江靜之

Chin-Chih Chiang

內容分析法、防疫記者會、記者提問、強勢性提問、對抗性

content analysis, COVID-19 news conference, journalistic questions, aggressive questioning, adversarialness

本研究針對2020年1月中旬至2022年7月中旬之COVID-19 CECC例行記者會，透過「強勢型提問」分析系統，使用內容分析法，以記者會隔天的《中國時報》、《自由時報》及《聯合報》三大報頭版相關新聞作為取樣標準，共蒐集214場，含7,921個問題，研究記者在跨越不同疫情時期之提問強勢程度，以及其與問題功能、內容主題、引述來源及同業代問之關係，藉此一探記者如何發揮新聞的看門狗功能。首先，本研究發現記者提問的強勢性與疫情嚴重程度成正比，即記者提問強勢性於「三級警戒」最高，「二級警戒」次之，即使到了「放寬防疫」時期，提問的強勢程度依然高於整體平均。其次，記者提問多聚焦於事實問題，且其強勢性低於請受訪者評論或表態之問題。其三，「防疫政策」、「防疫舉措」及「染疫個案」是每日記者會初期及二、三級警戒的共同提問重點，每週記者會時期特別關注「疫情政治」，而放寬防疫時期則著重詢問「防疫舉措」的改變。最後，約兩成的問題有引述來源，且其強勢程度高於無引述來源的問題。同業代問問題在三級警戒時期的數量最多，且其整體強勢程度同樣高於非同業代問之問題。

Press conferences have been regarded by governments as a key mechanism for communicating policies to the public ever since the 20th century. They also function as an institutionalized forum through which journalists hold political actors accountable by questioning them about their actions and policies. During the COVID-19 pandemic in particular, official press conferences not only provided an effective channel for authorities to communicate directly with the public, but also enabled journalists to ask government employees questions on behalf of the public, thereby taking on a watchdog role of the news media. These press conferences provide a valuable opportunity for researchers to examine the relationship between the government and the press and to analyze how journalists shift between competing professional norms of objectivity and adversarialness.

To explore how journalists in Taiwan fulfilled their watchdog role during daily Central Epidemic Command Center (CECC) press conferences across different phases of the pandemic, this research adopts the aggressive questioning framework of Clayman and Heritage (2009). Building on a literature review, it further modifies the framework in light of Mandarin Chinese’s linguistic characteristics and the institutional context of CECC press conferences in Taiwan. Employing content analysis to examine journalists’ use of aggressive questioning strategies and their levels of intensity, this study also incorporates variables including source attribution in journalists’ questions, delegated questions, question functions, and topics. Based on these dimensions, six research questions are proposed as follows.

RQ1: Are there significant differences among the dimensions of journalists’ aggressive questioning?
RQ2: Are there significant differences in journalists’ aggressive questioning and its dimensions across different phases of the pandemic?
RQ3: What is the distribution of delegated questions across different phases of the pandemic? Are there significant differences in journalists’ aggressive questioning between delegated and non-delegated questions?
RQ4: Are there significant differences in journalists’ aggressive questioning and its dimensions across different question functions?
RQ5: Are there significant differences in journalists’ aggressive questioning and its dimensions between questions with and without source attribution? Is there a significant relationship between the level of source attribution and journalists’ aggressive questioning?
RQ6: What is the distribution of question topics across different phases of the pandemic?

This study collects data from a total of 214 daily CECC press conferences held between January 22, 2020 and July 17, 2022 that were selected by Taiwan’s three major newspapers as sources for the following day’s front-page news articles. This enables the analysis to target journalists’ questions exhibiting significant newsworthiness.
For RQ1, correlation analysis reveals that initiative significantly and positively correlates with assertiveness and adversarialness, and that directness significantly and positively correlates with assertiveness. Assertiveness positively correlates with adversarialness, whereas directness negatively correlates with both initiative and adversarialness.

For RQ2, one-way ANOVA results indicate significant differences in aggressive questioning across pandemic phases. The level of aggressive questioning during the “initial daily news conference” period and the “weekly press conference” period is below average, but above average during the “Level 3 alert”, “Level 2 alert”, and “relaxed restrictions” periods. Further analysis of the four dimensions of aggressive questioning shows that initiative is below the overall mean during the “Level 2 alert” period, but slightly above the overall mean during the “Level 3 alert” period. Both directness and assertiveness are above the overall mean during the “Level 2 alert” and “Level 3 alert” periods, however, directness is below the overall mean during the “weekly press conference” period, whereas assertiveness shows the opposite pattern. During the “relaxed restrictions” period, both those two dimensions decline more sharply than aggressive questioning. Finally, adversarialness is above the overall mean during the “weekly press conference” period and below the overall mean during the “Level 2 alert” and “Level 3 alert” periods. In contrast to the other three dimensions, it increases sharply during the “relaxed restrictions” period and is above the overall mean.

For RQ3, among the 540 delegated questions analyzed, only two occurred during the “initial daily news conference” period and none were observed during the “weekly press conference” period. However, the proportion of delegated questions increased to 23.5% during the “Level 2 alert” period and 54.3% during the “Level 3 alert” period. Even during the “relaxed restrictions” period, the proportion remained at 21.9% of delegated questions. The level of aggressive questioning differs significantly between delegated and non-delegated questions, with the former exhibiting above-average levels and the latter showing below-average levels.

For RQ4, significant differences appear in aggressive questioning across question functions. “Requests for attitude statements” exhibit the highest level of aggressive questioning, followed by “requests for evaluation or comments”, whereas the other three functions are below the overall mean. The four dimensions of aggressive questioning differ significantly across question functions. “Requests for attitude statements” are above the overall mean in all four dimensions. “Requests for evaluation or comments” are below the overall mean only in adversarialness. Both “requests for new or specific information” and “requests for confirmation or clarification” are below the overall mean across all four dimensions. By contrast, “requests for suggestions” are below the overall mean in directness and adversarialness, but above the overall mean in initiative and assertiveness.

For RQ5, significant differences are in aggressive questioning between questions with and without source attribution. Questions with source attribution exhibit above-average levels, whereas those without source attribution are slightly below the overall mean. Across the four dimensions, initiative, assertiveness, and adversarialness are significantly higher in questions with source attribution than in those without it, whereas directness is higher in questions without source attribution than in those with it. The number of source attributions significantly and positively correlate with both aggressive questioning and three of its dimensions (initiative, assertiveness, and adversarialness). By contrast, that number significantly and negatively correlates with directness, indicating that questions with fewer source attributions tend to be more direct, as reflected in the use of direct questions and the absence of politeness markers.

For RQ6, the question topics consist mainly of “pandemic control measures”, “pandemic policies”, “domestic COVID-19 situation”, “COVID-19 cases”, and “pandemic politics”. “Pandemic policies”, including policy formulation, substance, and change, constitute one of the main focuses of journalists’ questions during the “initial daily news conference” period as well as the “Level 2 alert” and “Level 3 alert” periods. “Pandemic politics” appears more frequently during the “weekly press conference” period.