Home Teachers' Research Research Mentoring News & Events Reviews Links



National (UK)

How can I, as a Reception class teacher, collaborate with my teaching assistant to improve the learning of numerals to 20 with my pupils during mental mathematics sessions?

An assignment successully submitted as part of the Master in Education Programme "Methods of Educational Enquiry" at the University of Bath (2004)

Tutor; Sarah Fletcher
Researcher: Heidi Hughes


This essay aims to illustrate the issues that occur when planning and carrying out a small-scale educational enquiry. To thoroughly explain these it is far easier if the context of this assignment is based around a real problem.

My essay will start by explaining my personal context and ethical issues, then look at how to chose an appropriate methodology and ways to collect data and finally, how to analyse this data and evaluate the research process. Throughout, I will highlight any difficulties that may potentially occur and how these can best be resolved.


Hopkins (1993) discusses how, in more recent times, there has been a need to combine classroom research and whole school development, in order to improve teaching and learning in schools. Therefore it seemed appropriate to begin my MA in Education at Bath University as the teacher researcher in an enquiry based unit, looking at improving a specific area of classroom practice. I am hoping that this unit will enable me to understand various methods of educational enquiry, to consider their importance and when appropriate, to incorporate them into my research work. This assignment should enrich my understanding of educational research methodologies in preparation for my final dissertation.

I am currently employed as an Infant teacher by Tanglin Trust School, Singapore (TTS). The school teaches the British National Curriculum to over 1700 expatriate students aged from 3-18. Within each Infant year group there are eight parallel classes with 24 children in each. A qualified teacher conducts teaching, with a full time locally employed teaching assistant (TA) working alongside. For these Infant classes the emphasis is on the acquisition of skills and understanding in Literacy and Numeracy. Both the Literacy and the Numeracy Strategies have been incorporated within the school and these lessons are taught daily. When academic achievement is measured against UK averages TTS performs above UK averages .

For the past academic year I have been teaching a Reception class with the support of a Singaporean TA. Many of these TA’s have been employed by TTS for over 5 years and have adapted easily to working with a transient teacher population and a constantly changing British National Curriculum. For the majority, their only training is on-site and given by teaching staff or by a Teaching Advisor. Last year it was highlighted to Senior Management that morale was lower than perceived amongst TA’s . A working party investigated the situation further and found that TA’s felt they were not fully utilised during teaching sessions. Teachers were then encouraged to include TA’s far more effectively during taught lessons.

Within my own classroom the lack of TA involvement was more often prevalent during whole class teaching sessions. Rather than look at all such occasions, I decided that oral work and mental calculation sessions in Numeracy were a specific area of weakness. Positioning, interacting and collaboration between the teacher and the students are well established but the links between the TA, students and teacher are lacking and in need of improvement.

Thus the focus of my enquiry will be on the following: How can I, as the teacher, improve collaboration between myself, my TA and my students, to further their learning of numerals to 20 in my classroom during oral/mental mathematics sessions? Although I will only be researching this issue in my classroom, nearly thirty other teachers work alongside a TA in the Infant School and therefore, there is always the possibility that my findings/recommendations may provide assistance beyond the scope of this small-scale enquiry.

A question of ethics

Before starting any research, it is important to consider research ethics. Cohen and Manion (1994) remind us that at any stage during research issues relating to ethics could materialise. Kimmel (1998) sees ethical problems as both personal and professional. In my specific context, personal factors relating to cultural differences between my TA and myself will be at the forefront of my concern. It will be essential to clearly explain my research and also stress I am attempting to address my failures regarding appropriate TA involvement during whole class teaching. By doing this I aim to avoid my TA experiencing any ‘loss of face’, a common Asian-based phenomena. ‘Informed consent’ from my TA at the beginning of the project is vital, as is providing assurance of anonymity and confidentiality during all stages of the research (Cohen and Manion, 1994). In my given situation there would be no question of not sharing information and results as TA’s input would essential if change is to occur. In the Singaporean culture, there is often the feeling that there is only one ‘right way’ and that this must be found first time. It will therefore also be important to develop an understanding that we may not find the right solution immediately and this is acceptable. In terms of the interactions taking place between the Students, TA and myself, I draw upon the ideas of Frankfort-Nachmais and Nachmais (1992) quoted in Cohen and Manion (1994):

