On this page
India's largest school assessment is not a board exam. It is a sample survey run by NCERT, and its 2024 round tested more than 21 lakh students. This post explains what PARAKH Rashtriya Sarvekshan 2024 measured, what the national report found, and what students and teachers can take from it.
What is PARAKH Rashtriya Sarvekshan?
PARAKH is the national assessment centre at NCERT. According to the official PRS 2024 reports page, it conducted the PARAKH Rashtriya Sarvekshan (also known as NAS, the National Achievement Survey) across the country on 4 December 2024. It assessed Classes 3, 6 and 9 in government, government-aided, private recognised and central government schools.
The tests were competency-based. Class 3 students answered 45 questions in language and mathematics; Class 6 students answered 51 questions in language, mathematics and The World Around Us; Class 9 students answered 60 questions in language, mathematics, science and social science. Students, teachers and schools also filled in questionnaires.
How big was the survey?
The PARAKH Rashtriya Sarvekshan 2024 National Report gives the scale:
21,15,022 students assessed: 5,99,026 in Class 3, 6,63,195 in Class 6 and 8,52,801 in Class 9.
74,229 schools in 781 districts across all 36 States and UTs.
2,70,424 teachers and school leaders answered questionnaires.
Because the sample covers every district, the results can be reported at national, state and district level.
What did it find?
The national report gives the average percentage of questions answered correctly.
Class 3: language 64%, mathematics 60%.
Class 6: language 57%, The World Around Us 49%, mathematics 46%.
Class 9: language 54%, science 40%, social science 40%, mathematics 37%.
Two patterns stand out. Scores fall as students move up the grades, and mathematics is the lowest subject in both Class 6 and Class 9.
Which maths and science skills were weakest?
The report breaks scores down by competency. Some Class 9 maths figures:
Applying percentages to solve problems: 28%.
Using fractions in daily-life situations: 31%.
Logic and proof (forming and checking arguments in algebra, number theory and geometry): 29%.
Forming and manipulating algebraic expressions: 38%.
In Class 6, representing and comparing common fractions such as half and a quarter scored 29%.
In Class 9 science, describing how simple electric circuits work, including the heating and magnetic effects of current, scored 33%. Describing one-dimensional motion with speed and position scored 41%.
What did it find about different groups of students?
In Class 9, girls did better than boys in language and social science, while boys scored marginally higher in science and maths. Students in central government schools scored highest across all Class 9 subjects. Students from Scheduled Tribe communities had the lowest scores, with the gap most visible in maths. The report also lists states where each group did well; Punjab appears often among the top performers.
What does PARAKH not tell you?
It is a sample, not every student. It describes the system, not any individual child.
It is a one-day test. Scores reflect performance on 4 December 2024 with a fixed set of booklets.
Average percent correct is not a pass mark. A 37% average in maths does not mean 63% of students failed; it means the average student answered about 37% of questions correctly.
It shows where, not why. The data point to weak competencies but do not prove what caused them.
What does this mean for students preparing for boards, JEE or NEET-UG?
If the national average for using fractions and percentages in Class 9 is around 30%, many students carry that gap into Class 11 physics and chemistry, where almost every numerical problem uses ratios, proportions and unit conversions. A short, honest check of these basics can save months of confusion later.
Teachers can use the competency tables to plan revision. Picking two or three of the lowest-scoring competencies and revisiting them with mixed problems is more useful than repeating the whole chapter.
ASER, a household survey of rural children, tells a similar story at a more basic level; see our post on what ASER 2024 found. For the school system behind these numbers, see what UDISE+ 2025-26 shows.
How to read a national survey like this
Check three things: which grades and subjects were tested, what the number means (average percent correct, not percent passing), and whether a difference between groups is large enough to matter. Then look for the competency tables, which are usually more useful than the headline average.
If fractions, ratios or unit conversions are your weak spot, the Fermi Concept Library covers every concept in physics, chemistry, biology, mathematics and computing, so you can rebuild the idea before the next chapter depends on it.
This week, solve five fraction and percentage problems without a calculator and see which step slows you down.
