Why can't believe everything we see, case of the quality of datasets available in Africa

Even though I hate and was a bit traumatized with math and numbers, I do realize that the existence of numbers is really important not only in the scale of our daily life use such as having numbers to show how much should you pay for your coffee in Starbucks, but numbers are also used to monitor and control important things on a global scale such as countries GDP. Policy makers generally make policy adjustments or even new policies based on data they get from national and even international organizations that are regarded as reliable sources of accurate data, including those countries in Africa. However, can we rely on the reliability and accuracy of the data offered by these apparently trustworthy international organizations?

When purchasing food and other necessities, people will usually head to the closest supermarket, but where do they go to find out how much a country's GDP is? Up to this point, experts have typically used datasets of national income from the World Development Indicators, the Penn World Tables, and the datasets of Angus Maddison, all of which are based on national account files from the statistical agencies of each representative country. What differs here is how the method they use to calculate the total income of each country. If you think, ah, the data they produce will not be very much different, will it? You are completely wrong. The data generated from these three data sources produce different data about the GDP of countries in Africa such as how The World Bank places Mozambique as the 8th poorest country in Africa while datasets from Maddison place Mozambique into the 12 richest countries in Africa. Now, how is that possible?

There are many factors that cause the low quality of data available in African countries, such as lots of unrecorded important economic activities in these countries, not having up to date data available, and a lot of data that was not there to begin with. To produce a whole dataset of national income, a lot of regional data needs to be added together and calculated using a predetermined calculation method to become one clear and intact dataset to be able to see and review developments that occur within a country and matters other important. However, accurate data like this is very difficult to obtain in countries in Africa due to the unavailability of these data for various reasons such as lack of funds to conduct population surveys and censuses, difficulty accessing censuses, and so on. Then how can the state statistics service provide reports on datasets for a certain period of time? Your guess is absolutely correct if you think that making up numbers is the answer to that question.

Since there isn’t enough data to fill out the datasets required to evaluate the nation's development progress, experts typically fill in the gaps with their own figures. Making your own numbers, of course, can depend on a variety of variables, including modifying the data held by neighboring nations, using the number from earlier periods datasets with more complete data, or even using numbers produced by the formulas they develop. However, the statistics derived from these databases are still the outcome of expert assumption to create crucial national datasets and not the actual numbers that occur in the real world. Hence, it is very difficult to know the truth about many things in Africa, especially the accurate percentage of economic growth. It doesn't stop there, the existence of databases of countries in Africa who plays a very important role in making annual data is also not present. Making experts do not have the same database to base their calculations on and this makes the resulting datasets disastrous. Doesn't this sound bad?

The three prior data sources naturally overcame the lack of a database by using their own methods, including adding their own composition figures and databases to enhance the data they will display to the public. This explains why datasets about African nations is frequently different from one group to another. Making you hard to believe which data you should trust more isn’t it?