Sociology: Choice of method and the research process

- Argue that there is a measurable, objective social reality.

- Are interested in cause-and-effect relationships.

- Aim to find the cause of behaviour, patterns, etc.

- They believe sociology is a science.

- They prefer standardised research methods -> questionnaires, structured interviews, etc.

- They want data that is reliable and representative.

- They want quantitative data. 

- 'Interpreters'.

- They want to find and understand the meanings social actors (people) give to events.

- They use unstructured methods -> participant observations, unstructured interviews, etc.

- They want data that is valid.

- They want qualitative data.

Positivists -> science students who adore measuring things.

Interpretivists -> English students who love analysing quotes

Quantitative:

- Numerical

- Can be measured

- ♡  by positivists 

Qualitative:

- Words, thoughts, feelings.

- Cannot be measured.

- ♡ by interpretivists 

Some sociologists see advantages in both types of data.

Triangulation combines quantitative and qualitative methods.

The strengths of one method balance out the weaknesses of the other.

Sociologists use three key concepts to decide whether a method is useful:

- Reliability 

- Representativenss

- Valditity 

- Reliability;

A reliable method can be replicated to achieve the same results.

Standardised methods e.g questionnaires.

Favoured by positivists. 

- Representativeness;

Sociologists cannot study entire target populations as they are too big, so they study samples.

A representative sample mirrors the characteristics found in the larger group.

This allows generalisations to be made.

Favoured by positivists as it allows them to find patterns and causes.

- Validity

Valid data is authentic, it shows what something is really like.

It is gathered through unstructured methods e.g participant observation.

Favoured by interpretivists as they aim to find out what people really think.

Primary data:

Evidence collected by the sociologist themself for their own sociological purposes.

Collected through questionnaires, observation, interviews, etc.

Strength: can be tailored to discover exactly what the researcher is exploring.

Weaknesses: can be expensive + time-consuming + hard/ impossible to gather.

Secondary data:

Evidence collected by someone else for their sociological purposes can be used by the researcher.

Strengths: is usually free, is not time-consuming, can be on a topic the researcher cannot access themself, can span a large group/ long period of time.

Weaknesses: may not be tailored to what the researcher is exploring, may not be accessible.

Three main factors influence a researcher's choice of method:

- Their methodological perspective

- Practical factors

- Ethical concerns 

Positivists aim to find cause-and-effect relationships.

They favour structured methods that produce representative and reliable quantitative data.

E.g surveys, questionnaires, etc.

Interpretivists aim to understand people's thoughts and feelings.

They favour unstructured methods that produce valid qualitative data.

E.g unstructured interviews, participant observation, etc.

Time -> some methods are more time consuming than others.

Finances -> some methods are more expensive than others

Source of funding -> the sponsor may request a specific method to be used.

Personal factors -> e.g the researcher may have children and thus be unable to take part in lengthy or dangerous observations.

Research subjects -> some groups are harder to access than others e.g a criminal is unlikely to fill in a survey admitting to their crimes.

Research opportunities -> sometimes researcher might get lucky and not have time to prepare structured methods, thus they may resort to using unstructured ones.

Danger -> some methods are risky e.g covert participant observations. 

Research can have a negative impact on people's lives, sociologists must consider this and follow guidelines when researching.

Informed consent -> the sociologist must receive the fully informed consent of all individuals informed. People should not be manipulated or mislead about the research.

Confidentiality -> people must remain anonymous when studies are published to ensure that they are not at risk e.g of losing their job or being attacked.

Effects on research subjects -> sociologists must ensure that research does not harm participants physically or mentally. 

Vulnerable groups -> special care must be taken when researching vulnerable groups- e.g may be at greater risk of harm or struggle to provide fully informed consent. People may be vulnerable because of their age, gender, ethnicity, state of health.

Covert research -> raises many ethical questions are participants do not consent and are manipulated. Some argue it is justified if access is otherwise impossible e.g into criminal groups. 

Choice of research topic is influence by several factors:

Practical factors -> some topics are harder to study than other e.g high-level political decision making is likely to be inaccessible to the sociologist.

Funding bodies -> sponsors may choose the research topic.

Society's values -> changes in attitudes, values and opinions in society change the choices of research topics.

The sociologist's theoretical perspective -> this will affect what topics they choose to study e.g feminists will likely research gender issues whereas a Marxist would focus on class conflict.

Chance -> sometimes the sociologist may get lucky and find a topic e.g hospitalisation may allow to study hospital wards. 

Aims = most studies have a general aim or try to test a specific hypothesis.

An aim is a statement that states what the sociologist is planning to research.

A common aim is simply to collect data on a specific topic.

Hypotheses = other studies plan to test one (or more) hypotheses.

Hypotheses are more specific than aims.

A possible explanation for something is tested by collecting evidence to prove whether it is true or false.

Operationalising concepts = this takes place before research begins.

The sociologist has to define sociological concepts in a way that can be measured (this is called operationalisation).

E.g 'educational achievement' may be defined as having passed 5 GCSEs. 

Different sociologists define ideas differently and this can make it harder to compare findings.

The pilot study = a test study on a small group before the actual research takes place.

Used when conducting research via survey methods.

This is done to clarify questions and wording before money and time are dedicated to the actual study.

It is hard to study an entire target population due to: its size, lack of time, money, resources, etc.

Instead, sociologists study a sample of this group = a smaller part of the whole group.

Sampling frames = a list of all the members in a research population. From these people, a sample is chosen.

e.g of SF -> an electoral register.

Representative = must be a cross-section of the whole group (have characteristics of the whole group).

Positivists believe it is important that samples are representative as then generalisations can be made about the entire target population (then patterns and causes can be found)>

Some samples are too small to be representative.

Interpretivists do not need representative samples as they are not trying to find causes for behaviour, they're trying to understand the meanings of social actors.

If the sample is not selected from a complete and up-to-date sampling frame, it is not representative.

If it is not possible to obtain a representative sample, sociologists use:

Snowball sampling - it is collected by contacting key individuals who then suggest others that can partake. This continues until a group is formed and is not representative.

Opportunity sampling - 'convenience sampling' is when a researcher chooses from individuals that are easiest to access e.g people on the street or children in a classroom.

Random sampling -> everyone on the sampling frame has an equal chance of being chosen. It eliminates bias when choosing. If large enough of a group is formed, it can be representative however not all random samples are always large enough.

Systematic or quasi-random sampling -> every nth (e.g every 3rd) person in the sampling frame is chosen. This eliminated bias when choosing.

Stratified sampling -> the sociologist stratifies (breaks down) the frame by age, class, gender, etc and then creates a group based on the same proportions. e.g if 20% of the target popular is female, 20% of the sample also has to be female.

Quota sampling -> the target group is broken down and the sociologist is given a quota (e.g 10 men, 3 children aged 6, etc.) and they continue picking until the quota is filled.