Televised Political interviews: A Paradigm for Analysis

One surprising feature of the public opinion literature in Asia is the scarcity of references on the nature and effects of political interviews, especially televised political interviews. In the broad research literature on political communication, one can find a growing number of articles on the general function of the media and its effect on the construction and shaping of public opinion in such countries as Thailand, Malaysia, and Korea (Lewis, 2006; Sern & Zanuddin, 2014; Woo-Young, 2005). However, there is very little literature on the explicit role of political interviews in shaping public opinion, or on how political affairs talks are organized in this type of media discourse.

In Asia in particular, along with the disappearance of traditional values and with an increased exposure to political information through the media, particularly television (e.g., Thomas, 2005; Zhang, 2011), broadcast interview programs provide a valuable channel for citizens to make sense of politics. Interview programs allow citizens to follow public policy developments, to recognize emerging candidates and distinguish between them and their orientations, and to evaluate the various political alternatives.

Analyzing broadcast political interviews is important for several reasons. First it enables us to understand how talk in political interviews is achieved and what conventions govern the interaction between journalists (interviewers) and public official s or experts (interviewees). Specifically, it provide us with the details to such questions as: What are the nature of the roles, functions, responsibilities and rights of the interviewers and the interviewees; what shapes their discourse’s basic rules and structure; What is the nature of the questions, their structure and tone; How biased are the interviewers; To what extent are the interviewees’ replies evasive; and How do interviewees try to construct a certain image of themselves and others through their replies. Second, analyzing political interviews facilitates the recognition on how talk in political interviews differs from ‘ordinary’ talk, and in particular how talk in political interviews is constrained by its particular setting, including culture and political culture, and social issues and problems, including hierarchy, unequal power, race, and gender in a particular society. It also enables us to further look at the beliefs that underpin the interview and examine issues such as power, social truth and the motives of the participants to convey particular ideologies.

To the extent that cross-national, cross-cultural information is needed to establish a general theory of political communication and on the role and function played by the news media in politics, data from non-Western Asian countries is not only desired but also an essential in this regard.

In this paper I aim to call attention to the importance of detailing televised political interviews in Asia, and to propose, in a highly selected manner, a paradigm that will facilitate such a study. The paradigm is based on a series of studies I conducted in Japan during 2000, 2010~2011, and 2012~13 (Feldman, 2004, p. 76-110, 2015; Feldman et al., n.d., 2015). This paradigm introduces what I see as the core aspects and questions that should be addressed when detailing political interviews in a particular society. Use of the coding proposed by the paradigm will enable researchers to gather information on the nature of the interaction between interviewers and interviewees, their communicative style when addressing or replying to questions, and the strategies employed by the participants to pursue their goals.

The paper begins with a short discussion on the nature of political interviews.

Political Interviews

Broadcast political interviews aim to test public officials and subject-matter experts on policy affairs and questions of concern to the general public. As citizens have a right to be fully informed about political issues and the political agenda, public officials, including politicians, are consequently required to explain their actions and intentions. With this goal in mind political interviews enable the interviewers–journalists, social critics, or scholars and researchers from different fields acting on behalf of the public–to ask questions and challenge answers. Through questioning, interviewers may disagree with, argue, criticize, or otherwise confront interviewees in their attempt to unmask the truth about policies and political problems in which public officials or their groups are involved.

As such televised political interviews are designed to produce face-to-face confrontational and challenging interactions between journalists and officials. They have their own distinctive features, and a defined set of rules and norms structured by the functions of the interviewer(s) and the interviewee(s). First, these interviews are staged performances that take place with the participation of journalist(s) and political officer(s) or expert(s) and in which the ultimate addressee is absent from the actual event. The interview is enacted for the benefit of an “overhearing audience” (Heritage, 1985) whose expectations shape what is being said and how. Both the interviewer and the interviewees (politicians or experts) will have the general viewers in mind. The interviewer will simultaneously consider both the consumers of their talk show and colleagues in their organization; success or failure in their performance can determine their future career and their standing in the eyes of their peers and the corporation. For politicians, interview programs provide their best tool to speak “directly” to hundreds of thousands of people, an opportunity to advance their ideas and thoughts to the electorate; an occasion for enhancing positive images of themselves and their political groups; and a ground for attacking their political opponents and challengers.