‘The principle of informed consent should not…..be made an absolute requirement of all social science research. Although usually desirable, it is not absolutely necessary to studies where no danger or risk is involved’ (p 351)

My research primarily revolves around collaboration between the teacher and the TA and although the children are involved, I perceive there to be no real danger or risk to them. However, as part of my regular classroom practice, I would explain to the children what was taking place, especially if the methods involved them directly (e.g. a questionnaire or a video recording). This is supported by the work of Fine and Sandstrom (1998), who stress that, regardless of the age of children, they have the right to an explanation at their own level of understanding. Projects that involved children for longer and more involved periods of time would require informed consent from the children’s parents and the school management. In line with good professional practice, I would ensure I gained official permission from senior management and inform my line manager of any project developments that would involve the students further. On completion of my research, naturally all involved will be thanked and the results shared openly between us.

Choosing a methodology

When looking at which methodology to adopt it is important that you clearly define what your enquiry is about and why you are doing it. In very broad terms, there are two principle methodologies used in educational research today. The first being the scientific methodology, where the researcher takes an objective or hypothesis and, through the creation of laws or by finding the right answer, aims to find the truth in a given situation. This should provide a model/theory that can be used repeatedly to predict and control events. The second methodology is the naturalistic approach. Here the views, behaviour and opinions of people, on and within the environment, are of far more importance. The aim here is to look more closely at attitudes and to finding a solution to a given problem. There may be more than one solution depending on the context in which the problem is set. My research question could have a number of solutions given that I am dealing with two different teachers interacting with a class of 24 children. The various possible solutions will undoubtedly be linked to the behaviours of all involved, making it unlikely that a universally applicable model/theory is appropriate. This led me to adopt a naturalistic methodology.

Within the naturalistic approach, both action research and case studies are commonly adopted. Case studies, as defined in Bell (1987), are useful when attempting to clarify what is going on in a given situation, Cohen and Manion (1994) go onto explain how the case study researcher focuses simply on observing the characteristics of an individual (such as a child, teacher, school or LEA) and by doing so aims to make generalisations about other similar individuals. Adelman et al (1980) comments on how case studies are ‘a step to action’ and provide individuals with a wealth of data to interpret, however collecting a substantial number of observations in this manner can be a time consuming and costly process. Action Research as described below by Cohen and Manion (1994), is

‘essentially an on the spot procedure designed to deal with a concrete problem located in an immediate situation: This means that ideally, the step by step process is constantly monitored over varying periods of time and by using a variety of mechanisms (questionnaires, diaries, interviews, case studies, for example) so that the ensuing feedback maybe translated onto modifications, adjustments, directional change, re-definitions, as necessary, so as to bring about lasting benefit to the on-going process itself’ (p 192)

Whilst this definition interrelates case studies with action research, McNiff, Lomax and Whitehead (1996) explain that:

‘action is at the centre of your action research…action to which you, the researcher are committed by your personal and professional values, action that is informed by your careful considerations about it’s appropriateness, and action that is intentional and undertaken by you to achieve the objectives you have set’(p 71).

From my teaching experience, many of us complete action research without even realising it (McNiff, 2002). My research question centres around me as the teacher changing the way in which I teach, thus an action research approach seems more appropriate than a case study approach. However having made this choice I would also like to state that action research is carried out systematically which highlights the fact that although naturalistic in some ways there are many characteristics of scientific research within this approach. McNiff, Lomax and Whitehead (1996) add to this argument by discussing the problem-solving dimension as similar to scientific methods and McKernan (1991) suggests that action research ‘contributes to the science of education (p 33)’.

