Indian Economy-News & Updates

[TopCat]

Well thanks for the UNESCO stats. UNESCO does get the data from mostly govt source. Yet the definition of literacy hugely varies country to country. Same as for poverty.
For instance in Bangladesh the definition of literacy means standard 5 level of education who can still remain literate at the age of 15, considering discontinuation of education after primary level at the age of 11.
So the 49% of literacy you see is the test conducted by the bureo of statistics (this is done yearly basis taking random samples from all across the country and conduct actual test with question and answers) who will score 50% or above marks on the test of 5th standard education. They are called basic literate. those who scores 25-50% (merely pass the test) are called semi literates on the same test questionaries and the percentage of people above this category 66%. those who scores 25% or less are called non literate.
We dont have any statistics for standard 1 level of education who can only read and write.

 

[wtf]

Huh I knew it...
Do you have the statistics of 15 years above age group with basic literacy which includes reading/writing/comprehending/arithmatic/visual understanding like maps times and some IQ related visualiztion etc.

Yes, we call it primary school education

A comprehensive picture of primary education is difficult to obtain.
Government funding for primary education means that mostly anybody who wants to study gets to study, moreover most states chip in for free food at schools. Nevertheless, though almost all (96%) Indian students go to school, their situation is pretty bad.

The economist magazine says this
At least almost all Indian children now go to school: a survey of 16,000 villages carried out last year by ASER, an NGO, put the enrolment rate at 96%. But it also pointed to the appalling quality of education on offer. Half of ten-year-olds could not read to the basic standard expected of six-year-olds. Over 60% could not do simple division. One reason is that, according to a World Bank study, only half of Indian teachers show up to work. Half of Indian children leave school by the age of 14.

That is the universal definition of literacy.

It is not. No literacy test includes IQ. More on this at Wikipedia
Literacy - Wikipedia, the free encyclopedia and a more detailed one on the metrics used in India here Literacy in India - Wikipedia, the free encyclopedia

 

[wtf]

Well thanks for the UNESCO stats. UNESCO does get the data from mostly govt source. Yet the definition of literacy hugely varies country to country. Same as for poverty.
For instance in Bangladesh the definition of literacy means standard 5 level of education who can still remain literate at the age of 15, considering discontinuation of education after primary level at the age of 11.
So the 49% of literacy you see is the test conducted by the bureo of statistics (this is done yearly basis taking random samples from all across the country and conduct actual test with question and answers) who will score 50% or above marks on the test of 5th standard education. They are called basic literate. those who scores 25-50% (merely pass the test) are called semi literates on the same test questionaries and the percentage of people above this category 66%. those who scores 25% or less are called non literate.
We dont have any statistics for standard 1 level of education who can only read and write.

That is a great method of measuring literacy. The full paper about the survey in 2008 is here.
The difference between Indian census and the Bangladeshi one is the addition of arithmetic. (It is also the difference between Indian census and NLM) . The survey says about 35% are (absolutely) illiterate, pretty much same as Indian stats. about 49% are literate while another 16% fall somewhere in between.
Bangladesh Literacy Assessment Survey 2008


National literacy mission data here. They use the same definition of literacy as Bangladesh but then quote 2001 census data which only includes those aged 15 and above. Bangladesh on the other hand focuses on 11-14 year olds.
National literacy mission . Indian data that I could find did not break up the data into literate/semi-literate.

So in summary, yes, the 49% literacy of Bangladesh might be under representing the actual literacy of Bangladesh which by India's measures could be anywhere from 50%-66% range and comparable to India's 65%.

 

[Screaming Skull]

Well thanks for the UNESCO stats. UNESCO does get the data from mostly govt source. Yet the definition of literacy hugely varies country to country. Same as for poverty.
For instance in Bangladesh the definition of literacy means standard 5 level of education who can still remain literate at the age of 15, considering discontinuation of education after primary level at the age of 11.

I did some research of my own and found this report http://www.campebd.org/download/PopularEW2002English.pdf titled "Literacy In Bangladesh-NEED FOR A NEW VISION"

According to this report Bangladesh defines literacy as- "Possession of skills in reading, writing and numeracy related to familiar contents and contexts and the ability to use these skills in everyday life in order to function effectively in society." This is by and large the definition of literacy world over, even in India.

It is only the method of ascertaining this, which is different in Bangladesh as compared to India.

