Observation & in context

• Overt advantages - Avoids the ethical problem of obtaining information by deceit and joining deviant group activities.
• It allows the observer to ask the kind of naive but important questions that an outsider would ask, e.g, why do you rob & steal? Whilst taking notes openly. 
• Disadvantages - The group may refuse the researchers permission to observe them, or prevent them from seeing everything. Punch; two of the Amsterdam police officers he was researching told him that 'we only let you see what we wanted you to see'.
• It risks creating the Hawthorne Effect which undermines the validity of data. 
• Covert practical issues - It may be the only way to avoid the Hawthorne Effect & obtain valid data on activities people would rather keep secret. Humphreys; studied gay men's sexual encounters in public toilets, it preserved the naturalness of their behaviour. 
• You have to keep up an act & may need detailed knowledge about a group's life before joining it. Patrick; bought a jacket with cash instead of credit, which the gang never did.
• You can't openly take notes, Ditton; had to use the toilet to take notes when studying theft among bread delivery men, which aroused suspicion. Or ask naive yet important questions.
• Ethical issues - They're lying about why they're there & why they left.
• May have to take part in illegal/immoral activities & may have report it to police after. 

• Validity - Obtains rich qualitative data about how people really live.
• Achieves 'verstehen' because researcher gains empathy through personal experience, living as a member of the group. 
• Flexibility - Rather than a fixed hypothesis the researcher enters a situation with a relatively open mind about what they will find, as they encounter new situations, new explanations are forumulated, allowing the sociologist to followup or change the direction of the research then and there. Whyte; he learned answers to questions that he would not have had the sense to ask if it was an interview.  
• Practical - May be the only way to study certain groups, Yablonsky; a teenage gang is unlikely to welcome researchers as they represent authority,
• It enables the researcher to build a rapport & trust with the groups like; football hooligans, thieves, drug users, religious sects and other 'outsider' groups. 
• May be effective in situations where questionning won't work, Cicourel; study on whether police make unconcious judgements about the 'criminal types', interviews or questionnaires wouldn't work because they aren't aware they are making assumptions. 

• Practical - It is very time consuming, Wythe; his study took four years to complete. 
• The researcher needs to be trained to recognise aspects of the situation that are sociologically significant & worth further attention. It also requires interpersonal & observational skills that not everyone has. 
• It can be personally stressful & demanding, especially if covert.
• Personal characteristics such as age, gender or ethnicity may restrict the kinds of groups that can be studied. Downes & Rock; not everyone can fit into punk rockers or Hell's Angels.
• Some groups may not wish to be studied, which is why participant observation often focuses on powerless groups who are less able to resist being studied, e.g, petty criminals. 
• Ethical - In covert observation may have to decieve people or participate in illegal activities. 
• Representativeness - The group studied is very small, Downes & Rock; it is doubtful how far these "internally valid" results are "externally valid" of the general population. 
• Reliability - It is difficult to replicate or quantify, Whyte; his method was unique to him.
• Bias & lack of objectivity - There is a risk of "going native", or researchers loyalty to the group may lead to them concealing sensitive information.
• Invalidity is caused by Hawthorne Effect or researcher imposing their own views.

Structured observation 

• Practical issues - The Flanders system of analysis categories (FIAC) are used to measure pupil-pupil & pupil-teacher interaction quantitatively, they have behaviour categories & measure how much each behaviour occurs. Flanders; found that in the typical American classroom, 68% of the time is taken up by teacher talk, 20% by pupil talk & 12% lost in silence or confusion.
• They are simplistic, quicker, cheaper & require less training. 
• Reliability - They're easily replicated because FIAC uses only ten categories of classroom interaction, making it easy to compare with other studies. 
• Validity - Interpretevists crticise it for a lack of validity, Delmont; simply counting classroom behaviour & classifying it into a limited number of pre-defined categories ignores the meanings that pupils/teachers attach to it. 

• Practical issues - It may be easier to get permission to observe rather than interview because it doesn't distrupt lesson time, Fuller; the head of the London school she was studying decided she didn't need parental consent to observe normal school behaviour. 
• It may be difficult to write notes because schools are busy places, Hammersley; he had to quickly & covertly write notes about staffroom conversations, may lead to mistakes as he used his own interpretation to decide what they said. 
• Personal characteristics affect the process of observation, Wright; there were few black teachers & that produced agnostic (doubting) responses from white teachers. 
• Ethical issues - A covert approach isn't ethical because of the vunerability of young people, they are unable to give informed consent. 
• Delmont; the researcher may see something that could get a student in trouble, it is ethical to report it but may hinder the research as trust will be broken. 
• Extra care needs to be taken to ensure that the schools, pupil's or teachers identites are hidden, especially when image is so important (marketisation).
• The Hawthorne Effect - Teachers & students may take part in impression management.
• Unrepresentative - Usually a small number of students studied, also researchers might only associate with teachers they have things in common with. 
• Unreliable - Observers experiences are unique, also based on their characteristics.