Second, the “turn-taking system” noticeably defines the conflicting functions of interviewer(s) and interviewee(s). Both parties attempt to generate discourse for the “overhearing audience” in a two-way process. Thus, the interviewers are responsible for determining the topic for discussion, monitoring the duration of the discourse, and adhering to specific ritualistic patterns including introducing interviewees and concluding the interview session. At the same time, interviewers also pose questions and challenge interviewees to specify and explain their positions on a variety of issues. They are expected to do so by keeping a balance between adversarialism and objectivity, maintaining a stance of neutrality by not favoring specific politicians or political groups. The interviewees’ task is to reply to these questions to best represent both themselves as individuals and their political groups or institutions (e.g., Clayman & Heritage, 2002). Challenging these roles allocations would violate the primary rules that structure the political interview. However, deviations from these normative expectations may not necessarily be acknowledged, sanctioned, or repaired. Given the advantages that the interview offers politicians to speak to a large audience and promote their own and their groups’ agenda, they may strive to exert control over the interview. Thus, they may break the talking procedure, intentionally change the subject before or after giving a response, disregard the questions they are asked by repeating statements (irrespective of whether they have any relation to the interviewer’s questions), and shift the agenda and topic selection, a phenomenon which is termed “agenda shifting procedures” (e.g., Heritage & Greatbatch, 1991).

Another distinctive feature of political interviews is interviewees’ vagueness, evasiveness, or equivocal communication style as they hedge from providing direct answers to questions they are asked. Various strategies used by politicians to avoid giving direct answers were identified, and the low proportion of questions which receive their adequate replies have been reported in televised interviews broadcast in the UK, Taiwan, and Japan (Feldman et al., 2015, p. 67).

With these characteristics of the political interview in mind, I now turn to the proposed paradigm.

The Paradigm to Examine Political Interviews

The paradigm to examine televised political interviews utilizes the Conversation Analytic approach for analyzing questions and replies in political interviews. Conversation Analysis is a way of analyzing talk as a social phenomenon; it is used to analyze talk-in-interaction which views language mainly as a form of social interaction. Conversation Analysis seeks to determine how social practices take place within language and between participants and hence, how social order is formed (Hutchby & Wooffitt, 1998). It examines how participants mutually understand each other during turns, assuming that “ordinary talk is a highly organized, ordered phenomenon” (Hutchby & Wooffitt, 1998, p. 13). For the purpose of analysis, Conversation Analysis studies naturally-occurring data which has been recorded rather than data that is generated in laboratories. This recorded data is then transcribed and examined without any pre-conceived idea about what one wants to find, or what is in the data. Researchers let the data speak for itself monitoring what is done by the participants through their conversation.

Thus methodologically speaking, in the case of broadcast political interviews, questions and replies of selected interview sessions are recorded, transcribed, and examined without any detailed assumptions on their form, content, tone, and interrelationship, as the analysis focuses on the utterances of the interviewer(s) and interviewee(s) as they verbally interact with each other.

The paradigm to examine televised political interviews consists of four distinct elements at the core of the interaction between interviewers and interviewees. Each of these elements includes specific aspects and questions that should be included in a coding sheet designed to decode the information by trained coders.

The four elements are:

  1. (1) The Interviewers’ questions;

  2. (2) The interviewees’ replies;

  3. (3) The questions/replies sequences; and

  4. (4) The setting or background to the interviews’ sessions.

Identifying Questions: At the base of this paradigm is the need to identify questions and distinguish them from other utterances:

“Questions” are regarded as utterances made by interviewers in order to elicit information or opinions from interviewees. Questions may or may not utilize interrogative syntax.

