote in the consultation or not.
As a result, the framing paradigm could be applied to the study of public opinion and voting behavior in political science (Entman, 1993), and consequently explains the relationship between the media coverage of the new U.S. administration regarding Puerto Rican issues and readers’ views on the status of their island. This is particularly relevant at this point in time since Puerto Ricans will be asked to go to the polls in June to express their opinion regarding their position within or outside the U.S. Ultimately, the stronger the effects, the more relevant media framing will be in this context.
Moving on to attitudes, they are defined as an individual’s reaction to a particular aspect of her surroundings, be it people, objects, behaviors or political measures -such as holding the referendum in Puerto Rico or changing the status of the island- (Ajzen & Fishbein, 1977). Building on this, Eagly and Chaiken (1998) state they are “a psychological tendency that is expressed by evaluating a particular entity”
favorably or unfavorably (p. 1). In short, attitudes are the result of the evaluation of an object or thought (Malhotra, 2005). Nevertheless, while some authors conceive them as a reaction to a stimulus, Baron and Byrne (1984) focus on their stability and define them as a relatively permanent group of behaviors, beliefs and feelings towards people, objects and thoughts. Following this tradition, Hogg and Vaughan (2005) argue that they are a rather stable set of “beliefs, feelings, and behavioral tendencies towards socially significant objects, groups, events or symbols” (p. 150). When combining these different definitions with the issue being studied in this paper, attitudes towards the status of the island would result in an individual’s reaction to a political measure or decision related to the political status of Puerto Rico that is made explicit after
8 evaluating the issue, and which is affected by said person’s set of rather stable beliefs, feelings and behaviors.
Since the aim of the study is to see how Puerto Ricans’ attitudes are affected by media framing, knowing what attitudes consist of is important when measuring them. The ABC model of attitudes as explained by McLeod (2014) offers a theoretical ground for this. According to the model, attitudes are formed by three elements -the affective, the behavioral, and the cognitive components- which altogether determine the degree of positivity or negativity with which people react to a stimulus (Jain, 2014). This could be used to measure how much a positive media frame on the new U.S. administration influences positivity towards statehood in Puerto Rico, for instance. In other words, the positivity-negativity reaction could also be understood as the level of support for a certain proposition linked to the initial stimuli. Therefore, at least some of the following elements should be measured when looking at attitudes towards the status issue in the island: a) people’s feelings and emotions2, b) actions and intended actions3 (e.g.
whether people will vote in the referendum or not), and c) beliefs and knowledge4 (e.g. I believe the best option for the island is to maintain its current status).
Besides media framing effects and attitudes towards the status of the island, the three different status options will be defined due to their relevance in this study.
E.L.A., Statehood or Independence/Free Association?
On February 3rd, 2017, Puerto Rico Governor Roselló signed a law that will allow citizens to state their opinion on the status of the island in a referendum. As already said, this consultation offers Puerto Ricans three options to choose from. It is, therefore, vital that the terms stated on the referendum are properly defined, since concepts such as statehood and independence can seem very similar at first glance.
2 Affective component. 3 Behavioral component. 4 Cognitive component.
9 Let us begin with the most complex one of them all, the current status, Free Associated State or E.L.A. (Estado Libre Asociado) for its acronym in Spanish. This term was defined in the referendum of 1998 as the “permanent union between Puerto Rico and the United States,” which grants American citizenship to Puerto Ricans, a common market, currency and defense with the U.S., cultural identity and fiscal autonomy (Malavet, 2004, p. 85). In the upcoming referendum, the Caribbean
government added that this option also means that the island will remain under plenary powers of Congress and hinges upon Article 4 of the U.S. Constitution, which states that American territories like Puerto Rico are subject to any rules and regulations passed by Congress (Constitution of the United States, 1787, p. 9). For the purposes of this project, the E.L.A. will be understood as the permanent association between Puerto Rico and the U.S., granting islanders a common citizenship, market, currency and defense with the American people, protection of the Puerto Rican cultural identity, as well as the right of the U.S. Congress to regulate the island’s domestic affairs when deemed necessary.
Moving on to the next concept, statehood is related to becoming the 51st state of the U.S. Meléndez and Meléndez (1993) define this option as a combination of
“sovereignty” and “equality”. Sovereignty meaning that the island will have “control over local affairs,” as well as it will “protect the cultural and ethnical identity of its citizens” (p. 47); and equality referring to being treated like any of the other 50 American states in both economic and social terms (p. 47). Malavet (2004) stresses having “the same rights, benefits, and responsibilities” as any other U.S. citizen (p. 84). The initial proposal for the referendum of June states that statehood would mean the citizens’ support for legislation that guarantees equal conditions, representation in Congress and presidential vote for Puerto Ricans (Senado de Puerto Rico, 2017, p.
9-10 10). Taking all this into account, in this paper statehood is the integration of Puerto Rico as a state of the U.S., guaranteeing its citizens the same rights, benefits and responsibilities as their American counterparts, as well as the recognition and protection of their culture and ethnicity; likewise, it implies Puerto Rican support for any legislation that aims at granting equal conditions with the other 50 states.
Lastly, the concepts of independence/free association will be clarified. Malavet (2004) explains that independence supporters favor the “separation from the colonial power and […] the right of people to rule themselves in their own land” (p. 50, 86). This would additionally, mean “support of Puerto Rican citizenship” (Senado de Puerto Rico, 2017, p. 11-12). Therefore, here independence means the complete separation of Puerto Rico from the U.S., granting the island full autonomy and sovereignty over its political, legal, social and cultural affairs, as well as support of the Puerto Rican
nationality. Regarding free association, the Senate of Puerto Rico (2017) defines it “as a complete and unencumbered Independence” (p. 11), based on political association between the island and the U.S. to be agreed upon by the two sovereign nations. In this case, the people of Puerto Rico would also determine the extent to which jurisdictional powers are conferred to the U.S. (Senado de Puerto Rico, 2017, p. 11). Consequently, in this paper free association refers to the complete independence of Puerto Rico in
political association with the U.S., which bears the transfer of certain jurisdictional powers to the Union when agreed upon by both sovereign nations. In line with the structure of the status referendum taking place in June5 and in order to reflect the Puerto Rican situation as fairly as possible, the options of independence and free association have been combined in this study.
