![Sample our Urban Studies journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110826/TOC-Subject-Sample-2021-Urban-Studies.jpg"})
ABSTRACT
How is the air pollution issue in Delhi framed by the news media and narrated by nonprofit organizations? To study news media framing, we employed an inductive approach based on automated text coding of news coverage of the issue. To study nonprofit organization narrations, a deductive approach guided by the Narrative Policy Framework was used to focus on the stories told via online documents as found on nonprofit websites. The findings confirm existing theory and empirical research regarding the leading causes and effects of air pollution; however, perceptions are mixed regarding the government’s ability to implement policy. The combined deductive and inductive approaches provide a systematic and multi-method research study for understanding perceptions of air pollution in one of the largest cities in the world. The result is a depiction of the priorities that influence public opinions, political decisions, and eventually public policies.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. The methods used in this study draw upon different sets of literatures, and consequently a distinction was needed to separate frames and narratives within the two literatures. The automated coding method lends itself to the exploration of frames, while the manual-coding method is better suited to the exploration of narratives. There is a distinction by definition and in application.
2. The sample size of the automated coding document set was constrained by a need to match the timeframe of the manual coding document set. In addition, as this is not a longitudinal study, the timeframe for the documents needed to be restricted.
3. Word sense disambiguation was avoided by employing a manual review of the entire concept list of the text sample and eliminating any value-laden descriptive concepts, such as “harmful” or “suffer.”
4. A sensitivity analysis was performed with varying window sizes. The window size was both doubled and halved, with variance as low as 0.33% and at most 15.69%. For this application, variance sensitivity is sufficient to test the stability of the results (Homma & Saltelli, Citation1996, p. 14). The proportions of the concept co-occurrence did not change.
5. Interviews were conducted with two government officials, a member of an environmental nonprofit organization, and two academics active in the issue area.
6. “Delhi,”, “India,” and “pollution” were removed from the sample as these were the original search terms.
7. For example, within a window size of 10, the following sentence could occur: “Busses contribute to air pollution.” This would prompt a co-occurrence. However, the following sentence could also occur: “Busses do not contribute significantly to air pollution.” This would also prompt a hit, but the meaning is entirely different. The word “bus” can be categorized as “transport,” but cannot be assumed to be a “cause” frame within the news article.
8. The concept “people” is comprised of: public, non-wealthy, vulnerable people, pedestrians, and transit users.
9. Effects were not included in the analysis of character relationships primarily to maintain comparability with the automated coding, which is limited in its ability to validly identify effects (unlike cause and solution concepts).