SR - Research Designs and Methods Lesson

 

Research Designs and Methods Lesson

As you know, the scientific method that is used by all sociologists has five steps.  There is not much variation in the practice of the first three steps- but there is quite a lot in the fourth. There are many research designs and methods for sociologists to choose from and many reasons to choose different ones. The first explanation for the choice of a research design or method stems from the research orientation used and the topic to be studied. As each orientation has a different style, there are specific research designs and methods that are more suitable for each. The same is true for the research subject. Some questions are more suitably answered by one design or method over another. A second explanation for the choice behind a research design and method is funding. Some designs and methods cost more than others and budgets are always a factor. A third explanation is a time. Some forms of research take longer than others. Lastly, there is the question of ethics. It is very important for the sociologist to choose methods of research that comply with professional ethical guidelines.

Research Design

First, you should understand that there is a difference between research designs and research methods.   Research designs are the models that research methods are used within and determine the samples to be studied and the length of time required. There are three types of research designs:

Research Methods

Research methods are the specific procedures used to conduct research and there is much more variety within this category.

Once sociologists have collected the data from their research, they have to analyze it. This is the step in which sociologists give their data significance. By constructing a theory using the data collected, sociologists give meaning to the raw facts they have discovered. They do this through logical thought. There are two types of logical thought: inductive logical thought and deductive logical thought. In most cases, sociologists use both types within the same study.

Inductive logical thought is the reasoning that turns specific observations into a general theory.  For example, if you were to visit a new town and had a week's worth of warm weather (specific observation), you might conclude using inductive logical thought that the town has a warm climate (general theory.)  Naturally, this type of logic, where you increase your understanding of something larger based on smaller facts, has a flaw.  Perhaps those seven days of warm weather were an anomaly and the town is usually much colder.  Therefore, inductive logical thought might provide the hint of the truth but it does not provide absolute certainty. The findings from inductive logical thought are meant to be considered as probable.

Deductive logical thought, however, is the reasoning that turns general theory into specific hypotheses that are ideal for testing. In this case, the thinker is reducing information from the general to the specific.  For example, if you were to visit a new town and were warned that its climate was warm (general theory), you would be able to test a hypothesis that the town experienced more sunny days than cloudy ones.  Upon counting the number of sunny days and comparing them to the number of cloudy ones (specific observation,) you could prove whether or not your hypothesis was correct and create a theory, or conclusion, that a town with more sunny days than cloudy ones is warm.

Now that you have the basic idea of what sociologists do with their data, convert it to a theory or conclusion using either inductive or deductive logical thought, let's look specifically at what the sociologist is looking at to complete these thoughts.

Sociologists work within the constructs of "concepts." These concepts are the labels sociologists apply to elements of social life- family, media, religion, gender, class, race, etc. These concepts or labels have value in our society.  When the value of a concept changes from case to case, it is referred to as a variable and this is when things get a little tricky within the field of sociology. When you hear the word "value" you probably think of the word "measurement."  One cannot determine the value of something without measuring it. In math, determining the value or measuring something is fairly easy, but is that the case within a field like sociology that is dependent on people? The answer is no. People differ as to how they value or measure variables. Therefore, sociologists have to create an operational definition at the beginning of any study. In so doing, the sociologist specifies exactly what is to be measured before assigning a value to a variable. If a sociologist was conducting a study on race in America, he or she would need to create an operational definition of race. Prior to the U.S. Supreme Court's ruling in 1954 to end segregation in the nation's public schools, many school systems defined race based on the variables of white and non-white. Obviously, this system of measurement has changed over the decades. By the time the U.S. government conducted the 2010 census, the number of variables offered under the concept of the race included six main variables with an extra fifty-seven multi-racial variables.  

In order for an operational definition to be useful, it must be reliable and valid. Reliability depends on consistency in measurement. If repeated measurements, or values, result in the same information time after time, then it is considered reliable.  Again using the concept of race as an example, if the variables used in multiple government censuses to classify race result in consistent counts of the diverse population in America- then the measurement used is reliable.  However, consistency is just one aspect of the usefulness of a measurement. The operational definition must also have validity. To have validity, the operational definition must measure exactly what is intended to be measured. Studying race in America can reveal just how difficult it is to measure exactly what is intended to be measured. People are placed in racial categories for different reasons that vary from biological to sociocultural. This is why the 2010 U.S. Census included an explanation that the race variables used were based on "a social definition of race recognized in this country and [were] not an attempt to define race biologically, anthropologically, or genetically."  By explicitly stating what was to be measured and then by measuring race within those parameters, the U.S. Census Bureau was establishing validity for its research.

