Agricultural Crisis Communication Using Social Media

Methodology

This study was developed and conducted within the constructivist paradigm, which recognizes that each person has his or her own method of making sense of the world. This approach led to a qualitative research design utilizing case study methodology to investigate the research objectives. A case is a bounded system, selected because it is of interest to the research objectives. Creswell and Poth state that a case can be defined with parameters such as the timeframe to be studied and the place where the case occurred. For this study, the system was three agricultural crises, bound by time (2011-2013) and location (High Plains region). "Your case study is defined not so much by the methods that you are using to do the study, but the edges you put around the case". To study a case, there is no specific form of data collection methodology.

Purposive and snowball sampling were used to select a total of six individuals as participants for this study. This sample size was sufficient for a qualitative study of this nature as in-depth information from a small number of people can yield the data needed, especially if the cases are information-rich. Also, having varied perspectives on each crisis, which differed in magnitude, typology, and nature, helped the researchers establish a maximum variation sample in which cases are selected to illustrate the broadest range of information possible.

One-on-one, semi-structured interviews were conducted with each participant selected for this study. A researcher-developed interview guide consisting of 17 open-ended questions was used to guide the interviews. Each interview was audio-recorded to aid in transcription and data analysis. Researchers also took field notes during each interview.

Interviews were transcribed verbatim, and data were analyzed using a constant comparative method with open, axial, and selective coding. Following an initial review of all interviews, the researcher used NVivo 10 software for Windows in which key concepts, phrases, or terms were organized into broad preliminary categories through open coding. From these initial codes, data were further divided into several axial codes. Interview transcripts were evaluated several times as the researcher continued to select additional information to include. Preliminary codes and corresponding axial codes were then refined into a more cohesive system through several phases of comparing, combining, and refining the codes until finally major themes were created. To form a model, memoing, in which the researcher wrote down thoughts, feelings, insights, and ideas about the research project, was used.

Trustworthiness and rigor were established through ensuring credibility, transferability, dependability, and confirmability. Researchers used triangulation of multiple sources of data, peer debriefing, and member checking to establish credibility. Transferability was established though a maximum variation sample and thick, rich descriptions of the data. Dependability was ensured through the use of an audit trail of all data collected, and confirmability was established through the use of a researcher's journal in which researchers were reflexive and acknowledged their our experiences may have affected their view of the research.

The nature of this study resulted in several limitations. Some potential participants were unwilling or unable to participate which narrowed the number of possible participants. The limited number of participants may make it easy to identify some of the individuals in this study, though efforts were made to disguise their identities. Further, the location of the researcher compared to that of the participants led to the use of telephone interviews for the majority of the participants. According to Charmaz, face-to-face interviews are preferred, but telephone interviews are more practical and allow greater uniformity in interview delivery making them an acceptable form of data collection. Finally, data collection was limited to what participants were willing and able to provide and what was accessible to the researchers online.