THE HAWTHORNE effect is one of the sins that are commonly committed by researchers, and this includes those who are quite seasoned. The Hawthorne effect can be quite subtle, especially when the subject being investigated is complex.
The Hawthorne effect is described as “the modification of behaviour by study participants in response to their knowledge that they are being observed or singled out for special treatment. In the simplest terms, the Hawthorne effect is increasing output in response to being watched.” It is a yes minister syndrome in public affairs. Give Mr Minister the answer you presume he wishes to hear.
In survey methodology, sampling theory becomes representative provided the universe or the sample space from which the sample is taken is arranged so that the units can be randomly selected, be it with probability proportionate to size or disproportionate to size. The latter, which is disproportionate, is compensated for by post-stratification weighting of the data where the disproportionate allocation had to be executed to have an analysable cell size of the variable of investigation.
But what is more important is to be aware of the terrain of investigation especially as relates to questionnaire design and the sought after outcomes in order to limit the possibility of ecological fallacy effects creeping into the results. To this end a literature review of policy designs is crucial.
In the current environment of social media, misinformation can be quite rife. This is so in many areas, especially in policy design where contestation over policy is driven through a complex system of interactions. When survey methods are deployed to understand societal temperature, the heat can represent a different measure caused by leading questions or the Hawthorne effect.
A leading question could be in the form of a statement like “you are unhappy, aren’t you?”. This is instead of asking the question “how are you feeling?”. Another way of asking a leading question is by setting the scene such that when a particular question is asked the scene is already set to derive a particular answer. This happens in particular when the issue at hand is complex and does not lend itself to common sense.
Policy issues are often very complex and belong to contested schools of thoughts. Their adjudication is not a yes or no answer but required to be driven first through a process of analysis and knowledge of impacts. Impacts are always differentiated along who loses and who gains or by how much these losses or gains outcomes are manifest.
On May 31 Afrobarometer conducted a survey the conclusions of which reflect ecological fallacies from a point of view of divergence between policy considerations, its design parameters, answers to the questions and the milieu of monumental failure of public facing services.
One such is the output from the question asking, “Should Eskom be privatised?”. Answers to this question are that 59% of the sampled population generalised to South Africa said, “Yes, it should be privatised”.
Afrobarometer established that 95% of the population have access to the grid – obviously to date performing disgustingly poorly. Then 87% of the population is dissatisfied with the government's performance in the supply of electricity.
READ OUR LATEST ENERGY TODAY MAGAZINE BELOW:
Given that the poor areas are usually subsidised, and almost 59% of the population live in poverty, if a subsidy for electricity is not provided, notwithstanding that government has completely failed and reversed the gains of almost universal coverage of supplying electricity, then the poor will face the wall, which they are already facing anyway.
The answer, “yes” to privatisation by an overwhelming majority, omits a critical question of how in the US, Enron, a private entity, wreaked havoc in electricity markets. The ecological fallacy of this answer is in its truncating the here and now of an inept government that has been good at destruction and the populist view of privatisation that is devoid of knowledge of dramatic failures in the capitals of privatisation, namely in the US.
If the public were made aware that France’s electricity is run by the state and citizens access it at affordable rates, or in Muammar Gaddafi’s Libya electricity was at the time provided for free, while Iran’s electricity is dirt cheap and run by the state, as is the case in several other jurisdictions, the data misdemeanour of ecological fallacy would not be tolerated as science including through common sense.
No doubt, the utterly shameful performance of the South African government on almost all fronts gives citizens cold comfort over the notion of what a government is. But it would be foolhardy to then argue that there is no need for a government.
The mischief by Afrobarometer on this privatisation narrative may just reflect narrow interests of capital in South Africa, where Reprivatize South Africa has only 6 million clients to milk dry, the rest of the Republic of South Africa are not only dry, but are udderless.
Beware of ecological fallacy, induced the Hawthorne effect and poor sampling methods. These do not only damage the practice of survey methods, but can poison the water of Hammanskraal with devastating effects. If the privatisation of Eskom were to be based on this flimsy air thin evidence, then God have mercy on the new Reprivatize of South Africa.
Dr Pali Lehohla is the director of the Economic Modelling Academy, a Professor of Practice at the University of Johannesburg, a Research Associate at Oxford University, a board member of Institute for Economic Justice at Wits and a distinguished Alumni of the University of Ghana. He is the former Statistician-General of South Africa.
BUSINESS REPORT