Through this site, I hope visitors will increase their familiarity of and commitment to action research (AR). The site contains rationale and methods for AR, along with links to various websites regarding it. Information and resources available through this site are linked to text at right. This page is, along with my Research & Development Links page, one of two sections of my Research & Development Resources, which, in turn, is one of four sections of my Educational Resources. Please feel free, also, to send links, suggestions, comments, etc. to me @ E-mail to Larry.
AR WWW Links.
is action research?
Action research is a special kind of research. It is, simply, research people conduct to determine effectiveness of actions they take to improve a situation. In education, for example, action research is a way for educators to attempt to improve teaching and learning and, as they do so, to conduct research into those efforts. The table at right represents a way of depicting this view of educational action research. In essence, educators attempt to improve the sorts of outcomes listed on the right by changing conditions like those listed on the left. In general, the research component attempts to document and explain actions taken to improve educational outcomes. Educational action research also has a strong political agenda. It is about who controls teaching and learning; i.e., teachers and students or those outside of actual teaching and learning. Often, outsiders attempt to overly direct what happens in teaching and learning, despite not being familiar with many of the myriad factors that can determine 'success' in any teaching and learning situation. Therefore, educational action research can be a way to empower teachers and students to take more control over what happens in educational situations. This is exceedingly important, especially in the current market-driven environment, in which business interests have more control over curriculum and instruction than those being educated and those charged with responsibility to help learners achieve what is best for them.
are benefits of action research?
For its originator (Kurt Lewin, 1946; also see: Lewin), and for many others since, action research is a process of empowering those on the 'front lines' of action to conduct research into their own practices, rather than relying on and - more importantly - being subject to - research findings of 'outsiders.' In education, for example, teachers are encouraged to conduct research and curriculum development into their own teaching activities. Educational action research is, in a sense, an educational technology, in which teachers take action to improve student learning and, at the same time, gather data to demonstrate possible reasons for and usefulness of that action. It is an ongoing quest by teachers to develop the best possible learning experiences for students. Traditionally, teachers' main job has been to implement what others tell them needs to be taught, using methods others develop. However, those not closely involved in teaching and learning often miss many important factors affecting education and, perhaps more importantly, sometimes do not have the best interests in mind of all students being educated. Action research is, therefore, an opportunity for teachers to take control of curriculum research and development, as well as their traditional roles; i.e., that as implementers of curriculum and teaching strategies. Indeed, as Carr and Kemmis (1986) suggest, action research is "a form of self-reflective enquiry undertaken by participants in social situations in order to improve:
Action Research Methods
Teaching & Learning
POSSIBLE DOMAINS OF INTEREST/CONCERN
Educators can begin action research by thinking about "Possible Result Variables" that are important to them and to societies. There are many views about what 'outcomes' are important in science and technology education, for example. Some suggestions are provided at: of S&T; e.g., laws, theories & inventions, such as law of magnetism, the particle theory of matter, and how pulleys work; ii) Nature of S&T; e.g., what it is like to do science investigations (e.g., it can involve trial-and-error), how science & technology relate to each other (e.g., they depend on each other), and how inventions may affect societies and environments (often negatively, as well as positively); and iii) Skills of S&T; e.g., skills, strategies & habits of mind required for scientific inquiry & technological design, such as an ability to conduct and analyse experiments. Many also would add various attitudes. For example, they may want students to acquire a greater desire to promote a healthy balance of materials and processes in ecosystems.
Once educators have identified some 'outcomes' (Result Variables) that are important, they could (should?) attempt to identify some outcomes that are not being well achieved. For example, as shown in the table at right, they may note that students' "ideas about natural phenomena" (e.g., theories about living things) are inadequate. Or, they may note that students' "skills" (e.g., abilities to design controlled experiments) could be improved.
