Methods in Context

Harvey & Slatin (1976)- Examined whether teachers had preconceived ideas about pupils of different social classes. Used a sample of 96 teachers and each teacher was shown 18 photographs of chuldren from different social class backgrounds. These photopgraphs were equally divided in terms of ethnicity and gender. Teachers were asked to rate their children on their performance, parental attitudes to education, aspirations and so on. Found that lower class children were rated less. Showed that teachers label pupils from different social classes and use these labels to pre-judge pupils' potential.

Charkin et al (1975)- Used a sample of 48 university students who each taught a lesson to a ten-year-old boy. One third were told that the boy was highly motivated, one third were told that he was poorly motivated with a low IQ and the other third were given no information. Charkin et al found that those in the high expectancy group made more eye contact and gave our more encouraging body language than the low expectancy group.

Ethical Problems- Those that do not involve real pupils have fewer ethical problems. Harvey and Slatin's for example did not use real pupils so no child suffered any negative effects. However, others, such as Charkin et al have used real pupils and this raises ethical concerns. Their vulnerability and limited ability to understand what is happening means there are greater problems of deception, lack of informed consent and psychological damage.

Narrow Focus- Usually only examine one specific aspect of teachers expectations such as body language. Can be useful because it allows the researcher to isolate and examine this variable more thoroughly. However, this means that teacher expectations are not seen within the wider process of labelling and the self-fulfilling prophecy.

Practical Problems- Schools are large, complex institutions in which variables may affect teachers expectations such as class size, streaming and type of school. It is impossible even to identify, let alone control, all the variables that might exert an influence on teachers expectations.

Artificality- They tell us little about the real world of education, for example: Charkin used university students rather than teachers, Harvey and Slatin used photographs rather than real pupils. 

Rosenthal and Jacobson (1968)- Pygmalion in the Classroom, carried out their research in a California primary school. Pupils were given an IQ test and teachers were told that this had enabled the researchers to idenitfy the 20% of students who were likely to 'spurt' in the next year. In reality, the test did no such thing and the pupils were, in fact, selected at random. They had two aims: To plant in the minds of the teachers a particular set of expectations about their pupuls and to see if this had any effect on pupil performance. All the pupils were re-tested eight months later and over these eight months, the pupils gained on average 8 points but the 'spurters' gained 12. Indication that the teacher paid more attention to the progress of those they believed to be spurters.

Ethical Problems- Pose major ethical problems, for example while the 'spurters' benefited from the study, the remainining 80% didn't and some may even have been held back educationally because they recieved less attention and encouragement from teachers. Field experiments work best when those involved dont know they are part of an experiment. However, this requires deception.

Reliability- As a result of the many differences between school classes, in terms of the age of pupils, teaching styles and so on, it is hard to create an exact replica of a field experiment.

Validity- Rosenthal and Jacobson claimed that teachers' xpectations were passed on through differences in the way that they interacted with students. However, they carried out no observation so they don't have the data to prove this.

Broader focus- Rosenthal and Jacobson did look at the wider labelling process from teacher expectations through to the effect on pupils, rather than just examining single elements in isolation. Their study was also longitudinal, whicb enabled them to identify trends over time. 

Operationalisation of concepts- Involves turning abstract ideas into a measurabe form. Can be difficult when creating a questionnaire for pupils. Their grasp of absract concepts is generally less than that of adults, it may be more difficult to turn sociological ideas such as 'deferred gratification' into language that pupils will understand, This may produce answers that are based on respondents' misunderstanding of what the questions mean.

Samples and sampling frames- Schools routinely keep lists of pupils, staff and parents. These can provide accurate sampling frames from which the sociologist can draw a representative sample. However, schools may not keep lists that reflect the researcher's interests and schools may deny access to such confidental information even if the relevant sampling frame does exist. Schools also need to give a researcher premission to give out questionnaires and pupils may find a questionnaire off-putting as it is a formal document.

