By Mohamed Sankoh (One Drop)
I’m starting today’s One Dropian dropping by asking few questions. Why has the Institute of Governance Reform (IGR) never thought of conducting a survey on the human rights record of the Sierra Leone People’s Party (SLPP) government despite three international organizations (Amnesty International, the European Parliament, and the United States of America’s State Department) have recently taken the Bio-led administration to the abattoir for that?
Why has the IGR never thought of conducting a survey on the SLPP government’s management of the country’s economy? And why is the IGR yet to conduct a survey on the ID cards issued by the Electoral Commission of Sierra Leone (ECSL) for the June 24 general elections?
It seems to me that the IGR appears to be mainly concerned with issues that appear to either present the SLPP government in a favourable light or those that seem to attempt to launder the image of the Bio-led administration. Take for example its recent survey titled: “The June 2023 Elections: A Forecast of the Electoral Map of Sierra Leone”; just a perusal of it, one gets the feeling that it might have been conducted with a preconceived result in mind.
I’m assuming that the IGR’s latest survey appears to have been conducted with a preconceived result in mind simply because of the methodology used. Random Sampling is a method used by researchers to make generalizations about a population, which at the end of the day might have “the possibility of bias under certain circumstances”.
And it is from that angle of “possibility of bias under certain circumstances” that I intend to look at the IGR’s survey. First and foremost, IGR’s use of Random Sampling, as a methodology, might suggest to the critical mind that the survey might not have removed an intended bias from the entire process. Those who have been engaged in research work know that some researchers would choose targeted places for random sampling where they believe definite results could be attained to back up their own hidden bias or biases.
Let me assume, for the sake of an intellectual debate, that the preconceived result in the minds of some of the IGR’s “boots on the ground”, who conducted the interviews, was to predict a “59% win for SLPP” (as interpreted and reported by The Global Times newspaper of Thursday 1 June 2023). All they need to do is to select places where they know SLPP supporters are in large numbers and do their interviews. For example, if a pro-All People’s Congress (APC) Civil Society Organization (CSO) wants to do a survey that is preconceived to predict an overwhelming APC victory in the June 24 general elections; all that CSO will have to do is to skillfully carry out a random sampling at Abacha Street or Brookfields in Freetown and other APC strongholds, countrywide, where they expect six out of ten respondents to likely be pro-APC. With this random sampling; the expected result will be obvious!
Again, let me be somehow intellectual here. In their book: “Mass Communication Research Methods”, Hansen et al posit that, “There is nothing sacred or inviolate about any particular [research] method. Whatever methods we advocate, they will not have been conceived, developed or applied in a cultural or political vacuum…” So, with this in mind, I’m postulating that the IGR’s “boots on the ground” who conducted the “face to face interviews” with the “2,428 registered and 19 unregistered voters”, according to IGR, might not have done that in a political or cultural vacuum. They have their political leanings or sympathies and cultural backgrounds which might have influenced the manner in which they conducted those interviews.
And if I may ask: Were the “boots on the ground” certified by the IGR as political neutrals before they were sent on the ground? Were those “boots on the ground” randomly selected? Were they selected on certain conditions? Can I be provided with a list of those “boots on the ground” so that I will be able to do a tribal and regional audit? I’m asking these questions simply because if they are honestly answered, they might elucidate on certain issues which might assist me to determine whether the survey itself is “…subject to a wide range of extraneous influences” (to quote Hansen et al again) or not.
And it is interesting that the IGR accidentally tells me that I should not trust its figures simply because of this critical fact: under its bullet point of “Number of respondents”; I’m told that there were “2,448” respondents. But when the IGR breaks down this figure into “2,428 registered and 19 unregistered voters”, the final figure should have been 2,447 not “2,448” as stated! So, if the IGR can be very sloppy with compiling figures how then can I be sure that the entire process was not sloppy? How can I also trust the “+1.98%” margin of error in that survey?
Another issue of interest in the IGR’s latest survey is the number of respondents. With a voting population of 3,374,258 for the upcoming elections, according to ECSL’s Voter Registration, only “2,448” voters were interviewed by the IGR’s “boots on the ground”. Those who have engaged in research work will tell you that those “2,448” respondents cannot accurately represent the current 3,374,258 voting population of Sierra Leone. According to one researcher, “Small sample size can lead to inaccurate results. This is because a small sample is not representative of the population as a whole and it can more easily be influenced by chance or by selection bias”.
Yet the IGR is bold enough to tell me that it conducts “opinion polls to…support political parties with reliable data to plan their campaigns….” This reminds me of what Hansen et al put forward that, “There are still researchers who claim to be value free and objective—although over the last quarter of a century there has been a gradual realisation that research too, the questions asked and the methods used, may be subject to a wide range of extraneous influences.” Perhaps, just perhaps, those IGR’s predictions, in the survey under review, might have been influenced by “a wide range of extraneous influences” due to the political leanings and cultural backgrounds of the “boots on the ground” who conducted the interviews!
The only part where I slightly agree with that IGR survey is where it states that, “Also it is useful to note that as the campaign unfolds, some grounds might shift. Thus, a final prediction is required at least a week before election.” This tells me that the IGR might be aware that after 26 May 2023 (when the “boots on the ground” left the ground) to a week before the elections; some important or damaging events might have taken place in Sierra Leone which might make nonsense of their earlier predictions. So, they have to insert a straw which they might need to cling unto if the need arises.
Having gone through IGR’s “The June 2023 Elections: A Forecast of the Electoral Map of Sierra Leone” survey; I then spoke randomly with some citizens about it. Most of them dismissed it as unrealistic considering the current state of affairs in Sierra Leone.
[email protected]/+232-76-611-986