01.03 Sociology Research

Watch More! Unlock the full videos with a FREE trial

Add to Study plan
Master

Included In This Lesson

Access More! View the full outline and transcript with a FREE trial

Transcript

Today we’re going to be talking about research and research methods in sociology.


 

In today’s lesson we will cover 3 things. First, it’s important to understand why we do research. For some this will be obvious, but for others this is a legitimate question. After all, how much research do any of us actually do in the course of our everyday lives? We seem to do fine without it, right? Second, we will discuss 4 different research methods that sociologists use to study our social worlds. We will discuss experiments, surveys, ethnography, and secondary data analysis, highlighting the advantages and disadvantages of each. We will conclude the most important takeaways from today’s lesson.


 


First off let’s define research. Research refers to the systematic process of collecting data for the purpose of testing theories and producing knowledge. The research methods we will discuss later are all specific ways to collect data for the purpose of answering questions we have about our social world. But let’s step back for a moment and think about why we even need research? Many of you have probably heard common sayings such as “nothing is a substitute for common sense” or “trust your gut” and other similar claims about value of common sense and intuition. For the most part, our common sense and intuition serve us well in that they allow us to get by and survive in our social world. Think of them as shortcuts that allow our brains to deal with an overwhelming amount of information. But without rigorous research, we have no way of knowing whether or not our common sense and intuition are actually right. History tells us that before Nicolaus Copernicus came around in the 1400’s, it was common sense that the earth was the center of the universe. Prior to the late 1800’s, many people were ridiculed for putting forth the the claim that diseases can be caused by small organisms that live in our bodies. In the early 1900’s a common view of proper child-rearing held that children are “to be seen but not heard.” This view held that human children are simple creatures who need little more than food and water to grow and prosper. Of course, we now know that this view is incorrect and frankly dangerous due to a vast amount of research that has overturned such as idea. Again and again, scientific research has been shown to be our best way to understand the universe, and our social universe is no exception. Yes, common sense and intuition have their place. In the world of science they are perhaps most useful in generating ideas worth testing. But make no mistake, common sense and intuition do not qualify as evidence in the world of science. Let us now turn to the various methods by which sociologists carry out research.


Experiments are a research method that examines cause and effect in controlled conditions. The strength of experiments is their ability to examine cause and effect. For this reason they are largely considered to be the best method that scientists use in all scientific disciplines. Their primary weakness, specifically geared toward the social sciences is their lack of generalizability. Generalizability refers to the extent to which results from an experiment can be applied to a population as a whole. While experiments are the most used research method in most scientific disciplines, in sociology they are not. This is due to a number of reasons including training, publishing practices, ease of data collection, and occasionally moral concerns.


There are a number of concepts that are important to know when it comes to experiments. Let’s use an example to highlight these concepts. Let’s say that I am interested in testing whether or not eating a candy bar before taking a test improves test performance. I have 100 students in a class and on test day I flip a coin when each student enters the classroom. If the coin lands on heads, I give the student a candy bar to eat, if it lands on tails I do not. In this example, I have 2 conditions; the experimental condition in which the students eat the candy bar and the control condition in which students do not get the candy bar. Control conditions are vital for experiments. If I gave every one of my students a candy bar I wouldn’t be able to find out if the candy bar had an effect or not. There must be at least one control condition that serves as a reference for what happens without the manipulated variable.  Now, the independent variable is the variable which is manipulated in the experiment. In this example, the candy bar is the independent variable as that is what I, as the experimenter am manipulating. The dependent variable is the outcome which you are predicting the independent variable will impact. Therefore in our example, test performance is the dependent variable. Then lastly and most importantly, random assignment is the process by which experimenters assign subjects to conditions. Random assignment is just what it sounds like, it means that you randomly assign subjects to conditions meaning that subjects have an equal chance of being in any condition. Since I flipped a coin to determine whether or not students got the candy bar, each student had an equal chance of either getting the candy or or not getting the candy bar. If I had simply given the first 50 students who arrived to class a candy bar, my experiment would have been flawed. The reason is that the first 50 students who arrive are likely different in many way from the last 50 to arrive. Random assignment allows all the ways that people differ to “wash out” or be equated across conditions. Therefore, if the students who do eat the candy bar actually perform better on the test, the experimenter can conclude that it was the candy bar that improved performance and not some other variable that already differentiated the experimental condition and the control condition.


Our second method is survey research. This is the most common research method employed by sociologists. All of us are familiar with the basics of what a survey is. A survey simply involves asking people to answer a series of prepared questions. Today, this is most often done online however it can still be done via snail mail and in-person handouts.


 

One of the advantages of doing a survey is the generalizability of your results. As long as the people you survey are a random sample of people from a larger population (we will talk about this more shortly), you can talk about your results from this small sample as representing the answers from the population as a whole. Another advantage is that surveys are relatively easy to do and therefore allow researchers to gather tons and tons of data in a pretty quick time frame. One big weakness of surveys is getting at cause and effect. Surveys do not have the controlled conditions like experiments that allow you to tightly manipulate 2 different groups so that they only differ on one variable. Another weakness is something called responder bias. If you are asking people about subjects that are taboo or uncomfortable people may choose to simply not answer your questions. Also, surveys only get at what people are wiling to say or admit, not necessarily what they actually think. If you ask “are you racist?”, people are unlikely to openly admit they are so a researcher, or anyone really.


