How Research is Done: Finding a Research Topic

One of your assignments for this class is to survey your specific field of study to find a researchable idea for your thesis or major project.  You'll need to pay attention to some of the rules below.  With a little training, anyone can perform basic research in the areas that comprise Family & Consumer Sciences.
Some rules when you are first starting out: Basic Steps in the Planning and Conduct of Research
Traditional Survey Research
  1. Identify the problem area.
  2. Survey the literature relating to it.
  3. Define the actual problem for investigation in clear, specific terms.
  4. Formulate testable hypotheses and define the basic concepts and variables.
  5. State the underlying assumptions which govern the interpretation of results.
  6. construct the research design to maximize internal and external validity
    1. selection of subjects
    2. control and/or manipulation of relevant variables
    3. establishment of criteria to evaluate outcomes
    4. select or develop measures
    5. Specify the data collection procedures
    6. Select the data analysis methodology
  7. Execute the research plan
  8. Evaluate the results and draw conclusions.
Traditional Historical Research
  1. Identify an area of interest.
  2. Initially survey the literature relating to it.
  3. Define the actual problem for investigation in clear, specific terms.
  4. Begin literature search in depth, including historical records, "data" in the sense of various archives.
  5. Define all the basic concepts and important theories/events.
  6. State the underlying assumptions which govern the interpretation of results.
  7. construct the research design to maximize internal and external validity
    1. selection of archival information
    2. selection of interviews and specific historical searches.
    3. establishment of criteria to evaluate outcomes
    4. select or develop timelines and other measures
  8. Execute the research plan
  9. Evaluate the results and draw conclusions.

It is important that you keep notes. I always use a loose leaf binder with dividers on each section of the research process.
A typical project will have these sections:
Each of these sections will have subsections as the topic requires.  Often students will read research reports from page 1 to the end - however, the sections of a good research report are never written in order.
The Introduction should be written last.
The next two sections (literature review and methodology which are the real work of any research project) will probably develop together (hence the loose leaf notebook which allows insertion of notes as they are discovered or revealed.  During the composition of these two sections, the researcher will be thinking about findings and how these will be concluded and discussed.. Typically researchers will run across unfamiliar terms. Don't forget you have the entire internet at your disposal for finding definitions and explanations of any term for which you need more information (like canonical correlation).
There is an easy way to think about all this. Suppose you wanted to find out something about people. How would you carry out an investigation so that you could be pretty sure of your findings? You'd want to use the scientific method. Here is:


Wallace, Walter 1971. THE LOGIC OF SCIENCE IN SOCIOLOGY. Chicago: Aldine Press.

1. Theory - is used to get a researchable idea. The Library is the place where all Theories live in print.
2. Hypotheses - once you find a theory on which to base your arguments, and once you've sufficiently mined library information, you must derive hypotheses from the theory using Deductive Logic. Hypotheses must be developed from theory - they won't be fully formed in theory - your job is to construct them within the tenets of your theory. Hypotheses are testable statements of relationships between theoretical concepts. Begin to think about how you are going to measure all the parts of each hypothesis.  Try to think like this:
   
A concept lives in the ideal, theoretical world.  In order for a concept to be detected in the concrete, real world, you'll need some measures.
    For example, height is an ideal, theoretical concept that refers to the "distance from the base of something to the top".
                         weight is an ideal, theoretical concept that refers to the "vertical force exerted by a mass as a result of gravity".
          The relationship between these two ideal, theoretial concepts are usually related to each other, meaning this:
                As the distance from the base of something to the top of the thing gets larger, the vertical force exterted by the thing as a result of gravity gets larger."
                This hypothetical statement remains untested until we can apply concrete, real measures to the concepts.
In order to test this hypothesis, we'll need measures for height (say inches) and weight (say pounds). So we find, in the review of literature, studies that define height in inches and weight in pounds, and additional research that shows that the two concepts are related when measured by inches and pounds. 

