«LEARNING TO TEACH PHYSICAL EDUCATION IN PRIMARY SCHOOLS: THE INFLUENCE OF DISPOSITIONS AND EXTERNAL STRUCTURES ON PRACTICE by Ian Pickup, BA (QTS) A ...»
The findings are presented as relevant to the particular relations of the participants taking part directly in the study. A commitment was maintained, however, to ensuring internal generalisability; in other words, through appropriate sampling and research design, it was possible to generalise findings so that trainee teachers studying at the same institution at the same time could recognise as valid the accounts being presented in the findings. By presenting research findings in this way, it was also an intention to provide scope for future research to consider trainee primary teachers in different contexts at different times in order to validate the claims being made more broadly.
Ethical Issues An ethics application was completed and submitted to the School of Education Ethics Committee in 2004. Upon receiving feedback, the case study institution was selected and compliance with ethics procedures within that institution sought and confirmed following attendance at a local Ethics Board. This Board raised specific concerns regarding the nature of interview questions, in particular seeking assurance that deeply personal information relating to the sexuality of trainees would not be sought. Full ethical approval was granted by each institution in advance of the research commencing. The research was planned and conducted in full compliance with ethical guidelines provided by my academic institution (Roehampton University, 2004), the case study institution within which the research took place (anonymous institution, 2000) and with reference to national guidelines (British Education Research Association (BERA), 2004; British Sociological Association, 2002). As encouraged by BERA, the research was planned and conducted with an ethical respect for people, knowledge, democratic values, the quality of educational research and academic freedom. Of particular relevance are BERA’s guidelines for the researcher’s responsibilities to participants and to the research community. In this research, broad ethical responsibility included respecting all participants and non-participants within the research context. This included all trainee teachers who voluntarily took part in each stage of the investigation, other trainees who did not form part of the sample, academics and tutors working with the trainee teachers and professionals
working with trainees in school contexts. Every care was taken to:
operate within an ethic of respect for any persons involved directly or indirectly in the research they are undertaking, regardless of age, sex, race, religion, political beliefs and lifestyle or any other significant difference between such persons and the researcher themselves of other participants in the research (BERA, 2004, p.6).
Of particular importance was my responsibility to the participants, in which a balanced research relationship was sought to encourage disclosure, trust and awareness of potential
ethical issues (Orb, Eisenhauer & Wynaden,2000). I acknowledged a responsibility to:
ensure that the physical, social and psychological well-being of research participants is not adversely affected by the research. (Researchers) should strive to protect the rights of those they study, their interests, sensitivities and privacy, while recognising the difficulty of balancing potentially conflicting interests (British Sociological Association, 2002, p.2).
I also attempted to anticipate possible outcomes of interviews to guard against potential harm to the trainees. This was thought possible when exploring personal dispositions and experiences which may have triggered painful memories. Although I acknowledged that some trainees could benefit through the interviews being a form of catharsis and a time in which reflection and self-awareness could be developed (Hutchinson, Wilson & Wilson, 1994), I ensured that my moral obligation would be upheld at all times through awareness of the need to place concern for the individual ahead of any need to collect exciting data. Such ethical dilemmas arising from interviews are difficult to predict, although I made every possible attempt to prepare myself for eventualities of this kind, thinking through a variety of scenarios in advance. I also acknowledged the need for a continuous renegotiation of consent (Kvale,
1996) at each new data collection episode.
At the outset of the study, a detailed explanation of the research was provided, which included an account of the need for the research to take place and how and where the research would be reported. This explanation was given verbally, in person, to a whole cohort of first year trainee primary teachers at the outset of their course and at every subsequent data collection occasion.
The distinction between academic course requirements and voluntary participation in the research was highlighted, ensuring that the trainees understood that involvement in the research would have no bearing on academic or professional progression. Written voluntary informed consent was received from each participant before the start of the research process (appendix 8) and assurances were provided to each participant regarding the right to withdraw, to levels of confidentiality and data protection. All data was kept securely throughout the duration of the investigation and emerging findings and draft reports shared with respondents at appropriate times.
Data Collection Selection of participants and their context Participants were initially recruited as a voluntary, purposive sample. The trainee teachers were embarking on a three-year (2004-2007) BA undergraduate programme in Primary Education leading to QTS, studying at a Higher Education provider of ITT in the Southeast of the United Kingdom. Access to the cohort was negotiated with course staff before the start of the first semester (Autumn 2004), and consent requested from participants at the end of a whole-cohort lecture in the first week of the course (appendix 8). I explained the study verbally to the Year 1 cohort in person. Trainee teachers were not placed under pressure to participate, and it was made clear that withdrawal from the study or the withholding of specific information was possible at any time. Anonymity was guaranteed, with pseudonyms used throughout the study and in presentation of data.
The initial 83 respondents were aged between 18 and 47, including only 6 males, reflecting the gender imbalance typical of the primary teaching profession in the United Kingdom (Capita Teachers’ Pensions, 2004). A degree of natural wastage from the original sample (trainees leaving the course, not volunteering for on-going involvement) was anticipated. From the original sample of 83 participants, 24 volunteered to continue in group interviews and 15 then participated in individual interviews during the first stage of the investigation (see appendix 9). The number of participants taking part in years two and three decreased, partly due to a natural wastage where trainee teachers decided not to carry on, or in some cases left the course. This reduction in numbers was also planned for, in that a smaller number of participants were identified for further investigation as they were representative of the emerging findings. The reduction in the number of participants involved was from fifteen to fourteen and then six in Year 2 and 3. The small group of trainees interviewed in Stage 2b were developed as ‘in depth’ case studies in Stages 3a and 3b, whilst two trainees were returned to the sample to enable detailed consideration and testing of emerging findings. The interviews enabled the generation of ‘thick’, descriptive data (Geertz, 1973), the analysis of which enabled emerging findings to be continually developed and refined.
