The main topic for this week was data analysis. Similar to almost everything so far in research, data is analyzed differently according to the type of research design. The link provided in the lecture to the University of the West of England's site, presented a clear overview for quantitative data analysis, specifically in relation to statistics: "This is the process of presenting and interpreting numerical data. Descriptive statistics include measures of central tendency (averages - mean, median, and mode) and measures of variability about the average (range and standard deviation). These give the reader a 'picture' of the data collected and used in the research project. Inferential statistics are the outcomes of statistical tests, helping deductions to be made from the data collected, to test hypotheses set and relating findings to the sample or population." Numbers, numbers, numbers! Qualitative data seems to be analyzed with more of an inductive process. Once the data has been organized, trying to identify and analyze it to gain insight into the research, is a great qualitative method. Miles and Huberman's process of data analysis was my favorite and I think will prove most helpful. Their three steps are actually ongoing throughout the research process: 1. Data Reduction - This seems to be an analysis of initial data where one is almost brainstorming on the research question using this data. 2. Data Display - This is where the data becomes more organized in order to bring about conclusions. 3. Conclusion Drawing & Verification - Lastly, the researcher decides what his data means and then verifies those concepts. Throughout the data analysis progression, I think that within my organization steps I would want to have everything in tangible form in front of me. It's how I work now on research papers, and believe it to be the best idea for a real research process - both qualitative and quantitative.
The YouTube video and Slideshare presentation, gave over a few great hints. The YouTube video mentioned a website called counselingtechnology.net, which can be useful in creating surveys. I found it to be a simple interface, and there is free membership sign up for certain professional groups. There was a data analysis program for Excel mentioned a lot in the video called EZ Analyze, which may come in handy at some point. The lecturer discussed three interesting hints for survey data analysis. First, code the paper surveys with identification numbers to help input data later for easy analysis. Second, use a code that also contains information relevant to the survey, such as date or area surveyed. Third, every row within a data analyzing program is reserved for a new person's data - not a new idea to an Excel user like me, rather a verification of a sneaking suspicion. The Slideshow brought up the great question of "Why is sampling important?" This is something I ask myself when reading research that utilizes samplings; why is this sampling important to this study, and was it conducted successfully? According to the lecturer, sampling is used to test hypotheses which usually become 'law-like' in the sense that the sampling allows the researcher to infer certain facts from the sampling about the wider population. This inference requires the smaller sampling to validate the wider population correctly, or the internal validity must validate the external validity. Essentially, this sampling is important in both qualitative and quantitative studies, and I thought the Slideshare really put this concept in perspective.
This week I read two of Creswell's chapters on beginning to write a research paper. The first chapter on writing strategies covered some basics that were fleshed out for a clearer understanding with research design applications. What I found most applicable to me was the idea of writing an outline. I've always found this helpful in my writings, and still utilize this strategy today. Creswell writes, "specify sections early in the design of a proposal. Work on one section will often prompt ideas for other sections. First develop an outline and then write something for each section rapidly, to get ideas down on paper. Then refine the sections as you consider in more detail the information that should go into each one." This method was very similar to Franklin's three state model of first developing an outline, then writing a draft and more ideas surrounding the outline, and lastly editing and polishing. These methods are closest to how I write now, but with clearer steps and will therefore hopefully help me become a better writer in general. The one point in this chapter that I have never heard of before, but is fantastic advice, is the idea of a writing warm-up period as a starting exercise for both the mind and the fingers. Maybe this would be a great way to catch up on emails to friends.
The other Creswell chapter I read this week was about writing introductions. Ultimately, all research designs follow the similar pattern of "announc[ing] a problem and just[ifying] why it needs to be studied". When writing this introduction, Creswell's advice is ensuring that the first sentence is a 'narrative hook'. This is very similar to something my husband has been working on in reference to his songwriting. All songs have a hook, a part of the song that makes it memorable to the audience - so memorable in fact, that it's the part of a song that one would sing when thinking of that song. Although a song hook is usually not the first beats or the first stanza, it encapsulates a similar idea to the narrative hook of an introduction. The deficiencies model is a method of writing two pages of the introductory ideas. It contains five points that I discovered to be succinct and a great model to follow. 1. Write about the research problem. 2. Discuss studies that have addressed this problem. 3. Talk about the deficiencies in the aforementioned studies. 4. Discuss the significance of the study for particular audiences. 5. Write the purpose statement.
No comments:
Post a Comment