In protestant theological institutions: a critical appraisal of contextual challenges in kerala, india jessy jaison b b s., M d

4.7.1 Non Probability-Purposive Technique in Sampling

Leadership of seminaries, women and men students and theologically trained women were selected for the generation of data to evaluate the training of women in seminaries. “To represent the population, the sample must include all the different kinds of people in the group.”³⁸⁴ To reduce the risk of bias that may be caused by the option of accessibility sampling, the research chose a purposive sampling technique, in which was an attempt to find the most representative sample. A statistical sampling was not viable, because theological institutions, the academic programs and the type of training and all aspects of seminary life differed greatly from one institution to the other. Moreover, use of powerful statistics was not a primary commitment of the study. A non-probability sampling theory³⁸⁵ was therefore preferred to suit the explanatory and analytical task of the study. While serious quantitative researchers avoid the accidental/ convenient or quota samplings that are of non-probability, many are in favour of purposive sampling for special occasions in descriptive/exploratory research.³⁸⁶ Purposive sampling is considered very appropriate in three situations³⁸⁷ such as,

1) when the researcher uses it to select unique cases that are especially informative

2) when it is used to select members of a difficult-to-reach, specialized population, and

3) when a researcher wants to identify particular types of cases for in-depth investigation.
A well known version of this strategy is in Glaser and Strauss’s classic ‘The Discovery of Grounded Theory’³⁸⁸ that deals with the theoretical generalization generated by an inductive approach. However, the purposive sampling could be criticized for “being ad hoc and vague if not employed systematically.”³⁸⁹
In its more general form, theoretical sampling means selecting groups or categories to study on the basis of their relevance to your research questions, your theoretical position, and analytical framework, your analytical practice, and most importantly the explanation or account which you are developing. Theoretical sampling is concerned with constructing a sample (sometimes called a study group) which is meaningful theoretically, because it builds in certain characteristics or criteria which help to develop and test your theory and explanation.³⁹⁰
The sampling of seminaries was done according to this. In order to guard against subjectivity and bias the researcher looked for maximum possible differences in the sample cases. The methodological details were formulated with suspicion of the level of openness on the part of institutions towards the variety of information required in this study.
4.7.2 Sample Institutions
Twelve institutions were selected out of a comprehensive list of 120.³⁹¹ Therefore, to use the most representative criteria to ensure the most reliable sampling, institutions that have been in existence for a minimum of 15 years with good reputation, evidence of development in various aspects of training and those that offer up to M Div/B D were selected. The population included theological institutions within the wide variety of titles such as seminaries, Bible schools, Bible Institutes and Bible Colleges, putting them all into the major two categories – Episcopal and non-Episcopal.

0. 
   4.8 Data Gathering and Processing
   

Data gathering was done with prepared schedules for focused personal interviews³⁹² and themes for focus groups. Responses were fully transcribed in the space provided on each of the interview schedule concurrently with interviewing. The talks in the focus groups were also fully transcribed by two people simultaneously and the contents and summary were cross-checked with the participants at the end. Due to the controversial nature of the topic under investigation and as per the ethical commitments made to the sample seminaries, every attempt was done to conceal the identity of institutions and each of the respondents in this thesis. The pilot study (prior to the formal gathering of data) done in a seminary where both men and women were being trained jointly in all academic programmes resulted in minor revisions of the initial design. Changes resulted were as follows:

• 
  Select two men students instead of one from each seminary; let them take the interview in pairs
  
• 
  Reduce the number of questions in all schedules (each schedule had 15 questions earlier) and make each question focused; let respondents talk in depth on each of the themes represented in the questions
  
• 
  Conduct a focus group with women students apart from the earlier plan to have just one focus group, which was with theologically trained women.
  
• 
  Add certain questions to get the internal cultural attitude of informants on the topic. For example, questions such as, ‘What is your definition of a virtuous woman?’
  

4.8.1 Description of Sample Seminaries

Twelve seminaries representing Baptist church, Church of South India, Marthoma church, Pentecostal church, Brethren church and the Syrian Orthodox church constituted the sample, which was grouped into four, based on their nature of function. They are categorized below according to the time of founding to specify the earlier institutions and those founded later. For the purpose of grouping, the denominational affiliations of seminaries and the style of function were considered.

