Topic 3 - Experiments and MIC Flashcards
Laboratory experiments
- An experiment is a form of observation where variables are controlled in order to tesg hypothesis. A lab exmperiment takes place in an artificial envrionment where most variables can be effectively controlled
- Examples: The Stanford Prison Experiment, Milgrim’s Obedience Experiment, Bandura’s Bobo Doll Experiment
Practical Issues
- Open systems: Keat + Urry - lab experiments are only suitable for studying closed systems. Society is an open system - impossible to researchers to control the variables
- Individuals are complex: not possible to match members of the control and experiment group exactly
- Cant be used to study the past
- More useful for small samples: difficult to investigate large scale social phenomena
- The Hawthorne Effect: behaviour that occurs is artificial, they know they are being observed
- Expectancy effect: experiment bias, reseracher can influence outcome with what they expect is going to happen.
Ethical Issues
- Informed consent
- Harm to subjects: some may argue that minor harm may be justified ethically if the results yield significant social benefits
Theoretical
- Positivists: favour lab experiments due to reliability
- Interpretivists: lab experiments lack validity
Reliability and hypothesis testing.
- Positivists say they are hihgly reliable as they can be tested time and time again, can be easily repeated through the same steps
- Produces quantitative data
- Detatched and objective. Variables are manipulated.
- Researchers subjective feelings have so effect on outcome.
Representativeness
- Positivists see representativeness as important in order to make generalisations about wider social structures
- With lab experiments, you cannot be sure it reflects the wider population
- Small samples = unrepresentative, findings cannot be generalised beyond experiment
Validity
- Lack of external validity due to high level of control = less like the world outside of a lab situation
- Artificiaity may encourage the hawthorne effect: subjects react to being studied and produce invalid data
Interpretivism and free will
- Humans are fundamnetally different to subjects studied by scientists
- Humans have free will and choice
- Actions can only be understood by the choices we make based on the meanings we give to events
- LAB EXPERIMENTS = fundamentally innapropriate method to study human beings
Lab experiements and teacher expectations
MIC
- Harvey and Slatin
- Charkin et al
- Mason
Harvey and Slatin (1976)
- Examined whether teachers had preconceived ideas about pupils’ social class
- Sample of 96 teachers - each of them were shown 18 pics of children from different SC backgrounds
- Pics were equally divided in terms of gender/ethnicity (control variable)
- They were asked to rate children on performance, parental attitudes to education and aspirations
- WC children = rated less favourably
- Study indicated teachers label pupils from different SC backgrounds
Charkin et al (1975)
- Used a sample of 48 university students who taught lessons to a 10 year old boy
- 1/3 were told that the boy was highly motivated and intelligent
- 1/3 were told that the boy was poorly motivated with a low IQ
- 1/3 were given no information
- High expectancy group made more eye contact and had more ecnouraging body lang than low expectancy group
Mason (1973)
- Looked at whether pos/neg expectations had a great effect
- Teachers were given pos/neg/neutral reports on a pupil
- Teachers observed pupils taking a test to see if any errors were made
- They were able to predict the pupils EOY attainment
- Mason found that negative reports had greater impact on teacher expectations
Practical issues
- Schools are large institutions that have many variables that may affect teacher expectation
- Impossible to control the variables that might influence teachers expectations
Artificiality
- Lab experiements are articifical
- Charkin: used uni students not real teachers
- Harvey and Slatin: used pics of pupils rather than real pupils
Ethical issues
- Don’t involve real pupils: fewer ethical issues
- Mason/Harvey and Slation: no use of real pupils
- Charkin et al: used real pupils
Narrow focus
- Only examine one aspect of teacher expectations such as body lang
- Useful: allows researcher to examine this variable more thoroughly
- Not useful: teacher expectations not seen within wider process of labelling and SFP - Charkin didnt investigate how pos/neg body lang affected pupils’ performance
Field experiments
- Much more likely to be used in sociology than lab experiments.
- They are more valid but risk being unreliable because of the problem of controlling variables
- Takes place in objects natural surroundings
- Those involved dont know they are subjects of an experiment
Actor test and correspondence test examples
- Brown and Gay: studied racial discrimination in employment. Sent white and black actor for interviews for the same posts. Subjects were matches in every other way other than ethnicity
- Wood et al: Similar study, sending applications to over 1000 jobs apprently from three applicants of different ethnicities
Advantages/Limitations of actor and correspondence tests
ADVANTAGES
- More natural and valid for real life
- Avoids artificiality
LIMITATIONS
- Less control over variables: cannot be certain of the true cause
- Unethical: experiment is carried out without the knowledge and consent of subject
Comparative method
Famous study
- Durkheim’s stduy of suicide: analysed official stats
- Carried out only in the mind of the sociologist: thought experiment
- Not experimented with real people
- Re analyses secondary data
- Still designed to discover cause and effect relationships
Advantages/Limitations of comparative method
ADVANTAGES
- Avoids artificiality
- Used to study past events
- Avoids ethical problems or harming/deceiving subjects
LIMITATIONS
- Less control than field experiments
Field experiments and teacher expectations
MIC
Rosethal and Jacobson
- Pupils were given an IQ test
- Teachers were then told they identified 20% of pupils who were likely to ‘spurt’, however, pupils were actually chosen at random
- Rosethal and Jacobson had two aims: plant a particular set of expectations in teachers mind about their pupils and see if it had an effect on pupils performance
- Teacher’s expectations were therefore identified as the independent variable
- Pupils retested 8 months later
- 1st 8 months = average of 8 IQ points, ‘spurters’ = average of 12 IQ points
- Greatest improvement found in youngest children, however after 12 months only seemed to effect 10-11 y/o’s
Ethical issues
- ‘Spurters’ benefited from the study. 80% did not
- Potentially could have been held back academically due to teacher expectations, lack of attention and encouragement
- Unlikely to be carried out today where children have more rights and schools duty of care
Reliability
- Relatively simple research design = easy to repeat: within 5 years of the original study, it had been repeated 242 times
- However, given the differences between school classes it’s unlikely that the original could be replicated exactly