So the 49% of literacy you see is the test conducted by the bureo of statistics (this is done yearly basis taking random samples from all across the country and conduct actual test with question and answers) who will score 50% or above marks on the test of 5th standard education. They are called basic literate. those who scores 25-50% (merely pass the test) are called semi literates on the same test questionaries and the percentage of people above this category 66%. those who scores 25% or less are called non literate.
We dont have any statistics for standard 1 level of education who can only read and write.

When I first read your above statements I was amazed at the ability of the GoB to conduct a nationwide test for a population of 150 million Bangladeshis (a worthy acceptable sample size IMO should be at least 10% of the population if not more). This would be a bigger exercise than election itself. But again I was disappointed! Guess what? The sample size to determine the literacy rate in Bangladesh for the year 2002 from the above cited report is only 13,145!

How can a sample of 13,145 individuals ever portray a true picture of the state of literacy in a country of 150 mil people? I guess the GoB adopted such a method only to save the cost of a national level census. I mean this so absurd. I still can't digest that GoB is adopting such an absurd method. It is like compensating for elections with just exit polls or SMS polls.

Correct me if I am missing something here or if your views differ from mine but personally I would never accept such an unscientific methodology to determine something as important as the literacy rate of a country.

 

[wtf]

I did some research of my own and found this report http://www.campebd.org/download/PopularEW2002English.pdf titled "Literacy In Bangladesh-NEED FOR A NEW VISION"

According to this report Bangladesh defines literacy as- "Possession of skills in reading, writing and numeracy related to familiar contents and contexts and the ability to use these skills in everyday life in order to function effectively in society." This is by and large the definition of literacy world over, even in India.

When I first read your above statements I was amazed at the ability of the GoB to conduct a nationwide test for a population of 150 million Bangladeshis (a worthy acceptable sample size IMO should be at least 10% of the population if not more). This would be a bigger exercise than election itself. But again I was disappointed! Guess what? The sample size to determine the literacy rate in Bangladesh for the year 2002 from the above cited report is only 13,145!

How can a sample of 13,145 individuals ever portray a true picture of the state of literacy in a country of 150 mil people?

Statistically speaking a survey of 13,000 should provide an accurate picture due to central limit theorem (Central limit theorem - Wikipedia, the free encyclopedia). India conducts its National Sample Survey and publishes those results and the actual results are within 10% accuracy.
Bangladesh is certainly doing better than the 49% claim.

The problem here (comparing Indian and Bangladeshi data) is not with the survey method but with sampling itself.

• Bangladesh surveyed 11-14 year olds. I can't find the equivalent Indian NSS data for 11-14 year olds. Indian census focuses more on 15+ year olds (it treats all 0-6 year olds as illiterate) while enrollment rate for Indian primary schools is around 97%.
  
• Bangladeshi data is from 2008. Indian ones are from 2001.

 

[wtf]

I did some research of my own and found this report http://www.campebd.org/download/PopularEW2002English.pdf titled "Literacy In Bangladesh-NEED FOR A NEW VISION"

That data you have is different from the Bangladesh survey of 2008. It shows 49% as illiterate - I think this is the data from a survey conducted in 2001 in Bangladesh. The link I had posted earlier had that number at a lower 35%.

 

[Screaming Skull]

Statistically speaking a survey of 13,000 should provide an accurate picture due to central limit theorem (Central limit theorem - Wikipedia, the free encyclopedia). India conducts its National Sample Survey and publishes those results and the actual results are within 10% accuracy.
Bangladesh is certainly doing better than the 49% claim.

I am fully aware of the central limit theorem and also acknowledge that the sampling is scientific and statistically correct, as mentioned in the report that I cited. But my point is that the applicability of such a statistical method for something like literacy or population estimation is unscientific. As you know about the central limit theorem and statistical processes you would also know the traditional areas in which they are applied and the reasons for their applications. TRP ratings, exit polls etc. are good examples.

IMHO in most cases such statistical surveys usually throw up results that are grossly contrary to the reality. So your comment that "Bangladesh is certainly doing better than the 49% claim." holds no ground as it could have been "Bangladesh is certainly doing worse than the 49% claim." too, as the error could be positive or negative.

The problem here (comparing Indian and Bangladeshi data) is not with the survey method but with sampling itself.

• Bangladesh surveyed 11-14 year olds. I can't find the equivalent Indian NSS data for 11-14 year olds. Indian census focuses more on 15+ year olds (it treats all 0-6 year olds as illiterate) while enrollment rate for Indian primary schools is around 97%.
  
• Bangladeshi data is from 2008. Indian ones are from 2001.