1. 
   Episcopal Seminaries (a)-These were seminaries of Episcopal churches i.e., churches that follow the ruling of Bishops and the hierarchy. All three in this category were post-independence seminaries; strongly church based; functioning to assist their own churches; purpose was equipping people for church’s ministry; and academic orientation therefore, was secondary. The training system was not much flexible. Leaders all were church’s key functionaries; seminaries in this category are referred as ES(a)-1, ES(a)-2 and ES(a)-3 (This method is applicable for the following cases too).
   
2. 
   Non-Episcopal Seminaries (b)-All three in this group started before 1930; existing for own churches; preserving church’s values and culture; not appreciative of greater academic focus or any radical agenda; leaders were all people in key leadership roles in the church.
   
3. 
   Non-Episcopal Seminaries (c)-All three in this category were established between 1970-1990; interdenominational faculty, administrative staff and student body; associated with a church but not under absolute control of the church; academic excellence was the priority; seminary leadership could exercise relatively more freedom than in ES(a) and NES(b); students on completion left the seminary with their own plans and some to their own churches; charged fees from students
   
4. 
   Non-Episcopal Seminaries (d)-The three seminaries started between 1990-2006; consisted of interdenominational faculty, staff and students; claim/project some differences in values and functions from other seminaries; prefer students with higher academic qualifications and financial stability; excellent facilities on campus especially huge buildings
   

The six seminaries in the first two categories were founded before 1947 i.e., before India became an independent nation. They were places that prepared men for the priestly function of the church. These Episcopal and non-Episcopal seminaries focused to assist their own churches and they still carry on the same task with the same commitment they had at the beginning. In 1970s Kerala saw the mushrooming of Bible Colleges with both denominational and interdenominational outlook.

Larger Categories-2

1. 
   Denominational Seminaries-This consists of the first two groups (six seminaries) in the above list- ES(a) and NES(b). All claim long historical heritage with their strong church foundation. Though ES and NES differ in dogmas and function, these early seminaries seem similar in their relation with the church and traditional function; one seminary in this category was exceptional in certain information due to its better approval of women.
   
2. 
   Interdenominational Seminaries-This consists of the last two groups (six non-Episcopal seminaries from 1970 till date) NES(c) and NES(d). All of them are related to their church denominations but the seminary faculty and student body comprised people from different Episcopal and non Episcopal churches; one seminary among this group provided data obviously differing from others because of its significant attempts to affirm women in the seminary context
   

4.8.2 Data Generation Tool-1
Samples and conduct of interviews: I personally interviewed all respondents listed below in the months of October and November, 2006.

Seminary Leadership (12) – One each from twelve seminaries; all men- seven of them strongly represented their church denominations and five, not so. All, except one were from Kerala and one Principal worked in the United States for four decades. Each interview took about 45 minutes. I used a schedule containing ten questions to gather data about what the leaders thought the cultural and ecclesiastical challenges faced by women were and how the seminary went about coping with these and what was their personal attitude to various aspects of the issue. The questions were open-ended, providing them opportunity to speak out their ideas without restrictions. All respondents were very cooperative in providing the data throughout the interview, which apparently offered me full access to students for data gathering.
Men students (24) – Two from each seminary; all from Kerala; random selection by the institution from final year B Th., BD and M Div programmes, 20 bachelors and 4 married and 10 pastors and 14 laity. Instead of my initial plan to interview one male student from each seminary, I had two students talking to me together, as decided from my pilot test. Each interview took about an hour and a half or two hours. Some stayed even longer to have a personal discussion on the subject. They were friendly, appreciative of the topic and were more at ease, confident and open in pairs rather than as single respondents. Their assessments of the situation, practical knowledge of the issues and depth in reflections were invaluable sources for my interpretation of the data. Having 10 pastors among the sample also enabled me to attain a general feel of churches about women’s ministry.
Women students (18) – Two each from eight seminaries, one each from two ecumenical seminaries and two seminaries had no women. Each interview with an open ended schedule lasted an hour. Although most of the interviewees started off with some apprehension to talk, they soon developed confidence along the way with the simple conversational style adopted. They relaxed and at the end, thanked me for the opportunity to openly speak about their struggles and aspirations. All seminaries assured an uninterrupted situation for interviewing the students. The talks revealed the helplessness and insecurity of the students in many aspects of their life during and after seminary life.