In televised political interviews, some questions, usually asked at the beginning of an interview, are merely “small talk” to make interviewees “comfortable” upfront. These include “softball” questions, i.e., non-challenging questions that function as greetings before the actual interview begins, and often invite interviewees to brag about themselves or their work. Such questions should be excluded from analysis.

Hereupon are the four segments that should be coded:

1. The Questions

The goal of this section is to detail the attributes (e.g., the type, style, mode) of questions asked during televised political interviews.

There are six ways to classify or identify these questions:

1. Classifying Questions Based on Syntactic Expression

Interview questions are classified into two large groups according to their syntactic expression. The first consists of prefaced questions and the second of non -prefaced questions.

Prefaced questions are in most cases prefaced by a main clause in which the main prepositional content of the question appears in indirect form in a subordinate clause (in non-prefaced questions there is no such preceding main clause). This group include s questions such as “What do you think?” “What do you feel?” “Are you saying …?” “Are you suggesting …?” “Will you explain …?” “Could you say what …?” And, “Can I ask you …?”

Non-prefaced questions may be further subdivided according to whether or not they take interrogative syntax:

(1) Interrogative syntax appears in three basic question forms:

(a) Yes/no questions that request a ‘yes’ or ‘no’ response in relation to a certain matter. Responding with “Yes,” “No,” or any other clearly affirmative or negative reply (using words or expressions such as “certainly,” “of course,” “not at all”) is seen as constituting a reply.

(b) Interrogative-word questions, or the WH-questions, i.e. questions that start with what, why, who, when, where, or how; and,

(c) Disjunctive or alternative questions that present interviewees a choice between two or more alternatives. If the interviewee chooses one of the alternatives, then this can be seen as constituting a reply. It is also possible to present an additional alternative, which can be regarded as a reply. If the interviewee does not choose between the alternatives offered by the interviewer, nor offers another alternative, then response is regarded as a non-reply.

(2) Non-interrogative syntax questions include (d) declaratives, imperatives, or questions lacking a finite verb (i.e., moodless questions). They are identical in form to declarative statements in writing, but are generally spoken with a final rising, questioning intonation. Some declarative questions may not be accompanied by rising intonation, but the expression still clearly functions to elicit information.

Coders decode each question first as either (1) prefaced question, or (2) non - prefaced question. If (2) non-prefaced questions is selected, the coders further decode the sub-category of the question as either (a) yes/no questions; (b) interrogative-word questions, or the WH-questions; (c) disjunctive or alternative questions; or (d) declaratives, imperatives, or moodless question.

2. Distinguishing the Syntactic Structure of Questions

During interviews, interviewers assess the responses given by interviewees to determine whether they have answered the question or hedged it. If interviewers consider the response to be an answer, they can either expand on the same topic or shift to a new topic, called “topic extension” and “topical shift,” respectively. An interviewer who views the response as a non-answer can follow up with a “reformulation” or a “challenge.”

(1) Topic extension: the interviewer raises a minor implication of a prior statement in order to encourage an interviewee to reconfirm and/or expand on prior remarks.

(2) Topical shift: when shifting the topic, the interviewer doesn’t relate back to the preceding answer but rather brings up a new aspect of the general topic under discussion.

(3) Reformulation: the interviewer “reformulates” when he or she accurately restates an interviewee’s declared position, with the implication that the preceding answer was not adequate and needs clarification or expansion.

(4) Challenge: the interviewers directly challenges an answer by making explicit the possible implications or presumptions of the answer, with the goal of testing some aspect of the interviewee’s intentions, actions, or attitudes.

Coders decode here either (1) Topic extension; (2) Topical shift; (3) Reformulation; or, (4) Challenge.

3. Examining the Pragmatic Means of Weight Reduction

There are two ways in which interviewers could challenge an interviewee’s point of view without appearing to violate the required perception of neutrality: quotation of critics and questions accompanied by accounts.

(1) Quotation of critics: When interviewers base their questions on actual or hypothetical critiques by an interviewee’s opponents (political or other), or cite other actual or potential challengers (e.g., news media). Drawing on the words of others allows interviewers to openly disagree without appearing to do so personally.

