[UnpackED - 2] Why do governments do what they do regarding education?

The first post discussed the reasons for unsuccessful translation of governments' efforts to outcomes, in education domain. We discussed that it's primarily due to the way governments pursue reforms. This post discusses the factors that drive governments to pursue those modes of reform.

On Quantity - Reasons for fragmented and piecemeal reform

The first post identified fragmented and piecemeal approach to reform trying to address only one constraint at a time, as one of the reasons. There are several factors that drive governments to pursue this form of reform.

First, a combination of scale of the problem and lack of conducive atmosphere for widespread reforms. Each of the individual constraints in education are significant enough, especially when done at scale. Pursuing multitude of such reforms huge bandwidth of attention. In contexts, where a conducive environment is not present for this, often a result of weak political will, reform is often bureaucrat driven. Given these constraints, bureaucrats choose only one or two things to pursue and focus all their energies on them.

While the first reason has to do with the constraints in the system that lead to piecemeal policies, the other following reasons are to do with the mental models used in analysing reform and language used in communicating reform proposals.

Second, information feeds to policy that use diagnostic framework to analyse education. Have you ever heard of statements like "the main problem is lack of motivation in teachers", "good teacher training and headmaster training are lessons that we can learn from xyz country." This is what I call, a diagnostic framework of analysis. In diagnostic framework, root cause is attributed to one or two specific components (notice the emphasis on "main problem" in "main problem is teacher motivation"). Sometimes, it becomes a laundry list of suggestions that are to be done.

On the face of it, there seems to be nothing wrong with these statements. No one can deny that lack of teacher motivation affects their performance. No one can deny that a good teacher training and headmaster training are essential for a high performing system.

However, the problem with phrasing constraints in this manner is that it leads to piecemeal reforms. If one says that "xyz is the main problem", all the attention in discourse is driven only to this particular issue, leaving other issues unattended. In reality, we need many other along with xyz.

Further, there are two more issues with diagnostic mode of analysis:

i) it emphasises symptoms rather than causes. For instance, diagnostic only points out xyz as a constraint but it doesn't ask, what's the reason for this, why isn't addressed till now? Often the underlying causes for emergence of xyz constraint are also responsible for emergence of many other constraints. By emphasising only this particular xyz, we are crowding out other constraints. 

ii) it communicates necessity to initiate efforts on an issue but conceals the how aspect of it. "We should do teacher training", "we should do headmaster training" etc are the common suggestions, one notices. The point is that the issue is not with the lack of recognition of necessity of teacher training but with the way we do it. If the suggestion is to do teacher training, a teacher training is done to tick a checklist without regards to the way its done.

Thus, by feeding information into policy phrased in a particular manner using diagnostic framework, we are concealing the real issues.

Few other factors make people to phrase policy recommendations in this particular manner. Experts have what I call, expertise bias. It means that experts point only those things as constraints which they are experts in, and suggest recommendations on only those things which they are experts in. Someone tells to do teacher training, some tells to change pedagogy and so on. But no one tells to do everything!

Sometimes bureaucrats also constrain people to recommend policies in a diagnostic framework. Suggestions related to strategy and mental models leading to inappropriate policies is discarded as theory that's not useful for policy purposes. They demand "concrete actionable policy suggestions". It results in a policy recommendations in form of laundry list of individual components of education that are to be focussed upon.

On a related noted, some times the misconceptions regarding policy constraints like teacher salary, guest teachers also feed into policy diverting the attention from others to these specific individual items. Karthik Muralidharan notes that sometimes people just visit few schools and if they notice something like lack of basic things like textbooks, it registers strongly in their memory. A "need for closure" tendency that forces brain to quickly come to a solution aids this process. They then come back and popularise the notion that lack of textbooks is the single biggest constraint in education. Such discourse also drives one piece at a time approach (FYI: Many studies proved that textbooks aren't necessarily the major constraint).

Third, selling problems vs. selling solutions. Lant Pritchett illustrates it using a simple exercise . Simply ask yourselves - "what is the problem you are trying to solve?" Most people come up with statements like "lack of teacher training". Then write down a solution to your problem. If you defined problem as "lack of teacher training", your solution will be "teacher training". Now, write other solutions to your problem. If you have defined teacher training as the problem, you can't write anything else.

Lant Pritchett calls this as selling solutions. You wanted to sell the solution of teacher training, hence you defined the problem as lack of teacher training. The actual problem with education is lack of outcomes. If you define problem in this manner, you can list multiple solutions, apart from teacher training.

One can observe that the solution selling phenomenon leads to emphasis only on few aspects, crowding out others. It finally ends up in fragmented, piecemeal approach to reform.

Four, focus on intermediate outcomes. One of the effects of diagnostic framework of analysis and selling solutions is that governments tend to focus on intermediate outcomes and not final outcome. It is common for many district collectors and state governments to start efforts with "reducing dropout" as final motto. When your final aim is defined as reducing dropouts, it reduces your efforts on all the other things required for final outcome.