Don't work up yourself with so many different reports and surveys. Just read my post 76. It has UNESCO data for the period 1990-2006. Provides a good comparison. Again IMHO this is the most credible data that you can possibly access as the sole objective of UNESCO is to improve education standards world over and I am sure that they would never publish unverified data. So whatever the methods used by India or Bangladesh, if a figure has surfaced on the UNESCO report, then rest assured it has been normalized scientifically, considering all the variations the methods of different countries.

Also, remember no country in the world wants to project itself in negative light! So, if a certain method of evaluation proves to be disadvantageous to a country then it will certainly scrap it and adopt the general assessment standards.

 

[Screaming Skull]

That data you have is different from the Bangladesh survey of 2008. It shows 49% as illiterate - I think this is the data from a survey conducted in 2001 in Bangladesh. The link I had posted earlier had that number at a lower 35%.

The objective of posting that report was to highlight two things, one the definition of literacy in Bangladesh and two the method of survey to determine the same. The Bangladeshi member was confused between the two and assumed the definition of literacy in Bangladesh to be completely different from what it is in other parts of the world.

 

[TopCat]

I am fully aware of the central limit theorem and also acknowledge that the sampling is scientific and statistically correct, as mentioned in the report that I cited. But my point is that the applicability of such a statistical method for something like literacy or population estimation is unscientific. As you know about the central limit theorem and statistical processes you would also know the traditional areas in which they are applied and the reasons for their applications. TRP ratings, exit polls etc. are good examples.

IMHO in most cases such statistical surveys usually throw up results that are grossly contrary to the reality. So your comment that "Bangladesh is certainly doing better than the 49% claim." holds no ground as it could have been "Bangladesh is certainly doing worse than the 49% claim." too, as the error could be positive or negative.



Don't work up yourself with so many different reports and surveys. Just read my post 76. It has UNESCO data for the period 1990-2006. Provides a good comparison. Again IMHO this is the most credible data that you can possibly access as the sole objective of UNESCO is to improve education standards world over and I am sure that they would never publish unverified data. So whatever the methods used by India or Bangladesh, if a figure has surfaced on the UNESCO report, then rest assured it has been normalized scientifically, considering all the variations the methods of different countries.

Also, remember no country in the world wants to project itself in negative light! So, if a certain method of evaluation proves to be disadvantageous to a country then it will certainly scrap it and adopt the general assessment standards.

Well you have the report here and that already explained how we conduct the survey in ground level. This is also UNESCO funded report so there is no contradicton on UNESCO website and this report only that this report is most current.
The definition of literacy is already explained there and it based on the primary education (formal eductaion). Bangladesh never conducted any informal adult literacy program which is grossly of no use that India did through mobilization of Army and volunteers.
I asked in my earlier post the definition of Indian literacy, and it is claimed that only reading and writing is covered in your 66% literacy.

We exactly had the same mechanism (I am not sure) of calculating literacy through census and in year 2000 AL govt claimed 66% literacy based on those mechanism but was outright rejected by civil society and NGO's. Then they had to move to some more scientific and accurate mechanism with the help of donor agencies. What you see in that report is the real fact of Bangladesh now. Is it better than India or worse.. I dont know. I dont know Indias method of calculation neither I have or you have any report which clearly shows the break down of the literacy and the standard that follows.

So your comment that "Bangladesh is certainly doing better than the 49% claim." holds no ground as it could have been "Bangladesh is certainly doing worse than the 49% claim." too, as the error could be positive or negative.

That 49% is a solid literacy no question about it. 66% is semi literate. There can not be positive and negative as we conduct the survey every year. Its not just one single year and we are sitting idle with all gratitude. If in any year it was wrong could have been corrected the next year.

If we just take only recognizing alphabet I am pretty sure the literacy rate will hit more than 75%.


Thanks

 

[TopCat]

Statistically speaking a survey of 13,000 should provide an accurate picture due to central limit theorem (Central limit theorem - Wikipedia, the free encyclopedia). India conducts its National Sample Survey and publishes those results and the actual results are within 10% accuracy.
Bangladesh is certainly doing better than the 49% claim.

The problem here (comparing Indian and Bangladeshi data) is not with the survey method but with sampling itself.

• Bangladesh surveyed 11-14 year olds. I can't find the equivalent Indian NSS data for 11-14 year olds. Indian census focuses more on 15+ year olds (it treats all 0-6 year olds as illiterate) while enrollment rate for Indian primary schools is around 97%.
  