(2) Questions accompanied by accounts: the interviewer provides his or her own logic or opinions as the basis for challenging questions.

Coders decode either (1) Quotation of critics; or (2) Questions accompanied by accounts.

4. Identifying Grammatically Complete and Incomplete Questions

Questions can be differentiated between

  1. (1) Grammatically complete questions, and

  2. (2) Grammatically incomplete questions.

Questions categorized as grammatically complete follow the syntactic rule that a question particle (such as ka in Japanese) must be attached to the end of the sentence accompanied by rising intonation. Grammatically incomplete questions, considered less formal and less direct utterances, lack the question particle but project turn -yielding despite their incomplete structure.

Coders decode either (1) Grammatically Complete Questions; or (2) Grammatically Incomplete Questions.

5. Identifying Questions that Seek Personal Opinions versus Prevalent Views within a Group

Here we distinguish between questions that explicitly seek from interviewees

  1. (1) Personal, private information such as their own opinions, thoughts, beliefs, and feelings; as opposed to

  2. (2) Information, ideas, or views of a group that the interviewee belongs to, e.g., political parties and government ministries.

In other words, there is a differentiation between questions posed to interviewees according to the required perspective–personal perspective or that of a social or political group. The general assumption is that politicians would be asked to share the viewpoints of their groups, while non-politicians would be asked for their private, personal views; but this is not always the case.

Coders mark either (1) personal, private information; or (2) information, ideas, or views of a group that the interviewee belongs to.

6. Assessing the Questions’ Level of Threat

Interviewers’ questions vary in terms of their toughness and neutrality. Questions may be formulated in such a way that interviewees constantly run the risk of making face - damaging responses, i.e., responses which make themselves and/or their political allies look bad, and/or constrain their future freedom of action (Bull, 2008). Questions are distinguished between those that entail “no necessary threat” (least face-threatening) and those with a “high level of toughness.” Politicians tend to find the latter problematic, especially when they are potentially face-damaging to themselves, their political party, and to significant others (Feldman, n.d.). A high proportion of the latter questions would therefore constitute a tough style of interviewing.

The coding sheet in this part includes two related questions:

(1) A six-point scale, according to which each question is rated in terms of toughness (from (1) “no threat at all” through (6) “very tough” (“neutral” is not included in these six possible responses in order to force coders to make a selection on the relative degree of equivocation); and,

If the answer was not (1), then coders respond to another question:

(2) Indicate at whom the threat was aimed: (a) the interviewee’s face; (b) the party’s face; (c) a group to which the interviewee belongs; (d) the government (the administration, the cabinet); (e) others (specify).

2. The Replies

The aim of this section is to examine the manner that interviewees handle the questions posed at them.

Responses are analyzed in a series of four questions. The questions reflect a modification of the four Bavelas et al. (1990) dimensions of sender, receiver, content, and context in the Theory of Equivocation. Equivocation is regarded as a form of indirect communication characterized as ambiguous, contradictory, and tangential, which may also be incongruent, obscure, or even evasive (Bavelas et al., 1990, p. 28).

Each of the four dimensions is measured on a six-point Likert-type scale.

1. Sender

Assessed by the question “To what extent is the response the speaker’s own opinion (intention, observation, ideas)?”

The scale consisted of six options, ranging from (1) “It is obviously his/her personal opinion/ideas, not someone else’s,” to (6) “It is obviously someone else’s opinion/ideas.”

If the answer was not (1), then coders respond to another question:

(2) “Whose opinion does the speaker seem to express?”

Coders select from the following list: (a) Political parties; (b) The general public/public opinion; (c) The Government (the administration); (d) Mass media; (e) Economic and industrial circles (public and private sectors) and the third sections (organizations that are not for-profit and non-governmental); (f ) Refers to facts/historical flow of events/common sense.