3. 
   Data Generation Tool-2
   

Samples and conduct of Focus Groups

1. 
   Theologically Trained Women (5) – All with Master of Theology degree; all wives of those who are already in theological education; invited to one place from various regions of Kerala
   

The purpose was to gather the perspectives and experiences of theologically trained women who live in Kerala after their education. It was relatively difficult to get this group for many reasons including their suspicion of the purpose of the meeting and travel constraints.

2. 
   Women students in seminaries (5) –All from M Div and BD final year class, all spinsters; between 21-27; selected from five seminaries in central Kerala
   

This talk produced the most emotionally vibrant discourse of all. I saw students weeping, complaining and trying to gather themselves to talk about the realities of life. It lasted for two hours. As already stated, this focus group was not in the initial plan; but added to have better confidence in my data, along with the fact that the personal interviews had only 18 women students.

4.9 Sampling and the Question of Bias
There are a number of samplings needed for this study such as, sampling of institutions, women and men students, leadership of seminaries and theologically trained women. Since methodological procedures of a non-probability technique could be questionable at times, it was necessary to develop principles to defend the questions of bias, reliability and validity. My sampling techniques consisted of open sampling (where as many potentially relevant categories as possible are selected), relational and variational sampling (which is expected to help uncover and validate the relationship between various categories) and discriminate sampling (which is deliberate in selecting the relevant categories to enable the study to reach building up of theories), which are foundational for Grounded Theory’s data procedures.³⁹³ The following conceptual developments supported this task.

4.9.1 Analytical Induction

Purposive sampling raises the question of bias on the part of researcher. Therefore, the sample should not only let the researcher develop the theory or explanation, but also enable him/her to test it, for which an appropriate mechanism had to be built in. A classic way to do this, derived from procedures of ‘analytical induction,’³⁹⁴ was to seek out negative instances or contradictory cases in relation to the developing analytical ideas. In other words, ‘the researcher should use the sampling strategy not simply to acquire units from which she will generate data which support the analysis or explanation, but also to show that she has rigorously looked for cases or instances which do not fit with her ideas or which cannot be accounted for by the explanation which she is developing.’³⁹⁵ The choice of institutions that were different in their doctrinal stand point, the courses offered, and their practice of accrediting the institution was made on this methodological principle. The study was not in opposition to the quantitative method that used representative statistics, but held that the sample in the qualitative part of the study did not necessarily require being statistically a perfect representation of the whole. As Jennifer says, “If you are using a theoretical or purposive sampling strategy, then whether or not the sample is big enough to be statistically representative of a total population is not your major concern.”³⁹⁶

The logic of theoretical or purposive sampling is that you select units which will enable you to make meaningful comparisons in relation to your research questions, your theory and the type of explanation you wish to develop.³⁹⁷
Therefore, a criterion was set to select the particular units (sample institutions) out of the many, by giving more attention to ‘understanding the process’ rather than ‘representing a population’ while deciding on the number of institutions.

2. 
   Combination of Methods
   

In order to capture features of the status of women in theological education, the study gathered some quantitative data on men-women enrolment in 2005-’06. Logic in practice was relatively messy with more ambiguity and therefore the methodological process was carefully determined to ensure reliability of the information used to formulate concepts. The researcher took advantage of her personal insights, feelings and perspective to understand the life of women in theological education which she has associated with for more than a decade. Neuman termed this as “the informal wisdom of the researcher.”³⁹⁸ However, the informal wisdom of the researcher was not used in isolation anywhere in the study but was checked against the field data to provide maximum objective information.³⁹⁹ The task involved was mainly to provide a reflexive account of how the data were gathered from the field (reliability of the samples and tools of data collection) and how concepts were drawn from them (interpretation of the data).

