PRIMARY RESEARCH METHODS

  • QUALITATIVE DATA - non numerical, text, videos, photos, audio recording, collected using diary accounts and interviews and analysed with a grounded theory and thematic analysis

  • QUANTATIVE DATA - numerical data, statistically analysed, experiments, observations, correlations, closed rating scales, questionnaires

  • PRIMARY DATA - experiments, interviews, questionnaires (self report techniques), observations

  • SECONDARY DATA - historical documents, personal documents (letters/diaries), government statistics, media content, research by other sociologists

  • HYPOTHETICO-DEDUCTIVE METHOD - standard now, created by Karl Popper who decided how we should use the scientific method. 1) Choose a topic, issue or problem to negotiate. 2) Gather background research. 3) Develop a hypothesis or question, 4) Decide on methodology and sampling frame. 5) Conduct a pilot study. 6) conduct your research and collect data. 7) Analyse the data. 8)Draw conclusions. 9) Evaluate the study.

  • Pilot studies must contain different people to the actual study and can be used to check for issues, material as well as the dependent or independent variable. They are small scale and they check for issues before the expensive main study, they ensure the research works in practice, ensures the data collected is useful.

  • needed and generally check how the research will be conducted.

  • Factors affecting choosing a topic: personal interests, political and theoretical perspectives, funding, society and in vogue topics, ethics and sensitivity of spirit.

  • Gathering background research is important as it helps to formulate a hypothesis, find out what has been done before so they can add to it, helps to determine research methodologies

  • A research question sets out plans to investigate and a clear focus and a hypothesis is an informed guess, a written statement which is then tested and then either supported or refuted. Nothing can ever be 100% proven due to the fluid and changing nature of people, societies and time.

  • Methodology is which research method you are going to use to collect your data.

  • A sampling frame is how you are going to choose your dta (eg random, opportunistic, stratified, etc)

  • A sample of a small part or quantity intended to show what the whole is like, they an’t use everyone as it would take too long this a sample is representative

  • PET (practical, ethical, theoretical considerations)

QUESTIONNAIRES

  • a set of written questions that participants fill in for themselves, can be web or paper based. A form of self report techniques and participants report on their own feelings and opinions.

  • Self report techniques allow for a respondent to report their own experiences directly, qualitative data can be gathered through open questions. These techniques increase validity as they should collect information that is a true refelction of participants, thoughts, behaviours and opinions.

  • CLOSED QUESTIONS - numerical data, have a fixed choice of response, likert scales (1-5), rating scales and fixed choice options, facts, easy to answer, quick to answer, keep control of the conversation with the questioner

  • OPEN QUESTIONS - non-numerical data, no fixed response and respondents can answer as they wish, think and reflect, gives you opinions and feelings, control of the conversation to the respondent.

  • social desirability bias can warp data as questions are answered incorrectly as they modify answers in order to be seen in a better light.

  • RESPONSE BIAS - in a pattern of answering the questions in the same light

  • VALIDITY OF QUESTIONS - questions should be operationalised so they are clear for all respondents (increase face validity). Pilot studies are needed to check for issues, avoid confusing double negatives, avoid double barreled questions, avoid overcomplicated language (jargon), the way that questions are written can also introduce bias and create leading questions, emotive language also suggests authors views (bias) and can introduce social desirability bias

  • STRENGTHS - easily accesible, quick to make, can be representative (depending on sample), cheap, quick and easy to complete, if coded answers can be fairly easily collated and summarised (eg charts), huge amounts of data can be collected fairly quickly, if worded carefully data should be reliable and representative

  • WEAKNESSES - questions may be difficult to operationalise, SDB can occur, low response rate, can constrain/distort the wat respondents answer, respondents can be influenced to answer in a certain ways by the researcher/question, postal questionnaires have a high level of people who decline to answer (can only go to people with addresses), questions may be interpreted by responders in ways the researchers did not foresee or intend

  • PERVERT (practical issuess, ethical issues, relibaility, validity, examples, representativeness, theoretical issues)

SAMPLING

  • POPULATION - an entire group with specified characteristics

  • TARGET POPULATION - is the desired population’s subgroup to be studied, and therefore want research findings to generalise to. A target group is usually too large to be studied in its entirety so sampling methods are used to choose a representative sample from the target group.

