Action Research in Education: Meaning, Steps, a Worked Example and a Template for Teachers

What is action research in education?

Every teacher has a problem that keeps coming back. Half the class cannot do word problems. Students come late to the first period. Nobody speaks in the English speaking class. Action research is the usual instinct to try a fix, done properly: write down the problem, choose one clear change, measure students before, run the change for a fixed time, measure again, and think about what the evidence says.

NCERT's module on the steps of action research puts the idea simply. Action research is done by the teacher who is part of the problem. A person who is not part of the problem cannot do action research on it. The aim is to find a solution to a classroom problem so that the most students benefit. It is not about proving a big theory. It is about your class, this term.

Educators abroad describe it the same way. A short note from the National Council of Teachers of Mathematics (NCTM) in the United States quotes Parsons and Brown (2002): action research is a form of investigation designed for use by teachers to attempt to solve problems and improve professional practices in their own classrooms. The same note quotes Emily Calhoun's plain version: a fancy way of saying let's study what's happening at our school and decide how to make it a better place.

The term itself is old. Kurt Lewin, a social psychologist, published a paper titled "Action Research and Minority Problems" in the Journal of Social Issues in 1946. In India, NCERT's CIET first started an online course on action research in 2010, and it now offers its action research course for teachers through SWAYAM.

The characteristics NCERT lists

NCERT's module gives eleven characteristics. In short, action research is:

• a process with a set of activities, not a single event;
• systematic, so you follow steps and write things down;
• aimed at an immediate problem the teacher faces;
• cooperative, because it works best with colleagues and students;
• flexible in method;
• self-evaluated, because the teacher judges the result;
• not meant to be generalised to other schools; and
• cyclic, not a straight line: one round often leads to the next.

Action research vs formal research: the difference

The word research makes many teachers think of a PhD. Action research is much smaller. The table uses the points NCERT's modules make, so you know what is enough for a first project.

Two cautions from NCERT: your group is not a representative sample, so do not claim the method works everywhere; and a person working for a degree should not do action research for it.

Action research is close to two ideas you may already use. Formative assessment gives you quick evidence during teaching. Remedial teaching is the extra help for students who are behind. Action research is the wrapper that turns either of these into a planned test of what actually works.

Three ways to do it: alone, with a colleague, or as a school

Emily Calhoun's 1993 ASCD article, "Action Research: Three Approaches", is still a useful way to choose the size of your project. She describes three kinds.

Calhoun says the key to choosing is the purpose of the inquiry. A single tutor or new coaching faculty member should start alone and small. Where two teachers share a batch, a two-person project gives you a second pair of eyes without the meetings of a whole-school project.

The steps of action research, as NCERT teaches them

NCERT's module lists five steps: identification of the problem, formulation of hypothesis, method of research, data collection, and data analysis. Its later modules add report writing as the last step. Here is each step in teacher's language, with the decision you face at each one.

Step 1: Identify the problem and write the title

Pick one weekly problem that matters for learning and that you can change yourself. NCERT's examples include students coming late, not paying attention, failing in mathematics, not being interested in some subjects, and making punctuation mistakes. Then narrow it. "Students are weak in maths" is too wide. "Class 9 students cannot turn word problems into linear equations" is something you can work on in three weeks.

NCERT gives a clear rule for the title: it must tell the reader the variable, the group, and the type of study. Its sample titles follow the pattern "A study of the effect of X on Y of class Z students of School S", for example "A study of effect of Dictation in English on Achievement in English of class V students" of a named school. If the title does not name the group, NCERT treats it as an incomplete title.

Step 2: Write one sharp objective

The objective says exactly what you will compare. NCERT's model objectives all start the same way: to compare mean scores of something before and after the change. NCERT notes that researchers often write vague objectives, which is a problem because they do not tell you what data to collect.

Step 3: State your hypothesis

A hypothesis is your tentative answer. NCERT's modules suggest the null form when you have no strong base from earlier studies, for example: there is no significant difference in mean scores before and after the change. If many earlier studies point one way, you may write a directional hypothesis instead. For a first project in a tuition class, the null form is fine and keeps you honest.

Step 4: Plan the method: group, tool and design

Group: your own class or batch. NCERT says purposive sampling is the right choice, and usually the whole group takes part. Tool: a short test or checklist you make yourself. NCERT accepts that teacher-made tools are normal in action research, while noting their validity and reliability are unknown. Design: NCERT recommends the pretest-posttest single group design, written as O1 X O2, where O1 is the pre-test, X is the change (the treatment) and O2 is the post-test with the same tool.

Step 5: Collect the data

Give the pre-test, run the change for a fixed time, then give the same test again under the same conditions. NCERT's examples run the change for a set number of days at a fixed time each day, such as 30 minutes a day for 15 days, and say the test time must be the same before and after. The treatment should be long enough for an effect to show.

Step 6: Analyse, reflect and write it up

Compare the before and after scores. NCERT's module names the paired samples t-test (also called the correlated t-test) as the formal method for this design and says SPSS software can be used. For a classroom decision, a simple comparison of averages and of how many students improved is a good start; a section below shows how. Then write a short report so that other teachers can repeat what you did.

A worked example: action research in a Class 9 maths batch

Here is one complete plan, written the way NCERT's modules suggest. The setting is an evening tuition batch, but the same plan works in a school section. Everything below is a plan to copy and change; the results table is left for you to fill in with your own students' scores.

