Lab 2- Research Skills!

6/5/2009

Lab 2- Research Skills, from Nathan Morrow.

Tacit knowledge is knowledge that is held in someone’s brain and only freed when asked about. How do we retrieve, analyze, and write about tacit knowledge?

- Wee must come up with a research question. How are research questions different from other kinds of questions? The basic idea is that it must be researchable.

There are two elements to think about.

Where on the spectrum of broad to specific do you want to be? (Do they eat food in Italy? to, What kind of cherry tree do you have in your backyard and how old is it?) In most cases, you want something you can write a page about in a couple of hours.


Avoid overly subjective or difficult to define terms. Avoid the indefinable. This is a big trap when it comes to food, as food is inherently subjective. Words or phrases like “the best” are dangerous. Don’t ask “What is the BEST ice cream in Bolsena?". Instead ask, “What are the differences between 10 ice cream shops in Bolsena?”

These qualitative themes are hard to benchmark, and might require a doctoral thesis to cover in full!

Who, What, Where: The Three “Ws".

List your questions and get rid of those that are less important. Once you decide on your research questions, brainstorm. A good number of questions for a real research project is 100. In our case, you may want to create 50. Group your questions by who, what, where. Develop an iteration of your research questions. You want a conceptual framework or a concept frame.

A concept frame is a diagram of how you think a question works. A two by two matrix is all you will probably need for this course. Let's use an example. Say we want to look at the introduction of kiwi fruit to the Bolsena area, and its effect on the Bolsena economy and food economy.

Put across the top of the matrix "economy," and at the bottom, "local cuisine". Assemble a pro/con or yes/no construct. You want something that is both negative for local cuisine and negative for the local economy elsewhere. For example, kiwi flowers stink. Then list what is a pro for the local cuisine in regards to kiwi. As an example, kiwi is harvested in December when nothing else is. This allows workers to do something in the off season. Kiwis are high in nutrition and fiber, which is good for local health.

How about the lake here in Bolsena? Local transport puts kiwi fruit into markets, and has some sort of feedback into the cash economy, which has some sort of feedback into preserving the resources of the lake. Many diagrams are circular, and feed into one another.

Produce meaningful comparisons. Refine your research questions using this technique. Perhaps while considering kiwis, you might realize you need to compare picking techniques for kiwis and olives to understand the local economy. This also helps you realize what kind of question you are asking.

Not all methods or techniques are the best for every question. You may have: how many, how much, who, where, when, how, why.

What is the best way to measure how many and how much? Find local statistics on, say, olive or kiwi workers. Find or produce a survey with a decent sample size. A healthy sample size provides accuracy.

What about finding out Who and Where? A quantitative technique works well for this. You want surveys and samples, as well as multiple interviews.



What about How, Why, and When? These questions try to describe something that is dynamic, not a static snap shot. For these, the best approaches are case studies, key informants, and focus groups. These interviews are less random then those used in quantitative research. You want to find someone with specific knowledge and know-how. You need a key informant. You also must describe how and why you chose the person you did. How did you find out who knew the most about the topic you are researching, and what was your criteria?

Case studies. A case study is: looking at a specific time bound series of events and describing them on a time-line.

Focus Groups A focus group is six to twelve people, defined by a criteria (which can be multiple). Age, gender ratios, or employment can all work as criteria. You, the researcher, talk about a single topic from multiple perspectives. A group interview is not necessarily a focus group, as is commonly believed. A good focus group needs criteria about who is in and who is out, and a single idea or object from various perspectives.

More on Methods:

Single – An interview or a case study.

Two – Doing two interviews, or doing multiple interviews.

What would you do a single interview for?

Perhaps only one person really understands the topic well. Research is all about resources, including budget and time. This keeps you at the level of the descriptive analysis or the narrative analysis.

You can compare and contrast different kinds of views using two different interviews. For example: what would a Cajun chef think about fish for lunch, and what would an Italian chef in Bolsena think?

If you are trying to generalize anything, you need multiple interviews. If you want to test your conceptual model, compare similar and different components of different case studies or interviews. In your 50 questions, you want to organize them to provide a narrative analysis.

Practical Considerations:

Do you do a transcript or not? What are the advantages of a transcript?

You can do iterations of your analysis. It will provide repeatability. You can repeat the activity. The downside of a transcript is that you have to transcribe it. If you are not going to use the transcript, don’t make one.

Skype has a transcription plug-in. It is great for evaluations of, say, a 25 million dollar program!

The Instrument:

If you are not going to do a transcript, you need the an instrument. An instrument is the tool you use to record the responses. It is a form that can be used again. The more specific your questions are and the more constrained and clear your instrument is, the better off you will be. If you want to compare two things, use two columns. A form needs to exist to record your data. It is hard to just take notes when talking to someone. You can practice an instrument on someone and see how long it takes, finding out what questions work and what questions do not work.

Analysis:

Triangulation means having at least three sources corroborate the information. This is the test of the validity of basic qualitative research. If three different perspectives tell you pretty much the same thing, you should record this in your research. You want to group similar ideas: go through the interview and look for themes or similar ideas.

Coding. You want to go through and code your transcript or your responses. These are all the things about cheese, these are all the things about Bolsena. You are going through the data and analyzing it.

Some Other Things to Keep in Mind:

Everyone loves a representative quote. The more quotes used in your research, the better. Oppositional quotes are also great. You can have a common view and then the opposite view.

Feelings! How does lunch in Bolsena make people feel? Code for feelings on lunch, score them on one to five from sort of depressed to ecstatic, then come up with a way to organize the information.