Action research in practice

Action research has been a popular model used in educational research since the 1960’s. Various action research models looking at the improvement of an individual through the practice of self-reflection are found in the work of Kemmis and McTaggart (1982), Cohen and Manion (1980), Ebbutt (1985), Elliot (1995), McKernan (1991), Bassey (1998) and McNiff (2002). Although similar, they do have differences. Elliott (1993) argues that Kemmis (1982) saw action research as a linear model with the initial problem remaining static, whereas he felt that the initial problem could be shifted during the process of the enquiry in a somewhat more flexible approach. This debate is then expanded upon by McNiff (2002) who discusses how it is important to both keep action research open ended but also to remember to frequently check and ensure we are still striving to solve the initial problem. Elliott (1993) very much focuses on the benefits that can be derived from the process of action research. His opinion, based on curriculum research, being that action research can be used as a strategy to maximise pupils’ achievements and thus assist in reaching National Curriculum targets. However it should be noted, that in his opinion, this strategy doesn’t take into account the ethical dimension of teaching and learning (p 52). Kemiss and McTaggart, Hill and Kerber and Lewin, amongst others, all see action research as purely a collaborative activity. Winter (1996) goes on to explain that collaboration

‘is intended to mean that everyone’s view is taken as a contribution to understanding the situation (p 13)’ Cohen and Manion (1994) further add to this stating:

‘Advocates of action research believe that little can be achieved if only one person is involved in changing her ideas and practices. For this reason co-operative (action) research tends to be emphasised and encourages’ (p 190)

In contrast, Stenhouse and Whitehead see this as limiting and argue that action research can also be individualistic and still be beneficial (Hopkins, 2002). Whitehead (1989) discusses the importance of the word ‘I’ being significant in the action research title and emphasises the importance of the researcher seeing them self as a ‘living contradiction’ in the question they are asking. That is to say, each of us has our own set of educational values and by reflecting and looking to improve our own practice, we are not adhering fully to these values. We are contradicting ourselves, which makes it fundamental to put the ‘living I’ right in the centre of the enquiry. From here we are act as the both the practitioner and the reporter in the given enquiry. Whitehead then explains how this, combined with the ideas of others including Collingwood (1978), Lomax (1987) and Polanyi (1958), can be drawn upon to produce what he calls a ‘living educational theory’.
From here several authors have attempted to combine these two schools of thought, collaborative and individualistic. Kemiss and McTaggart (1988) discuss how action research becomes collaborative as individual actions are examined. McNiff (2002) sees the focus of action research as you and your learning but also as a process of understanding this influences the learning of others (p 87). She goes onto clarify that:

‘Action enquiries begin with an individual’s question, ‘How do ‘I’ improve my work?’ When the enquiry is shared with others, and they wish to be involved – possibly by critiquing or by deciding to do something similar, or by offering new ideas for new enquiries - then the question changes to ‘How do we improve our work?’ (p 10)

In my opinion, when deciding on the most applicable model, you need to look for the model you find easiest to work with. Having considered both my research question and the available methodologies I have decided to conduct my research using a collaborative action research approach. Although theoretical models are a useful reference tool and assist with the construction of an initial framework they will never be able to be followed precisely (Elliott, 1993). I feel the model put forward by McNiff (2002) shows the action research cycle most clearly and also comes with the useful warning that although appearing clear cut it will be subject to the unpredictable nature of the real world. It attempts to both address a particular professional problem and implement a solution but also continues on as further problems arise or a new area of practice is identified for investigation.

* Identify an area of practice to be investigated;
* Imagine a solution;
* Implement the solution;
* Evaluate the solution;
* Change practice in light of the evaluation………

Adapted from McNiff (2002, p 11)

Before starting any action research it is essential to systematically plan the process you intend to go through in order to complete your research. The first part of the planning process is to decide upon the duration of the research cycle. Given my research question I envisage my action research can be completed in one term and my plan can be found in Appendix 4. Highlighted in this plan are the processes that need to be taken into account when completing this type of research including, reviewing appropriate literature & seeking constructive criticism and advise from colleagues. Reviewing literature when conducting ‘action’ research in which the focus of enquiry is your own practice may at first not appear appropriate. However it can be very beneficial and important to compare and contrast your experiences with previous research results. Constructive criticism and advise from an appropriate colleague can also be beneficial in assisting your research when discussion with those collaborating in your action research may be leading you to a narrower focus or complete rethink. This person, known as a critical friend, can be anyone who you feel has the time, the interest and the ability to be critical if and when necessary. They should acts as a:

* Witness - in terms of knowing how the research process took place
* Helper – ensuring the researcher gives a good account of what has happened
* Evaluator – providing support and criticism as appropriate
* Supporter – providing praise, sympathy and encouragement
(Adapted from McNiff, Lomax and Whitehead (1996) p 85)

It is important that as stated by McNiff, Lomax and Whitehead (1996), action research originates from ideas the practitioner has taken from their own strong values. In this way, often the fact you have highlighted an issue as a problem illustrates you are in fact not abiding by your own professional values and standards. Bart McGettrich (2000) provides a framework for Scottish teachers undergoing initial teacher training that combines our values with our knowledge, skills and learning. It is this model that I would like to use to initially to highlight and address this important area.

Professional values and commitment

Professional skills Professional knowledge
and aptitudes and understanding

Methods of data collection

When collecting data and finding evidence your aim should be to show an improvement in your practice and that you are having an influence on others (McNiff, Lomax and Whitehead, 1996). Regardless of whether the data is collected using quantitative or qualitative methods it is important to be systematic. Cohen (1994) quotes Merton and Kendall (1946) stating that

‘Social scientists have come to abandon the spurious choice between qualitative and quantitative data: they are concerned rather with the combination of both, using the most valuable features of each (p 40)’

All good data should be dated, labelled and stored in terms of context and authenticated with a signature. Various data gathering techniques are appropriate, including diaries, questionnaires, video and audio recording, observations, interviews, sociometry and documentary evidence. The methods to adopt depend on the research question being asked. All involved should decide upon the methods, help set the performance indicators and then aim to see if the parameters set have been fulfilled. To avoid personal bias at this stage triangulation (comparing two or more views of the same event/idea) of data sources and individual perceptions is vital.

Action research must involve keeping a diary to monitor both progress and behaviour (Kemiss and McTaggart, 1992). Collaborative diary keeping, although highly sensitive in ethical terms, allows for vital triangulation of this data source. Although subjective and not rigorous in nature diaries can provide evidence of change over time. Entries should be made following meetings, lessons, discussions or any other relevant times. Meetings should be carefully documented and due to their subjective nature analysed alongside other data such as lesson plans, which are a documented form of data showing what has happened over time. Critical conversations should also be recorded as they can assist in highlighting changes in practice, the motivation for doing so and also provide evidence that the validation process has been continuous (McNiff, Lomax and Whitehead, 1996).

Given that my research is classroom based, observations will be important but the techniques utilised to record the observations must also be appropriate. Whichever recording technique is adopted it is important that the person doing the recording familiarises themselves with the classroom, arrives before the lesson begins and if possible studies a series of lessons to allow children to behave as normally as possible (Wragg, 1994). Various approaches to the process of recording observations can be found in literature on action research including systematic, descriptive/narrative or technological (Simpson and Tuson, 1997). Descriptive/narrative recording involves making notes on the whole lesson allowing behaviour contexts and the sequence of events to be recognised. However such a method is usually best completed over an extended period of time, which would not suit my purpose. Technological recording (e.g. video or audio taping of lessons) provides the researcher with permanent data and a true, unbiased version of events. Initially as a recording tool this method appears most valuable but children can get in the way of the video, not speak clearly enough to be heard, perform for the camera or become intimidated. Transcribing the video or audio can also be time consuming. Systematic observations, based on procedures and categories carefully worked out in advance, can provide an objective account of the lesson (Croll, 1986). They are also less likely to be biased and provide quantitative data. Various published methods, such as the Flanders Interact Category System (FIAC), can be adopted if they suit the research question or as Croll (1986) states:

‘researchers should not be reluctant to devise their own systems if their research problem demands it (p 47)’ but ‘will almost certainly be of value to incorporate some aspects of a well established schedule (p 50)’.

