Copyright Sociological Research Online, 2000
T Marcus and D Manicom (2000) 'Consciousness in Transition: A Case Study of Social Identity Formation in KwaZulu-Natal Study Description and Methodology'
Sociological Research Online, vol. 4, no. 4, <http://www.socresonline.org.uk/4/4/marcus.html>
To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary
Received: 26/8/1999 Accepted: 18/10/1999
- The aim of this article is to describe the Class Race and Gender (CRG) Research Programme. The CRG research programme aims to explore the development of consciousness in South Africa, to understand how we come to be the black and white, rural and urban, rich and poor and men and women who make up our stratified and differentiated society and to identify and assess the impact of changes over time. This complex problem is being investigated through a study of class, race and gender identity formation in the first generation of children entering the new, compulsory education system. This article specifically tries to document the research process; its methodology and the instruments which were used and developed in order to engage with the issues under investigation. The article also tries to explain the rationale informing the choice of the sample and methods and describes how these research methods were implemented. Research with people is always interactive and reflexive, even if the researchers do not concern themselves with what the research might contribute to respondents. Yet, in questions there are ideas and information which people think about and learn from. Research is or can be a learning process for respondents. For respondents (and researchers) there is a continual tension between the limits of research (finding out) and the possibilities of intervention (acting out).
- Children; Class; Consciousness; Describe; Gender; Methodology; Race; Rationale; School
- The aim of this article is to describe the Class Race and Gender (CRG) Research Programme, more specifically to document the research process; its methodology and the instruments which were used and developed in order to engage with the issues under investigation. The article also tries to explain the rationale informing the choice of the sample and methods and describes how these research methods were implemented.
- The CRG Research Programme is an interdisciplinary research project bringing together specialists from Sociology, Psychology and Dietetics and is based at the Community Agency of Social Enquiry (CASE), Pietermaritzburg and the University of Natal, Pietermaritzburg. The aim of the programme is to explore the development of consciousness in South Africa, to understand how we come to be the black and white, rural and urban, rich and poor and men and women who make up our stratified and differentiated society and to identify and assess the impact of changes over time. This complex problem is being investigated through a study of class, race and gender identity formation in the first generation of children entering the new, compulsory education system.
- The research team is led by Prof. T. Marcus, together with two other principal investigators Prof. E Maunder (Department of Dietetics and Community Health) and Ms B Killian (Department of Psychology). Prof Ben Parker and Mr Volker Wedekund (Department of Education) participated in the conceptual process, but were forced to withdraw from the programme due to pressure of work. The team also included three doctoral, four masters, one honours and several post-graduate diploma students working in the fields of gender, race, transport, locality and place, self esteem and anthropometric indicators.
- The concept of the study is to follow grade one entrants as they progress through the ten years of compulsory schooling, beginning with the 1997 intake. Empirical investigations are to be done with them in grade one, grade four, grade seven and grade nine which is the final year of compulsory schooling. In addition, each field work episode will study pupils who are in their sentinel grades. Thus, it is intended that pupils in all four grades are studied every time new empirical work is carried out, and the timing of new empirical work is determined by expected or actual passage of the new entrants into the next sentinel grade. While there is likely to be overlap between grades four and seven, the only intentional follow through will be conducted with the 1997 grade one cohort.
- Given constraints in both human and financial resources, and to ensure that the programme remains manageable, the study does not extend back into the children's homes nor does it engage with the teachers and curricula which make up the schooling environments.
- The study is sited at ten schools - five primary and five high schools - within a radius of 45 kilometres of Pietermaritzburg. The schools were selected because of their locations; both with respect to their physical places and the communities and people they have historically served. The choice of schools was also determined by the central focus of the study, namely the children who were in grade one in 1997 and who would remain in the primary education system as presently structured during at least three of the four research episodes of the programme. The choice of high schools was more difficult, especially in the urban context. Generally, they were selected for their proximity to the selected primary school and as a likely feeder. All the schools, with the exception of one high school (Epworth High), are co-educational. A pair of schools; primary and a high are private (Epworth Primary and Epworth High); a pair are historically white, urban schools (Carter High and Northern Park); a pair are historically black, urban schools (Georgetown High and Sanzwili Primary); a pair have historically served black children living within the white-owned commercial farming sector (Jabula High and Lidgetton Primary) and a pair of schools have historically served rural communities living within the former Bantustans (Mtholangqondo and Henley).