2. Receiver

In broadcast news interviews, there are multiple receivers. Thus, when an interviewee responds to a question, it is not always clear whether the intended receiver is the interviewer or some other party, such as the general public, a particular segment of the public, or another politician or group of politicians. These other parties can be referred to as the “overhearing audience.”

To address this issue the following questions are posed:

(1) “To what extent is the message addressed to the person(s) who asked the question?” (i.e. the interviewer(s)). Possible recipients are assessed on a six -point scale, ranging from (1) “Obviously addressed to the interviewer(s)” to (6) “Addressed to other people.”

If the answer is not (1), coders respond to another question:

(2) “To whom does the message seem to be addressed?”

Coders indicate whom they think is the intended target of the message by selecting from the following list: (a) Another interviewee politician(s) that appear(s) on the show in the studio; (b) Decision-makers (at the center of politics or Government officials); (c) Voters in general (d) Specific prefecture/prefecture residents; (e) The public in general (f ) Economic and industrial circles (public and private sectors) and the third sections (not for-profit and non-governmental organizations); (g) Other (specify).

3. Content

Assessed by the question “How clear is the message in terms of what is being said?”

The six options evaluate the various degrees of equivocation, ranging from (1) “Straightforward, easy to understand, only one interpretation is possible,” to (6) “Totally vague, impossible to understand, no meaning at all.”

If the answer is not (1), then coders should specify what made the reply unclear or difficult to comprehend by selecting from at least one of the following four options:

(a) It consists of long /complex sentences that are difficult to follow; (b) There is run-around, double talk (circuitous talk); (c) It includes difficult terms, or professional jargon; (d) It consists of multiple arguments (sometimes inconsistent).

4. Context

Assessed by the question “To what extent is this a direct answer to the question?”

The six options range from: (1) “This is a direct answer to the question asked,” to (6) “Totally unrelated to the question.”

If the answer is not (1), i.e., the interviewee fails to reply to the question, then the response is to be coded in terms of the following 12 Categories of Non-Reply (based on Bull & Mayer, 1993):

(a) Intentionally ignores the question (typically while launching into another discussion); (b) Acknowledges the question without answering it; (c) Questions the question (reflecting a question back to the interviewer(s); (d) Attacks the question; (e) Attacks the interviewer; (f ) Declines to answer a question; (g) Makes a political point; (h) Incomplete answer; (i) Repeats answer to previous question; (j) States or implies that the question has already been answered; (k) Apologizes; (l) Other (specify).

These 12 Categories are not mutually exclusive as more than one category can be used at a time.

A message can be equivocal on any of the four above dimensions. So the content may be perfectly clear (unequivocal in terms of content), but not a direct answer to the question (equivocal on the context dimension).

3. Subjects of Enquiry/Response

The aim of this section is to identify the key topics at the center of interviewer enquiry and interviewee response, and to distinguish the issues at stake in each question - response:

Two questions are posed:

1. “What is the main content of the question about?”

2. “What is the main content of the response about?”

Each of these two questions is sub-divided (and coded) in respect to six criteria detailed below, all mutually exclusive, on the content of the questions and answers:

(1) Knowledge, or lack of knowledge, of a certain topic or a fact.

(2) Human affairs/Significant others (i.e., others’ performance at work, impressions on their activities, evaluation of their ability, characteristics, personality, attitudes, and thoughts, and human relationships).

(3) Political and social institutions (e.g., impressions, opinion, and judgments on the activities, attitudes, views, thoughts, and ideas within political parties, party factions, and the media).

(4) Political process (i.e., involving procedures of decision -making and course of action in the government, the bureaucracy, and within or between political parties).

(5) Political commitment (promises regarding courses of action, pledges, and public obligation).

(6) Issues (opinions, stances, and views on policy issues, on social, economic, political and other problems, and topics on the public agenda).

If category “(6) Issues” is selected, coders answer an open-ended question:

(7) “What is the issue (problem or topic) at the core of the question/response sequences?”

Coders identify the issues at the focus of the discussion between the interviewers and the interviewees and list them one by one. It is later clustered in related categories.