When completing a systematic observation either time sampling, continuous or event recording occurs. Time sampling sees an observer record at set time intervals providing data on the specified events but this method cannot provide an indication of true frequency. For event recording, the recorder looks for the predetermined events and records their frequency not their duration. This approach would seem more appropriate for my research, with the critical friend acting as observer and recording the number of interactions between my TA and myself. Such an observation would be similar in nature to sociometric methods where marks are made on interaction lines, however McNiff (2002) argues you cannot draw rigid conclusions from such a diagram, as they do not account for the behaviour of others within the classroom. To account for this, the observation sheets would show agreed categories for interactions (e.g. nod, talk, and smile) but would also allow observers to add additional comments on behaviour after the observation. A diagrammatic classroom plan would also be included showing the position of people being observed.

Questionnaires are a quick and easy method of obtaining information. However, the composition of them can be challenging and extensive preparation is required in order to gain an unbiased opinion. Question structure needs to be carefully addressed, with the avoidance of leading or hypothetical questions and assumptions being particularly important. For my research I need to ascertain the personal values of all individuals and as my action research is collaborative to find this via interview would require an external interviewer asking all parties involved the same questions. Although interviews can provide in-depth responses a questionnaire would, if worded well, prove less threatening. If needed another similar questionnaire could be conducted at the end of the research and comparisons made. Pupil’s opinions are often interesting and I would propose to give children a simple multiple-choice questionnaire to analyse their perception of our interactions. At all stages in the writing of questionnaires, critical friends should be involved to avoid bias. If the research was at a larger scale a sample and pilot should be undertaken.

By using questionnaires and an initial systematic classroom observation, analysis of data can be triangulated and a baseline formed at the start of the research. Diary entries, lesson plans and minutes from meetings will provide data during the research and at the final stage a second observation, and repeating of the children’s questionnaire will aim to show how action has led to change. Although for my research question these are the most appropriate methods, for other questions they may not be.

Data analysis and evidence

The various strategies that are going to be employed to analyse the data collected need to be established prior to data collection (Croll, 1986). For quantitative data, well-defined statistical methods of analysis can be adopted. However as discussed by Miles and Huberman (1984), lacking are clearly defined methods of analysis that assist with interpretation and validation of any qualitative data that may be present. Both Hopkins (2002) and Miles and Huberman (1984) have looked in more detail at how such qualitative research can be interpreted. Whether looking at quantitative or qualitative data to find evidence all relevant people must be involved to avoid bias and clarify interpretations (Lomax, 1991). In action research you are looking for changed thinking over time as a direct result of the action. If you can support your claim with data, then it becomes evidence (McNiff, Lomax and Whitehead, 1996). Once a researcher has found some evidence it is important to validate it by finding consistency within the results. This can be achieved through setting up a validation group drawing upon a wide circle of interested parties including your critical friend. Through sharing meaningful personal experiences the ability to validate any evidence is increased. Such meetings make the work public, allow fair and accurate judgements to be made and may also provide further insights into the research question. Such gatherings can occur throughout the research and should be transcribed.
The reliability of the data collected is often somewhat reflected by the generalisation of findings. That is to say, if this research were to be carried out at another time or by another researcher would their findings be the same? Given the relatively unique classroom situation in which this action research will be conducted I would somewhat question the reliability of my data. Which is not to say the research will not be useful, far from it. I would be more than willing to put forward recommendations to colleagues based on the data and my findings.

Research evaluation

Upon the completion of any action research the authors must insure both themselves and their critical friend carry out a thorough examination of their work. Bell (1999, p 221) lists a series of questions to help first time researchers evaluate their research. If the research was successful, practices will have changed, thinking will have been modified and professional development furthered. As a final mention McNiff, Lomax and Whitehead (1996) state that ‘if their lives (research participants) are better, the research may be evaluated as effective (p 43)’. However this being action research, undoubtedly a new research question will have materialised and the whole process could begin again.