  • SAMPLE - a group os people who take part in a research investigation drawn from the population

  • REPRESENTAtIVE SAMPLE - a subset of the target group with a similar distribution of relevant characteristics, in turn allowing us to generalise from the sample to the target group with some justification. An unrepresentative sample is one that does not reflect the distribution of characteristics of the target group, cannot be generalised and thus is bias

  • BIAS - when a particular group of people are under/over represented in your sample, amkes it hard to generalise findings

  • SYSTEMATIC SAMPLING - a systematic method is chosen for selecting from a target group (eg every fourth person in a list is used in the sample (often representative, fair and biased, cost effective but can can lead to small biased sample groups)

  • STRATIFIED SAMPLING - sample divides people into groups (each with a key characteristic) which should be present in the final sample, then each of those sections is sampled individually (unbiased, highly effective but cost and time ineffective)

  • OPPORTUNITY SAMPLE - participants who are both accesible and willing to take part are targeted ( easy to access, quick and affordable but often biased and not representative)

  • VOLUNTEER SAMPLE - sample consists of people who have volunteered to take part in the study

  • QUOTA SAMPLING - population is stratified, as it is for stratified sampling and then the researcher is given a quota

  • SNOWBALL SAMPLING - you contact someone to interview, they contact someone elso, who contects someone else etc (good for contacting people who would be hard to contact otherwise, not representative and small sample)

  • RANDOM SAMPLING - each person has an equal chance of being selected (often representative, more unbiased and cost effective but small sample can lead to bias)

POSITIVISM AND INTERPRETIVISM

  • POSITIVISM - prefer quantative methods as it is scientific and objectivem eg Durkheim. Like to use questionnairres, structured interviews, experiments and social statistics. Wish to uncover the general laws about behaviour and use quantitave methods to study large numbers of people in a reliable and representative way.

  • INTERPRETIVISM - prefer qualitative data, questionnaires with open questions, case studies, participant observation, eg Weber and Verstehen (standing in someone else’s shoes). Wish to understand with individual people’s experiences. gaining in depth insights into their lives using qualitative methods for greater validity and better rapport.

  • both need to consider whether research is theoretically reliable and valid, researcher atittudes will skew this idea.

  • reliability: can the research be replicated effectively

  • validity: are the findings telling the truth about the situation

  • SOCIAL FACTS - concepts and institutions in society that are scientifically and objectively verifiable (according to positivits). Durkhiem said that social facts ‘consist of manners of acting, thinking and feeling external to the individual with a coercive power by virtue of which they exercise control over him’. Eg Durkhiem’s study of suicide, he researched official statistics on suicide in several European countries and found that the suicide rate was influenced by social facts such as divorce rates, religion and the pace of social and economic change, he argued that the suicide rate increased when there was too much or too little intergration or regulation.

  • HISTORY OF POSITIVISM - emerged out of the industrial revolution and the enlightenment (1650-1800, authority of the church being in decline and was the start of modern science. The core beliefs were: underlying laws explain how the world and sociology works, all men could understtand this, aws could be applied to society to improve it.

  • BIRTH OF SOCIOLOGY - 18-19th century, provided a new discovery in physics, chemistry and biology which led to new technology and industrialisation. This caused a social transformation due to urbanisation, and thus the growith of marxists, hundreds of thousands of people flooded to expanding cities (eg Manchester), with new urban centres being plagued with new social problems such as poverty, unemployment and social unrest. Comte is considered the founder of sociology.