- The research is sited in the schools, although it is not a study of the education system per se. The schools were not selected for their representativeness in the KwaZulu-Natal Province but rather, they are seen as locales which reflected both in combination and particularity; urban\rural, public and private schooling thus bringing together regularly over years a range of children from different backgrounds. The schools facilitate access to a cross section of school-going children.
- Data was captured using the Paradox programme and a relational data base was set up. The time frame of actual capture used three people from July 1997 to February 1998, which amounts to 4500 person hours. This time was used to clean, code and re-code the data. Each respondent was ascribed a unique identifier which assists in the process of linking the different information they provide us with.
The Study Population
- This study involves approximately 1474 respondents, 388 of whom are the primary cohort of grade one scholars. In the first episode of field work carried out over the four months March to June, 1997 a further 345 respondents were in grade four, 319 were in grade 7 and 422 in grade 9.
- Although initially a smaller sized sample had been envisaged, it became increasingly apparent that in order to meet the comparative objectives of the study and to do so in a way which would satisfy quantitative research requirements, the size of the overall sample had to be increased. Part of the reason for this lay in the wide range in size of the schools and therefore, the number of pupils in each grade at each school. In the end, all pupils in grades one, four, seven and nine at all participating schools became part of the study sample, with the exception of three -the primary and secondary urban, historically black schools (Sanzwili, Georgetown and Lidgetton Primary) and the urban, historically white high school (Carter High). At these, the number of pupils in each grade were sufficiently large to draw a sample.
- However, the final size of the sample was modified due to parental or respondent unwillingness to participate in the programme, absenteeism on the days when field work was done and incomplete questionnaires.
- The final number of respondents for the study per grade in each school was:
Northern Park Primary
- The gender composition of the sample is fairly evenly divided between the sexes, although the overall balance tends to favour girls and young women (53% female and 47% male), albeit only slightly in grades one and seven (50.5:40.5 and 52:48 females to males respectively) but more markedly in grades four and nine (54:46 and 55.5:44.5 females to males respectively).
- Overall, this is largely a study of black (African) and white school-going children, a racial composition determined by the schools selected for the study. Six of the schools are exclusively black (Henley Primary, Georgetown High, Mtholangqondo, Lidgetton Primary, Jabula High, Sanswili) while four of the schools (Epworth Primary, Epworth High, Carter High, Northern Park) have a minority of black scholars, most of whom are African. The programme was not able to include schools historically set up to service Coloured and Indian children because of a lack of resources. 77 % of the respondents are black (African) and 18% are white, 2% are coloured and 3% Indian. In the primary cohort; grade one - 82% of the sample are Black, 10% are White, 4 % Indian and 3% Coloured. In grade four; 80% of the sample are Black, 15% are White, 3% Indian and 2% Coloured. In grade seven; 77% were Black 20% were White and 3% were Indian. In grade nine 68% were Black, 26% were white , 2% were Indian and 4% were Coloured.
- In terms of locational distribution, 22% attend rural schools within the former Bantustans. 12% of the respondents attend at farm schools, located in the historically white owned commercial farming sector. 38% of respondents attend at public urban, historically black schools located in the black townships, formerly segregated from the white run municipalities. 19% of respondents attend public urban schools which were exclusively white until 1991 and 9% attend private schools - which although predominantly white had opened access to black school-goers about a decade sooner than the state.
- Part of the motivation for selecting children by grade was to demarcate groups of scholars within relatively narrowly defined age bands of three or four years. As it turns out, the spread of ages across each grade is wider, with evident racial, sexual and locality characteristics.
- The overall study population spans the age range 6 - 27. In grade one, 23% of the sample are seven years old. 23% are in grade four where the average age is 11, 22% are in grade seven where the average age is 14 and 29% are in grade nine, where the average age is 16. However, it needs to be noted that within each grade there is a proportion of respondents who are over the age norm. For the most part, the reason for respondents being over age can be attributed to repeat years. Overall, 64 % of respondents said that they had never failed, 27% report failing once and 9% have failed twice or more. Some are likely to have started late and others may have dropped in and out of schooling as circumstances outside the schooling context dictated.
