I only had a twenty minute lunch break today, but I thought I'd better use it to try another sample answer...a bit rushed, but hopefully of use of some things to potentially include...
Using the pre-release material and your wider sociological knowledge, explain and evaluate the use of semi-structured interviews in researching motherhood and social class 
The aim of Vincent et al’s longitudinal study was to explore the ways in which mothers of different social class backgrounds experience life. This was a qualitative study, using semi-structured interviews. This suggests that Vincent et al are working from an interpretivist perspective, gaining rich, in-depth and valid data in order to fully understand the lives and experiences of the mothers in the study. The study took place over a five year period, though the pre-release material suggests that several different projects on several different groups was conducted in this time, so this may not have been strictly a longitudinal study (which would usually focus on a single group over a prolonged period).
A semi-structured interview are a cross between structured and unstructured interviews. Like structured interviews, they tend to include lots of closed questions – however, unlike structured interviews, they can also include open questions and the interviewer can change the order of the questions for different interviewees. This means that, like unstructured interviews, they can feel more conversational and can be used to gather qualitative data. Because of the flexibility and conversational nature, semi-structured interviews can be a lot more time consuming and unpredictable than structured interviews.
Vincent et al used two sampling methods: Opportunity sampling and snowball sampling. Opportunity sampling involved Vincent at al directly recruiting participants at relevant venues, while snowball sampling involved these participants finding more participants like themselves. This gave Vincent et al control over the sample size, but both methods are non-random, which reduces their representativeness and makes them open to bias, because Vincent et al were (directly and indirectly) picking the participants that they knew best matched their research. The total sample was 126 mothers (55 working class and 71 middle class). Both of these are small samples considering the target population, although because semi-structured interviews are time consuming (in this study, they last between 1-2 hours) and unpredictable, a smaller sample is beneficial.
The questions were based around the mothers’ work choices and use of childcare. The pre-release material suggests that qualitative data only was gathered. This data is presented in the form of quotes from participants and Vincent et al’s interpretations of the collated findings. Qualitative data focuses on meanings and feelings, expressed usually through words, which is ideally suited to Vincent at al’s research as they wanted to explore the experiences of mothers. Quantitative data (expressed in numerical form) would not be able to offer any depth or meaning to people’s experiences. Qualitative data is high in validity as it offers us the truth of people’s thoughts and feelings, directly from their own mouths. Researchers can build a rapport with participants, enabling them to open up more and enabling the researcher to gain empathy and understanding (verstehen). However, qualitative data is much more difficult to analyse and interpret, and interpretations may be subjective. Methods that gather qualitative data tend to be low in reliability. Semi-structured interviews, for example, are very difficult to reliably compare to one another, as every interview is different and not all interviewees get exactly the same questions, or in the same order. With semi-structured interviews, there is also a risk of researcher imposition, whereby the interviewer influences the respondent’s answers through tone of voice or manner of questioning.
Overall however, semi-structured interviews were an appropriate method for this study. As interpretivists, Vincent et al may not have been overly concerned about reliability or representativeness, as the main goal for an interpretivist is validity – and semi-structured interviews offer high validity. There is never an assurance of perfect validity (as some respondents might lie, or bend the truth – particularly in a study like this, where social desirability means they want to come across as good mothers), but there remains still much richer validity than a quantitative method could ever capture, as numerical data offers no real meanings behind the statistics. Due to the low representativeness, Vincent et al’s study could not really be generalised to the wider target audience, but they have still achieved their aims in gaining insight into the experiences of the women in their study, and have further laid the foundations for a more comprehensive study in class and childcare which they recognise is needed.