Once the sociologist has established exactly what the study will measure, it is time to see how the variables involved are related. Ideally, the sociologist hopes to find a cause-and-effect relationship among the variables being studied. This would mean that a change in one variable would result in a change in another variable. Sociologists call the variable that causes the change an independent variable.  They refer to the variable that is changed as the dependent variable.  Therefore, the value of one relies on the value of the other. If a cause and effect relationship is evident, the sociologist will be able to predict future events. Let's draw a picture of this using something familiar- the amount a student studies for a test (the independent variable) affects the grade a student receives on the test (the dependent variable).

Unfortunately, when studying people there are many variables to consider and this dilutes the probability of establishing a concrete cause and effect relationship between two variables. Often sociologists encounter correlations between variables. This means that there is a relationship in which two or more variables change in unison. We know that a student studying for a test and the student's test score change, or rise, together. But there may be a third variable in effect, perhaps the student had a good night's sleep and a healthy breakfast before taking the test. This third variable could have impacted the student's performance as much as the studying. If that third variable ends up having more of an impact than the studying did, then the first conclusion that studying more raises test scores would turn out to be a spurious correlation. In other words, the conclusion drawn between studying and test scores would be an apparent but false relationship between the two variables. In order to avoid the formation of spurious correlations, the sociologist uses a control within his or her experiment. Control is the holding constant of all variables except the one being tested so that the effects of that one variable can be clearly identified. Thus, a sociologist would make sure that the student always had a good night's sleep and a healthy breakfast before every test was taken and would allow the only difference to be the amount of studying allowed. If a demonstrated correlation then appears between the identified independent variable of studying on the dependent variable of grade and there is no evidence of a third variable affecting the outcome, then the sociologist can draw a cause and effect conclusion that studying impacts test scores. That sounds like a lot of "ifs," doesn't it? It especially does when you consider how difficult it is to "control" people who do not live in a laboratory in order to eliminate the third variable. There is always the additional possibility of a Hawthorne Effect occurring.

The Hawthorne Effect refers to the change in a subject's behavior caused simply by the awareness of being studied. The special attention paid to a subject by a researcher can lead to the subject modifying his or her behavior. If a student knows he is being watched while taking a test it could result in the student paying closer attention to the test questions and thereby answering them more correctly than is usual for him. Sociologists discovered the Hawthorne Effect during a study on productivity in an electricity factory near Chicago during the late 1930s. Researchers tried to determine if extra lighting within the factory would improve employee productivity. Through their experiments, the researchers concluded that extra lighting did improve productivity. The experiment ended and the lighting returned to the pre-experiment dimness. That is when the researchers noticed that the productivity levels of the factory workers did not drop back to the pre-experiment levels. It soon became apparent to the researchers that the factory workers had changed their behavior due to the simple fact that they were being watched and they dubbed this phenomenon the Hawthorne Effect after the name of the factory.

In the last few paragraphs, we have only discussed the variables surrounding the persons being studied by a sociologist. The sociologist is a person too and therefore brings his or her own personal issues to the research's table and must be wary of a personal slip-up that could ruin the study. With this in mind, the sociologist must make every effort to maintain objectivity during a study. Any intentional or unintentional bias can affect how data is collected and analyzed with the result that the study could become worthless. Even Max Weber, who argued that sociologists should employ verstehen in their studies, promoted the idea that sociologists should remain detached from their subjects so that their findings would reflect the reality of society and not their own feelings. The only way, really, to be sure that the findings of a study are legitimate and not biased by the researcher is to replicate the study. If the same study is carried out using the same research methods time and again but by different sociologists with the same results, we can be fairly certain that the conclusions drawn are legitimate.

Sociological Research Key Terms Review Activity

In order to prepare for your key terms quiz, work through the crossword puzzle below.

 

IMAGES CREATED BY GAVS OR OPENSOURCE