5
Puerto Ricans will have to choose between a) E.L.A., b) statehood and c) independence/free association in June. In case the last option is the most voted one, another referendum will take place after the summer to choose between a) independence and b) free association.
11 Research Questions and Hypotheses
At this point, I will draw the four research questions that drive the course of this study, which englobe all the concepts defined above and aim to unveil how the different theories described before adapt to this particular case.
RQ1: How has media framing of the new U.S. administration regarding Puerto
Rico influenced Puerto Ricans’ attitudes towards the political status of the island?
RQ2: How has media framing of the new U.S. administration regarding Puerto
Rico influenced Puerto Ricans’ attitudes towards the status referendum? Since all the questions target the relationship between media framing and Puerto Ricans’ attitudes towards political events, hypotheses should follow the same pattern in all cases. According to Entman (1993), public opinion is affected by the framing of a news story, and therefore readers would hypothetically be influenced by media articles. Scheufele and Tewksbury (2007) stress that salience or framing influences how people perceive an issue, while Nelson et al. (1997) specify that attitudes are affected by media framing. In sum, theory shows that people are affected by the information they receive on a particular matter. For this case, it means that a biased political article could make readers take the same stand as the one defended in it. Taking all this into consideration, the following results can be expected:
H1A: Compared to people exposed to the negative frame, those exposed to the
positive frame will be more favorable to statehood.
H2A: Compared to people exposed to the negative frame, those exposed to the
12 H3A: Compared to people exposed to the negative frame, those exposed to the
positive frame will be more optimistic regarding the definite solution of the status issue.
H4A: Compared to people exposed to the negative frame, those exposed to the
positive frame will be more likely to vote in the referendum in June. Likewise,
H1B: Compared to people exposed to the positive frame, those exposed to the
negative frame will be more favorable to independence/free association. H2B: Compared to people exposed to the positive frame, those exposed to the
negative frame will be less likely to support the referendum.
H3B: Compared to people exposed to the positive frame, those exposed to the
negative frame will be more pessimistic regarding the definite solution of the status issue.
H4B: Compared to people exposed to the positive frame, those exposed to the
negative frame will be less likely to vote in the referendum in June. Methodology
General Design
In order to answer the proposed research questions, a cross-sectional, online survey-experiment was carried out. It was designed in three different stages -an initial questionnaire, a stimuli and a post-test- and shared online with potential participants on April 26th, 2017. Each person was randomly assigned one of three articles that
independently represented the different news frames: positive, negative and neutral. Since the topic being studied here is very particular, external validity only applies to countries in which sovereignty in not fully granted such as the U.S. Virgin Islands or Guam, among others.
13 Target Participants and Final Sample
The target group for this experiment were people who were at least 18 years old at the time of completing the questionnaire and born in Puerto Rico. In order to reach the right respondents, a link to the survey-experiment was shared on Facebook with 42 Puerto Ricans that met these characteristics. They were asked to take part in the
experiment, as well as to send it on to other people in their network. Although this technique –snowball sampling- is not the ideal one, it was the most feasible for this study considering the lack of funds and the physical distance between the researcher and the targeted participants.
The goal was to reach at least 50 respondents per condition and a total of 150 in order to observe weak framing effects as well. After a week gathering responses the goal had already been reached and over 300 Puerto Ricans had submitted their answers. Nevertheless, some of the experiments had to be discarded because they were filled out by minors or people who had not been born in Puerto Rico. For identification purposes, two questions were placed at the beginning of the survey acting as gatekeepers, along with the informed consent: are you over 18? And were you born in Puerto Rico? A negative answer on either of these sent the participant directly to the end of the survey. Lastly, experiments that had not been fully completed were also discarded for the analysis.
Once non-valid answers were deleted from the raw dataset, the sample consisted of 204 participants ranging from 19 to 70 years old. Despite the wide age range, young people between 19 and 26 years old submitted 64.2% of the responses. The experiment was completed by almost double the amount of women (64.7%) than men (34.8%), and
14 most of the participants (70.6%) had attended bachelor courses6. As for the social class, the vast majority placed themselves in the three central categories (98%). Also most of the participants (89.2%) lived in the island when completing the survey and 62.7% had never lived in the U.S. for at least 3 months. Respondents felt mainly Puerto Rican (78.4%) and had a center-left, center-liberal political profile. Lastly, 66.2% of them believed the status of the island was a very important issue.
A one-way between groups multivariate analysis of variance was performed to see whether the three condition groups differed in terms of gender, age, residence and so on. The test showed that there were no statistically significant differences across these groups, F (20, 384) = .59, p = .917; Wilk's Λ = 0.94 (see Appendix 3). Therefore,
participants were evenly distributed across conditions and randomization was successful for demographics.
Stimuli Development
Before moving on to the explanation of the stimuli, it is important to take into consideration the media landscape in Puerto Rico. As mentioned in the introduction, the main newspapers in the island create their content based on their point of view on the island’s status. It is precisely because of this that I decided to avoid taking the news story for the experiment from any Puerto Rican newspapers, and rather write it from scratch. By doing so, a) participants would not remember having already read the news story they were exposed to, and b) they would react to the story alone rather than to the newspaper where it came from. Moreover, besides U.S. and Puerto Rican newspapers, the status issue is not as relevant for media in other countries and therefore coverage is not as detailed. Consequently, the articles were completely designed for this study, taking into account the different angles Puerto Ricans newspapers had emphasized on
6
Slightly less than 20% of the respondents had started or obtained a master’s degree, and over two thirds of the sample answered the three questions measuring political knowledge correctly (69.6%, 74.5%, 72.1%, respectively).
15 the selected topic and using Associated Press as an additional source: the U.S.’s reaction to the upcoming consultation.