4. The Setting of the Interview

The aim of this section is to detail selected factors of the setting or the background of the interview that may influence the nature of the interaction between interviewers and interviewees, and the content of the questions and the replies. These factors influence thus the scope of the information disseminated to the public. They include, but are not limited to, the following aspects:

(1) The nature of the television program in which the interview takes place:

(a) Nationwide program or local level program

(b) Regular news program;

  1. Regular daily/weekly/monthly interview program; or

  2. Special interview programs (e.g., during election campaign)

(c) What is the viewing rate (How popular is the program?)

(2) The interview’s format:

(a) One-on-one interview (a single interviewer vs. a single interviewee)

(b) One interviewer vs. multiple interviewees

(c) Two interviewers vs. a single interviewee

(d) Two interviewers vs. multiple interviewees

(e) In small groups

(f ) Questions asked by the interviewers and other “commentators” who are invited to present questions and participate in the discussion

(g) Questions asked by the interviewers and other interviewees who are invited to present questions and participate in the discussion.

(h) Other (specify)

(3) The broadcast time and the length of the interview:

(a) Morning/afternoon/evening/night

(b) The interview session duration: Minutes & Seconds

(4) Who asks the interview’s questions? (What are their social and political orientations?)

(a) Journalists

(b) Scholars

(c) Experts (in such areas as public policy, social affairs, and economics)

(d) Other (specify)

(5) Who are the interviewees? (What are their ranks, in government, for example?)

(a) National level politicians (e.g., presidents, prime ministers, ministers, leaders of parties)

(b) Local level politicians (e.g., prefectural governors, city mayors)

(c) Government officials (e.g., administrative vice-ministers, heads of sections)

(d) Military, Army officers and personnel

(e) Decision-makers from various social and economic sectors of society

(f) Nonpoliticians (e.g., university professors, experts in economy, social critics).

(e) Other (specify)

(6) The Political Culture and Media Organization

(a) Does the interview take place in a democratic, free society?

(b) How is the media system organized? Who owns/controls the media?

Conclusions: Gathering and Analyzing the Data

The proposed paradigm is only a preliminary one; other aspects and questions could be included to enhance its scope. A comprehensive paradigm will be developed after additional research with case studies in socio-culturally and politically diversified Asian countries has been conducted.

Collecting and analyzing data using the suggested paradigm can be seen as effortful and time consuming. Indeed, there are many stages to complete before analysis begins: selecting an interview on an appropriate televised program, recording the interview (using perhaps a DVD recorder), making a verbatim transcript, verifying the transcript against the recorded interview, training coders on how to decode the data, and checking inter-coder reliability. However, this process should not discourage ambitious researchers from such fields as social psychology, communication, and political science, as the sense of exploration of this unknown feature in political communication should be gratifying. Let a thousand flowers bloom.

Biographical Note

Ofer Feldman (PhD, University of Tokyo, Japan) is a professor of political psychology and behavior at the Faculty of Policy Studies, Doshisha University, Kyoto, Japan. He is the author of more than 90 journal articles and book chapters, and more than 100 encyclopedia items, in the fields of political psychology/behavior, communication studies, and Japanese politics. He is the sole author of seven books and monographs, including Talking Politics in Japan Today (2004) and Seiji shinrigaku ([Political Psychology], in Japanese, 2006), and sole and coeditor of six books, including Beyond Public Speech and Symbols (2000) and Political Leadership for the New Century (2002), and Politische Psychologie: Handbuchfur Studium und Wissenschaft ([Political Psychology: Handbook for Study and Science], in German, 2015).

He can be reached at the Faculty of Policy Studies, Doshisha University, Kamigyo-ku, Shinmachidori, Imadegawa-noboru, Kyoto, 602-0047, Japan, or by e-mail:


All correspondence concerning this article should be addressed to Ofer Feldman, Faculty of Policy Studies, Doshisha University, Kamigyo-ku, Shinmachi-dori, Imadegawa-noboru, Kyoto, 602-0047, Japan, or by e-mail: