1.2: How Do Sociologists Study Society Notes:

THE DISTINCTION BETWEEN POSITIVIST AND INTERPRETIVIST METHODS:

POSITIVISM:

• It is an approach that focuses on quantitative data (includes numbers and statistics). Remember by "+" sign so it includes math and numbers

• This approach is used by natural sciences (biology, physics, and chemistry) and goes back to sociologists Auguste Comte and Emile Durkheim

• Researchers aim to be objective/unbiased (their values and feelings should not influence the results) in order to get accurate results. However, critics seem to believe that it is impossible to be completely objective

• Positivists favor experiments but tend to use social surveys and questionnaires (all which produce quantitative data)

INTERPRETIVISM:

•  They believe that since humans are conscious beings that make their own choices, we cannot study them the same way scientists experiment on natural sciences

• Interpretivists aim to understand how people make sense of the social reality around them before understanding their actions. In other words, interpretivists have to put themselves in other people's shoes in order to understand why people behaved the way they did

• Interpretivists favor qualitative data that comes from unstructured interviews and participant observation

For example, a positivist may be interested in how many people failed an exam (statistics) while an interpretivist may be interested in why they failed an exam

THE MAIN STEPS IN DEVISING A RESEARCH STRATEGY:

1. RESEARCH AIMS AND SELECTION OF A TOPIC:

• Sociological research starts with the identification of a problem

• When deciding what topic to choose, sociologists may be driven by:

1. Their personal interests, experiences, or observations

2. What is already known and unknown about the topic

3. Social changes and developments

4. Whether funding will be available or not (since it can be costly - not just materials wise but also time). Money can come from the university, government, charitable organizations, etc.

5. How practical it will be when researching

6. Ethical issues that can arise

2. HYPOTHESIS SETTING AND REVISION:

• After choosing a topic, you will need to review the evidence available i.e. what is already known about the topic

• Evidence can be found in books or academic journals. This stage is often called literature review

• Next, you will come up with a hypothesis which is a theory/explanation at the start of the research that the research is designed to test (approve or disapprove). It can be a cause-effect relationship (example: when you heat water to 100°C, it will boil) or a correlation (if you study for 12 hours straight, you will get an A*)

3. PILOT STUDIES:

• It is a small-scale test of a piece of a research project before the main research

• Used to see if there are any problems with the research to save time, effort, and money in the long run

4. SAMPLING:

• If someone is researching about the number of people who are to pick up money dropped on the floor, it would be impractical to experiment on every person in the world. So, a sample is constructed which includes a smaller number of participants

• The people that the research is about are called the survey population

• A list of everyone in the population is called the sampling frame

• Examples of sampling frames: electoral roll (in elections), telephone directories, school registers, census (collect information about the entire population, not just a sample)

• The sample has to be representative so that the results can be generalized (findings can be applied to a larger amount of people)

EXAMPLE THAT INCLUDES ALL THE RESEARCH STRATEGIES: 

• You decide to research on whether boys do better in school or girls (selection of a topic). After doing background research and looking into known evidence (revision), you decide to prove that girls do better in school than boys (hypothesis). By setting up a quick test with both MCQ and written questions, you give it to a few students of the same year (pilot study). If the questions were phrased wrong, you will fix the issue in order to save time, effort, and money in the long run. 

• Let's say the year you are researching consists of 100 people, 60 girls and 40 boys. It is too expensive and time consuming to experiment on every person. So, your sample might consist of 10 people: 6 girls and 4 boys, 20 people: 12 girls and 8 boys, or 50 people: 30 girls and 20 boys, etc. The sample must be in the same ratio in order to achieve generalizable results.

TYPES OF SAMPLING:

1. RANDOM SAMPLING:

• Each person has an equal chance of being chosen to be part of an experiment however it may not be representative

• Example: drawing names out of a hat

2. STRATIFIED SAMPLING:

• When the sampling frame is divided by age or gender and then a random sample is taken from each frame

• In the example given above that includes all the research strategies, the sampling frame was stratified into boys and girls (60 girls, 40 boys)

3. SYSTEMATIC SAMPLING:

• When you choose names based on a pattern but it isn't random since not everyone has an equal chance of being chosen

• Example: choosing every tenth name, choosing every other name, etc.  

4. CLUSTER SAMPLING:

• Used when the population is spread over an entire country so only a few cities/rural areas are chosen as a sampling frame then random samples are taken from those

• It saves time and money

5. OPPORTUNITY SAMPLING:

• They include the people who are available during the time of the research but it is not random

• It lacks a sampling frame

• Example: when someone stops people on the street and asks them questions

6. QUOTA SAMPLING:

• Choosing a sampling frame based on certain characteristics in people and it is often used in market research (in business)

• It lacks a sampling frame

• Example: choosing 15 students who take 3 A levels

7. SNOWBALL SAMPLING:

• Involves finding one respondent and have them put you in touch with other similar respondents

• It lacks a sampling frame

• Example: when interviewing drug addicts/gangsters

DIFFICULTIES IN IMPLEMENTING A RESEARCH STRATEGY:

ETHICAL ISSUES AFFECTING THE CHOICE AND IMPLEMENTATION OF A RESEARCH STRATEGY

• Ethical issues are issues that have a moral dimension i.e. making decisions based on what's right

• Some of the main ethical guidelines include:

1. The participants must not be harmed (includes physical or emotional harm)

2. The participants' consent must be given (they must agree to the experiment and be aware of the details)

3. The researchers should respect the participants' privacy (avoid invading questions such as their earnings, religious beliefs, sexual activity, etc.)

4. The participants must not be deceived (avoid lying to the them about the purpose of the experiment)

• The researcher must also make sure that the research is:

1. Anonymous: anything that can identify the individual such as their name or any other characteristic does not appear on the survey form

2. Confidential: make it impossible to trace the individual's answers back to them

THE MAIN METHODS USED IN SOCIAL INVESTIGATION:

1. QUESTIONNAIRES AND SOCIAL SURVEYS: 

• The main types of questions include:

1. Closed/pre-coded questions: The researcher would have a question written down with several answers for the respondent to choose from. This makes it easier for the researcher to analyze the results but it limits the response creativity/accuracy

2. Scaled questions: usually "on a scale from 1-5" or "strongly disagree, disagree, neutral, agree, strongly agree". Lately, researchers avoid odd numbers since if the respondent does not understand the question/is confused, they are more likely to choose the middle value.

3. Open questions: they involve written questions and, in return, give qualitative data instead of quantitative which is harder to analyze in tables/statistically

WAYS OF ADMINISTERING SURVEYS:

1. SELF COMPLETION QUESTIONNAIRE:

• The respondents answer the questions on their own without the researcher present

• Example: postal questionnaires (sent out and returned by post) but they have low response rates

• To improve response rate: 

1. Follow up with people who do not respond + remind them to

2. Offering the respondent an incentive of money if completed

3. Sending a stamped, addressed envelope

4. Sending the questionnaire with the respondents' names to make them feel more involved

5. Making the questionnaire short and easily understandable

2. STRUCTURED/STANDARDIZED INTERVIEWS:

• It includes the same questions that would be asked on a self-completion questionnaire in the same order. The only difference is that the researcher reads the questions and choices out to the respondents then records their responses

• Structured interviews can be done through a telephone (telephone questionnaire) or face-to-face. If it is face-to-face, then the time and location must be specified to make the respondent feel at ease

GUIDELINES FOR A GOOD QUESTIONNAIRE:

1. A questionnaire should be as short and straight forward as possible

2. Instructions for completing it should be easily understood and easy to follow up

3. There should be only as many questions as necessary to obtain all the information needed

4. Start with the short questions

5. Give the participants alternative answers to express their views

6. Questions should not be leading

7. The questions should avoid words that could not be understood by everyone

8. It is best to leave personal information to be filled towards the end

RELIABILITY VS VALIDITY:

1. Reliability: when the research can be repeated and similar responses would be obtained but they are not necessarily valid

2. Validity: when the findings accurately represent the reality that it intended to capture

Example: Sociologists did a social survey with children and asked them what horror movies they had seen. Many of the responses concluded that children had in fact seen these horror movies. However, to check the validity (the accuracy), some of the movies in the list did not even exist, yet children claimed to have seen it. 

2. INTERVIEWS:

• They include:

1. Unstructured interviews: interviewer only has a few prompts and the interviewee speaks freely

2. Semi-structured interviews: interviewer has a list of questions/topics to be covered. Order can vary and questions not on the guide can be asked

3. Focus groups: a group brought together to be interviewed on a specific topic. There is an interview guide and interviewees can speak freely. It is good since they decide on what's important but irrelevant discussions can arise and some speakers may be more dominant than others

• Advantages of qualitative interviews:

1. They provide detailed and valid information

2. Flexibility of the interview allows more questions asked in order to probe deeper

3. Interviewers are able to assess the honesty and validity of the answers as they are given

4. They can bring out information for further investigation

• Disadvantages of qualitative interviews:

1. Time consuming

2. Difficult to make generalizations since standardized questions are not available

3. Less reliable since it is difficult to replicate

4. Interviewers need to be highly skilled

5. Responses can be different due to interviewer bias or interviewer effect

• These interviews differ from structured interviews since:

1. They are more flexible

2. Interviewers do not need to keep track of the questions

3. Interviewers can follow up on questions to get more details

GUIDELINES FOR A GOOD INTERVIEW:

1. Make the interviewee feel comfortable and reassure them about their anonymity

2. Create a certain amount of order

3. Make the wording clear

4. Avoid leading questions

5. Make sure the interview is conducted in a quiet and private setting

6. Use good recording equipment to transcribe the interview if necessary 

3. EXPERIMENTS:

• Usually done by positivist method (to produce quantitative data) to find cause and effect or correlations

• The aim is to test the hypothesis (either accept or reject it)

• Experiments have high level of validity

• The experiments involve manipulating the independent variable and noticing the change in the dependent variable (remember by the fact that manipulating has an "i" just like in independent) while all the other factors are kept constant

• Experiments can be either:

1. Laboratory Experiments: the experiment takes place in a lab where all possible external factors are eliminated. However, this is deeply flawed since people live in societies and not labs controlled by researchers thus their behavior will change. This changed behavior is known as the Hawthorne/Observer Effect i.e. when people's behavior changes due to the researcher's presence

2. Field Experiments: the experiment takes place in a natural setting in the real world and is often used by non-positivists. Although it is a very effective way of getting inside a group and understanding their behavior, it presents ethical issues, it is harder to control, and involves a lot of risk

4. CASE STUDIES:

• It is a detailed, in-depth study of one group or an event

• Can involve any method/combination of methods like quantitative and qualitative

• Advantages:

1. Allows different aspects of the case to be explored using appropriate methods

2. They can provide deep and detailed account of the case

3. If the case is carefully chosen it can be possible to draw wider conclusions

4. It can produce findings that can be tested by other researchers elsewhere

• Disadvantages:

1. The findings may only apply to the case so generalizations cannot be made

2. The findings cannot be replicated

3. Deep involvement of the researchers in the case can lead to their feelings influencing the findings

5. LONGITUDINAL STUDIES:

• They are surveys taking place at intervals over a long period of time

• Often used by government-funding organizations to ask basic questions about lifestyle, health, etc.

• Panel studies are examples of longitudinal studies where the same people are used for the research

• Advantages:

1. Shows how people's lives change over time in a movie-like manner

2. It is possible to show what factors change people's lives

3. Valid data can be achieved since the participants are committed

• Disadvantages:

1. Needs commitment as it is time consuming

2. There is an inevitable drop out rate during the research (also known as sample attrition)

3. Being a part of this research can change the participants and the Hawthorne Effect can be present

6. PARTICIPANT OBSERVATION:

• It is used to understand the world from the subjects of the experiment's perspective

• Participant observation involves joining a group and living life the way they do. There are 3 stages of participant observation:

1. Getting in: it happens when the researcher joins a group. S/he can adopt an overt role which involves being open about your identity and letting the subjects know you're researching them or a covert role which involves covering your identity and intentions. Covert usually used when studying dangerous people (drug dealers, etc.)

2. Staying in: the observer will have to develop a role that will gain the trust and cooperation of those being observed. This can involve participating in their same activities 

3. Getting out: includes leaving the group without breaking/ruining relationships and making sure the members of the group cannot be identified

7. NON-PARTICIPANT OBSERVATION:

• When a researcher observes a group but does not take part of their actions

• Can be used when groups are unwilling to cooperate in the research

• Advantages:

1. Eliminates the Hawthorne Effect

• Disadvantages:

1. Researchers will have to assume the meanings of activities that people do which can raise issues over validity and reliability

8. CONTENT ANALYSIS:

• It is a method of studying communication and the media which includes classifying content and counting frequencies

• Example: counting how many times a teacher said "okay" in a class

• Advantages:

1. It is reliable

2. Avoids ethical issues

3. Provides information about content media in statistical form which can be used to test sociological theories

• Disadvantages:

1. Produces quantitative data because numbers do not give us the reason why a media text is the way it is

2. It can be difficult to decide what categories to use

9. TRIANGULATION:

• The use of two or more methods in the same research project

• Advantages:

1. It can allow researchers to support quantitative data with qualitative data which provides validity and reliability

2. Can be used to check validity and reliability

3. Cross-referencing can be used to check the accuracy of the data

4. Provides a balance between methods

• Disadvantages:

1. Using several methods is time consuming and expensive

2. The researcher needs to be skilled in several research methods

3. Positivist and interpretivist methods have different approaches so it may be difficult to combine both

THE IMPORTANCE OF ANALYZING AND EVALUATING RESEARCH:

1. VALIDITY:

• Refers to the accuracy of the research that reflects reality

• Examples of methods high in validity would include participant observation and unstructured interviews

• One can say, in a scientific manner, that validity and reliability are inversely proportional i.e. when one increases the other decreases

2. RELIABILITY:

• Refers to the extent where the findings of the research can be confirmed by repeating the study

• Examples: science experiments and surveys if similar results were obtained

3. REPRESENTATIVENESS:

• The degree to which research findings about one group can be applied to a larger group/similar groups

• This means that the sample used should be of the same proportions of society (like the example given about the girls being smarter than boys above)

• Sampling error i.e. the difference between the results of the sample and the results of society can occur but can be reduced by having a large random/stratified random sample

4. RESEARCH BIAS:

• Bias may come from the researcher's values such as their religious or political views

• Positivists believe that researchers should be neutral however interpretivists believe this is impossible because no one can be completely unbiased

• The way researchers can influence the findings is known as imposition problem which can happen through social characteristics like age, gender, etc.

NOTES DONE BY FARIDA SABET

CLICK HERE TO GO TO THE PREVIOUS TOPIC

CLICK HERE TO GO TO THE NEXT TOPIC

CLICK HERE TO GO BACK TO THE NOTES MENU