  • POSITIVIST WEAKNESSES - treats individuals as if they are passive, people’s subjective realities are complex and demands in depth qualitative methods, ‘social facts’ may be invalid, provides a detached, shallow understanding of human behaviour

  • INTERPRETIVISM AND OBJECTIVITY - will always have a form of bias, it will inevitably be viewed differently, social reality is constructed through the meanings and interpretations individuals give to their actions and interactions. Meanings are subjective and vary between people and contexts thus researchers must understand them from the perspective of their participants. Interpretivists argue that researchers are also a part of the social world and this they use their own values and assumptions in their research process.

  • According to Interpretivists, lab experiments are not suitable for the study of society as the social world differs to the natural world. People are not chemicals, objects or forces they behave differently and are more complex with everyone having different opinions and thoughts.

  • Interpretivists also take a micro-approach to research, in order to gain a clearer more ind depth understanding to why they behave in the way that they do, but this can be time consuming and cost ineffective, argue society is constructed from the bottom up through everyday interactions

  • Interpretivists also argue that sociological research cannot be objective, as there is always some form of bias through the researcher and the participants with SDB. Individuals are also not the same and ever changing so research will inevitably be viewed in different ways, people have free will so may not participate in the norms and values of society (so can’t generalise), social reality is constructed through the relaity constructed through meanings and interpretations.

  • INTERPRETIVIST WEAKNESSES - a small number of people makes the sample unrepresentative, not reliable as its hard to replicate the evidence, verstehen sociology assumes that all individuals engage in rational behaviour, understanding leads them to believe that individuals understand the motivation behind their own actions, interpretations of events can be biased and lack objectivity

  • INTERPRETIVIST STRENGTHS - gives us a greater depth of understanding of motivations for behaviour and how people experience the world, examine micro-approaches, implies individuals have complete agency, enables is to see how socoal reality is controlled through meanings and negotiations

OBSERVATIONS

  • eg Rosenhan ‘being sane in insane places’. Looked at diagnoses of schizophrenia in the 1960s, observations and a field experiment, was naturalistic (in the mental health institution) covert and participant

  • observations are non-experimental as they do not test an IV

  • NATURALISTIC OBSERVATIONS - behavipur is studied in a natural setting, kept the same as it normally is

  • CONTROLLED OBSERVATION - variables are controlled, participants are likely to know they are being observed, may occur in a lab

  • PARTICIPANT OBSERVATION - researcher is involved in the behaviour being studied

  • NON-PARTICIPANT OBSERVATION - researcher is not involved in the behaviour being studied

  • OVERT OBSERVATIONS - participant is aware of the study, increased ethical but often creates demand characteristics or SDB

  • COVERT OBSERVATIONS - participant is unaware they are being studied (less ethical but more valid

  • event sampling us when an observer record the number of times a certain behaviour occurs

  • inter-observer reliability (trained, 2 or more, clearly defined behavioural categories) measures the consistency, agreement, or correlation between two or more observers recording the same behavior or event.

  • STRENGTHS - don’t need specialist equipment, low researcher skill (if have a good behavioural tally chart), naturalistic is likely to be cheaper, covert is easy to gain consent, non participant of controlled is not as decieving, naturalistic has higher ecological valifity, less demand characteristics, covert has lower hawthorne effect, gain detailed data,

  • WEAKNESSES - time consuming, may need multiple observers (expensive), covert is likely to involve deception, if participant then may involve law breaking, may invade privacy, not always representative (hard to have a large study), , may be hard to replicate, overt may have SDB or hawthorne effect

INTERVIEWS

  • structured interviews - interviews follow an interview schedule (a set list of questions that the interviewer sticks to) cost effective, little training needed, produce easy to analyse data (can check is reliable), quicker, gather in depth qualitative data is open question used, little training needed, not possible to deviate from interview schedule so may have limited info