- Since it is inappropriate to presume age from grade, the data is analysed by age. As the chart shows, 39% of respondents are 6-10 years old, 14% are 11-12 years old, 32% are 13-15 years old, 12% are 16-18 years old and 3% (n=47) are 19 years old.
- Three observations can be drawn from the study population.. Firstly, while the study's assumption that grade one is the first contact children have with the formal education system, the ages of between 8% and 16% of respondents suggest that for approximately one tenth of the children interviewed in 1997, their first contact is a repeated contact. This needs further investigation.
- Secondly, while White, Indian and Coloured children are most likely to move through the education system more or less in compliance with the age norms, the passage of black children through the system is characteristically much more bumpy and uncertain. This is an important aspect which we will have to address in our future investigations.
- Thirdly, for black school going children, the peer environment of the majority of younger grade members is influenced, if not shaped by a group of older girls and boys who share their school day. This has potentially pervasive consequences for gender identity formation and is an issue which warrants further research.
- Consciousness is at the interface between structure and agency. Eye (1978) captures this positioning of consciousness in asserting that it cannot be separated from the objective world; it is caught up in this world at the very moment that it takes possession of it. He (ibid) goes on to explain that consciousness is neither completely objective nor exclusively subjective, but rather is ambiguous in form. As such, developing an approach to explore social identity across disciplines poses enormous theoretical and methodological challenges. The team engaged in discipline specific literature reviews as well as intensive debate about appropriate ways of approaching the research both in terms of theory and methods.
- A preliminary scan of the literature in South Africa reveals that there is a general paucity of research into social identity or consciousness in children and teenagers. The discipline which has developed theories and tools for these age sets is psychology, although mostly around the areas not of our concern, except for social identity theory in social psychology which provides a useful starting point.
- For the purposes of the study, whilst there was and continues to be theoretical exchange across disciplines and perspectives, the programme did not seek theoretical consensus nor did we anticipate only one outcome. Rather, a lower order agreement around critical issues was preferred; which could be informative across disciplines and within different theoretical frameworks. This approach made co-operation practical and applied. At the same time it created the expectation and possibility for synergy -where outputs would not only be bound by the limits of each discipline or particular theories/hypotheses.
- Working from the assumption that understanding consciousness needs research into both structure and agency. Middens (1977) view is that the social world is constituted as meaningful by the meanings ascribed to it by humans in the course of their interactions with it while at the same time these interactions act to constitute and reconstitute the actors as well. It is within this context that the team decided that our efforts needed to focus in on the children's comprehension, experiences and sense of place in their worlds. This decision meant that in developing research questions and instruments we often had to resist approaches which would take the study into their homes, communities or the education system.
- In order to delimit the concepts of class, race and gender, the team identified themes or areas which we considered to be useful and important signifiers of each concept and through which it would be possible to explore social identity. In no particular order, they were poverty, work, food eating habits and well being, gender and sexual identity, racial identity, self-esteem, prejudices and preferences, hierarchy and values, and sense of place & space. These were then developed in a range of instruments as concrete questions or problems.
- The CRG Research Programme combined survey, experimental and interpretative research methods in order to generate a complex set of data which would capture the complexities of life across a wide range of children whose ages and experiences are very diverse. The possibilities of using qualitative research techniques were substantially constrained by financial, scale, time, and data capture limitations. Nevertheless, and although much of the data was captured through quantitative techniques, it was also possible to experimentally modify qualitative methods.
- Fifteen instruments were developed to collect anthropometric and dietary, psychological and sociological information. These were driven from discipline specific hypotheses and practices, although appropriately, there is overlap.
- A group of three instruments focus on anthropometric, diet and habitual physical activity. The first is an anthropometric data record which measures the height, weight and arm, waist and hip circumferences of each of the respondents.. The second is a 24 hour dietary recall worksheet\exercise where respondents are asked to recall and record everything they ate or drank the previous day. The purpose of the exercise is to capture food types as well as the regularity of food intake. No attempt was made to establish quantities largely because of the difficulty of getting reliable responses. The third instrument is an observation sheet which captures the habitual physical activity levels and preferences of respondents, as observed by teachers. Given the brief presence of the research team in each of the schools, this exercise had to draw on the teachers' accumulated knowledge and observations of learners. All three instruments are standardized and fairly routine procedures. While the physical information serves to create a more complete picture of respondents health and well being, the information gathered in the dietary recall exercise is expected to yield social and cultural insights.
