Irrespective of which place in the world we are living, our morning everywhere starts with news and current affairs taking place across the world assisting us in being informed about the highlights ‘changing world’ is witnessing and the dire issues which the world is dealing with. The problems our world is facing today are highly varied in terms of size and depth. Ranging from climate change to poverty to education to health, they contribute to major chunk of the societies’ and countries’ development and economy. Solving such big problems involve huge finances, strong leadership and a scientific approach. One such approach I came across lately is shared below.

I lately attended a workshop on Randomized control trials also known as RCT, a highly scientific and data driven method to evaluate the impact of interventions. The model talks about picking up any problem( literally any problem) ,designing a solution or intervention to solve it, executing it on field and evaluating the impact created by different types of interventions and then deciding which solution is best for scaling up. The purpose for doing so will be to identify the best possible solution for a given problem and then scaling up further.

For eg: In last decade we have seen the rise of smart classes in schools. In fact it has become such a common thing that it plays a big role in informing parent’s decision in selecting school. However, how much impact in learning has happened due to this massive change? Can we say that the financial and human investment being put will result in proportional results in learning? To answer such questions, RCTs are done. So that, after putting in years of humongous efforts we shouldn’t realize that there was a simpler, cost effective and better result producing method available.


The steps to do an RCT:

Identifying the problem:

Something said repeatedly becomes concrete and unquestionable in our minds. Like how we know that poverty is a big issue, as we see and read about it every now and then, so is a well stated fact. But before beginning to solve a problem, identifying the real problem and its root cause is highly crucial and this is what is done at step

1. Proving a problem is a problem backed by data from research done and statistics collected is highly important to have clarity on the depth, breadth and the causes of the issue. It also provides clarity on which part of the problem are we targeting if it is a huge one.

Identifying an intervention:

To solve the chosen problem, the selection of the solution (intervention) which we feel will solve the issue is done at step 2. There can be one or multiple interventions that we want to check but this can be decided based on the problem and the budget available. Also, one important point while doing multiple interventions is to see the impact of each intervention stays exclusive when multiple interventions are done with same group of people.

Identifying a target and a control group:

To compare our solution is working or not, we will keep two groups under observation. One group- target group– will be provided with the intervention and the second group – control group – will not be given any intervention. The selection of the group of people, number of people, location will be dependent upon the problem, applicability of solution and cost effectiveness.


Once the basic work of identifying the problem, solution and groups is done it calls for the time to execute the same on the field. The effectiveness of the execution plays a very important role as the effective dissemination of the intervention to all the target group will contribute in evaluating the effect of intervention rightly.

Evaluating the data:

The data is collected before and after the intervention from both control and target groups. The collection of the data and then deriving the insights out of it will lead to the conclusion of whether intervention is working or not. Assist in comparing which intervention is better and why. Is the program cost- effective when applied on a large scale? Is the solution providing desired results so as to apply at a larger group? Thus, using answers to these questions the decisions at policy level can be made.

The method explained above is an introductory version of RCT method, which otherwise is highly detailed, scientific, has multiple layers to it. Though RCT in first place requires time, finances and efforts but can in-turn play an important role in saving them by protecting non-impactful interventions to go at scale. It can help problem-solvers distinguish between feelings and facts & ideas and reality.



By Ritika Arora – Educational Specialist


Ritika Arora

Ritika Arora

​Ritika Arora, who is working as an Educational specialist in Large scale assignments at Educational Initiatives,is a highly enthusiastic girl with a zeal to do something substantial in education sector. She is born and brought up in Delhi and has completed her Engineering in ​E​lectronics and ​E​lectrical in 2011. She has worked in Ericsson on networking projects and then joined Teach For India fellowship in ​P​une. She took up various projects during her fellowship, and taught English , Maths and Science to 7th and 8th grade students in a Government school in ​P​une. Besides working she likes to hangout and meet new people​,​ ​i​s a movie buff and has a fair interest in adventure sports.
Ritika Arora

Latest posts by Ritika Arora (see all)