To guard against the hazardous subjectivity in terms of the reliability of research, this study also relied on the combination of research methods. For example, the basic quantitative information on women in theological education in Kerala, which has not had any common, authentic record so far, can be a source to supplement or complement the qualitative data used for the interpretation. Such combining attempts have been accepted as credible by social researchers such as Sprague and Zimmerman, who remarked,
We do not have to reject quantitative methods to approve of qualitative methods. Posing one against the other is presenting a false choice, especially from the perspective of feminist and other sociologies of knowledge which recognize that each way of doing research is a construction and has its biases.⁴⁰⁰
The inclusion of quantitative data was not necessarily because logically they were crucial to establish truth but to ensure the data collected from the field could go well with the statistical data to form a blend of reliable information to explain the social reality. Having identified the fact that seminaries were not always willing to open up their policies and practices with regard to women, the researcher considered what Neuman calls ‘triangulation of research methods,’⁴⁰¹ which assumed that different methods reveal different perspectives, which would produce significant findings while exploring a social reality. For instance, interviews with theologically trained women, women currently on campus, men students and chief functionaries of seminaries may provide differing perspectives on the topic under consideration. In spite of fearing the variation in results, I intentionally chose to present those variations, if any, as important pieces of knowledge about the social reality. Lever noted, “Variation in results, yielded by different methods, far from being an unwanted source of error or bias, can be an additional source of data.”⁴⁰²

2. 
   Triangulation of Methods/Cross Checking
   

Generation of data only from women students would not have provided the overall picture of the situation of women in theological institutions with the church-related and social elements involved. It was, therefore, decided to interview men students and the leadership of theological institutions so that through the ‘triangulation of methods’, a much more ‘rounded portrayal of the situation’ might well be gained.

…efforts should be made to cross check findings, and in a more extensive study, to use more than one method of data collecting. This multi-method approach is known as triangulation, as is described in Open University Course E811 as ‘Cross checking the existence of certain phenomena and the veracity of individual accounts by gathering data from a number of informants and a number of sources and subsequently comparing and contrasting one account with another in order to produce as full and balanced a study as possible’.⁴⁰³

The triangulation of methods is found especially helpful in studies related to education. Cohen and Manion take it a stage further by their evaluation of the study of comprehensive schools, “Multiple methods are suitable where a controversial aspect of education needs to be evaluated more fully… a much more rounded portrayal of these [specified educational] institutions is required and here is a clear case for the advocacy of multiple methods.”⁴⁰⁴ They suggested the multiple methods could measure and investigate factors such as academic achievements, teaching methods, practical skills, cultural interests, social skills, interpersonal relationships and community sense. Triangulation, which has been an increasingly accepted method in social research with its specific objective to control bias, enabled the researcher to cross-check the findings, to make comparisons and contrasts between views and to have a comprehensive view of the situation. There are also issues relating to reliability and validity of the empirical data.

2. 
   Reliability and Validity
   

Some measures are necessary in this research to ensure the reliability and validity of the data gathered. Wragg, while studying the conduct and analysis of interviews raised questions such as, “Would two interviewers using the schedule or procedure get a similar result? Would an interviewer obtain a similar picture using the procedures on different occasions?”⁴⁰⁵ “Whatever procedure for collecting data is selected, it should always be examined critically to assess to what extent it is likely to be reliable and valid.”⁴⁰⁶ Judith Bell defines these concepts, “Reliability is the extent to which a test or procedure produces similar results under constant conditions on all occasions….Validity…. tells us whether an item measures or describes what it is supposed to measure or describe.”⁴⁰⁷