On April 13th, the U.S. Department of Justice sent a letter to Puerto Rican Governor Roselló stating that they would not provide $2.5 million to the status referendum, and left the possibility of financing it if a current status option was included in the consultation. Taking this story as a base, I wrote three articles that had the same structure, sources and content, but each one of them stressed a different side of the story: the positive article focused on the will of the U.S. to help Puerto Rico finance the referendum as long as all status feelings were represented in the consultation, including the E.L.A. The negative story put the spotlight on the denial of the U.S. to economically support the referendum unless the current status was included as an option in the consultation. The neutral article combined quotes from the stories above,
balancing arguments in favor and against the reaction of Trump’s cabinet. The purpose of this last story was to have a control group in order to see the effectiveness of the experiment7.
As a result, the manipulation consisted in changing the frame of the news stories in order to have a positive, a negative and a neutral frame on the U.S. administration’s reaction to the referendum. Once the articles were written, they were checked by several native Spanish speakers for language, realism and consistency. They were also
formatted similarly to other news stories published online to make them look as genuine as possible.
Survey-Experiment Procedure
When conducting the experiment, participants first answered a set of questions that tackled the problems the island is facing, their national feeling, their political views
7
16 and their political knowledge. Immediately after, they were shown one of the three manipulated news articles; the allocation to the story conditions was randomized. This was followed by another set of questions targeting participants’ opinions on the status of the island, on the referendum itself, their expectations regarding the effectiveness of the referendum in solving the status problem, as well as their voting behavior in the
consultation. At the end of this section, a manipulation check and demographic data collection were included8. All this was done online using Qualtrics as a mean to distribute the survey-experiment and to gather participants’ responses.
Variables and Measures
Similarly to other experiments measuring framing effects, the dependent variables included items that dealt with the issue discussed in the stimuli, as well as with the broader topic (Lecheler et at., 2015, p. 347). For the latter I used the following statements: “I believe the best option for the island is to maintain its current status,” “Puerto Rico should become the U.S.’s 51st state,” “the island should be a completely independent country,” “this referendum will solve the status issue” and “the status of the island will not be solved in the coming years.” The first three measured attitudes towards the status of the island (Status Options), while the last two dealt with attitudes towards solving the issue (Solving Status). Also dealing with the broader topic,
participants were asked whether they would vote in the referendum or not. In the case of attitudes towards the referendum itself -targeted by the articles- (Status Referendum), I used the items “I agree with holding this referendum” and “no more referendums should be held in the island.” All respondents rated their (dis)agreement with the statements in a five-point scale, ranging from “totally disagree” to “totally agree.” In order to check the reliability of the scales for further analyses, some of the items were reversed.
8
17 The reliability of the scales turned out high but not ideal. The fact that each item measured a slightly different idea (current referendum vs. future ones, for instance) could explain this reliability problem. Cronbach’s alphas were .604 for ‘Solving Status’ and .617 for ‘Status Referendum’ (see Appendix 4). In the case of ‘Status Options’, only after disregarding the item referring to the current status and leaving the extremes alone (independence and statehood) did I achieve a moderately reliable scale .652. As a consequence, it was feasible that analyses of the scales led to non-significant results and items had to be analyzed independently.
Analyses and Results Attitudes towards the Status Options
In line with what literature states, I predicted that people who read the positive article would be more favorable to statehood than those exposed to the negative one (H1A). To test this hypothesis I carried out a one-way ANOVA to see whether framing
affects Puerto Rican perceptions of status. I first analyzed agreement with statehood (M = 2.44, SD = 1.483, σ = 2.198) and saw that there were no significant differences between groups as shown in Table 1 (F (2, 201) = .557, p = .574). Raw data already hinted at this since the means in each condition did not differ much from each other (Mpos = 2.61, SDpos = 1.584; Mneg = 2.39, SDneg = 1.419; Mneu = 2.35, SDneu = 1.465).
The opposite hypothesis, which claims that people exposed to the negative article would be more likely to support independence (H1B), was not supported by the
evidence either: when looking at agreement with independence (M = 2.98, SD = 1.421, σ = 2.019), data shows that there are no significant differences across groups (F (2, 201) = .252, p = .778; see Table 1). Once again, the means in each group were close to each other (Mpos = 2.97, SDpos = 1.414; Mneg = 2.90, SDneg = 1.465; Mneu = 3.07, SDneu = 1.397).
18 The same occurs with the statement on E.L.A. (M = 1.97, SD = 1.146, σ = 1.314), for which there are no relevant differences (F (2, 201) = 1.621, p = .200; see Table 1). Means did not show much variation here either (Mpos = 2.02, SDpos = 1.147;
Mneg = 2.11, SDneg = 1.214; Mneu = 1.78, SDneu = 1.064). This data could be used as proof that the neutral manipulation was successful. However, since none of the two other cases showed any significant results, H1A and H1B can be rejected.
Table 1
ANOVA Framing Effects on Attitudes towards Status Options
Attitudes towards the Status Referendum
Results were significant for the second set of hypotheses on whether people exposed to the positive article would be keener on holding the referendum than those exposed to the negative one (H2A), and vice versa (H2B). A one-way ANOVA test of the
item “I agree with holding this referendum” (M = 2.83, SD = 1.365, σ = 1.864) revealed that there are statistically significant differences between groups (F (2, 201) = 6.002, p = .003). As Table 2 illustrates, a Tukey post-hoc test showed that support for the referendum is significantly higher among people exposed to the positive article (M = 3.33, SD = 1.235) than those exposed to the negative (M = 2.61, SD = 1.459, p = .006)
Items Sum of
Squares df
Mean
Square F
I believe the best option for the island
is to maintain its current status
Between Groups 4.233 2 2.117 1.621
Within Groups 262.527 201 1.306
Total 266.760 203
The island should be a completely independent country
Between Groups 1.025 2 .512 .252
Within Groups 408.897 201 2.034
Total 409.922 203
Puerto Rico should become the U.S.’s
51st state
Between Groups 2.460 2 1.230 .557
Within Groups 443.835 201 2.208
Total 446.294 203
19 and the neutral ones (M = 2.64, SD =1.282, p = .009)9. In this case, there was no
significant difference between those exposed to the negative and neutral articles (p = .988).
