I will present a detailed description and critique of the research process in which I conducted a frame analysis to compare ten news stories published in New York Times.

The first step to conduct a textual analysis is to determine the topic that interests me and what analytical method I will employ to examine the materials. Looking back on the numerous events happening in the passed few weeks, I am interested in how the mass media in U.S. cope with the public crisis in such different social aspects as nature, economy, politics and national security, say, the devastation of a series of hurricanes which plow through the Caribbean and southern Atlantic. The roles of mass media in the prevention and warning of crisis (pre-disaster), confrontment with large-scale disasters (during the disaster) and measures taken to deal with problems arising from catastrophe (after-disaster) can be derived from their continuous news reports on the topic of the disasters. The attitudes of the specific social institutions, such as the federal and state government, people who directly or indirectly suffer from the disasters, and people being away from the disaster but paying intense attention to it can be reflected in the news stories. In the consideration of the limitation on the numbers of news reports to be analyzed in this exercise, I selected ten news stories which reported on the hurricane Ivan and its effects in the southeast area as my research subjects. All the articles were originally published on the section of National Desk in New York Times and retrieved from Lexis-Nexis Academic Universe database by using “Hurricane Ivan” as keywords in searching. The time range was set to “the past month” and articles were picked from the results arranged in the chronological order. Ten reports about Hurricane Ivan from Sep 11th to 20th appearing on National Desk section were finally chosen as sample.

Frame analysis is employed as the method to examine how the journalists constructed the overview of natural crisis during the 10 days within which the hurricane was predicted to make landfall in the US and cause massive losses and what perception of the whole event readers might obtain from the series of reports. I think the method is appropriate to carry out “to describe the evolution and trend of the news discourse concerning an issue.” (Pan & Kosicki, 1993, p. 69) However, the caveat of textual materials selection is obvious in that to achieve the goal of examining the mass media as an entire institution, a limited number of news stories from one single news source is far from enough regardless of how representative and authoritative the source is.

Bearing the pitfall in mind, I started off by fleetly skimming over all the articles in their original time order to obtain a general impression about “what the stories talk about exactly and how they talk about”. The scanning was proved to be effective and efficient in that it set a relatively high perspective from which I maintained a universal impression about “what’s going on” instead of getting myself overwhelmed by the details. During the process, such notable elements aroused my attention as the repetition of certain themes, the regular reference to certain organizations, and the change of tone over the time…etc.

To better analyze the text, I divided the reports into three categories which are characterized by their respective emphases on the specific aspects of the event. The categorization was based on the stages on which the hurricane Ivan was near, on, and away from the southeast. The “prediction—confrontation-- sequel” model can best generalize the changes of framing in the ten days. The differences on the timetable determine the diversity of reporting keystones. In the continuous reports regarding the same news event, the five Ws and one H change along with the development of the event. Nevertheless, in this case, the discursive elements were endowed different properties as the hurricane moved around. The grouping of the reports avoids the discrete analysis of each report and puts them into a shared background for the convenience of conducting comparisons between and within groups. I assume this categorization be conducted in the future similar research.

In the broad sense, most articles follow the dominant inverted pyramid structure as Pan and Kosicki states in their paper (p. 65). The headline exhibits the main point of the whole article, and then followed by lead providing more details to gain the readers’ attention. However, quite different from the news stories concentrating on a single or a series of related news event, the hurricane reports lack in continuous and dramatic episodes so that the newsmakers employed a cinematic technique “montage” which is defined by Merriam-Webster’s collegiate dictionary as “the production of a rapid succession of images in a motion picture to illustrate an association of ideas.” In the case of hurricane Ivan news report, a succession of different characters’ behaviors in change settings is presented to illustrate the multidimensional influences of the hurricane on the area. That’s how I came up with the comparison.

Having set the framework of the analysis, the next step is to do a word-by-word reading on those articles and mark down the key words or phrases on the coding sheet. This coding is regarded by Pan and Kosicki as a method that “integrate structural and lexical features into a coherent frame.”(p.65) Having read through the articles and done a sketchy analysis on the text beforehand, I felt the careful reading parts and coding section were not as difficult as they seemed to be. I think the skimming as a preparatory step is of great necessity, especially to me whose first language is not English.

The final step is to write the assessment of data. In the process of quoting certain excerpts from the reports, I ran into the dilemma that whether I should quote those words and phrases which convey the strongest meanings to exemplify the intensity of the frame or I should quote those more frequently used expression to exemplify the ubiquity of the frame. If I use the first approach, will the research be regarded as manipulated by the subjective judgments of researcher? If I choose the second method, will the effects of my analysis be weakened? I finally made the decision to quote the most representative parts in the consideration of the significance of research.

In this attempt of frame analysis, sampling of data was more based on the convenience of research instead of its potentials to answer the research question. Thus, the preliminary assessment of data is inevitably crude and fragmented. However, the section urges me to think of how to organize words to present my analysis in a systematical way. I think the core question is how to make text speak for itself. To put another way, it is about how to elicit the hidden meanings out of the visible discourse by clarifying the signifiers and signified.

The practice of frame analysis is indeed a significative process during which I make effort to combine the knowledge on books with the practical aspects of research. The self-report on the process is more of significance to me in that the procedure of reflecting, self-examining and recapitulation offer me a unique third-party angle to comb my thinking and actions. If time and energy permit, I’ll maintain the journal as a part of my future research.