An unreliable datum might lack validity; but not all reliable data may be valid. Therefore, even when the most representative sample of respondents was selected, the research needed to set the questions rightly to make the study valid. The study, hence, was conscious of the fact that the Kerala picture cannot represent the women in theological institutions in India as a whole or even part of South India, such as Bangalore. For example, it would be worthless to mention what percentage of my sample is against the theological education of women. To ensure maximum objectivity, the sample included seminaries that were very much part of churches with various doctrinal and ritualistic differences and institutions that were affiliated by different accrediting agencies and institutions that have relatively long standing history to highlight their foci and emphases. The data generated from such variety of a sample was expected to ensure reliability as far as the whole of Kerala scenario was concerned. The issue of reliability, nevertheless, raised a few practical concerns in the research. Each institution had a different structure and maintained different focuses and concepts in the training of women. The research had difficulty in assessing the variety of information received. Three institutions of ecumenical category differed among themselves due to the doctrinal and practical background of the churches behind them and the way each looked at the society and the traditional practice of the church in Kerala. The same was true with the non-ecumenical institutions. However, in this task to make a serious effort to ‘place the issue on desk’; the researcher attempted to truthfully present what was gathered from various sources, using the sources and tools that seemed most appropriate.

4.10 Data Analysis Method
The Critical-Interpretive Social Science method employed in this research was essentially a qualitative, action-oriented method, intending to help people better their lives. The cultural and ecclesiastical challenges of women students in seminaries were investigated and the unidentified or ignored realities that require attention were brought to the surface, with a view to advocate change. The procedures of data generation and analysis follow in general-not scrupulously- the guiding principles of the qualitative inductive Grounded Theory method developed by Glaser and Strauss.⁴⁰⁸ The theory in the current study derives from the challenges faced by women students in seminaries. The data generation tools were formulated with the guidance of relationship among concepts, proposed in the hypothesis- specifically in its stated underpinnings. Concurrently with data generation, the meanings and patterns of relationship in attitudes and practices of people regarding the topic were identified and compared.
4.10.1 Data Description and Interpretation
Data from men and women students, leadership and theologically trained women were analysed separately to ensure clarity in the views each represented. Similar data from the schedules were grouped and given conceptual labels by their patterns of relationship. These themes or conceptual labels were derived from both actual words and conceptualizations of the data. Whenever such themes were displayed, actual quotes from respondents followed to ensure openness and credibility of the data. This helped to increase the ‘theoretical sensitivity’ (awareness of subtleties of meanings) of the data, a key factor in Grounded Theory. Due to the multifaceted nature of the topic under investigation and as per the ethical commitments made to the sample seminaries, the identity of institutions as well as of each respondent was kept confidential.
Coding of the data was done through three steps- open coding (breaking down and categorizing of the raw data into its major dimensions), axial coding (putting the data back to make connections between the major categories by identifying conditions, consequences and outcomes of action/interaction) and selective coding (process of selecting the core category and systematically integrating it with other categories).⁴⁰⁹ Data from all categories of respondents were compared and verified against each other in the interpretation section. Though generalization on each theme was not the purpose or a viable result of the topic, precise theories on certain aspects of the research were inevitable results of the study. For the presentation of data, tables that display the responses of the interviewees are used. All significant themes and their inner meanings are presented in tables or diagrams mostly with the frequency of responses with further descriptive note on each. In doing this, an attempt is made to limit the reporting of comments by presenting only the representative quotes which will best illustrate the abstract ideas shown on tables. Whenever the numerical presentation seemed not of central significance, such ideas are shown in diagrams instead of tables. While presenting data that are of general cultural significance, institutional categories are not shown on tables but only the frequency of responses.

4.11 Summary

I have set out in this section to explain my proposed study, the methodology, some of the potential limitations and how I gained access to a suitable and valid sample. This empirical research followed a qualitative, inductive approach where the critical interpretive social science method was employed. It reflected on the feminist critique of the power assumptions inherent in theological education, although not titled as exclusively a feminist research. The Grounded Theory method and the combination of the interpretive and critical social scientific approaches with underpinnings philosophically associated to feminist approaches was crucial to maintain a balanced stance between the feminist methodological priorities. While the basic sampling frame and data analysis employed guiding principles from grounded theory, various tools such as personal interviews, official document analysis and focus groups were used to gather primary data from the twelve sample institutions. The following chapter presents and analyzes the empirical data.

CHAPTER 5: SCENARIO OF WOMEN’S THEOLOGICAL TRAINING IN KERALA: PRESENTATION, ANALYSIS AND FINDINGS OF THE DATA