Often times in quantitative research, we can confuse reliability and validity since they can be quite interwoven in their implications toward each other. From Creswell (2012), I have found the following delineations:

Reliability – scores or indicators need to be consistent and found similar when they are testing the same things (p. 159).

Validity – is the evidence that the test is measuring what it is supposed to measure (p. 159).

In the big picture, we need to have consistent scores (i.e. reliability) that assist with evidence that we are measuring what we intend to measure (i.e. validity).

1. Test-Retest Reliability – this indicates the stability of testing over a time period.

2. Alternate Forms Reliability – use of two-test testing the same variable.

3. Inter-rater Reliability – this is used in observations in behavior. I.e Observer scores are compared to each other for similarities and differences.

4. Testing of Responses for Reliability Suggestions:

• internally consistent (i.e. the subject’s answer to question #1 is the same answer as found in question #10)
• Kuder-Richarson Split Half test (KR-20 and KR-21)
• Spearman-Brown formula
• Co-efficient alpha testing

—

As outlined by various researchers (Phelps et al 2005; Gall et al 2007; Phillips 2007), providing creditability and reliability of research methods will promote strength of data findings as well as strengthen the application of findings to other music research.

—

Gall, M. D., Gall, J. P. & Borg, W. R. (2006). Educational research: an introduction (8th edition). Boston: Allyn & Bacon.

Phelps, Roger, Sadoff, Ronald H., Warburton, Edward C. and Ferrara, Lawrence. (2005). A Guide to Research in Music Education. 5^th ed. Lanham, MD: Scarecrow Press, Inc.

Phillips, K. H. (2008). Exploring research in music education and music therapy. New York: Oxford University Press.

Firmly based in context-dependent knowledge (i.e. rule-based knowledge to virtuoso/expert) Flyvbjerg (2006) presents 5 challenges to our misunderstanding of case study methodology.  While large samples of research data can develop a breadth of knowledge, individual case studies can present a larger work of depth on a specific point of study.  Various established researchers in the field of research methodology (—Campbell, 1975; Eysenck, 1976; Ragin & Becker, 1992) have seen through the challenges of reliability, theory and validity posed to case study research and begun to establish the meaningful research outcomes of the case study methodology.

The main contention found by various researchers is that the case study methodology presents issues with validity of research. As we can see from the visual reminder below, the virtuoso, or expert, comes to be an expert in a field through individual case knowledge. From specific cases that demonstrate both individuality for a study and possible connections to other research studies, a person becomes an expert in a knowledge field with the increased number of individual case studies that contain knowledge connections.

By challenging previous ideas of mis-understandings on case study research, it can be posited that like the natural sciences, the case study and how it is chosen is the objection really for generalizing for validity. Therefore, it is problem dependent and can use “falsification” (Popper, 1959) to generalize. Once we introduce the use of falsification to our research problem, a stronger case (no pun intended) can be made for the strength of case study research.

I think it can be safe to say that writing comes in waves; some days are better writing days than others. Thankfully, when we have a troublesome writing day there are a couple of ways to move through the mental block and, hopefully, onto some great writing output. To that end, I’ve added a few helpful items from my reading this week (Creswell, 2011) to remind myself of how to navigate those waters when they come.

1. Find a template and use that as a guide.

Creswell (2011) has a great template example on how to write a five paragraph statement of the research problem. By referencing other literature, statistical trends and a quote or two a template can quickly become a well developed piece of writing that may inspire you onto the next writing section. 🙂

2. Read other research studies.

What better way to write well than to immerse yourself in other well-written research studies. The more we surround ourselves with great models, the more likely we will begin to take on similar traits and habits. Needing a place to start? I have found a wealth of great doctoral studies from the AECT.org site as well as through Proquest doctoral searches.

3. Just write.

We can always find a way to procrastinate but in the end it really doesn’t make us feel any better about not having gotten any writing done for the day. Just grab your computer (or pick up a pen and feel the old-school vibe!) and write something. Writing in a stream of consciousness for a few minutes might just trigger those research writing skills back in action.

Ok, speaking of procrastinating, I need to get back to my own research. 🙂

Happy researching!

—