Data Powered Positive Deviance for climate change mitigation

The DPPD Approach

The Data Powered Positive Deviance (DPPD) approach, which has been developed by the DPPD Network, focuses on benefitting from non-traditional data to identify positive deviants and in a second step explore their solutions in order to potentially scale them. The DPPD Network was initiated in 2020 by the GIZ Data Lab, the UN Global Pulse Lab Jakarta, the UNDP Accelerator Labs Network and the University of Manchester’s Centre for Digital Development.

The DPPD approach follows a very similar methodology to the Positive Deviance (PD) approach. The PD method has been developed around the observation of the existence of individuals in each community or organization who attain better results than the rest of their peers, in spite of the fact that they have similar preconditions and resources. The DPPD method adds to the PD approach the use of new and unconventional data sets: DPPD utilizes non-traditional data – data that is digitally captured (e.g. mobile phone records and financial data), mediated (e.g. social media and online data), or observed (e.g. satellite imagery) – for the quest to find outliers.

Überblick über die fünf Phasen des DPPD-Ansatzes

The DPPD approach is especially beneficial as it reshapes development cooperation and aids moving away from the traditional ‘top-down identification and tackling of development challenges’. The DPPD method searches for pre-existing solutions in the target communities of development projects. By doing so, these communities’ resources and competences can be involved in tackling problems and contextual variables that are well accounted for. Thus, these bottom-up outcomes are more likely to be well tailored and therefore acceptable for the target communities.

Best Practices for Climate Change Mitigation

The DPPD method has until now been deployed in seven pilot projects around the world. Two of these projects are specifically focused on climate change and mitigating its effects, namely in Ecuador and in Somalia (for more detailed information, click on this video).

Climate change has been causing droughts in many parts of Africa. Pastoralists in Northern Somalia, who form the core of the population, are especially affected by climate change. Their livelihood depends on the availability of water and grazing resources which they need for their livestock. Livestock and products derived from livestock makes up for 80% of Somalia’s annual exports. With the droughts and the population growth, pressure is put on these resources and can potentially lead to conflict in the region. Rangeland management has been key in conserving grazing resources. The GIZ Data Lab, the GIZ Project on Sustainable Land Management, the UNDP Accelerator Lab Somalia and the University of Manchester Centre for Digital Development are working to identify positive deviants in Somaliland, namely communities with healthy rangelands since the 2016 – 2017 drought. Thus, understanding and pinpointing good practices that can be adopted by other communities in the region. The project team chose the West Golis livelihood zone as the study subject as the majority of pastoral community can be found there and because it offers eased access. The data used here includes:

Remote sensing data,

settlement location data,

observation data,

focus group discussions, and

semi-structured interviews.

Positive deviants have been identified and ‘rangeland rehabilitation activities’ could be observed. However, the pilot is still ongoing and the validation of the solutions still continues. After concluding the research, the team hopes to be able to recognize practices that can affect rangeland health and how these practices could contribute to the fight against climate change.

Further information about the project:

Rangelands Defying the Odds: A DPPD Inquiry in Somalia (external website)

Positive Deviance: Somali Rangelands Defying the Odds (Youtube video)

The DPPD Handbook Launch Event Min. 18:08 – 24:30 (Youtube video)
The Amazon region suffers from vast deforestation and in Ecuador hundreds of hectares of rainforest disappear yearly almost entirely for agricultural use. For the most part, the rainforest is converted to pastureland. The GIZ Data Lab together with the UNDP Accelerator Lab Ecuador, GIZ Ecuador and the University of Manchester have been working on using the DPPD to identify positive deviants, in this case farms, with low deforestation rates. The project focuses on two counties in Ecuador, Joya de los Sachas and Sucúa, over the span of five years (2015 – 2020). Farms that have proven to have a low yearly deforestation rate (below 1,3%) for three consecutive years (2018 – 2020) would be categorized as positive deviants.

The data used for this project is a mix of traditional and non-traditional data and include the following;

remote sensing data,

vaccination data,

cadastral data,

official statistics,

land use data,

observation data, and

semi-structured interviews.

The team was able to identify 53 potential positive deviants in both counties combined. In order to achieve the aim of the project, the project team is conducting interviews with farmers from the identified farms. This qualitative data results provide a better understanding on the practices of these farmers that led to their identification as positive deviants. Eventually, these practices could be translated into policies and thereby be spread among other farms resulting in overall reduced deforestation.

How to apply the DPPD method

The DPPD Handbook provides practical guidance on the implementation of the DPPD approach for specialists of any institution working in development cooperation (check out the DPPD Handbook launch event). It offers detailed step-by-step advice on measuring suitability of the DPPD method for projects as well as on how the method should best be employed and how to benefit from the findings.

Some of the most important things to consider when applying the DPPD method can be summarized in the following points;

When it comes to the planning and data analysis part:

Data readiness: Make sure that non-traditional data sets exist and are sufficient for finding conclusive results.
Data accessibility: Make sure you can access these data sets and are allowed to utilize them. Data access agreements can also be negotiated in order to gain access to non-public data.
Data attainability: Make sure the accessed data is suitable for the scope and resources of the project.

When it comes to the part when we discover underlying factors:

Concept communication: Make sure you understand completely how target communities will benefit from applying the method and make sure to communicate these benefits with the target community in order to achieve high engagement.
Incentive model: Think of creative ways to incentivize people from the target communities to participate in further data collection, such as surveys.
Gatekeepers: Identify individuals who can act as ‘shortcuts’ to interviewees.
Resources: Optimize the study design based on the resources (time, financial and personnel) you have.
Ethical considerations: Respect for the community and its culture as well as interaction with the community are essential. Make sure that applying the method will not put anyone from the community at risk. The target community should be at the center of your work.

When it comes to designing policy interventions based on the finding:

Make sure you identify the policy areas and levels, persons from the target communities you can formulate policies with as well as exactly how you can transfer your findings into policies.

Further information

The DPPD Handbook - Data Powered Positive Deviance DPPD
check the handbook
The Data Powered Positive Deviance Handbook Launch Event
link to video

Data Powered Positive Deviance for climate change mitigation

The DPPD Approach

Best Practices for Climate Change Mitigation

How to apply the DPPD method

Further information

The DPPD Handbook - Data Powered Positive Deviance DPPD

The Data Powered Positive Deviance Handbook Launch Event

You might also be interested in

Saving forests with the help of algorithms and data

UN Climate Change Conference initiates Digitalisation Day

Paying extra for the gig: What is the value of my work?