Table 2
Tukey HSD Comparison Framing Effects on Attitudes towards the Status Referendum across Conditions
When looking at future consultations, analyses determined that the item “no more referendums should be held in the island” (M = 2.37, SD = 1.259, σ = 1.585) did not show any significant difference between groups, as indicated in Table 3 (F (2, 201) = .286, p = .751). The means in the three conditions did not vary much either (Mpos = 2.41, SDpos = 1.296; Mneg = 2.28, SDneg = 1.289; Mneu = 2.43, SDneu = 1.208).
Consequently, evidence supports H2A and H2B, but only when they refer to the
referendum process taking place in June. People are not affected by frames when it comes to the possibility of future consultations.
9
All the items analyzed here were coded from strongly disagree (1) to strongly agree (5), therefore a higher score means more support for the particular statement.
Dependent
Variable (I) Frames (J) Frames
Mean Difference (I-J) Std. Error 95% Confidence Interval Lower Bound Upper Bound I agree with holding this referendum Positive Negative .722 * .233 .17 1.27 Neutral .689* .232 .14 1.24 Negative Positive -.722 * .233 -1.27 -.17 Neutral -.033 .223 -.56 .49 Neutral Positive -.689 * .232 -1.24 -.14 Negative .033 .223 -.49 .56 * p < .05
20 Table 3
ANOVA Framing Effects on Attitudes towards the Status Referendum
Attitudes towards Solving the Status Issue
The last two statements measured attitudes towards definitely solving the status issue. The objective was to see whether participants exposed to the positive frame would be more confident in the solution of this problem compared to people exposed to the negative one (H3A). Likewise, they aimed to see if respondents who read the
negative article would be more pessimistic regarding the solution of this problem (H3B).
The first item, “this referendum will solve the status issue” (M = 1.84, SD = 1.054, σ = 1.112) showed attitudes towards this occurrence were significantly different between groups (F (2, 201) = 5.338, p = .006). When carrying a Tukey post-hoc test (see Table 4), results showed that confidence in the solution of the status issue is higher among those exposed to the positive frame (M = 2.20, SD = 1.181) than the ones who read the negative article (M = 1.65, SD = .987, p = .007) or the neutral one (M = 1.72, SD = .938, p = .024). When looking at the relationship between the negative and neutral frames, however, evidence shows that there was no significant difference between these groups (p = .903).
Items Sum of
Squares df
Mean
Square F I agree with holding
this referendum Between Groups 21.322 2 10.661 6.002* Within Groups 357.011 201 1.776 Total 378.333 203 No more referendums should be held in the island
Between Groups .913 2 .457 .286
Within Groups 320.773 201 1.596
Total 321.686 203
21 Table 4
Tukey HSD Comparison Framing Effects on Attitudes towards Solving the Status Issue across Conditions
The last statement referred to Puerto Rican confidence in a future solution of the status of the island. Table 5 shows that for the item “the status of the island will not be solved in the coming years” (M = 3.98, SD = 1.098, σ = 1.206) frames did not
significantly affect Puerto Ricans’ attitudes (F (2, 201) = .797, p = .452), as seen in the little difference across means in each group (Mpos = 4.02, SDpos = .922; Mneg = 4.07,
SDneg = 1.234; Mneu = 3.85, SDneu = 1.096). As a consequence, H3A and H3B are
supported by the results. It is important to bear in mind, however, that once again this only applies to the upcoming referendum solving the matter and not to future solutions. Table 5
ANOVA Framing Effects on Attitudes towards Solving the Status Issue Dependent
Variable (I) Frames (J) Frames
Mean Difference (I-J) Std. Error 95% Confidence Interval Lower Bound Upper Bound This referendum will solve the
status issue Positive Negative .549 * .180 .12 .97 Neutral .474* .180 .05 .90 Negative Positive -.549* .180 -.97 -.12 Neutral -.074 .173 -.48 .33 Neutral Positive -.474 * .180 -.90 -.05 Negative .074 .173 -.33 .48 * p < .05 Items Sum of Squares df Mean Square F This referendum will
solve the status issue
Between Groups 11.381 2 5.690 5.338* Within Groups 214.281 201 1.066
Total 225.662 203
The status of the island will not be solved in the coming
years
Between Groups 1.927 2 .963 .797
Within Groups 242.951 201 1.209
Total 244.877 203
22 Attitudes towards Voting in the Referendum
When looking at Puerto Ricans attitudes towards voting, literature suggests that framing can affect behavior and therefore I expected a variation across frames, with people exposed to the positive article being more likely to vote in the referendum than those exposed to the negative one (H4A), and vice versa (H4B). To do so, I created a new
variable that responded to the question “will you vote in the upcoming referendum?” where one equals “yes” and zero equals “no,” and run a Chi-Squared test to see if there is a relationship between frames and voting behavior. The results showed that more people will vote (86.8%) than not (13.2%), and that those exposed to the neutral article are more likely to vote than the rest (Posframe = 86.9%; Negframe = 84.5%; Neuframe = 88.9%). However, as Table 6 illustrates, there is no statistically significant association between frames and voting behavior (χ(2) = .599, p = .741), which leads to the rejection of the last two hypotheses.
Table 6
Results of Chi-Square Test and Descriptives for Attitudes towards Voting in the Referendum
Analysis of Scales Combined
As previously stated in the methodology section, the reliability of the scales was not ideal; I nevertheless conducted an ANOVA test for ‘Status Referendum’ and
‘Solving Status,’ where I obtained significant results in one of the items, and disregarded ‘Status Options’ due to the lack of them. I created a new variable that
Voting behavior Frames
Positive Negative Neutral
Will vote 53a (86.9%) 60a (84.5%) 64a (88.9%)
Will not vote 8a (13.1%) 11a (15.5%) 8a (11.1%)
Note. χ2 = 0.599, df = 2. Numbers in parentheses indicate column percentages. Each subscript letter denotes a subset of Frames categories whose column proportions do not differ significantly from each other at the .05 level.
23 computed the results of the statements in each scale into an average10and the outcome was not statistically significant. In the first case, the combination of “I agree with holding this referendum” and “more referendums should be held in the island” (M = 3.23, SD = 1.117, σ = 1.247) showed that frames did not affect present and future attitudes at the same time (F (2, 201) = 1.888, p = .154). Likewise, the average of “this referendum will solve the status issue” and “the status of the island will be solved in the coming years” (M = 3.23, SD = 1.117, σ = 1.247) also proved the absence of significant effects in attitudes regarding the same matter at the different points in time (F (2, 201) = 1.812, p = .166).