Access and response rate- Response rates are often low, schools may be reluctant to allow sociologists to distribute questionnaires because of the disruption to lessons that it may cause or because they object to the topic. 

Practical issues:

Very useful for gathering large quantities of basic factual educational information quicjly and cheaply. For exmaple, Michael Rutter (1979) used quetionnaires to collect large quantities of data from 12 inner London secondary schools. From this, Rutter was able to correlate achievement, attendance and behaviour with variables such as school size, class size and number of staff.

However, the data generated by questionnaires is often limited and superficial. 

Written questionnaires involve participants being able to read and understand the questions, Thus, they are unsuitable for those who cannot read reasonably well, such as young children or those with certain learning difficulties.

Children generally have a shorter attention span than adults so questionnaires need to be relatively brief if theya re able to stand a chance of being completed. This limits the amount of information that can be gathered. 

Children's life experiences are narrower and their recall from different from those of adults.

Anonymity and detachment:

Can be particularly useful when researching sensitive educational issues such as bullying, where their anonymity may overcome pupils' embarrasment or fear of retribution from bullies. As a result, response rates may be higher and pupils may be more likely to reveal details of their experience of being bullied This may produce more valud data than would a face-to-face structured interview.

However, much depends on whether pupils are reassured that their anonymity will be safeguarded. This reassurance may be difficult to achieve with such a detached method as a questionnaire.

Interpretivist sociologists emphasise the importance of developing rapport with research participants and so they reject questionnaires as a means of researching pupils. Because of the lack of contact with respondents makes rapport difficult to establishm young people may be less likely to give full and honest responses.

Practical Issues:

Young peoples linguistic and intellectual skills are less devloped than those of adults and this may pose practical problems for interviewers. Young interviewees may: be less articulate or more reluctant to talk, not understand long, complex questions or some abstract concepts, have a more limited vocabulary and use words incorrectly or differently from adults, have a shorter attention span and poorer memory.

However, children may also have more difficultly in keeping to the point, especially in unstructured interviews. As Janet Powney and Mike Watts (1987) note, young children tend to be more literal minded and often pay attention to the unexpected details in questions. Training therefore needs to be more thorough but this adds to the costs of research.

Unstructured interiews can often take an hour or more to conduct. Given the time constraints that most teachers work under, interviews with them would probably have to take place outside school hours.

For young children in particular, there is also the ethical issue that they may be unsettled by strange situtations such as an interview, so researchers need to take particular care that the interview does not distress them

Reliability and validity:

Structured interviews produce reliable data because they are standardised: each interview is conducted in precisely the same way, with the same questions, in the same order, tone of voice and so on. 

However, structured interviews may not produce valid data, since young people are likely to respond favourably to such a formal style- perhaps because it makes the interviewer appear too much like a teacher.

Access and response rate:

Schools are heirarchal institutions and this can cause problems when seeking to interview teachers or pupils. E.G. to interview a teacher, they may need the headteachers permission and a student may need a parents permission.

Schools may be reluctant to allow sociologists to conduct interviews during lesson time because of the disruption it causes, or because they object to the researchers chosen topic such as drug use.

Improving the validity of interviews with pupils:

Researchers can adopt strategies to improve the validity of interviews with pupils and young people. For example, Sheila Greene and Diane Hogan (2005) argue that interviewers should:

Use open-ended rather than closed-ended questions, not interrupt childrens answers, tolerate long pauses to allow children to think about what they want to say, recognise that children are more suggestible and so it is particularly important to avoid asking leading questions, avoid repeating questions since this makes children change their answer because they think it was wrong.

In general, unstructured interviews may be more sitable for overcoming barriers of power and status inequality. Their informality can put young interviewees at their ease and establish rapport more easily. This can be particularly useful when dealing with sensitive topics such as bullying.

Has both strengths and weaknesses:

Pupils and young people are often strongly influenced by peer pressure and this may reduce the validity of the data gathered in a group interview, where individuals may conform to peer expectations rather than express what they truly think. In additon, the free-flowing nature of group interviews make it impossible to standardise the questions and this wull reduce the reliabilty of the method and the comparability of findings.

