Public policy wonks and aficionados of
bureaucracy swear by evidence-based decision making. However, it is also
undeniable that critical public policies seldom achieve intended outcomes. More
often, public policies are associated with either unintended consequences or
ambiguous outcomes. Whenever policies succeed or fail, it is unclear for whom
and why.
For example, free primary education
policy implemented in all EAC member states has not produced access everywhere.
Moreover, learning outcomes are uneven and contested. Various agricultural
policy provisions have not improved national food and nutritional security or
eliminated the need to import primary food commodities. Moreover, policies and programs
aimed at improving maternal, neonatal and child health have produced mixed
results.
Sound evidence is necessary but not
sufficient to produce desired, equitable wellbeing outcomes as consequence of
policy or program implementation. The abundance of misalignment of policy
objectives and policy outcomes suggests that existing evidence-based approaches
to policy making do not address the critical elements across the policy cycle,
which are necessary for effective program or policy implementation and results
or impact
Moreover, the context of policymaking
has become more complex. Public participation and bottom up actions are
becoming increasingly important, often required by law. Citizens are more
literate, well informed and actively engaged in the policy formulation and
implementation process as well as government decisions and actions. More and
more, with the ubiquity of social media platforms and with devolved levels of
government, citizens have become agents of good governance and not passive beneficiaries.
Big data – the burst in variety,
velocity and volume of data, including growing prominence of unstructured,
non-sampled data available on social media – is providing new opportunities for
innovation in the policy cycle, and especially with a huge potential to improve
policy effectiveness. The emergence of big data has made possible the
development of tools, including statistical
and geospatial analysis, sentiment analysis and visualization, which aim to
support key stakeholders at every stage of the policy cycle hence facilitating
hindsight, deepening insight and informing foresight.
While
traditional policy analysis has been devoted to retrospective (ex post) analysis, the combination of
complexity thinking and big data analytics will enable a critical addition to
the toolbox, prospective (ex ante)
analysis, which essentially informs foresight hence, defining expected policy
impacts, while enabling near real time policy/program evaluation and refinement
during implementation.
Policy
or program design and implementation is on the cusp of a fundamental leap; from
decision-based approaches to an exciting new dawn of intelligence-based models
that are powered by big data analytics.
Big
data therefore provides an invaluable companion for program development and
design, identification of indictors for program monitoring and evaluation as
well as stakeholder mapping.
In
my view big data analytics presents an opportunity for policy makers and
bureaucrats, as well as big donor agencies to think and design policies and
programs as testable change hypothesis. Moreover, implementation locations
provide unique geographic, socio-economic and institutional arenas for real world
experimentation, testing and learning innovation in policy formulation and implementation.

This work is licensed under a Creative Commons Attribution 4.0 International License.