Analysis Technique: Significant use of quantitative with qualitative analysis technique

Analysis Technique: Significant use of quantitative with qualitative analysis technique

            One essential insight is that big data analytics and artificial intelligence have a contrasting impact on the evaluation challenge. This factor implies that the level at which the element of causality can credibly be assigned between aid-funded projects and the implications for the donor projects. Conversely, new data analytics tools can bring forth new insights into the behavior and the living standards of individuals. However, the feedback loops might lead to further complicate causal inference (Letouzé and Stock 2019). With so much data to analyze, it is imperative to choose a careful scientific design. In response to this, a consensus has emerged for the use of mixed methods. The latter includes both qualitative and quantitative analysis. These two analysis techniques are increasingly embedded within the daily analysis to allow more dynamism (Letouzé and Stock 2019). Using both qualitative and quantitative methods provides a more vibrant set of indicators as well as sustained feedback, which facilitates monitoring and evaluation of the development projects. Hence, guidelines should be developed to integrate qualitative and quantitative analysis into big data analytics and artificial intelligence into monitoring and evaluation.

Case Study

Poverty: Poverty maps using a mobile phone and Satellite Data

The traditional approaches used in monitoring and targeting poverty areas by donor agencies rely heavily on the data from the census. Most of these data might be unavailable or outdated in many middle-income countries. However, alternative approaches can be used to provide such information (Letouzé et al., 2019). Recent research indicates that the use of innovative information sources, like satellite, as well as the use of analytical methods, can contribute immensely towards monitoring and evaluating the effectiveness of the development interventions established in such countries.

Data Sources. Remote Sensing, as well as the geographical mapping information systems, can be utilized thttps://studygroom.com/patients-health-needs/o assess the distances to roads and cities, thus reflecting access to market and information. The mobile operators call detail records can be used to track the movement of the mobile phones at an aggregate level. The geographical reach of the calling networks of different individuals can be correlated with the available economic opportunities (Letouzé et al., 2019). Both remote sensing and call detail recording capture distinct elements that can be used to provide an insight into the living conditions and the behaviors of individuals in the recipient countries. These timely data can be used in the monitoring of the recipient countries by the donors.

Approach. Different approaches, such as the one-dimensional and multi-dimensional indices, can be determined to evaluate the indicators of the living standards of individuals in the recipient countries (Letouzé et al., 2019). Additionally, monetary-based metrics can be utilized to determine the level of consumption and determine whether individuals’ level of income fall below the established poverty line.

To illustrate this, Steele et al. (2017) choose Bangladesh to utilize different sources like remote Sensing and call detail recording, as well as traditional data collected from surveys. The author aims at determining the level of accuracy that can be portrayed by each source to assess the poverty line (Steele et al., 2017). The approach recommended demonstrates a methodology for poverty modelling that is flexible. It represents an attempt to build maps using a combination of the two methods. The maps were generated using the CDR features, remote Sensing, as well as the geostatistical models. The dark color indicates the more impoverished regions and higher error in the uncertainty maps.

In differentiating the estimations for rural and urban areas, some features like night-time lights and the time taken in transport to the closest metropolitan areas were utilized in developing the rural areas models. The distance to roads as well as waterways was considerable in both rural and urban strata (Steele et al., 2017). The methodology provides for a robust alternative used in poverty indices that were obtained from the census data. Information collected from satellite images facilitates developing global measures of human well-being at the national level.