Thursday, October 7, 2010

Monitoring attendance and educational outcomes

A substantial proportion of academic research and policy focus in the education sector revolves around ensuring teacher attendance. This is understandable given the shocking levels of teacher absenteeism that bedevils education in many states across India. However, in so far as the objective of education is the achievement of bench-marked learning outcomes, it may be a simplification to assume that ensuring teachers attend school is enough.

In their now famous randomized experiment of 120 single teacher schools run by Seva Mandir in Rajasthan, Esther Duflo and Rema Hanna had monitored the impact of teacher attendance, captured using a tamper-proof date and time function camera, on learning outcomes. This camera was provided to teachers, along with instructions to have one of the children photograph the teacher and other students at the beginning and end of the school day, and their salary was linked to attendance.

They found that absence rate (measured using unannounced visits both in treatment and comparison schools) changed from an average of 43 percent in the comparison schools to 24 percent in the treatment schools. Learning levels too improved - a year after the start of the program, test scores in program schools were 0.17 standard deviations higher than in the comparison schools and children were 43 percent more likely to be admitted into regular schools.

The reality though is more nuanced. Getting a teacher to school is just the first step, undoubtedly important in so far as it increases the likelihood of achieving learning outcome objectives. Though, as the aforementioned study and others point to the fact that teacher attendance improves learning outcomes, as it should, the extent of improvement is debatable. And many of these studies reflect a very low baseline (especially true of many North Indian states where teacher attendance is the exception) and may therefore point to an exaggerated impact on learning levels.

In fact, attendance related effectiveness is conditional on two subsequent actions - going to the classroom and communicating effectively with the students. And as we know, both, especially the later, are formidable challenges in themselves. If we think that ensuring morning attendance solves the problem, be prepared for a shock to know that leaving the school early is an equally big problem. In view of all this, ultimately any reliable assessment of school performance will have to come from objective measurement of learning outcomes.

In other words, the returns (in terms of achievement of learning outcomes) from merely getting teachers to school, while significant, do not constitute a large enough step in ensuring the achievement of learning outcome objectives. This is because this objective is so intimately and exclusively dependent on the cutting-edge action - the process of teaching and learning - and this itself is not exclusively or (for some categories, as baseline levels improve) maybe even substantially (as is being imagined) dependent on the mere physical presence of the teacher. In fact, class-room teaching (and the quality of its outcomes) is more a function of the individual skills, capability, and motivation of the teacher than any other exogenous intervention.

In contrast, atleast to the extent of primary health care and first referral facilities, just getting the doctor to the hospital is more closely related to the achievement of final (hospital-related treatment) health care objectives. There is only a remote likelihood of a doctor turning away waiting patients or be lackadaisical and less rigorous in his treatment advice.

Does this mean that it will deliver greater bang for the buck to have expensive devices like biometric attendance readers in place for monitoring the attendance of doctors and nurses as against teachers, atleast to start with? Or does the reality of very high absenteeism and low learning levels in many Indian states mean that there are considerable low hanging fruits to be plucked with teacher attendance itself?

Finally, in view of the fact that improved learning outcomes necessarily require teacher attendance, there a very strong case to be made out that irrespective of what is done to improve teacher attendance, objective (by an independent third party) assessment of learning outcomes is a fundamental requirement.


Jayan said...

Teachers are probably more motivated than others - more so with teachers in primary-high schools. There must be some reason why they go absent. Was there any study done on why teachers go absent? Reasons could be many - start with School Infrastructure
- Do they have a building in first place?
- Connectivity by public transport?
-Water, toilets, electrivity?
- bench, desk and a block board?
- Are we hiring right?
- How much do we pay?
- How regular the payments are

The schools in Kerala are far better. Main reasons could be very good pay and consistent pay, work life balance. It will not be difficult to see why teachers do not come to school.

An independent assessment of teaching-learning activities at all schools will be good starting point.

Anonymous said...

what about teachers doing their job for their pay alone ?
Do they need all of these for them to do their basic job ?
What about doing making do with whatever is available ?

sai prasad said...

That assessment is a requirement, is beyond dispute. How to measure ? And what to measure ?
These questions are so complex that answering them might turn out to be like the proverbial dog running after its tail.

Jayan said...

To anonymous.
The question has bigger scope than the discussion. Unmotivated work force is problem to be solved with long term measures -- no silver bullets. In that sense teachers absenteeism is not much different from any other govt officer's absenteeism. I believe it is better to focus on find what is good and working and then replicating/adopting that to other places is better than focusing on the weak links..