Discussion and Conclusion Puerto Rican Attitudes
The aim of this paper was to see how framing effects apply to Puerto Rican attitudes towards the status of their island, holding another status referendum, whether to vote in June, and the definite solution of this issue. The analysis showed that news framing did not have any effects on citizens’ opinions on the different status options and their voting behavior. A possible explanation could be that people do not construct their political point of view nor take political action based on information alone, but also based on their own predispositions, such as values, party identification, and ideology (Zaller, 1992, p. 554). Political knowledge also allows people to create consistent views on different issues and shape their attitudes towards them (Delli Carpini & Keeter, 1997). Therefore, despite academia’s statements on the malleability of public opinion (Entman, 1993), this study suggests that people’s inclinations and knowledge seem to play a big role in political issues deeply linked to the recent history of their country.
10
24 Contrary to the previous case, results did show that people were affected by news framing when it comes to support for the referendum and the definite solution of the status issue. However, in both cases results were promising only when applied to the referendum that is to take place in June. Those who faced the positive stimuli were considerable more supportive towards the consultation than the rest, but this was not the case for future referendums. An explanation for this might be that the articles focused on the status process of June, its funding, Trump’s cabinet’s reaction to it, and therefore there was no information in the articles that could affect attitudes towards the future. These results could also hint at the possibility that framing effects only influence people’s attitudes towards the topic being discussed; moreover, it could indicate that these effects are especially relevant for upcoming events but disappear in the long-term. It was also noticeable that the neutral and negative frames did not vary much between them. This could be a consequence of a weaker manipulation in the negative or neutral cases than in the positive one.
Likewise, confidence in the solution of Puerto Rico’s status gave similar results: citizens exposed to the positive frame were more optimistic towards solving the issue than the other two groups. Their opinion was also only influenced in the short-term; solutions of the status issue other than the coming consultation were not affected by the stimuli. Once again, there were no differences between the negative and neutral frames. If the manipulation was indeed weaker in these cases than in the positive one, it is logical that the results show the same pattern as before. One last explanation as to why this little difference between the control and negative groups is that it could have been caused by external factors. The day after sending the survey-experiment, Donald Trump published two tweets on Puerto Rico –the first ones in months- that were not keen on helping the island: “Democrats are trying to bail out insurance companies from
25 disastrous #ObamaCare, and Puerto Rico with your tax dollars. Sad!” (Trump, 2017) and “the Democrats want to shut government if we don't bail out Puerto Rico and give billions to their insurance companies for OCare failure. NO!” (Trump, 2017). Since the articles were supposedly written over a week before Trump’s tweets, the effect of the latter could have prevailed for those who read the president’s statements.
Limitations of the Study
Although analyses showed that some of the hypotheses are supported by the evidence, it is important to acknowledge that this study had some limitations that could influence the results.
Firstly, the fact that the survey-experiment was carried out at once did not allow measuring variations in the dependent variables before and after being exposed to the stimulus without participants noticing or being influenced by their own answers in a hypothetical pre-test. Therefore, a suggestion for further research in this topic would be to carry out a longitudinal study, as it is compatible with pre-tests. Furthermore,
longitudinal studies could be used to measure, not only how media framing affects attitudes, but also how long the effects last. On the other hand, by combining the stimuli and the questionnaire into one single survey, it was easier for participants to complete the experiment at one go. Furthermore, by avoiding leaving time between tests,
respondents were not lost in the course of the study (e.g. people forgetting to answer the second part, losing interest in the subject, or simply not feeling like finishing it).
A second challenge faced in this project was that people completed the experiment online. Although this is a very useful way to recruit participants, it is also impossible to control how they respond to the survey: whether they are completely focused on the tasks or not, whether they read the stimuli in full or not, whether they look for additional information while filling the survey or not, or whether they respond
26 individually or along with others. Snowball sampling was also an issue because it was the only feasible way to reach the desired amount of people to carry out this study. As stated above, most participants were middle-class, young, educated people, which could potentially skew the results. Further studies should aim for samples that represent Puerto Rican society as a whole.
Lastly, the different variables being measured here did not assemble into one single scale. The limited amount of statements and the fact that each statement within a topic measured slightly different points in time (present vs. future) made it very difficult to create a reliable scale. Moreover, although an experiment allows control of multiple variables, it was impossible to control for external input on the referendum. Holding the consultation was one of Governor Roselló’s main campaign promises -he stated that “the transition to statehood ha[d ]started” already in his victory speech (Guillama Capella, 2016),- but the measure has become increasingly controversial due to the difficult situation Puerto Rico is going through. The island faces a public debt estimated in over $70 billion; migration to the U.S. has increased in the last decade –which
translated into a net immigration rate of -64,000 people in 2015, according to the island’s Institute of Statistics (Velázquez-Estrada, 2015). On top of that, the University of Puerto Rico will have to carry out cuts that are expected to reach $512 million by the year 2025 (El Nuevo Día, 2017). Since the experiment took place when all these other problems started arising it is feasible that they have affected citizens’ attitudes towards the issues studied here.
Conclusion
In line with previous research, this paper has shown that media framing can indeed have an impact on attitudes towards political matters and towards specific events like the status referendum in Puerto Rico. However, it is important to notice that the
27 findings of this study also point at other directions in relation to framing theory.
Although a slanted news article can affect people’s attitudes and/or behaviors, results suggest that this influence has a limited effect timewise and only applies in the short-term. Accordingly, it is interesting to think about how long the effects of framing last and how strong they are. This could, for example, explain the difference between the items measuring the present Puerto Rican status referendum and a future one. Lastly, some scholars argue that additional factors, such as knowledge or preferences can play an important role in attitudes; therefore, further studies should take these into account as additional -perhaps mediating/moderating- variables.