Physical Self Perception and Perceived Importance Profiles The PSPP and PIP scales were used at the start of the investigation for two key purposes.
Firstly, the scales enabled me to meet with and introduce the research to a relatively large number of potential participants. The scales were completed by the trainee teachers during this first meeting, enabling immediate capture of data relevant to the research questions in a relatively unobtrusive fashion. Secondly, the data gleaned from completion of the scales was intended to support the larger and on-going qualitative elements of the investigation.
Although I acknowledged from the outset that conclusions resulting from analysis of the quantitative data were to be treated with caution as a result of the overall sample size and the uneven number of males and females within the sample, the emerging patterns and trends indicated by the quantitative analysis were used to support the development of the interview schedules and lines of enquiry. Thus, the quantitative results were intended not to be viewed in isolation and only tentative findings were drawn from Stage 1a research. A separate quantitative study would be necessary in the future in order to test findings with a substantially larger sample size, particularly including more male trainee primary teachers.
Following a successful small scale pre-test of Fox’s (1990) PSPP within my own HEI, this scale was used for the full study. The small scale pilot study was used to familiarise myself with the profiles, to rehearse verbal explanations and ethical statements and to test my view that these were appropriate tools to use in Stage 1 of the investigation. The instruments are thought to be theoretically grounded, psychometrically well-developed and measures of multiple facets of the physical self (Byrne, 1996).The PSPP was also chosen because of its original derivation from research on a student population (Biddle & Mutrie, 2001) and the relevance of the four sub-domains of the scale to the research questions, particularly in relation to research question 1 which seeks to identify trainee dispositions towards PE and sport. The focus on physical self-perception within the PSPP was also deemed to match closely to accounts of identity given in chapter 3 of this thesis in which teacher identity is thought to be closely linked to teacher development. The PSPP is a thirty-item self-report instrument, containing questions relevant to ‘sports competence,’ ‘body attractiveness,’ ‘perceived strength’ and ‘physical condition’ plus a global perception of ‘Physical Self-Worth’ (PSW). An example of the style and structure of a question (from within the sport competence sub-domain) and the possible range of response is given in figure 4.1.
Figure 4.1 Example question from PSPP
A range of measures of ‘physical self’, including Marsh et al’s (1994) Physical Self Description Questionnaire (PSDQ), Franzoi and Shields’ (1984) Body Esteem Scale (BES) and Richard’s (1987) Physical Self-Concept Scale were reviewed and discounted for reasons including specificity for particular populations (e.g. adolescents in the case of PSDQ) and the focus of questions (e.g. omitting sports competence in the case of BES). Furthermore, as Byrne (1996) contests, despite being originally designed for use with students, the PSPP has also shown evidence of reliability and validity with older adults (for example as applied by Sonstroem, Harlow & Josephs, 1994), suggesting suitability for use with both traditional entrants to Higher Education and mature participants. The participants in this research include those entering teacher training immediately after A- Level study, as a career change in midlife and after child-rearing career breaks (see summary of participants, appendix 10).
The data derived through completion of the PSPP is a numeric representation of respondent self-perception relevant to each of the sub-domains described above. The figures for each category were reviewed and grouped. Respondents within highest (maximum is twenty four for each sub-domain) and lowest (minimum is six) possible scores for each sub domain were identified. A nominal description of ‘negative self-perception’ (accounted for with scores between 6 and 12) and ‘positive self-description’ (accounted for with scores between 13 and
24) was used to group trainee teachers with similar profiles. This enabled me to identify cases across the full spectrum of results together with ‘deviant cases’ to ensure that examples for further investigation were sought to challenge original suppositions (Silverman, 2001).
Attention was paid to specific sub-domains and emergent themes, similarities, patterns and questions noted. Descriptive statistical analysis highlighted the percentage of scores present in each sub domain, along with calculation of mean, mode and standard deviation. In addition, the percentage of respondents within nominally positive and negative self-perception ranges was noted. Finally, gender sub group mean scores were produced and analysed through the use of appropriate non-parametric statistical tools.
The themes, similarities, patterns and trends identified through the PSPP were subsequently used to plan semi-structured interview schedules. Initial interviews conducted in Stage 1 focused on dispositions, and questions were framed in relation to the quantitative data derived from the scale. Such questions enabled validation of the statistical analysis, but also served to stimulate trainees’ reflection regarding the issues being investigated. The numerical values were a point of cross-referencing and referral throughout the analysis of interview transcripts and were used to check meaning, researcher understanding of participants’ viewpoints, and to monitor the characteristics of the reducing size of sample as the research progressed. On completion of the PSPP, respondents were also asked to complete the accompanying Perceived Importance Profile (PIP, see appendix 6). This was intended to provide a further source of descriptive data to illuminate patterns and trends. In particular, it was hoped that the PIP would provide information to identify trainee teachers for whom a negative selfperception could potentially be negated by a low perceived importance, those for whom a high perceived importance could create a discrepancy in behaviour if a low self-perception score was achieved, and those who attach a high level of significance to a sub-domain and rate themselves highly in that same category. The PIP contains eight questions, two in each of the four sub-domains in the Fox (1989) hierarchical model. The eight extra questions were not expected to take a great deal of time and are phrased in the same fashion as the statements in the PSSP, aiding ease and speed of response.