At this stage, there are - at least - two general ways action researchers can proceed. In a classic sense of their meanings, action researchers could conduct: i) 'science' or ii) 'technology.' If they were to begin with a 'scientific' goal, they would collect data to gain evidence and develop explanations for relationships between "Possible Cause Variables" and "Possible Result Variables." In other words, they would want to understand how natural (e.g., education) systems work. For example, they could gather data to describe and explain how "classroom layout" may be affecting students' development of appropriate "ideas about natural phenomena." They may find, for example, that the seating arrangements in their room hinder development of ideas, because not much conversation can occur among students. Once they have sufficient data to explain such relationships, they could then conduct a kind of educational "technology"; that is, they could take certain actions (e.g., rearranging seating and encouraging discussions) to improve students' development of ideas about natural phenomena. Again, they would collect data to support their claims that the actions they took did lead to improvements in student outcomes. However, it is important to note that action researchers could 'skip' or delay the 'scientific' goal and begin with 'technological' interest; that is, they could go straight into taking what they (or others) consider to be appropriate actions in order to improve student outcomes. There is legitimacy in educators taking action very quickly, in the sense that there is acknowledgement that educators have a wealth of practical experience and, from that, can make legitimate judgements about what aspects of teaching and learning may require improvement. In short, they often know what is not working and likely reasons for disappointing results.
educators choose to start with a 'scientific' goal, they
begin by brainstorming (and gathering data to support)
Variables" may be leading to poor achievement in
A result of brainstorming about such relationships could
in the table below:
For example, a teacher may suggest that the prescriptiveness of curriculum standards are limiting students' development of skills - because of the hypothesis that standard curriculum requires more teacher direction (and more teacher direction limits students' opportunities to develop skills on their own). While this may be a possible relationship, teachers should consider relationships about which they may be able to take positive actions. It may be difficult, in other words, to quickly change the prescriptiveness of the curriculum.
relationship worth exploring would be between certain
strategies (e.g., lecturing) and students' development
of ideas about
phenomena. This may be a worthwhile area for initial
it is something teachers can (eventually) more easily
change - although
that should not stop them from trying to change more
teachers may be quite anxious to get on with trying
to improve learning outcomes, letting the science of how
outcomes be derived later. They may want, for instance,
to develop ways
to affects students' attitudes towards energy
conservation. This is a
'technological' goal, since the aim is to improve education,
rather than just
understand it (which would be a 'scientific' goal).
For whatever goal educators plan to conduct research, it generally is a good idea to have various theories in mind to help explain the possible relationships. For example, to explain how a predominance of lecturing would limit students' development of ideas about natural phenomena, they could draw on constructivism. This learning (and knowledge development) theory suggests that learners learn by changing ideas that are already in their minds and, because many of these ideas are subconscious, it is important to encourage learners to explore and clarify their ideas before instruction begins. Considering such theories is very important and, indeed, may be among the first things that action researchers should do. Some other educational theories are at: Learning.
COLLECTION AND ANALYSIS
In order to make claims about teaching and learning that people might believe, educators often need to systematically collect relevant data. Such data collection and analysis often occurs several times during an action research project. It may occur, for instance, in the beginning in order to determine current conditions of teaching and learning. Then, after appropriate actions are taken, data collection and analysis likely will have to be repeated - in order to gain evidence to support the idea that the actions likely were the reasons for changes in outcomes.
educators in their research, this page has a special
methods. It is located:
& Developing New Actions
CONDITIONS TO CHANGE (i.e., Actions to Take)
Working with such broad questions, like "What teaching strategies can be used to improve students’ abilities to conduct their own experiments?," demands considerabe brainstorming, discussions with others, and reading of educational literature (for example). Moreover, this suggests that planning action also involves going back to the earlier stage of action research - i.e., "Reviewing Teaching & Learning" - and re-examining the concerns for which actions may be planned. Indeed, as the chart at right is meant to indicate, planning actions may involve further analysis of cause-result relationships between educational conditions and outcomes. The analysis may tell you that, if conditions are unsatisfactory (leading to poor outcomes), then a solution may be to reverse the conditions or develop new conditions that may lead to improved outcomes. So, for example, a teacher finding that students seem to lack skills for conducting experiments, may try to develop activities to specifically develop these skills. The chart at right lists several kinds of action teachers might take in a science and technology programme. Other ideas may arise through discussions with colleagues, with specialists in the field and through reading journals, magazines, books, websites, etc. An excellent place for action researchers to start is by examining reports of others' action research; e.g., at Queen'sU AR.