In practical terms, it is important to note that media coverage of the new U.S. administration regarding Puerto Rico seems to affect what Puerto Ricans think of the status referendum and the odds of solving this problem. Despite ignoring the strength of these effects, it is important that both citizens and journalists keep this in mind. For the first group, it is an opportunity to understand the power that daily news articles can have over people when only one side of the story is told or read, and a chance to be more aware of how our media consumption affects our own views on political issues. For journalists, I believe it is a golden opportunity to reflect on their job and the role they want to play in society. After all, as Edelman (1993) puts it, the world is “a
kaleidoscope of potential realities” (p. 232) and it is in the hand of journalists to choose which one to portray.
Acknowledgements
I would like to thank Camilo J. Miguel Bosques and Juan José Monroig for their useful background information about Puerto Rican newspapers, politics and statistics institutes, as well as for answering my interminable set of questions about their island. I would also like to thank Amanda Walters and Gabriela Morales Vila for sharing a piece
28 of the culture of the mofongo, bomba and coquí that inspired this paper. Lastly, I would like to thank all those who spread the survey-experiment and contributed to this project in any way.
References
Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological bulletin, 84(5), 888–918.
Baron, R. A., & Byrne, D. (1984). Social psychology understanding human interaction, Boston: Allyn & Bacon.
Bradner, E. (2016, Nov. 9). Five surprising lessons from Trump's astonishing win. CNN. Retrieved from http://cnn.it/2pxWkYL.
Chong, D., & Druckman, J. N. (2007). A theory of framing and opinion formation in competitive elite environments. Journal of Communication, 57, 99–118.
Claridad. (2016, Dec. 14). Trump, el Congreso y Puerto Rico. Claridad. Retrieved from http://bit.ly/2rwc816.
Craig, S. (2016, Aug. 20). Trump’s empire: A maze of debts and opaque ties. The New York Times. Retrieved from http://nyti.ms/2qpVgpY.
Delli Carpini, M. X., & Keeter, S. (1997). What Americans know about politics and why it matters. New Haven: Yale University Press.
Eagly, A. H., & Chaiken, S. (1998). Attitude, structure and function. In D. T. Gilbert, S. T. Fisk, & G. Lindsey (Eds.), Handbook of Social Psychology (pp. 269–322). New York: McGowan-Hill.
Edelman, M. (1993). Contestable categories and public opinion. Political Communication, 10, 231-242.
El Nuevo Día. (2017, Apr. 20). La presidenta de la UPR aclara el propuesto recorte de $512 millones. El Nuevo Día. Retrieved from http://bit.ly/2oYRyS4.
29 Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal
of Communication, 43(4), 51-58.
Gamson, W. A., & Modigliani, A. (1987). The changing culture of affirmative action. In R. G. Braungart, & M. M. Braungart (Eds.), Research in political sociology, 3, 137–177.
Guillama Capella, M. (2016, Nov. 8). Regresa el PNP al poder. Diálogo UPR. Retrieved from http://bit.ly/2p1vIyi.
Hogg, M., & Vaughan, G. (2005). Social psychology (4th ed.). London: Prentice-Hall. Hurley, P. A. (1994). The nature and origins of mass opinion (Book Review) [Review of the book The nature and origins of mass opinion]. The Journal of Politics, 56(2), 528-531.
Jacobs, B. (2017, Feb. 19). Trump attacks 'dishonest media' while making false claims at Florida rally. The Guardian. Retrieved from http://bit.ly/2luPy61.
Jain, V. (2014). 3D model of attitude. International Journal of Advanced Research in Management and Social Sciences, 3(3), 1-12.
Jefferson, T., Madison, J., Paine, T., & Adams, J. (1787). Constitution of the United States. Philadelphia. Retrieved from http://bit.ly/2qecrJZ.
Lecheler, S., Keer, M., Schuck, A. R. T., & Hänggli, R. (2015). The effects of repetitive news framing on political opinions over time. Communication Monographs, 82(3), 339-358.
Library of Congress. (2017). Foraker Act (Organic Act of 1900). Retrieved from https://www.loc.gov/rr/hispanic/1898/foraker.html.
Malavet, P. A. (2004). America’s colony: The political and cultural conflict between the United States and Puerto Rico. New York: New York University Press.
30 Malhotra, N. K. (2005). Attitude and affect: New frontiers of research in the 21st
century. Journal of Business Research, 58(4), 477–482.
McLeod, S. A. (2014). Attitudes and behavior. Retrieved from http://bit.ly/THTzmj. Meléndez, E., & Meléndez, E. (Eds.). (1993). Colonial dilemma: Critical perspectives
on contemporary Puerto Rico. Boston: South End Press.
Méndez Núñez, C. (2016, Dec. 15). ‘Denos las herramientas que terminaremos el trabajo.’ El Vocero. Retrieved from http://bit.ly/2qHkkMT.
Nelson, T. E., Oxley, Z. M., & Clawson, R. A. (1997). Toward a psychology of framing effects. Political Behavior, 19(3), 221–246.
Puerto Rican Constitution. (1952). Art. IV. Del poder ejecutivo. Retrieved from http://bit.ly/2pEFbws.
RealClear. (2016). Politics: Latest election polls [Data file]. Retrieved from http://bit.ly/1lC6WuE.
Regis, J. A. (2017, Feb. 14). Un siglo de la ciudadanía americana. El Nuevo Día. Retrieved from http://bit.ly/2pXKTN6.
Scheufele, D. A., & Tewksbury, D. (2007). Framing, agenda setting, and priming: The evolution of three media effects models. Journal of Communication, 57(1), 9– 20.
Schuck, R. T., & de Vreese, C. H. (2006). Between risk and opportunity: News framing and its effects on public support for EU enlargement. European Journal of Communication, 21, 5–31.
Senado de Puerto Rico. (2017, Apr. 17). Referido a la comisión sobre relaciones federales, políticas y económicas del senado de Puerto Rico. Retrieved from http://bit.ly/2pSX2D3.
31 Subervi Vélez, F. A., Hernández López, N. M., & Frambes-Buxeda, A. (1995). Los
medios de comunicación masiva en Puerto Rico. Comunicación y Sociedad, 24, 45-83.
Tani, M. (2016, Nov. 9). Donald Trump shocks world, wins presidential election in biggest upset in political history. Business Insider. Retrieved from
http://bit.ly/2qwiKsO.