On the other hand, Greene and Hogan argue that group interviews are particularly suitable for use with pupils. They create a safe peer envrioment and they reproduce the small group settings that young people are familiar with in classroom work. Peer support also reduces the power imbalance between adult interviewer and young interviewee found in one-to-one interviews.

They can also reveal the interactions between pupils, however peer pressure may influence individuals to given answers that conform to the group's values, rather than expressing their true opinions.

Practical Issues: An example of the structured observational schedules favourted by positivists is the Flanders system of interaction analysis categories. Used to measure pupil-pupil and pupil-teacher interaction quantitatively. Observations can thus be easily converted into quantative data simply by counting the number of times each type of behaviour occurs. These methods are quicker, cheaper and require less training.

Reliability: Techniques such as FIAC are likely to be easily repliated because FIAC only uses ten categories of classroom interaction, which makes it relatively easy for other researchers to apply in a standardised way. It also generates quantitative data, which makes the findings easy to compare with those of other studies.

Validity: Interpretivist sociologists criticise structured observation of classroom interaction for its lack of validity. For example, Sara Delamont argues that simply counting classroom behaviour and classifying it into a limited number of pre-defined catergories ignores the meanings that pupils and teachers attach to it.

Practical Issues: 

Schools are complex places and more time consuming to observe than many other settings. It took Lacey two months to familarise herself with the school, while Eggleston (1976) needed over three months just to set up his cover role for observations.

May be easier to gain permission to observe lessons than to interview pupils and teachers. 

Personal characteristics such as age, gender and ethnicity affect the process of observation. At the time Wright (1992) was carrying out her research, there were few black teachers and she found her African Caribbean ethnicity produced antagonistic reactions from some white teachers. On the other hand, she found that many black pupils held her in high esteem and would ask her for support.

Observation of interactions in school settings is limited by the restrictions of the school timetable, holidays, control over access, health and safety legislation and so on.

Schools are busy public places, so the observer may find it difficult to find the privacy needed to record observations. E.G. Hammersley had to use the back of a newspaper (to be covert)

Ethical Issues: Covert approach is not appropiate due to their greater vulnerability and limited ability to give informed consent so it normally has to be overt. Delamont points out that every observer in a school sees and hears things that could get pupils into trouble. In some cases, this may even involve the law e.g. stealing. What to do with this 'guilty knowledge' is both an ethhical and practical problem. He also notes that given the harm that can be done to pupils, teachers and schools, additional care should be taken to protect their identity.

Validity: Gives an authentic understanding of the world-views of social actions. This understanding is particularly important when researching issues such as classroom interaction or labelling in schools. However, the power difference between young people and adults is a major barrier to uncovering the real attitudes and behaviour of pupils. They may present a false image when being observed by an adult researcher, thus undermining the validity of the research. Observation is more likely than most methods to overcome this problem, because it gives the researcher the opportunity to gain acceptance by pupils. Further factor limiting validity is that teachers may be quite skilled at disgusining their feelings and altering their behaviour when being observed.

Very difficult to carry out covert observation of educational settings, especially classrooms. This is because there are few 'cover' roles the researcher can adopt and because he or she stands out as being much older than the pupils. This means that classroom observation has to be overt. However, this makes it very difficult to avoid the Hawthorne effect, where the prescence of the researcher influences the behaviour of those being observed.

For example, Ronald King (1984) tried to blend into the background in an infant school by initally spending short periods of time in the classroom to allow the children to become familiar with his prescence. So as not to be seen as a teacher, he avoided eye contact and politely refused their requests for help. In an attempt to be unobstrusive, he even used the classrooms Wendy House as a 'hide'. This example shows how difficult it is for an adult observer to reduce the effect of their prescence on pupils' behaviour, the children may have changed their behaviour 

Representativeness: Scale of the education system is vast, around 4,000 secondary schools and over 30,000 primary schools in England and Wales. By contrast, most observational studies focus on a small number of pupils in just a single school. For example, Willis (1977) studied a core group of 23 boys. This small scare of such studies results from the fact that it takes time to become familiar with the setting, gain the trust of pupils and teachers and carry out the actual observations. The limited scale of the typical observatinal study means that observing school interaction is unlikely to produce representative data.