Social and Psychological:
- Another cluster of three instruments are driven by social and individual psychological concerns. The first is a Social Status Technique derived from Central's Self-Anchoring Striving Scale where respondents are asked to place other individuals hierarchically. Rather than doing this pictorially with drawings, as the exercise had originally been developed, ladders were built and stylized figures were created and respondents were asked to place them physically on the ladder in response to questions asked. Two sets of six figures were created, each consisting of male and female "dolls" for white, black (African) and Indian people. One set of figures depicts economically better-off learners; the other, learners who are economically poor.
- The exercise does not involve a forced choice as more than one figure can be placed on any step. The Social Status Technique is a measure of preference, social stratification and personal identification built around a set of "who" questions. Given a choice of six stylized figures, children are asked 10 questions - who has the best food, who is the happiest with their life, who does best at school, who gets sick the most, who walks the most, who should sort out a problem, who will get the best job, who will become an important person in the world, who has the most friends and who is most like the respondent. Their responses are then scored. This exercise was implemented using the split-half technique, with half the respondents using one set of figures and the other half the other set of figures. To overcome interviewer bias, they used both sets of stylized figures alternately.
- A further two instruments which focus on individual self-esteem are used in order to get a sense of how respondents value themselves and perceive themselves in relation to others. The Self Scale and the Culture Free Self-Esteem Inventory (CFSEI), as standardized, internationally validated instruments, are used, unmodified. The sub-scales of the CFSEI give data on social, academic and parental-relationship issues.
General and Sociological:
- The next set of eight instruments are sociological - albeit, with evident general application for the whole programme. There are four questionnaires which explore experiences and attitudes. The first covers demographic and locational information about the respondent asking. questions about kin and household, dwelling type space and place, amenities and services, religion and literacy, domestic work, and residential and geographical mobility.
- The second questionnaire explores poverty and work, asking questions about food routines and hunger, clothes, attendance at school, educational aspirations and possible obstacles to achieving these, pocket and spending money, and the differences between rich and poor people as well as economic activity and the type of work of parents and respondent, employment aspirations and perceptions of the importance of work.
- The third questionnaire focuses on gender and sexuality, asking questions about the similarities and differences between boys and girls, their own play and leisure time preferences, gendered expectations about activities now and in the future, marriage and sexual preferences, pregnancy, contraception and sexually transmitted diseases and gendered responsibilities in regard to these, name calling and abuse and sense of self
- And the fourth questionnaire asks about race and colour, covering self-categorization, prejudices and preferences, colour and its impact on present achievements and future aspirations, inter-racial contact, name calling and bullying, and sense of self.
- The fifth instrument is a 12 hour activity recall, where with very limited markers, respondents are asked to remember and record what they did from the time the bell rang at the end of the previous school day until the time the school bell rang the following morning. The purpose of this exercise is to get an insight into after-school routines, what children and adolescents do and the social differences these activities may represent. As with the dietary recall, no attempt is made to quantify activities in time.
- The next exercise - "mapping the main meal"- is an experimental activity developed to get a sense of the respondent's place in his or her family or household. Modifying techniques developed in Participatory Rural Appraisal, respondents are asked to plot their previous main meal - in which room they sat, where they sat, who they ate with, who prepared the food, who served it and in what order, who got more or less food and why, whether they helped somebody else eat and why, whether there was a prayer or blessing, whether people talked, watched TV, read or listened to the radio during meal time, their sense of what the main meal means to them, etc. Their responses are then described and analysed in relation socio-economic and demographic information.
- The last two instruments are qualitative and are intentionally broad. In a "A Joke and Three Expressions I have Heard" respondents are asked to write down any joke that they have heard told about other people. Similarly, in recording three expressions or sayings they have heard, respondents are asked to reflect those that are used to describe other people. The purpose of this exercise is to get a sense of prejudice and difference and to examine the social differences in how these are presented.
- The final exercise in this group is an essay, the subject of which is "A Memorable Experience". Respondents are asked to write down a good or bad experience and to write about a mark or impression that has endured. The main aim is to get participants in the research programme to voice their experience, to locate it in a social context and to relate it to the way they come to be the people they are.