Thornburgh, D. (2007). Puerto Rico’s future: A time to decide. Washington D.C.: Center for Strategic and International Studies.
Trías Monge, J. (1997). Puerto Rico: The trials of the oldest colony in the world. Michigan: Yale University.
Trump, D. J. [realDonaldTrump]. (2017, Apr. 26). Democrats. [Twitter moment]. Retrieved from http://bit.ly/2s3M0I3.
Trump, D. J. [realDonaldTrump]. (2017, Apr. 27). The Democrats. [Twitter moment]. Retrieved from http://bit.ly/2r4aj9q.
Tuchman, G. (1978). Making news: A study in the construction of reality. New York: Free Press.
Velázquez Estrada, A. L. (2017). Perfil del migrante, 2015. San Juan, Puerto Rico: State Data Center de Puerto Rico. Retrieved from https://censo.estadisticas.pr/. Zaller, J. (1992). The nature and origins of mass opinion. Cambridge, UK: Cambridge
32 Appendix 1A: Original Survey
Estimado/a participante,
ante todo, nos gustaría agradecerle su colaboración ya que será de gran ayuda para el desarrollo de este proyecto de la Universidad de Ámsterdam. Con la siguiente
investigación queremos conocer la opinión de los puertorriqueños sobre una variedad de temas.
Le agradeceríamos que responda de forma individual y con total sinceridad; recuerde que no hay respuestas correctas o incorrectas. La duración estimada del experimento es de unos 10 minutos.
Haga clic en “siguiente” cuando esté listo/a para comenzar y, de nuevo, muchas gracias por su participación.
Un saludo,
***
Antes de comenzar, nos gustaría recordarle que su anonimato está garantizado y que ninguno de los datos que aporte serán entregados a terceros sin su previo
consentimiento. En caso de que los resultados de este estudio sean publicados, esto se llevará a cabo manteniendo su anonimato.
Así mismo, conserva el derecho de retirar su consentimiento sin justificación alguna y de abandonar la encuesta en cualquier momento. Esta investigación no supondrá riesgo alguno para su persona y no será expuesto a material explícitamente ofensivo.
Usted se compromete a completar el siguiente cuestionario SOLO si es mayor de edad (+18) y nacido en Puerto Rico.
Por la presente, declara que ha sido claramente informado de la naturaleza de esta investigación y acepta participar en la misma de forma completamente voluntaria. He leído el texto aquí presentado y acepto participar en el estudio.
*** ¿Es usted mayor de edad (+18)?
Sí No
Condition: No Is Selected. Skip To: End of Survey. ***
33 ¿Nació usted en Puerto Rico?
Sí No
Condition: No Is Selected. Skip To: End of Survey. ***
A continuación le pediremos que dé su opinión sobre varios temas de actualidad en la isla.
***
En esta pregunta le exponemos algunos de los problemas por los que está pasando la isla. Por favor, indique cuán importantes cree que son:
Muy importante 1 2 3 4 Nada importante 5 Deuda pública Estatus de la isla Éxodo de población a EEUU Leyes de cabotaje Recortes en la universidad *** ¿Cómo se siente usted?
Puertorriqueño/a Estadounidense Ambos
Ninguno
34 En política se suelen utilizar las expresiones izquierda y derecha. En esta pregunta le mostramos una serie de casillas que van de izquierda (con el valor 0) a derecha (con el valor 10). ¿En qué casilla se situaría usted?
0 1 2 3 4 5 6 7 8 9 10 ***
En política se suelen utilizar las expresiones conservador y liberal. En esta pregunta le mostramos una serie de casillas que van de conservador (con el valor 0) a liberal (con el valor 10). ¿En qué casilla se situaría usted?
0 1 2 3 4 5 6 7 8 9 10 ***
35 ¿A qué candidato/a votó en las últimas elecciones a gobernador/a de Puerto Rico? (Noviembre, 2016)
Ricardo A. Roselló Nevarez (PNP) Rafael Bernabe Riefkohl (PPT) David E. Bernier Rivera (PPD) María D. Santiago Negrón (PIP)
Alexandra Lúgaro Aponte (Independiente) Manuel Cidre (Independiente)
No voté
*** Display This Question:
If ¿A qué candidato/a votó en las últimas elecciones a gobernador/a de Puerto Rico? (Noviembre, 2016) Manuel Cidre (Independiente) Is Selected
¿Por qué decidió votar a este candidato? *** Display This Question:
If ¿A qué candidato/a votó en las últimas elecciones a gobernador/a de Puerto Rico? (Noviembre, 2016) Alexandra Lúgaro Aponte (Independiente) Is Selected
¿Por qué decidió votar a esta candidata? ***
Próximamente se celebrará un plebiscito de estatus en la isla, ¿en qué fecha tendrá lugar? (Por favor, indique el mes y el año)
***
Por favor, indique en qué año tuvieron lugar las elecciones a gobernador que ganó Luis G. Fortuño:
***
Según los estudiantes universitarios, ¿por qué los distintos recintos de la Universidad de Puerto Rico deben ir a huelga indefinida?
*** ARTÍCULO
36 A continuación le presentamos las opciones entre las que tendrá que elegir en la
consulta de estatus de junio. Si se celebrase el plebiscito de estatus mañana, ¿qué opción elegiría?
Estadidad
Libre asociación / Independencia Actual estatus territorial
No votaría
*** Display This Question:
If A continuación le presentamos las opciones entre las que tendrá que elegir en la consulta de esta... Libre asociación / Independencia Is Selected
En una segunda ronda de consulta, ¿qué opción elegiría? Libre asociación
Independencia
37 Por favor, indique su opinión sobre las siguientes frases:
Totalmente en desacuerdo En desacuerdo Ni de acuerdo ni en desacuerdo De acuerdo Totalmente de acuerdo Creo que la mejor opción para la isla es mantener el estatus actual Puerto Rico debería convertirse en el 51 estado de Estados Unidos La isla debería ser un país completamente independiente Estoy de acuerdo con la celebración del plebiscito No deberían celebrarse más plebiscitos en la isla Esta consulta va a resolver el problema del estatus de la
38 isla El estatus de la isla no se va a resolver en los próximos años ***
Si se celebrasen elecciones a gobernador mañana, ¿votaría usted o no al mismo candidato que votó en noviembre?