Reliability: Tend to lack reliability because data recording is often unsystemagtic and hard to replicate. For example, Hammersley found that on one occasion he had to write his notes on the back of a newspaper because he was observing staffroom conversations covertly. The personal characteristics of different observers may evoke different responses. 

Practical Issues:

Much of this data is published and thus avaliable to the sociologist, saving them both time and money. Educational statistics allow sociologists to make comparisons between the achievements of different social groups based on ethnicity, gender and social class. Also, because educational statistics are collected at regular and frequent intervals, sociologists can make comparisons over time. 

Governments gather statistics to monitor the effectiveness of their educational policies such as those dealong with the curriculum, subject choice, raising standards and reducing the inequality of achievement. Such issues are also of great relevance to sociologists.

However, governments collect statistics for their own policy purposes and these may not be the same as those of sociologists. 

Because of the sigma and peer group bullying sometimes associated with recieving free school meals, some pupils dont claim them even when entitled to do so. They also cannot tell the sociologist avout the interaction processes in schools that may lead to this under-claiming.

Representativeness: Some offical statistics are highly representative, for example all state schools have to complete a school census three times a year. This collects info on pupils' attendance, ethnicity and gender etc. 

Reliability: Positivists favour official statistics because their reliability means they can be used to test and re-test hypotheses and thus discover cuase-and-effect relationships. Generally very reliable but governments may change the definitions and categories e.g. when the Conservatives first introduced league tables of school performance in 1988, a schools position was based purely on its exam results. This generally meant that schools with middle-class pupils were placed higher.

Validity: Interpretivists question the validity of educational statistics, they argue that they are socially constructed. For example, they see truancy statistics as the outcome of a series of definitions and decisions made by a variety of social actors, such as parents, teachers and pupils. Schools may manipulate their attendance figures by re-defining poor attenders as being on study leave or additional work exerpeince. 

Practical Issues: Public documents on education are often easily accessible. For example, David Gillborn (1995) in his study of racism and schooling, was able to access a wide range of school documents, including school policy statements, local authority guidelines on anti-racism and the minutes of staff meetings and working parties. These documents gave Gillborn the 'official' picture of what was happening in terms of racism and anti-racism in the schools he studied. 

Personal documents can be more difficult to access. Valerie Hey (1997) made use of the notes girls passed to each other in class to understand their friendship patterns. However, the notes were not always easy to obtain, as the girls were experts at hiding them from teachers.

Some educational documents are confidental, such as teachers personnel files and pupils' disciplinary records; so sociologists may be unable to gain access to them.

Ethical issues: Few ethical issues concerned with using public documents as permission for their use is not required. However, there are more ethical problems with personal documents. For example, Hey collected in the notes that grls had pass to each other in class. In some cases, the girls offered her the notes freely but in others Hey collected them from desks at the end of the day and in this case, informed consent had not been obtained.

Representativeness: Some official documents are legally required of all schools and colleges, such as records of racist incidents. This makes it more likely that we can form a representative picture of racism in schools across the whole country. However, of course, nt all racist incidents may be documented. Personal documents are often less representative. For example, Hey collected about 70 notes but the unsystematic way in which she came by them makes it likely that her sample was unrepresentative. 

Reliability: Deliberate falsifications or accidental mistakes made when filling in registers reduce their reliability because teachers are not applying the measure of attendance consistently. 

Validity: Documents can provide important insights into the meanings held by teachers and pupils and can therefore be high in validity. However, all documents are open to different interprtations. For example, we cannot be sure that Hey's interpretation of the meaning of the notes was the same as that of the girls.