The spiral of silence: examining how cultural predispositions, news attention, and opinion congruency relate to opinion expression

Abstract

This study examines new explanatory predictors behind the spiral of silence theory, using the issue of legalization of same-sex marriage in Singapore as the context of study. Our results show that fear of isolation and saving face were negatively associated with individuals' willingness to express their opinion on the issue, whereas news attention and issue salience were positively associated. Also, fear of isolation was negatively associated with individuals' willingness to offer a rationale for their opinion, whereas news attention and issue salience were positively associated. Power distance had no effects on outspokenness. Notably, news attention moderated the influence of fear of isolation and saving face on public outspokenness.

Keywords:

• spiral of silence theory
• public opinion
• Singapore
• saving face
• fear of isolation

Acknowledgement

This work was supported by the Research Cluster Grant [WBS M5806003] and the Start-Up Grant [M58060008] from the Wee Kim Wee School of Communication and Information at Nanyang Technological University.

Notes

1. We compared our sample's demographic characteristics with the 2010 population census data released by the Singapore Departments of Statistics (2011). Singapore's population demographics are as follows: Gender – 50.6% female; ethnic groups – 74.1% Chinese, 13.4% Malay, 9.2% Indian, 3.3% others; median age – 37.4 years; median education – lower secondary qualification; median household income – $5000. Our sample was similar to the population in terms of gender (51.5% female), age (Median = 40.0 years), and ethnicity (75.6% Chinese, 10.6% Malay, 10.9% Indian, 2.90% others). However, respondents were slightly less affluent (median monthly household income between $3000 and under $4000), and slightly more educated (median = upper secondary qualification). Despite the slight differences, our sample statistics were generally a good reflection of the population parameters. We treated these demographics as control variables in our regression analyses, which should have compensated for the slight differences in our sample characteristics and population parameters. Moreover, we were primarily interested in assessing the relationships among the independent variables of interest in the spiral of silence theory, which should have made up for the slight differences in demographics.

2. The youngest male/oldest female (YMOF) method (Keeter & Fisher, 1997), a non-probability selection technique, was used in our study. We obtained a within-household sample by asking to speak with the ‘youngest male, 18 years or older, who is now at home’. If no eligible male was present, then the interviewers would ask to speak to the ‘oldest female, 18 years or older, who is now at home’. Although the YMOF technique has been effective in increasing cooperation rates and decreasing costs, it may not closely reflect the population parameter after taking into account correlational distributions of demographics, as compared with other techniques (Gaziano, 2005).

Future studies may consider using other within-household sampling techniques, such as the Kish method (Kish, 1949) or the last birthday method (Salmon & Nichols, 1983). For the Kish method, the interviewer lists all eligible household members (i.e. recording the names of all males in the household and their associations with others in descending rank of age, then likewise records all females), and then calculate the probability of selection (Gaziano, 2005). For example, interviews would ask:

How many adults live in this household and can be reached here? [Exclude those generally away, unable to be interviewed.] How many of those are [random fill: male–female]? [Generate random number from 1 to Num_adult. Identify selected household member according to method (age + gender): Example: ‘the youngest male’]. (Denk & Hall, 2000, p. 8)

The Kish method is commonly used as it is a probability method that yields representative sample (Gaziano, 2005). Although Kish questions are potentially intrusive, highly competent and capable interviewers may offset this limitation (Groves & Kahn, 1979; Lavrakas, 1993). The quasi-probability last birthday technique is a method in which the interviewer asks to speak to the adult household member who has the most recent/last birthday (Salmon & Nicholas, 1983). An example of the last birthday method would be: ‘According to our selection procedure, I need to speak with the adult living in your household who is 18 or older and has had the most recent birthday’ (Keeter & Fisher, 1997, p. 2). The last birthday method tends to yield higher cooperation rates and be less intrusive than the Kish method, and tends to generate more representative samples than the YMOF technique (Gaziano, 2005).

3. We would like you to imagine that you are at a face-to-face gathering with a group of people you do not know very well. Suppose the topic of legalization of same-sex marriages in Singapore came up and you discover that most of the people disagree with you. Keeping in mind that most of the people in the group do not support your point of view, we would like to ask you what you might do if you were in a situation like this. Using a 1–7 scale where 1 means very unlikely, 4 means neither likely nor unlikely, and 7 means very likely, please tell me how likely will you be to do the following: (a) express your own opinion and (b) offer a rationale for your own opinion.

4. To measure power distance among Singaporeans, we adapted items from the CVSCALE (Yoo, Donthu, & Lenartowicz, 2011), which were validated at the individual-level (Donthu & Yoo, 1998). The scale also obtained high internal reliability when administered in Malaysia, a neighboring country whose culture is quite similar to Singapore (Kueh & Voon, 2007).

5. Each interaction term was developed by multiplying the centered scores of the individual components to limit multicollinearity problems with the standalone terms (Cohen, 2003).