Sí No No lo sé
*** Display This Question:
If Si se celebrasen elecciones a gobernador mañana, ¿votaría usted o no al mismo candidato que votó en noviembre? No Is Selected
Por favor, explique brevemente por qué cambiaría su voto: ***
Display This Question:
If Si se celebrasen elecciones a gobernador mañana, ¿votaría usted o no al mismo candidato que votó en noviembre? Sí Is Selected
Por favor, explique brevemente por qué mantendría su voto: ***
¿De qué trataba el artículo que ha leído?
Las dificultades, obstáculos o inconvenientes de Estados Unidos de cara al plebiscito de junio.
El apoyo de Estados Unidos a la celebración de un plebiscito representativo. Ninguna de las anteriores.
*** ¿Cuántos años tiene usted a día de hoy?
39 ¿Cuál es su género? Hombre Mujer Otro ***
¿Cuál es el nivel de educación más alto que ha cursado (no es necesario haberlo completado)? Escuela media High School Bachillerato Máster Doctorado/PhD Otro *** ¿A qué clase social diría usted que pertenece? Baja Media-baja Media Media-alta Alta *** ¿Dónde reside usted actualmente?
Puerto Rico Estados Unidos Otro
***
¿Ha vivido en alguno de los 50 estados de los Estados Unidos por un período igual o superior a 3 meses?
Sí No
*** IMPORTANTE:
Antes de finalizar, nos gustaría indicarle que el artículo de prensa que acaba de leer ha sido creado y manipulado por los investigadores para poder estudiar cómo ha influido la
40 cobertura mediática de la presidencia de Donald Trump en las actitudes de los
puertorriqueños con respecto al estatus de la isla.
Le agradeceríamos que NO hablase del objetivo y contenido de este experimento con futuros participantes puesto que esto distorsionaría los resultados.
Muchas gracias por su cooperación.
***
Appendix 1B: Translated Original Survey Dear participant,
first of all, we would like to thank you for your cooperation, as it will be very helpful for the development of this project of the University of Amsterdam. With the following investigation we want to know the opinion of Puerto Ricans on several issues.
We would appreciate that you respond individually and with honesty; keep in mind that there are no right or wrong answers. The estimated duration of the experiment is about 10 minutes.
Click on “next” when you are ready to start and, once again, thank you very much for your participation.
Best regards,
***
Before starting, we would like to remind you that your anonymity is guaranteed and none of the data you provide will be delivered to third parties without your prior
consent. In case the results are used published, this will be done in such a way that your anonymity is completely guaranteed.
Likewise, you retain the right to withdraw your consent without any justification and to leave this questionnaire at any point. This experiment will not pose any risk to you and you will not be exposed to material that is explicitly offensive.
You agree to complete the following experiment ONLY if you are over 18 and born in Puerto Rico.
I hereby declare that I have been clearly informed of the nature of this research and I agree to participate in this project voluntarily.
• I have read this text and I agree to participate in this study. ***
Are you over 18?
If no: end of survey.
41 Were you born in Puerto Rico?
If no: end of survey.
***
In the following section, we will ask you to give your opinion on various current affairs taking place in the island.
***
In this question, we have selected some of the problems the island is going through. Please, select how important you think each one of them is:
Public debt, island status, Puerto Rican immigration to the U.S., cabotage laws, cuts in university funds.
*** How do you feel?
Puerto Rican, American, both, none.
***
In politics we usually use the terms left and right. In the following question we will show you a scale going from left (value 0) to right (value 10). Where would you place yourself?
0 - 10
***
In politics we usually use the terms left and right. In the following question we will show you a scale going from conservative (value 0) to liberal (value 10). Where would you place yourself?
0 - 10
***
Which candidate did you vote for in the last Puerto Rican governor elections? Name of each candidate + “I did not vote” (7 choices in total).
***
In the case they choose the independent candidates: briefly explain why you chose this candidate?
***
A referendum on the status of the island will soon take place, when is it going to be held? (Please, state the month and the year).
42 Please, state the year in which governor Luis G. Fortuño won the governor
elections in Puerto Rico:
***
According to the students, why the different campuses of the University of Puerto Rico should join an indefinite strike?
*** ONE OF THREE RANDOMIZED ARTICLES
***
Here are the different status options you will encounter in June’s referendum. If this referendum was to take place tomorrow, what would you choose?
Statehood, free association/independence, current status, I would not vote. ***
In a hypothetical second round, which option would you choose? Free association, independence.
***
Please, state your opinion on the following sentences (Likert scale: totally agree to totally disagree):
• I believe the best option for the island is to maintain its current status. • Puerto Rico should become the U.S.’s 51st
state.
• The island should be a completely independent country. • I agree with holding this referendum.
• No more referendums should be held in the island. • This referendum will solve the status issue.
• The status of the island will not be solved in the coming years. ***
If governor elections were to be held tomorrow, would you vote for the same candidate you did in November?
Yes, no, I do not know.
***
If the answer yes and no: please, explain why you would change/maintain your vote:
*** What was the article you read about?
43 Difficulties set by the U.S. regarding the referendum.
Support of the U.S. in holding the referendum. None of the above.
*** How old are you today?
*** What is your gender?
Female, male, other.
***
What is the highest level of education you have joined (no need to have finished it)? Middle school, high school, bachelors, masters, PhD, other.
*** In which social class would you place yourself? Low, mid-low, middle, mid-high, high.
*** Where do you currently live?
Puerto Rico, U.S., other.
***
Have you lived in any of the U.S.’s 50 states for at least 3 months? Yes, no.
*** IMPORTANT:
Before we finish, we would like to let you know that the article you just read was created and manipulated by the researchers in order to study how media coverage of Donald Trump has influenced Puerto Ricans’ attitudes towards the status of the island within -or outside- the United States.
We appreciate that you do NOT talk about the aim and content of this experiment with prospective participants to avoid biased answers.
Once again, thank you for your cooperation! ***
44 Appendix 2A: Original Stimuli
