Gender Approach in Project Evaluability Assessment
Gender mainstreaming is a complex process. Even with the will of authorities and teams, projects often fail to identify differential impacts, and thus also fail to anticipate gender risks. And when they are anticipated, these are often timid or vague initiatives, and generally with the wrong criteria in the design.
To understand why it is so difficult to introduce the gender perspective across projects, we must first understand the concept of evaluability.
General evaluability criteria are usually applied relatively systematically to projects at the start of their cycles. According to its simplest definition, it is about determining the possibilities of a project or policy to be evaluated during or after implementation.
But assuming that every project is evaluable based on the expected scope, the costs implementers are willing to cover, the willingness of stakeholders and the openness to various methodologies, evaluability becomes incidental to the identification of the project potential to be evaluated under optimal conditions, and produces proposals to generate those conditions.
In addition, evaluability allows:
- A more clear problem definition
- Agreed upon solutions
- Linking solutions to the assumptions behind the problem as it was defined
- Articulating the logic behind goals, components, activities, resources and results
- Producing the indicators needed to verify that logic
- Serving as input for feasibility analysis
- Ensuring measurement, monitoring, evaluation and accountability
In this sense, gender evaluability involves determining the optimal conditions that should be met in gender issues. This means analyzing the extent to which the project can be evaluated, based on valid and reliable information, depending on how it will have affected the lives and prospects of men, women, children, adolescents, people of sexual diversity from different generations.
There are still insufficient indicators disaggregated by gender and their relevance is not always obvious. For example, in the medical-pharmaceutical industry, clinical trials with animals traditionally use male mice, while females are excluded to avoid an additional variable: hormonal cycles. The conclusions are extrapolated to the overall population, which obviously includes females. At the human experimentation phase, two things have happened: It is unknown whether male-optimized drugs can work worse or have unexpected side effects on female subjects, and drugs that could potentially work better in women do not reach the testing stage.
As long as gender is not an integral part of evaluability criteria, the best designed experiment will meet all evaluability requirements without passing the first gender evaluability test: It may affect users differently depending on their gender, but with no data to support it.
Even when experiments include population of both genders, the male-centered bias is difficult to avoid. When we think of workers, we presume a male figure: someone whose main responsibility is to work outside the home to provide for his family, with no household chores and no need to care for dependents, with no time restrictions or limitations on his physical autonomy, someone who builds his persona around his work. When an intervention deals with universal workers, it essentially targets male workers.
Similarly, when we reflect on the impact of interventions on infrastructure, we think of households: units with collective needs and collective resources. We are unaware of the hierarchies that arise within; members have different possibilities for access and control of resources based on their gender, age, disabilities, etc. At the end of the project and based on the need to evaluate results, we are unable to discern the effects across population groups: We did not anticipate the need to do so.
As in the experimental sciences, evaluation should be a permanent process, performed throughout the lifespan of projects. The same should apply for gender and inclusion evaluation: It is not a question of parallel measurements, but of systematically performing informed evaluations about the potential effects of each initiative on the lives of men and women given the different roles that are socially assigned to them.
Therefore, there are some prerequisites for evaluability:
- Technical capabilities to understand the complexities and specific features of gender and scope articulation of projects
- Legitimation of those who interpret gender-relevant information, in order to request it from other stakeholders
- Resources, to apply those capabilities to information collection and analysis
- Relevant information for the gender approach: With proper planning, it can be requested from stakeholders (like any other indicator system, they must follow SMART criteria)
- Planning, because gender evaluation is not an alteration of established criteria to evaluate the pre-existing project, but it must integrate, be permeated by and ultimately transform the evaluation itself
- Participation, as a cornerstone for evaluability from a gender perspective. Only consultation with the protagonists will yield an accurate evaluation
Therefore, CAF is advancing a strategy of optimizing procedures and training staff in gender mainstreaming. To ensure the above prerequisites, knowledge of civil servants in Logical Framework for Results-Based Management will be updated; training in fundamentals will take place on various business issues, and their gender articulations; stakeholder engagement plans will be implemented; a series of guides will be published and content, tools and methodologies for implementation of the Gender Safeguard will be adjusted.
Thus, evaluability conditions will be significantly improved, i.e. a transformative equality perspective will be introduced in technical cooperation projects and investment for development projects.