• Bangladeshi data is from 2008. Indian ones are from 2001.

The statistics is fine with the error of margin should not more than 2/3%.
49% is literate with primary education. 66% is semi literate with primary education. there is nothing 11-14 year old age group in that report. But one sampling was done with the age group 11 and more and another age group 15 year and older. 15 Year and older is the subset of 11 year and older group.

Again the standard is based on 5 years primary eduction. Not based on only recognizing alphabet.

I would love to see Indian method of calculation and breakdowns.

 

[TopCat]

We conduct the survey almost every year and see the progress that follows. So the method is fined unless it would have been significantly different every year and would not have followed and straight line. It is improving 3-5% every year now as the formal education is coming to maturity now. I again say its not done through adult literacy program.

 

[TopCat]

The objective of posting that report was to highlight two things, one the definition of literacy in Bangladesh and two the method of survey to determine the same. The Bangladeshi member was confused between the two and assumed the definition of literacy in Bangladesh to be completely different from what it is in other parts of the world.

Yes the definition vary significantly.Even UNESCO report says that. The definition we followed in 80's and 90's is no more applicable in 2009.

 

[TopCat]

When I first read your above statements I was amazed at the ability of the GoB to conduct a nationwide test for a population of 150 million Bangladeshis (a worthy acceptable sample size IMO should be at least 10% of the population if not more). This would be a bigger exercise than election itself. But again I was disappointed! Guess what? The sample size to determine the literacy rate in Bangladesh for the year 2002 from the above cited report is only 13,145!

How can a sample of 13,145 individuals ever portray a true picture of the state of literacy in a country of 150 mil people? I guess the GoB adopted such a method only to save the cost of a national level census. I mean this so absurd. I still can't digest that GoB is adopting such an absurd method. It is like compensating for elections with just exit polls or SMS polls.

Correct me if I am missing something here or if your views differ from mine but personally I would never accept such an unscientific methodology to determine something as important as the literacy rate of a country.

Well this report is very scientific even you could argue on the sample size. But the distribution of sample is very deep as they conducted the survey at the Upazialla (police station) level and on all economic group. Also the sampling is a continuous process as they are surveying all year long every year. So the sampling size in a particular year may be small but it cumalates as year adds on.

 

[wtf]

Yes the definition vary significantly.Even UNESCO report says that. The definition we followed in 80's and 90's is no more applicable in 2009.

UNESCO is not using the same data as Bangladesh. Their data is for 2006 and are marked with ** meaning it is a UIS (Unesco Institute for statistics) estimate. National data is marked as *.

The data given there disagrees with Bangladesh government estimates. I would say that the study was conducted on different age groups and for different years.

Anyway, why are we bringing Bangladesh into a discussion about Indian states ? Bangladesh can do well on its own and we wholeheartedly congratulate you for the achievements. We'll start a fight about Bangladesh vs. India in education at some other point of time (The only fight worth having between neighbours).

Now can we get back to Bihar vs Mizoram fight please

 

[wtf]

IMHO in most cases such statistical surveys usually throw up results that are grossly contrary to the reality. So your comment that "Bangladesh is certainly doing better than the 49% claim." holds no ground as it could have been "Bangladesh is certainly doing worse than the 49% claim." too, as the error could be positive or negative.

Don't work up yourself with so many different reports and surveys. Just read my post 76. It has UNESCO data for the period 1990-2006. Provides a good comparison. Again IMHO this is the most credible data that you can possibly access as the sole objective of UNESCO is to improve education standards world over and I am sure that they would never publish unverified data. So whatever the methods used by India or Bangladesh, if a figure has surfaced on the UNESCO report, then rest assured it has been normalized scientifically, considering all the variations the methods of different countries.

I agree that Unesco data is comparable across countries -they seem to have done an independent survey. I won't say it is the most accurate, but it is pretty good indicator.

I said that Bangladesh is doing better than 49% by taking the maximum upper limit of 66% and subtracting 10 from it. I don't think adding a 10% to 66 makes any sense since complete surveys have put the number in 2001-02 at 50% for Bangladesh. I presumed that they could have increased it by 5-6% by 2008. I can't see a reason why the number would go below 49% since it is pretty well established that it was 49% in 2001 (except for one study in 2002 which said 41%).

Anyway, let's not put down Bangladesh even if we can't find the best data. They certainly are doing good compared to Pakistan or even their neighbouring Indian state of Bihar, and it is a great achievement to have.