- The last exercise, administered at the end of the process, is a Feedback sheet. Self evident in its purpose, this instrument is designed as an open-ended reflection by participants in the research process. It is innovative, in as much as it is built into the design of the programme and all respondents are asked to use it.
- Together, these instruments yielded a wealth of data, however, in working through the feedback sheets several important questions about process have come to the fore. An important issue which the study has raised is the issue of using multiple surveys. In capturing the data for each survey it was found that each survey had different totals due to either spoilt or incomplete questionnaires. It is therefore difficult to establish what the total population really is because the totals for different questionnaires do not correspond.
Operationalising the Study
- The biggest conceptual hurdle the study had to overcome was to operationalise the research; to find a way to schedule all 14 instruments within a time frame that a) would successfully accomplish the task; b) would be acceptable to the schools; and c) would do so in a manner which was as unobtrusive as possible. The solutions we found were several.
- Fieldwork was divided into two waves. The first wave of field work centred on capturing anthropometric information. A team of up to 17 diploma students drawn from the Department of Dietetics and Community Health at the University of Natal, Pietermaritzburg under the supervision of Prof. Maunder weighed and measured all respondents over one or two days at each school.
- At the tale end of the first wave, the second wave of field work began collecting data for the remaining 14 instruments. This entailed assembling a team of between 20 and 30 fieldworkers, comprising combination of under-graduate and graduate students, HSRC trained field workers, unemployed matriculants and graduates, a few lecturers, a field work manager, Bev Killian and myself.
- It was concentrated intensively within each school for between two and seven school days. Within the time-frame of often a week or less, in each school, work was divided by grade so that school work was interrupted for no more than two days for each grade, for the most part. This was viewed very positively by the schools since they were concerned about taking learners out of their routines.
- To get through the large number of instruments in the time allocated, 8 exercises were identified as being suitable for self-administration - but in groups and under supervision. The Feedback sheet was self administered during class time under the supervision of teachers although without the research team presence Aside from the anthropometric data record, the remaining five - the four questionnaires and Social Status Technique Ladder - were administered in face-to-face interviews. This approach was applied to all the grades except grade one, with some modification, as it was found to be more useful to absorb Meal Mapping into the face-to-face interviews schedule.
- For face-to-face interviews, each field worker was assigned two or three respondents for the day with whom they worked through all the instruments. They alternated respondents as they completed each exercise. And each exercise lasted for between twenty and thirty minutes. An initial attempt to group administer with grade ones, albeit modified into smaller clusters involving field workers assigned to two or three respondents, proved to be time consuming and unfruitful .In the end, all exercises conducted with grade ones were done in face to face interviews. The grade ones did not do three instruments - A Joke, Essay or Feedback.
- These procedures had also to be modified to accommodate colour and gender concerns. Thus, where male respondents were being interviewed by female field workers, the questionnaire on gender and sexuality was administered by a male field worker. Similarly and in so far as it was possible, White and Indian field workers were paired with non-African respondents.
- Fieldwork ran almost uninterrupted for over three months. The hospitality of the schools and good weather, especially where classroom space is a scarce resource, made a long work detail much easier. There were particular difficulties at the urban black schools. The timetable had to be rearranged because of local security conditions at one school. Then, the subsequently rescheduled programme coincided with the proposed COSATU national strike, which was postponed but still took hours of school time to renegotiate. In addition, a water crisis in the area saw the school we were working at closed for a day as teachers went to protest at the local authority offices. Prolonged work at these schools disrupted the schedules we had set up with the next schools, it added to the costs of the field work and wore down the field work team.
- Many research programmes are not designed with loops to feedback to respondents. While sometimes this is not practicable, where it is possible it provides a very real opportunity for the research to be reflexive and assist the researcher to learn about both process and content.
- The CRG Research Programme has built in several opportunities to interact with respondents and their host schools over the research. The study is designed to report both general and school-specific findings back to each of the respective schools. From the Feedback, it is clear that this will have to be to both staff and learners. Where requested, the Programme will also run a workshop in each school on how to do research. It has proposed a cultural exchange between the farm school and the private school around the theme "Who I am - Who I want to be", which will be realized if they wish to take it up.
- The Feedback sheets provided the opportunity for each individual to correspond personally with the research programme, with the exception of one school where teachers took it upon themselves to write "the answers" on the board for the children to copy, which they did!