Throughout this essay I have aimed to show that when conducting educational research, it is essential to do so by following a clearly defined and structured approach. Researchers must consider the most appropriate methodology, the most effective ways to collect the data and finally the best way to interpret it in any given situation. Consideration should also be paid to the words of Dadds and Hart (2001).
‘Perhaps the most important new insight for both of us has been awareness that, for some practitioner researchers, creating their own unique way through their research may be as important as their self-chosen research focus. We had understood for many years that substantive choice was fundamental to the motivation and effectiveness of practitioner research (Dadds 1995); that what practitioners chose to research was important to their sense of engagement and purpose. But we had understood far less well that how practitioners chose to research, and their sense of control over this, could be equally important to their motivation, their sense of identity within the research and their research outcomes.’ (p 166)

Here they explain how methodological inventiveness is not only appropriate but is important in leading to unique enquiry and boosting motivation and effectiveness at the individual level. However, justification for any given research choice must be explained, as should any limitations that may arise. For an educational researcher to successfully complete a given enquiry all these factors are important along with the ability to review their work throughout the process.


Adelman, C., Jenkins, D. and Kemmis, S., (1980) Rethinking case study: Notes from the Second Cambridge Conference’,in Simons, H. (ed.), Towards a Science of the Singular. University of East Anglia, Centre of Applied research in Education
Bassey, M. (1995) Creating Education through Research. Newark, Kirklington Press
Bell, J. (1999) (First pub. 1987) Doing your Research Project. Buckingham, Open University Press
Cohen, L. and Manion, L. (1994) Research Methods in Education. London, Routeledge and Kegan Paul
Croll, P. (1986) Systematic Classroom Observation. Lewes, Falmer Press
Dadds, M. & Hart, S. (2001) Doing Practitioner Research Differently. London, RoutledgeFalmer.
Ebbutt, D. (1995) Educational action research: Some general concerns and specific quibbles, in Burgess, R (eds.) Issues In Educational Research. Lewes, Falmer Press
Elliott, J. (1993) Action Research for Educational Change. Buckingham, Open University Press
Fine, G. A. and Sandstorm, K.L. (1998) Knowing children: Participant Observation with Minors, Qualitative research Methods 15 California, Sage
Frankfort-Nachmais, C and Nachmais, D. (1992). Research Methods in the Social Sciences. London, Edward Arnold
Hopkins, D. (2002) A Teacher’s Guide to Classroom Research. Buckingham, Open University Press
Kemmis, S. and McTaggart, R. (eds) (1988) The Action Research Planner. Victoria, Deakin University Press
Kimmel, A.J. (1988) Ethics and Values in Applied Social Research. Newbury Park, Sage
Lomax, P. (1991) Managing Better Schools and Colleges: An Action Research Way. Clevedon, Multi-Lingual Matters
McKernan, J. (1991) Curriculum Action Research. London, Kogan Page
McGettrich, B. (2000) The Standard for Initial Teacher Education in Scotland. Internet copy.
McNiff, J. (2003) Action Research for Professional Development: Concise advice for action researchers. Internet copy
McNiff, J. and Whitehead, J. (2002) Action Research: Principles and Practice. London, Routeledge Falmer
McNiff, J. Lomax , P. and Whitehead, J. (1996) You and Your Action Research Project Principles and Practice. London, Routeledge
Merton, R.K. and Kendall, P.L. (1946) ‘The focused interview’ in American Journal of Sociology 51 541-57
Miles, M and Huberman, M. (1984) Drawing valid meaning from qualitative data: Towards a shared craft. Educational Researcher, 13(5): 20-30.
Miles, M and Huberman, M. (1994) Qualitative Data Analysis. California, Sage
Polanyi, M. (1958) Personal Knowledge. London, Routledge and Kegan Paul
Schön, D. (1983) The Reflective Practitioner: How Professionals Think in Action. London, Temple Smith
Simpson, M. and Tuson, J. (1997) Using Observations in Small-Scale Research. Glasgow, Scottish Council for Research Education
Winter, R. (1996) Some principles and procedures for the conduct of action research, in Zuber-Skerritt, Z (ed.) New Directions in Action Research. London, Falmer
Whitehead, J (1989) Creating a living educational theory from questions of the kind, ‘How do I improve my practice?’ Cambridge Journal of Education, Vol.19. No.1, 41-52
Wragg, E.C. (1994) An Introduction to Classroom Observation. London, Routledge




Site owner contact details

Click here to join the

Discussion Group