Scientific research follows eight basic steps. They are explained below.
1. Selecting a Topic
The first step is to select a topic that a sociologist finds interesting e.g. homelessness, poverty, gender equality in an organization, etc.
Some sociologists choose a topic because they want to better understand a social problem and may be even solve it.
2. Defining the Problem
The second step is to define the problem, to specify what sociologists want to learn about the topic.
Homelessness is a vague problem that is common around the globe. Sociologists may examine specific aspects of homelessness e.g. how homeless people survive on the streets or how homeless people’s diet compares with others, etc.
3. Reviewing the Literature
Sociologists must read what has been published on their topic.
They need to narrow the problem, identify known areas and decide which area needs to be researched.
Sociologists might even find out that the question they are interested in may have been answered already so they don’t need to rediscover what is already known.
4. Formulating a Hypothesis
The fourth step is to formulate a hypothesis, a statement of what sociologists expect to find according to predictions from a theory.
“Men who are more socially isolated are likelier to abuse their wives than men who are more socially integrated” is an example hypothesis.
A hypothesis predicts a relationship between or among variables e.g. variables in the above hypothesis could be social isolation, social integration and spouse abuse.
These variables need to be precisely defined so they can be measured; we call these definitions of variables operational definitions.
5. Choosing a Research Method
Sociologists then need to decide how they are going to collect the data. Sociologists use seven basic research methods (or research designs).
They choose the method that will best answer the particular questions they are asking.
Most importantly they want to figure out the averages of a research so they can see whether their theory holds up with the statistical data.
6. Collecting the Data
When collecting data, sociologists have to take care to assure the data is valid and reliable.
Validity: operational definitions must measure what they are intended to measure. Spouse abuse is a vague term. It needs to be defined in detail so no one has any question what the sociologist’s operational definition is for spouse abuse.
Reliability: if other researchers use a sociologist’s operational definitions, their findings will be consistent with the sociologist’s findings. For example, you show a 10 percent rate of spouse abuse, but another researcher using the same operational definitions determines it to be 30 percent, the research is unreliable.
7. Analyzing the Results
Sociologists will have been trained in a variety of techniques to analyze the collected data — from those that apply to observations of people in small settings to the analysis of large-scale surveys.
If a sociologist has a hypothesis in their research, now is when they will test it.
8. Sharing the Results
To wrap up their research, sociologists will write a report to share their findings with the scientific community.
They will state their methodology, their operational definitions, comparisons with other reports and whether they agree with the theories of others.
Research Methods (Designs)
Let’s look in greater detail at the fifth step to see what research methods sociologists use.
Surveys
Survey: the collection of data by having a population answer a series of questions e.g. wives that are in an abused in the relationship will answer the questions.
Population: a target group to be studied e.g. the wives.
Sample: a representation of the population e.g. it is impossible to survey all the wives in the world, so we require a representative sample. The best way is to use a random sample.
Random Sample: everyone in your population (the target group) has the same chance of being included in the study. A random sample will represent the target population fairly and will allow generalizations.
Researchers need to establish rapport, a feeling of trust, with their respondents, especially when it comes to sensitive topics. Anonymous surveys also help in this regard.
Sociologists need to make sure that the questions are neutral in their surveys e.g. the question, “don’t you think that men who beat their wives should go to prison?” is subtly tilting the respondent to favor prison for abusers.
In surveys, you can have open-ended or close-ended questions.
Open-ended questions allow the respondents to share their full opinion but are hard to compare.
Close-ended questions are easy to compare in the study but the opinions of respondents may be lacking.
Observation (Fieldwork)
Participant observation: the researcher participates in a research setting while observing what is happening in that setting.
However, some issues you can’t just sit around and observe e.g. abusive relationships. Sociologists face the dilemma: How involved should you get in the lives of the people you are observing?
Case Studies
To do a case study, the researcher focuses on a single event, situation, or individual.
The purpose is to understand the dynamics of relationships and power, or even the thinking that motivates people.
For spouse abuse, a case study would focus on a single wife and husband, exploring the couple’s history and relationship.
However, case studies are specific and it is hard to generalize from them.
Secondary Analysis
If you were to analyze data that someone else has already collected, you would be doing secondary analysis.
For example, if you were to examine the original data from a study of women who had been abused by their husbands, you would be doing secondary analysis.
Analysis of Documents
The fifth method that sociologists use is the study of documents, recorded sources.
To investigate social life, they examine such diverse sources as books, newspapers, diaries, bank records, police reports, immigration files, and records kept by organizations.
The term documents is broad and also includes video and audio recordings.
Experiments
Experiment: the use of control and experimental groups and dependent and independent variables to test causation.
Experimental Group: the group of subjects in an experiment who are exposed to the independent variable.
Control Group: the subjects in an experiment who are not exposed to the independent variable.
Independent Variable: a factor that causes a change in another variable, called the dependent variable.
Dependent Variable: a factor in an experiment that is changed by an independent variable.
Causation: means that a change in one variable is caused by another variable. Three conditions are necessary for causation:
correlation: two variables that exist together.
temporal priority: one variable takes place before the second.
spurious correlation: two variables that exist together may not be correlated. There might be a third hidden variable.
As an example, lets experiment with a group of abusers to see if therapy works on them. Divide the group into two parts: one that gets therapy (experimental group) and one that doesn’t (control group). Therapy will be our independent variable because it may affect their behavior (dependent variable). If the wives report better behavior after therapy, you can conclude that the therapy worked and vice versa.
Unobtrusive Measures
In our technological society, we are surrounded by unobtrusive measures, ways to observe people who are not aware that they are being studied. The face-recognition cameras, tracking services, and coupons, which raise ethical issues of invasion of privacy, are part of marketing, not sociological research.
The unobtrusive measures used by sociologists are relatively primitive e.g. counting number of empty whisky bottles to gauge alcohol consumption.
Ethical Principle: To record the behavior of people in public settings, such as a crowd, without announcing that you are doing so is acceptable. To do this in private settings is not.
Gender in Sociological Research
Gender certainly can be an impediment in research.
There are certain cases such as the abuse example where abused women might respond better to women researchers than men. That’s because they were abused by men.
In the same way, men may open up more about their personal lives with a male researcher as compared to a female.
Ethics in Sociological Research
Research ethics requires:
honesty
truth
openness
no falsification of results
no plagiarism
informed respondents
anonymity of those who provide information
Humphrey’s Tearoom Example: sociologist disguised himself as a watch queen to look at people in bathroom stalls.
Leave a comment