 


Let’s wrap this up with 2  important concepts to remember. A representative sample is a subset of a population that accurately reflects characteristics in the population as a whole. To illustrate, let’s say that a researcher is interested in the drinking behavior of the student population at a university. Now, the researcher could go around and survey each and every student at that university but no one has the time nor resources for that. Instead, the researcher will look to survey a representative sample of students. In order to get a representative sample of students, each and every student at that school needs to have an equal chance of being in the survey. This brings up to our second concept to know, random sampling. Random sampling is exactly what we just mentioned. If you draw a random sample from a population, that means you are giving each person in that population an equal chance at being in your sample. This assures that the small subset of students that are surveyed have all of the same characteristics as the entire population. This allows the researcher to assume that data observed from the representative sample would also hold true for the entire population (the entire student body in this case). As an example of what not to do, imagine that the researcher just stands out front of the school library on Thursday, Friday, and Saturday nights and hands out the survey to students. Based on the data collected, the researcher concludes that students at this university don’t drink very much at all. What’s the problem? The researchers sample wasn’t randomly drawn and therefore isn’t representative. Students who are going to the library on weekend nights are highly unlikely to be large consumers of alcohol. This is why it is very important to give every person in a population an equal chance of being in the survey or else you will get results that do not hold true for the population as a whole. Imagine if instead the researcher administered the survey on weekend nights outfront of the popular students bars and clubs? The results would be very different and again, would be a very poor representation of student behavior as a whole.


Sometimes referred to as participant-observation, ethnography is our 3rd research method. Ethnography involves actively observing and participating in real world interactions to better understand a particular group or culture. Sometimes this involves being a simple spectator, taking notes, and observing behavior from a distance. More often, this involves actually engaging in the context of study and participating with the group that is being studied. Ethnography is commonly employed by anthropologists who study cultures around the world by living with them for months and sometimes years at a time. This involves taking rigorous written notes of anything and everything that is seen, heard, and experienced.


 

The primary advantage of choosing ethnography is that you are seeing interaction unfold in its natural context. You are not limited by the artificial nature of an experiment for example, nor are you limited by an impersonal survey administered over the internet. Ethnography allows researchers to gather rich data as every single interaction is documented. If something unexpected comes up you are free to explore it. 


On the other hand, the primary disadvantage of ethnography is researchers are limited by their own perspective. We know that 2 different people can see the same events and come to different conclusions. Researchers employing ethnography are no exception. This raises concerns of repeatability which is exceptionally important in science. In order for our knowledge to grow we must repeatedly test and re-test our ideas to make sure they continually stand-up to the evidence. This is difficult to do when relying solely on ethnography as the method of choice. Nonetheless, ethnography has its place as a tool in the sociologists tool-kit. Observing real-life unfold can often be extremely useful towards generating interesting research questions and gatherning novel data  as each interaction is likely to be unique in it’s own way.


Our 4th and last research method, secondary data analysis involves using existing materials for the purpose of answering one’s research questions. This method is often employed when collecting data is difficult or impossible such as gathering data about the historical past. Secondary data analysis is referred to as comparative-historical analysis when researchers use historical documents to develop an understanding of past events, people, and cultures. Researchers that use more modern data sources such as surveys administered by other researchers and data gathered by government agencies such as the census bureau are also doing secondary data analysis. 


 

The advantage of this method is obvious. Data collection can be an exceptionally time-consuming process. Why go to the trouble of creating and gathering new data if data already exists that can answer your questions? Of course, the big disadvantage is that the perfect data may not exist for any particular question. For example, there may be data available that shows the average salary for medical professionals in a particular region, but not separate figures for how average salaries compare between men and women. If you are just interested in average salary great! But if you need average salaries by gender then you are out of luck. All told, using data that already exists can be great as it saves time, but just because data is available doesn’t it mean that it can answer any specific question a researcher might have.


Alright, to conclude our common sense and intuition are limited abilities. While they work for us on a day-to-day basis, they are not acceptable when it comes to developing a scientific understanding of our world. To better understand our social world, sociologists employ 4 different research methods that each carry their own advantages and disadvantages. The experiment is best for getting at cause and effect, the survey is the most widely used method in sociology, ethnography involves going out and observing the real-world, and secondary data analysis involves using data that already exists.


 

Alright, to conclude our common sense and intuition are limited abilities. While they work for us on a day-to-day basis, they are not acceptable when it comes to developing a scientific understanding of our world. To better understand our social world, sociologists employ 4 different research methods that each carry their own advantages and disadvantages. The experiment is best for getting at cause and effect, the survey is the most widely used method in sociology, ethnography involves going out and observing the real-world, and secondary data analysis involves using data that already exists.


 


We love you guys! Go out and be your best self today! And as always, Happy Nursing!



View the FULL Transcript

When you start a FREE trial you gain access to the full outline as well as:

  • SIMCLEX (NCLEX Simulator)
  • 6,500+ Practice NCLEX Questions
  • 2,000+ HD Videos
  • 300+ Nursing Cheatsheets

“Would suggest to all nursing students . . . Guaranteed to ease the stress!”

~Jordan