Five, misinterpretation of RCTs. The fragmented approach existed much before RCTs but it's getting strengthened due to misunderstanding of RCTs. There are many RCTs in education that consider individual inputs to education (infrastructure, libraries, textbooks, SMCs) and test the effect of addressing them, in a controlled environment. They often don't show positive results. These results are often used by people to criticise efforts on these fronts. It's a misinterpretation of RCT result because the result says that working on libraries alone won't yield results, but it's interpreted as never construct libraries.

The RCT literature is also a part of the problem. Many define the utility of RCTs as follows "governments have limited resources, so we strive to find the initiatives that can maximise the outcome". Framing problems in this manner legitimises the limited resources condition of government. Most often, the limited resources is a function of governments' will and hence can be increased. RCTs take this limited resources as a given and go in search of one particular intervention that yields outcome. Phrasing solutions in form of one particular intervention dilutes the necessity to work on other constraints, leading to fragmented and piecemeal reform.

One piece at a time, diagnostic framework to identify constraints etc. are useful for questions like "What's wrong with my car?", where there are probably only one or two problems. It's not for complex issues like education that involve multiple inputs. Further, there are no meta-problems in cars, meaning there are no such fundamental issues that give rise to numerous other issues. Also, few problems don't mix up to lead to new problems. It isn't the case with education. The underlying meta-problems, a source for the visible constraints exist and few problems can mix up to create new problems. We need a more in-depth analysis here.

On Quality - Reasons for improper implementation 

The first post outlined four reasons for poor quality of reform - i) mismatch between nature of governance required and governance practiced; ii) overlaying scale-ups over weak systems; iii) premature loading of system; iv) focusing programme implementation focus over resolving constraints.

Translating traditional understanding of accountability is the factor that leads to first issue. Programme mode of thinking instead of capacity mode of  thinking is the reason for rest three.

Many problems with government are usually linked to lack of accountability. In daily life parlance, accountability means finding if someone has done something wrong and punishing them for that. This identify and punish approach presumes that we can identify and we can punish. This approach fails when applied to education domain because it's difficult to identify the wrong (outcomes only appear after long time), difficult to monitor because there's no one standard way of teaching against which teacher can be measured. Since this fails, all such dynamic, engagement-intensive tasks are converted to those amenable to rule based monitoring.

Some of this is also reflected in the way some think about the education governance drawing inferences from elections. Some remark that bureaucracy can implement elections because responsibilities are clearly assigned and accountability is fixed. They strive to implement the same in education. As discussed earlier, this usually fails.

All these together lead to a post-office style functioning in education bureaucracy.

The driving factor for the 2nd, 3rd and 4th reasons for poor quality of reform is the programme mode of thinking. Programme mode of thinking is when you immediately think of a programme that can address a particular constraint. This leads to premature loading of systems because this thinking disregards capacity constraints. It leads to repeated imposition of scale-ups because the diagnosis of failure is always attributed to programme. Finally, it leads to temporary workarounds because programme is important here and not resolving constraints.

Instead of programme approach of thinking, we need a capacity approach of thinking. When faced with a constraint, unlike programme approach that thinks of a programme, capacity approach asks the following questions - i) what's the reason behind this problem; ii) what's the nature of capacity required to address this constraint? (increasing SMC involvement requires different type of capacity, teacher training requires different type of capacity and so on); iii) what's the extent of capacity required?

If Finland is revamping curriculum to teach students in terms of themes instead of subjects, before prescribing it, you think of the nature of capacity and extent of capacity required to carry out that reform. While dealing with teacher training, instead of thinking of it in terms of a usual programme implementation, you think of the capacity required and take steps accordingly. This prevents it from being a rule based task.

Thinking in these terms can thus prevent the the trap of programme mode of thinking. Also, read Gulzar Natarajan's insightful post titled "Why do we gloss over state capability deficiencies?" for reasons that span across sectors.

Overall, we can observe that inappropriate mental models and world views are the factors driving the inappropriate nature of reform. Hence, disentangling these models should be the first step towards reform. Unfortunately, this doesn't involve actionable policy recommendations, the only requirement of some, but this is much more powerful than that.

Till now, we have discussed i) why governments' efforts to reform education not yield results?; ii) why do governments do what they do regarding education. Both of these dealt only with public education. The next post will deal with private education. It addresses the question - Why do our private schools not perform better despite not having all the issues of public sector discussed in 1st and 2nd posts.


Read my book: UnpackED  - The black box of Indian school education reform (pdf, free to download)
Follow on Facebook: @iterativeadapt
Follow on Twitter: @iterate_adapt
Email subscription or RSS Feed: Enter id in the "Subscribe" text box, on the top right of the blog.

No comments:

Post a Comment