Data Analysis Strategy
As described in Marshall & Rossman, data analysis "is the process of bringing order, structure, and interpretation to the mass of collected data" (Marshall & Rossman, 1999, p. 150). The data that is collected in the various methods that I plan to incorporate in this study (surveys, interviews and observations) will be analyzed using the analysis strategy described by Creswell (1998) for ethnographic research. Creswell recommends the three aspects of data transformation advanced by Wolcott (1994b), description, analysis and interpretation of the culture-sharing group. The description may be where the writer, through progressively focusing the "description of a day in the life" of an individual. In this study, it was obtained through interviews and observations (Creswell, 1999). Different perspectives are made evident through the views of those interviewed. Analysis involved highlighting specific material that was introduced in the descriptive phase and the display of the findings in tables, diagrams and charts (Wolcott, 1994b). Wolcott suggests searching for patterned regularities in the data, "comparing the cultural group to others, evaluating the group in terms of standards, and drawing connections between the culture-sharing group and larger theoretical frameworks" (Creswell, 1999, p. 153). Creswell continues with his list of what to include in the analysis strategy by suggesting that making an "ethnographic interpretation of the culture-sharing group is a data transformation step as well" (Creswell, 1999, p. 153). This interpretation of analysis strategy sounds intriguing in that the researcher actually goes beyond the database and questions what is to be made of the information (Wolcott, 1994). In this part of the analysis the researcher may speculate outrageous, comparative interpretations "that raise doubts of questions for the reader. The reader draws inferences from the data or turns to theory to provide structure for his or her own interpretation" (Creswell, 1999, p. 153).
In designing my own data collection and analysis strategies, I came across some major challenges. In creating an online survey, much time is spent in the actual html coding of the web page. One or two questions on a survey is a simple matter, but when it nears 50 in number the complexities involved in managing the data and having it relayed to the researcher within an easy to comprehend format is a time consuming task. My first challenge was to create the survey using phrases that were understandable yet not too wordy. A web page is limited by a variety of parameters. The rule of thumb in good web page design is to create a page that loads quickly (for the client) and does not require much, if any downward scrolling. Horizontal scrolling is a total 'taboo' option, so the questions in the survey all needed to fit on the smallest screen size, usually 600 pixels in width (Ladd, O'Donnell, 1998). Other considerations include color contrast for words and backgrounds, visual interest and easy to understand directions. The online survey was presented on http://www.MrsSinger.com, which is an art education site that I have developed and personally maintained for approximately two years. There is a repeat clientele that visits this site regularly so I am able to control the data collection to a greater degree than if the site were new and attracted a wide variety of visitors. This site of mine attracts many students, families and educators worldwide. I have a solid database of those who have visited the site in the past, along with their email addresses, so I was able to do a broadcast email to past visitors requesting their participation in a survey at "Mrs. Singer's" site. In the email I also asked everyone to see if they could ask their friends and family to fill out the surveys too. This proved to be very successful in gathering a large mass of participants in a short period of time. Since I have many visitors from around the globe, I was able to obtain a wide array of participants from culturally diverse populations. In creating the form fields for the surveys I needed to insert coding that would enable me to know which question was being answered and what choice the respondent had chosen for each question. I set up the form field to automatically submit the responses to an email alias that I created specifically for this research. I created two email addresses, one for each survey, thus enabling an easy manner for me to keep the data gathered from each survey separate from each other. Within the code, but not visible to the site visitor, I was able to place notes to myself so I would know which question was being answered and how without having to refer to a master. These data management considerations evolved out of necessity since a large amount of data quickly developed. I received more responses than I had initially predicted. I needed to re-code the surveys twice to accommodate the incoming data and make it easier to handle.