  • unstructured interviews - more like a conservation, on a theme with the interviewer asking questions as they are a part of the conversation. SDB may not occur as its more like a natural conversation, or may occur as they want the interviewer to like them. Also more valid but can increase in demand characteristics and SDB, allows more in depth info, develop rapport, expensive, tme consuming, not easy to replicate, easy to go off topic

  • in real life, most are semi-structured with an interview schedule that they can deviate from and can ask follow up questions

  • CLOSED QUESTIONS - fact given easy to answer, quick to answer, keep control on the conversation with the questioner

  • OPEN QUESTIONS - respondents can think and reflect, opinions and feelings, control of the conversation is with the respondent

  • interviewer bias, notably unstructured, less so in planned interview schedule in structure. Interviewers expressions, body language and tune can also influence answers, or the interviewer may identify too closely with the group being studied.

  • May also struggle with artifiiciality, even relaxed its still not a ‘normal’ conversation’ and people may behave differently if they are aware they are being studied.

  • Status and power inequalities between the inteviewer and interviewee may affect the interviewee’s willingness to answer honestly. - RICH showed thatnwhen adults interviewed children, the childrens need to please the interviewer affects the answer. Gender and ethnic differences can also affect interviewee relationships. GRIFFIN abadoned interviews for participant observations due to this.

  • FEMINIST CRITICISMS OF INTERVIEWS - GRANHAM - questionnaires and structured interviews are patriachal and give an invalid picture of womens experiences, Researcher is in control of the interview and decides the line of questioning which mirrors womens subordination in society, survey methods treat women as isolated individuals ignoring the context of the power relations that oppress them. Fixed choice questions impose the researchers categories on women, meaning they can’t express their experiences of oppression, meaning results wil conceal power imbalances.

  • BUT structured interviews don’t reveal how the interviewer feels about the situation.

  • GRANHAM - recommends using direct observation to allow researcher to understand womens behaviours, attitudes and meanings.

  • Other feminists favour unstructured interviews so researchers can build equal and collaborative relationships wth interviewees based on trust and empathy.

  • IMPROVING THE VALIDITY OF INTERVIEWS - KINSLEY asked questions about sexual behaviour very rapidly so that interviewees would have less time to think and lie their way around. Follow up interviews can be used to check the reliability of answers. BECKER used aggression, disbelief and ‘playing dumb’ when interviewing Chicago schoolteachers to extract sensitive information that they wouldn’t have otherwise revealed (need special skills so hard to replicate). Ensure the interviewer and interviewee match via class and ethnicity avoid power imbalances. Eg NAZROOS survey of health of ethnic groups was carried pout in the interviewee’s language of choice

  • OAKLEY AND THE SOCIOLOGY OF HOUSEWORK - done at a time when sexism was not a widely recognised subject (1974). Conducted 40 in-depth interviews with London housewives, they took 2 hours and she tape recorded the conversations, sample came from 2 seperate parts of london (when predominantly m/c, the other w/c), selected from medical records from 2 general practices, housewives were between 20-30 and had at least one child under 5. New idea of hers - qualitative data and value laden ideas

EXPERIMENTS

  • field are like lab experiments, they look to manipulate the IV and measure the DV, unlike Lab experiments, unlike lab experiments other (extraneous variables) are not controlled as the experiment is conducted in the ‘field’ (eg Rosenthal and Jacobsen’s pygmalion in the classroom).