The problem the teacher noticed

In unit tests, many students in a Class 9 batch solve linear equations correctly when the equation is given, but leave word problems blank or write a wrong equation. The teacher sees that students start calculating before deciding what the unknown is.

The change the teacher will try

A four-step routine for every word problem: read the problem twice; underline what is given and circle what is asked; write "Let x be ..." in words before any numbers; and check the answer by putting it back into the story. The teacher models two problems in each class, then students do two in pairs using the routine, for 20 minutes per class.

Why the plan is built this way

• The test marks the equation and the answer separately, because the real problem is setting up the equation, not the arithmetic.
• The same test is used before and after, as NCERT's design needs. If students may remember it, use a parallel version with the same structure and different numbers, and say so in the report.
• The exit slip adds a second kind of evidence. NCTM's note on action research suggests gathering data in different ways, from different sources and at different times.
• Three weeks is long enough for a routine to become a habit and short enough to finish before the next unit test.

The results table to fill in

How to read your results without fear of statistics

You do not need special software to make a sound classroom decision. Work through these questions in order, using the table you filled in.

1. What was the average before and after? Add all pre-test scores and divide by the number of students; do the same for the post-test.
2. How many students improved, stayed the same, or fell? An average can rise because a few students jumped while most stayed still. Count each group.
3. Did the part you targeted improve? In the example, compare the equation marks, not just the total.
4. What do the other sources say? Read the exit slips and your diary. If students say step 3 is hardest, that is your next change.
5. Was anything else going on? A holiday, a school test, or a new student can affect scores. Note it honestly.
6. Is the difference significant? If you want the formal answer, NCERT's module names the paired samples t-test for this design. The module says software such as SPSS can do it; a colleague who has studied research methods can help.

Then decide. NCTM's note ends the cycle with taking action: choose the best option, make short and long-term plans, and keep assessing. In practice there are three outcomes. Keep the change if it helped. Adjust it if it helped some students or some steps. Drop it if nothing moved, and try a different idea in the next cycle. All three are good results, because each one is based on evidence instead of guesswork.

A one-page action research plan template

Copy this table into a notebook or a document and fill in the right-hand column before you start. If you cannot fill a row, your plan is not ready. NCERT's modules treat the title, objective, hypothesis and methodology as the core of the plan, and the extra rows here cover the practical details teachers usually forget.

A six-week timeline that fits a normal term

Action research topics for Indian classrooms and coaching centres

NCERT's module gives real titles as models: the effect of teaching aids on achievement in mathematics in Class IX, the effect of yoga on stress in Class IX, the effect of classroom assignments on achievement in Hindi in Class VIII, and the effect of English dictation on achievement in English in Class V. Its Module 4 example studies an activity-based method for teaching fractions in Class VII. Below are more ideas in the same shape. Each one names a change you control and a result you can measure.

If you use a rubric to judge answers, keep it the same before and after. Our guide to rubrics for assessment shows how to build one in a few steps.

How action research counts towards your 50 hours of CPD

This part matters to teachers in India. NCERT's CIET notes that the National Education Policy 2020 expects each teacher to take part in at least 50 hours of continuous professional development (CPD) every year, driven by their own interests.

NCERT's Guidelines for 50 Hours of Continuous Professional Development (2022) list action research both as a suggested CPD theme for teachers and as one of the other CPD activities that earn hours. Under content development and other academic activities, the guidelines give "Action research/innovative projects/case studies: Three days (18 hours)". The same entry appears for teacher educators and institutional leaders. A face-to-face CPD session in these guidelines is 1 hour 30 minutes, and a training day can have four sessions, or 6 hours.

For teachers in CBSE-affiliated schools, CBSE's press release of 17 April 2025 says that, under notifications dated 1 April 2025, all teachers must complete a minimum of 50 hours of training per year: 25 hours through CBSE or government regional training institutes, and 25 hours through in-house or school complex based training.

How a particular school, board or state counts your action research towards CPD can differ. Ask your head of school or the training coordinator which format they need for the report before you start, so your write-up can be counted.

How to write the action research report

NCERT's report-writing module says the report should let any interested teacher repeat the study in their own class without trouble. It suggests writing the main part under headings, not separate chapters, because an action research report is short.

NCERT's module also lists writing rules. The main ones: write the report in the past tense; spell out an abbreviation the first time you use it; use "the teacher" or "the researcher" instead of "I" or "we"; give every table a number and a title; show results to two decimal places; avoid vague words such as less, more or beautiful; and use the same technical word throughout. It also asks for references in APA style.

Common mistakes, and how to avoid them

When action research is not the right tool

Action research is for problems you can change inside your own teaching. It is not the right tool for a problem with a cause outside the class, such as a student's health, a family crisis, or bullying; those need the school counsellor, parents or the head. It is not a degree thesis. And it is not worth doing as paperwork only. If you do not plan to change your teaching based on the result, choose a smaller lesson plan improvement instead and save the time.

Sharing what you learn as a teacher on Sikhami

Action research is more useful when other teachers can see it. NCERT's report module makes the same point: the purpose of writing reports is to help other teachers facing the same problem. Sikhami is a social network for people learning in India. Every teacher gets a free profile there, to share notes and short learning clips and to connect with institutes and learners. A short note on the routine you tested, or a clip showing how you model it, can help another teacher start their own cycle.

If you run or teach at a coaching institute or tuition centre, Sikhami's free LMS handles classroom batches and attendance, and its video courses can include tests. See Sikhami for teachers to set up your profile.