Once plans for action and research have been developed, it may seem trivial to mention the need to, essentially, carry out those plans; that is, in education, to teach (in a new way, perhaps) and collect data about it. However, because action research has its strength in allowing teachers to take control of curriculum development and research (as well as their traditional role as implementers of plans), and, crucially, because teachers are closer to teaching and learning situations than others interested in education, they may recognize needs to change the plans while the action is being carried out. Indeed, this is a unique feature of action research; that is, it is a naturalistic sort of research. The actions and research plans may be changed as teachers detect results that suggest new ways to act, new reasons to take different actions, etc. This is, in essence, the Formative Research mentioned above.
example, if the teacher had an action research project
under way that
using assessment rubrics for getting students to become
(to think about and take more responsibility for their
could be that, once under way with the action (e.g.,
with students), the teacher may find that, although
students are doing
well with constructing rubrics, classroom management has
This may force the teacher to re-examine the immediate
goals for action
research, returning to Reviewing
or Planning &
Developing New Actions.
In such cases, incidentally, it is clear that action
research can be
more 'fluid' than what is depicted above. On
occasions, on the other hand, action
research can be fairly one-directional, with the
teacher being able to
follow through, without interruption, on plans for
action and research
and, afterwards, develop reports of
This section describes, in significant detail, methods actually followed in order to achieve the findings discussed below. Overall, the 'action plan' would include both plans for action and plans for (diagnostic, formative and summative) research. Relevant sections of this web page dealing with these include:
This is, in essay form, a discussion of the results of action research. In it, one is basically trying to use evidence and arguments to defend claims about the extent to which the action(s) led to improvements in the outcomes studied. A ‘Reading Buddies’ programme may have, for instance, enabled students to learn terms used in Science & Technology and, as a result, become better able to express themselves in writing. There should be a detailed discussion of results of analyses of data; e.g.., What aspects of the actions were important? and What kinds of achievement resulted from the action(s)? One should, as well, discuss how confident you are in conclusions.
TYPES AND SOURCES
ENSURING RELIABILITY AND VALIDITY
It is important to ensure reliability and validity. The former term refers to how repeatable are the results. This can be ensured by repeating the method of assessment soon after the first attempt. Validity refers to the extent to which the conclusions of the assessments can be trusted. In other words, how well does the method of assessment judge what it is intended to judge? For example, how well does a test judge a person's understanding of a concept? One common way to ensure validity is to only draw conclusions where there are at least two sources of data. For example, if you wanted to claim that students understood a concept, it would not be enough to demonstrate they could answer a test question. They should also demonstrate their understanding in assignments. The most valid ways to assess students are, as well, the most natural ones; e.g., samples of their regular class assignments are likely better than how they respond to surveys. Again, to be able to say that conclusions are valid or ‘trustworthy,’ one needs to have two or (better still) three types of data that would lead to the same conclusion. For example, if students say they like activities, that is not enough. However, if they can be seen smiling during them and they spontaneously ask for more of them, then the conclusion may be more valid. Finding three data types that support the same conclusion (claim) is called 'triangulation.' As a final check on conclusions, a teacher might have an associate examine them, along with data used to make these claims.
Some further hints for classroom research include:
As data from multiple sources are collected, it is important to soon examine them, look for trends and consider conclusions. Suggestions for analyzing qualitative data are:
Action Research Resources
Action Research Bibliography