It is sometimes said that having a clear list of grade-wise learning outcomes is a critical part of improving learning levels in India. These, it is said, serve as clear goals and are hence useful for teachers and the education system as a whole.

Our own experience of conducting assessments and learning programmes for about 20 years with large government systems as well as private schools does not fully align with this. We say this partly based on the fact that such ‘learning standards’ have officially been endorsed in India ever since the Minimum Levels of Learning were introduced by the NCERT in the early 1990’s. Many states have had their own state standards and some even have tried to map these standards to questions in the exams. It is not very clear how these steps have helped learning levels or how the current push for learning outcome statements or competency-based education is different from what has been promoted and practised over the last 30 years.

An important reason why formal learning standards may be LESS important in India is that all states have had a long-standing tradition of single state-prescribed textbooks[1]. Thus, any teacher has to use this textbook almost exclusively for classroom teaching. While systems which do not prescribe a single textbook need learning standards to indicate what should be taught and what will be assessed, these prescribed textbooks themselves codify the ‘learning standards’ in Indian states! Further, these textbooks in most states (and definitely at the NCERT level) are of fairly good quality and have no major shortcomings in terms of grade-wise topic coverage. So, it is not clear why teachers would need learning standard documents when a textbook provides the same information in even more detail.

Of course, one could correctly argue that textbooks may wrongly convey that it is factual content that should be taught and assessed, so we need a document to convey that the focus should be on understanding rather than recall, and so on. While this may be true, such a document would not need to again list the topics and subtopics that students need to learn. It could instead explain the importance of developing understanding and application skills and give examples of questions that assess for such competencies.

Listings of competencies invariably say that students should be able to “differentiate between small and capital letters in print[2]” in English or “counts objects using numbers 1 to 9 and compares numbers up to 20[3],” in Maths. Clearly, a) the textbook already makes these points clear and b) this listing does not in any way emphasise understanding or deeper learning. Further all such competencies can equally well (and validly) be mapped to poor-quality questions that simply emphasise recall of information.

Let us see this with an example. Consider this competency: “children should be able to calculate the sum, difference, product and quotient of numbers”. It could be represented by both these types of questions:

While both questions are important, traditional tests use only questions like those on the left. The question on the right is one that requires deeper thinking and understanding (though it too would map to the same competency.) We have found, therefore, that providing samples of such questions can be more effective than merely statements like the above.

Most educationists would agree that children should become competent not only in the above two types of questions but many more including real life situations representing this competency.

So a good learning standards document could say (to take the example of a different competency): “It is important for students to have an estimate of how much units of measurement like a metre or kilogram actually represent. The teacher could do activities to convey this and also ask, for example, if a skipping rope would be 2 cm, 200 cm or 20 m long.”

Traditional listings of competencies or learning outcomes rarely contain such statements.

The other important information that learning standards almost never contain is lists of student misconceptions, alternative conceptions or common errors. These too would be extremely useful for teachers (and act as additional information that the textbook does not contain). Traditional learning standards documents do not contain these at all.

Further, traditional learning standards do not incorporate any performance data on how students fare on different types of questions. These would be very helpful for teachers to know how many children in the class have not understood a concept and possible reasons why.

Let us look at another example to illustrate that very different types of questions can be mapped to the same competency.

Competency – The student understands how shadows are formed

There is one, even more important problem with the common expectation that every question must be mapped to a single competency. This may actually eliminate more complex questions testing multiple competencies, and make learning shallower. Rather than focusing on ‘competency-based learning’, we make the case that we need to improve the quality of our questions so that they test for and require deeper learning from students. What is important is that the teacher must be able to ask the right questions in order to be able to assess and remediate for the competency.

As mentioned at the beginning of the note, there seems too little evidence to believe that states or schools that made better Learning Standards had improved learning outcomes. In the US, for example, researchers have ranked the quality of states’ educational standards. They do not necessarily correlate with actual performance. To quote one example, Connecticut’s English and Language Arts standards were rated a D[4], but “students in… Connecticut perform roughly the same on the PISA reading test as students in the top-scoring countries (i.e., Canada, Finland, and Korea)”[5].

Our company once worked with an Indian state, helping them create assessments annually over a 5 year period. After the 3rd year, there was a directive to create a list of ‘grade-level competencies’. The document was prepared. Note that assessments were done before the document was prepared and similar assessments were done after it also. After the document was available, the questions in the assessments were mapped to the statements in the document! Both to us, and other external observers, there was no other difference or impact from the availability of such a document.

As mentioned earlier, rather than a listing of competencies, information that can guide teachers on what their focus should be and examples of what would constitute good learning is the need of the hourWe have been impressed for decades by the Japanese Learning Standards documents[6], for example. They DO NOT list out most or all of what children should be able to do. Rather, they gave broad guidelines. They explained that the purpose of grammar is not to test rules in sentences, but that children should be able to construct correct sentences. The test of vocabulary, the documents explain, comes from checking if children can understand authentic texts they find in the world around them.

Details of actual student performance and learning gaps should also be shared with such a document, possibly as an easy-to-access website. A sample of such a site created by EI is shown below:

If we are able to make Learning Standards ‘documents’ like these, the imaginary use case described below could soon become reality:

A teacher is about to start teaching Polynomials in class 9. She refers to the Learning Outcome Database first reading the section on ‘Why Polynomials’, and then ‘Difficulties Students Face’. She notes the 15 commonly made errors and glances through some of the question examples. She downloads a 25 question test which she decides to use before her first class, noting that it actually tests the pre-requisites children need to understand polynomials. She also realises that the test will clearly show her where her class and individual children are. For every question answered incorrectly in the test, the database provides a series of graded questions and examples to help children getting those questions wrong. As usual, she feels a bit overwhelmed with the number of questions available, but knows from past experience that she can use as little or as much as she wants! Even in the one hour she spent, she got 3-4 question ideas which she decides to use in her introduction in class the next day.







[4] This note will not get into the merits and demerits of having prescribed compulsory textbooks

[5] NCERT Grade 1 English Learning Outcome #4

[6] NCERT Grade 1 Mathematics Learning Outcome #2

Latest posts by Sridhar Rajagopalan and Arvind Singh (see all)