  • STRENGTHS - observed in natural environment (less chance of the Hawthorne effect influencing their behaviour) and have a higher ecological validity (can be replaced), quantative data produced (easy to analyse), lab experiments can have fully informed consent, ecological validity, decreased hawthorne effect in field experiments, high internal validity, can control variables, positivists like them

  • LIMITATIONS - extraneous variables are factors that you can’t contol, can change result (lower validity) but also means that it is more realistic, cost of monitering the subjects is expensive as scientific equipment is require to record subjects behaviour, extranous variables means it cannot produce cause and effect relationships, only causation, ethical issues also arise between deception and informed consent. time consuming, inflexible, require skilled researchers, decreased validity is they know they are taking part,

  • HAWTHORNE EFFECY - behaving differently as you know you are being observed, which undermines validity, occurs due to human agency

  • ECOLOGICAL VALIDITY - setting in experiment mirrors real life

  • Experiments are not often used: tricky to control variables, hard to predict social trends, don’t ascertain feelings, interested in people’s real life social behaviour so experiments may mot be the best method to do this with, doesn’t work well with large populations (low generalisability)

  • gain control over variables, precise measurements, test hypothesis, establish cause and effect relationships

  • lab experiments are where the research takes place in an artificial setting and the variables are manipulated by the researcher

  • INDEPENDENT VARIABLES - the variable we manipulate deliberatly changing it to see what the effect is on the dependent variable

  • DEPENDENT VARIABLE - the variable that we measure to see whether the independent variable has had an effect on it, to make the expirement fair we need to control factors that may have affected the dependent variable unitentionally

  • EXRANEOUS VARIABLES - factors that could effect the the dependent variable unintentionally as a child.

  • CONFOUNDING VARIABLES - factors that have effected the dependent variable unintentionally, extranous will become confounding if not controlled, which ruins the validity of the experiment.

  • EXPERIMENTAL GROUPS - in an experiment we compare an experimental group to a control group to make sure that the independent variable is the factor that changes the dependent variable. The experimental group is the one we subject to the manipulated independent variable, the control group you keep any varibales constant and would give a plcaebo to.

  • Lab experiments aren’t used often in sociology (as they are considered artificial and can’t be used to measure large shifts in social trends. However, some social psychologists have used lab experiments to successfully investigate social behaviour, including Milgram’s obedience studies and Bandura’s experiments into the effects of the media on aggressive behaviour in children.

  • BANDURA - interested in studying the effects if violent images on the media on the behaviour of children, in one of his experiments he divided a group of kids into 4 groups. 1) saw real-life adults using a mallet to attack an inflatible bobo doll. 2) saw a film of adults attacking the doll. 3) Saw a film of a cartoon character attacking the doll. 4) Saw no violence at all. The kids were then placed in a room for 20mins with a doll, similar to the one in the image shown in the first 3 groups. The findings in this study were that agression towards the dol was higher in each of the groups hat had experienced prior violent imagery.

  • MILGRAM - participants in the study were instructed to adminster electric shocks to a learner, even when that obedience caused harm to the learner. The results of the study showed that the majority of participants continued to administer shocks to the maximum level, even when they believed that the shocks were doing serious harm. The study has been heavily criticised for lacking ecological validity and causing psychological distress to participants, as well as decieving them,

  • FIELD EXPERIMENT - PYGMALION IN THE CLASSROOM - Rosenthal and Jacobsen (1968) studied low educational achievement by mexican children. They found that when teachers were told some pupils who had been randomly chosen were ‘intellectual bloomers’, this caused the teachers to treat them differently and they performed better at the end of the year. By introducing a relatively controlled element into the classroom (interaction between teachers and pupils) they were able to test their hypothesis and demonstrate that the expectations held by teachers about their pupils was a significant factor in the intellectual development of those pupils

  • HEUSSENSTAMM - 1971, a sociology professor at california state had a theory about police prejudice. She chose 15 students from 3 erthnic backgrounds (black, white and latino. She chose students who routinely drove to and from campus along the Los Angeles freeway routes, and who’d had perfect driving records for longer than a year. Those were her control variables, these students also all had safe, up to date cars and signed a pledge to drive safely. Next she placed a black panther bumper sticker on each car, the first citation (for an incorrect lane change) was made two hours after the experiment began. One participant was pulled over 15 times in 3 days and quit the study. After 17 days the 15 drivers had collected a total of 33 citations and the funding to pay traffic fines had run out, the experiment was halted.