- By way of conclusion, the key issues raised in the Feedback sheets will be raised.
- The response to the programme is overwhelmingly positive. The respondents enjoyed participating in it for a range of reasons - they saw it as fun, exciting, different from what they had ever done, and for some a chance to miss school. Most strikingly, they found it a learning experience, an opportunity to reflect on their own circumstances, how they see themselves and how they relate to other people. They also particularly liked being asked to talk about themselves, to express their views and to be heard. Research with people is always interactive and reflexive, even if the researchers do not concern themselves with what the research might contribute to respondents. Yet, in questions there are ideas and information which people think about and learn from. Research is or can be a learning process for respondents.
- While most found the research process enjoyable and pleasant, there was a repeatedly expressed concern about the research's purpose, what it was being done for and if it would have any impact on their circumstances. In sharing details about personal experiences and needs, it is not surprising that respondents anticipate that once articulated, they will be heard and even that somebody will do something about them. For respondents (and researchers) there is a continual tension between the limits of research (finding out) and the possibilities of intervention (acting out). To state that the research will provide a better understanding of the issues being explored, that it will put the critical issues into the public debate and onto the policy table, is not a very satisfactory reply to a child or adult who has told you that they often go to sleep without eating or they have to miss school because they have no money for transport. But it is really all that the researcher can say and do.
- In terms of the issues being explored; many commented on one or another aspect which they described as difficult, uncomfortable, embarrassing, stupid or strange. One reading of such responses would be that it reflects badly on the research, that some aspects were unintelligible to the respondents. In some instances this may well have been the case. Some respondents could not understand why they were weighed or measured, for example. Similarly, some questions might have fallen into this category, for example, those that asked about eating patterns and preferences.
- However, these concerns could and should be read differently. They reflect on the social circumstances of different segments of our society. While some respondents found it stupid to ask about toilets and the space they occupy within their homes, -don't we all have flush toilets, our own bedrooms and our own beds? - others found questions about consumables or holidays stupid - you know we don't have cars, electricity or fridges and never go on holiday, why ask?
- Similarly with issues that are sensitive. Sex, sexuality and inter-personal relations is an issue "known" to be sensitive - something that is embarrassing, personal and difficult to talk about. Some felt it to be private, personal and not to be talked about. But it was not seen this way by all respondents. Many, in fact, said they welcomed the opportunity to talk and find out about these issues and what is appropriate behaviour.
- Food, what people ate and eating routines is known to be sensitive with respect to puberty and obesity. But, from our respondents it is clearly also a sensitive issue culturally, where it signifies difference or if you are very poor and you are food short. Race, for some, is sensitive, reflecting our racially complex context where race is a significant indicator of identity and where racial stereo-types are being challenged. Others found asking about family life and parental care to be sensitive.
- Depending on where you are socially placed, issues are or can be sensitive. But does this mean that such questions should not be posed? 1 doubt, rather it suggests that care has to be taken with how they are posed and respondents have to be respected and treated with care. Which in the end is about process. Many of the respondents commented about the ease with which they could communicate with field workers, the fact that they were made to feel comfortable and that the research was conducted in an entirely unthreatening way. Moreover, they felt reassured that they were could refuse to answer a question if they so chose
LEARNERS BY GRADE AND SCHOOL
Note: the table reflects the numbers for schools which were sampled and is based on the original lists given to the study before registration had been complete
2 This bias reflects the gender composition of the schools involved.
3 It should be noted that a fourth instrument, which was conceived as integral to this group, but which we were unable to obtain in time for the field work, is the Katz-Talk prejudice test. Developed in the United States to test racial attitudes in children, the material consists of a series of slides which depict ambiguous school situations in which a child is either the initiator or recipient of a positive or negative race-related interaction. The respondent is asked to select from the slides and the outcome of choices reflects individual in-group preference and out-group rejection. The Social Distance Scale was piloted as a possible replacement but it was not applied in the first round of field work.
4 These were Self-scale, Culture Free Self Esteem Inventory, Dietary Recall, Activity Recall, A Joke, Essay and Meal Mapping.
Copyright Sociological Research Online, 2000
EYE, H. (1978) Conciousness: A phenomenological study of being conscious and becoming conscious. London, Indiana University Press.
GIDDENS, A (1977) New Rules of Sociological Method. London, Hutchinson.