This isn't as easy as it might sound. Had these been unfamilar concepts, we might have chosen other, perfectly good, measures of the concepts that would make finding relationships more difficult.
We could have chosen attoparsecs for height and metric tons for weight.  The relationship would likely hold true, wouldn't have been as easily detected.
Here's a graphic illustrating the relationship between concepts and measures"
concepts and measures




Example of a testable hypothetical statement in social science: Past research might tell us that among adult Americans (over the age of 25), persons born later in the century are more likely to marry later, attain higher levels of formal education, and achieve higher degrees of socioeconomic status.  This is really several hypotheses in one (keep it simple).
For each concept in the hypohesis, the research will have to select measures.  The method used to turn concepts into measures is known as Operationalization of concepts.
By turning the language into a graphic representation of hypotheses, the researcher can avoid burying the crux of study in heavy text.  Like this:

In Hypotheses, for every Concept there should be a corresponding Measurement
Just as    Concepts are related to Concepts.
              Measures are related to Measures

The measure for:
Knowing exact measures allows for the creation of an Instrument Design which refers to the type of Observation Schedule you wish to keep. This entails:
Questionnaire construction
Observation Checklists
Experimental designs

Using questionnaires means putting the Measure in the form of a question:

"In what year were you born?   19___
"How old were you when you first married?" _____ years of age
and so on.
 
Sampling refers to the type of GENERALIZATIONS you wish to make after completion of your study...
Logically, if you wish your findings to be generalizable to the U.S. Population, you will have to sample from that population. If you are studying just Northeast Ohio, the eight counties around Akron will do just fine. If you are concerned only with the Population in Akron, Ohio, then .....

Sampling Akron would mean taking the entire population of citizens who are Americans and who are over 25 (and who have been married at least once), dropping their names into a "hat" and selecting a random number (you'd need about 200 for this study to have representativeness.
Now you are ready to make Observations

3. Observations are where you actually implement your research strategy on the people, places or things your study has been designed to cover. This part is also known as DATA COLLECTION - this is where questionnaires are administered.
Each type of data collection has its own strong and weak points. People can lie on questionnaires while Observers can have short attention spans.

After Observations - you are ready to use Analytical Methods on your data. All observations must be coded (either quantitatively or qualitatively) into some systematic, uniform pile of retrievible data. Computers are used most often for this purpose. 
 
Once data has been coded, the researcher may then apply some form of Evaluation, such as statistical tests, eyeball judgements, content analysis, or other method to see if the hypotheses should be supported or not.

4. Generalizations - Remembering your sample, we generalize the results of analysis back to the theory with which we started this whole process.

If our analysis "failed to reject the hypotheses", then we have supported our theory.
If we observed no evidence of the hypothestical prediction, and we are very certain of our research methods, then we must hold the theory in question - and modify it.

Types of Research:

Observational Techniques:
1.  Ecological fieldwork.  Jane Goodall watching gorillas.  Graduate Students observing toddlers at the CCD.
2.  Anthropological fieldwork.  Observing “primitive” cultures – e.g., rain forest tribes, teenagers interacting, etc.
3.  Sociological fieldwork.  Watching people at a specific event – political rallies, rock concerts.

Direct Observation -clipboard, white coat, checklists - can be behind one-way mirror
Indirect observation-pretending to do something else while observing - spying
Participant Observation. - observer is actually involved in the activities - studying motorcycle gangs.

Observing Physical Traces and Archival Records
1. Searching through Garbage Dumps and Landfills
2. Looking through the physical record
3. Library archives - special collections research.

Questionnaires and Interviews
1. Open-ended - guided interviews
2. Closed-ended/Forced Choice - questionnaires

Types of Measures
Research Strategies:
Your job - either now or very soon in your graduate career - is to isolate a researchable problem, begin reviewing the literature for your topic area, make your summary tables of the literature, and start to derive hypotheses from your literature review.

My advice, though not always appreciated by either graduate students nor my colleagues, is to begin reviewing the literature at  every possible moment.  Students who use term paper assignments and classroom activities as opportunities to add to their stock of knowledge on some researchable topic will be much further ahead of the thesis or project when the time comes to see that task through to completion. 

A thesis is at least a one year project - the beginning of the process can start any time while in your graduate program.
Imagine two graduate students, each one with a thesis or project ahead of them. One waits until the middle of their 2nd year to begin the literature review - which means they will likely be in their 3rd year by the time the thesis is completed.

The other student uses term paper assignments for EACH graduate class as an opportunity to review part of the literature for their thesis. They'll start at the beginning of the literature and, over their first three semesters, move to more specific research until - by the time they are ready to select a thesis committee -  they are practically finished with their literature review and are almost ready to write up a thesis proposal and have their first meeting.  This wiser, more efficient student will have only the data collection and data analysis portions to complete - and will likely be finished by the end of their 2nd year.

My advice is to start to become an expert in some aspect of your studies by picking an area of concentration that holds good promise of developing into a researchable idea.  If a term paper assignment is very specific, ask your instructor for some leeway in molding the assignments to your advantage.  You'll be surprised at the number of your professors who also see the wisdom of this approach.

And that is your assignment for this class - to identify a researchable idea in a general way.

Return to Syllabus