Throughout the world, doctor of philosophy is the last certificate anyone can earn from a university and feels being above others who do not have it. It is a certificate that confers enormous respect for the holders. Getting it requires a lot of processes. In some countries, PhD programme is being carried out using course work and research approach, while in others research approach is the most appropriate. In our experience, we have also seen countries where the programme is being done by publications. This means that students have to present a number of publications for the certificate.
However, this piece is not focusing on these approaches, it only presents some insights that emerged from a recent conversation on a digital platform of one of the African schools dedicated to skilling, reskilling and upskilling of early career researchers on the continent. Our analyst was part of the conversation, which was driven by an article shared by the founder and coordinator of the African-Early Career Researchers’ Academy, Professor Ayobami Ojebode of the Department of Communication and Language Arts, University of Ibadan, Nigeria.
Professor Ayobami Ojebode: Most of us have to deal with these statisticians from time to time …
Tekedia Mini-MBA edition 14 (June 3 – Sept 2, 2024) begins registrations; get massive discounts with early registration here.
Tekedia AI in Business Masterclass opens registrations here.
Join Tekedia Capital Syndicate and invest in Africa’s finest startups here.
Our analyst: It’s really important, especially when the researcher lacks proper understanding of those inferential statistics.
Professor Ayobami Ojebode: Look at you… why not, “especially when the statistician lacks proper understanding of the subject of research”
Our analyst: That is why researchers should employ action research approach even if the study doesn’t require it. As a statistician, I need to be carried along in your research design. I do tell most colleagues that without good statistical insights (qual or quant) research outcomes will not resonate with the intended purposes. Therefore, researchers need to understand some fundamentals of the statistics required for their researches.
Professor Ayobami Ojebode: Let’s discuss more after people have read the article… very interesting.
Rasheed Adebiyi: That some researchers lack proper understanding of inferential statistics is not debatable. However, what I found interesting is the angle of statisticians lacking a proper understanding of the subject of research. This happens when the statistician does not have a background in the discipline from where the research originates. There should be a balance somewhere.
Our Analyst: Yes. That should be a balance. That’s why researchers need to hire someone who has background knowledge of the field of the researchers instead of generalists. Despite this, researchers still need to engage the statisticians at the conceptualization stage not when data have been collected. I remember a case where a PhD candidate had collected data without knowing that a key demographic variable of the respondents is not included. To him, it’s not necessary. Unfortunately, he has a question that needs that. Another angle to the discourse is that understanding of statistics helps in knowing how to frame questions towards appropriate statistics application.
Rasheed Adebiyi: Yes. It is important to hire a statistician that understands the discipline of the researchers. It is as well good to have an understanding of statistical applications. For instance, anyone who wants to measure influence or the extent of the relationship between two variables should know that he or she would use correlation. Equally, it is critical too to know that first level analysis as reflected by frequency counts and descriptive statistics may not yield much. What is important is that some of us as researchers need to drop our phobia for Mathematics transferred to statistics and seek to know more about these feared but not too difficult statistical terms.
Our Analyst: Let me add a great addition here. Do we know that emphasizing theories driven research questions/studies is not enough without analysing collected data using theories as well? For instance, using social construction theory requires that researchers use multivariate statistics not univariate ones. The theory also expects researchers to interpret the outcomes using propositions and existing empirical evidences, leveraging cross-validating approach.
Mustapha Adeitan: I agree that there should be a balance and the researcher should explain the basic essence of the research to the analyst in the basic language that a layman would understand irrespective of the background of the researcher. However, I think the curriculum in Arts and Humanities should include the basics of statistical analysis that will be enough to help the researcher in crafting good and easy to analysis research questions and data collection items. With that, balance will be met and both the researcher and the analyst will have an easy ride on the journey to do the analysis.
Our Analyst: Yes. We need to include statistics in our curriculum. In the United States of America, UK and South Africa, which I know, you don’t finish your PhD without knowing relevant software for analysis in your discipline. Before the inclusion, our supervisors need to help us. Even if I failed to state advanced ones in my work, I expect my supervisor to challenge me. With that I will believe that including it [when I know its usefulness to my study] will not be misconstrued. I have had experience, where a PhD student told me that his supervisor does not want advanced statistics when it is obvious that first level analysis is nothing in the current post-positivism research era.
Mustapha Adeitan: One thing I have seen recently is that some of our lecturers in Nigeria are yet to wake in the post-positivism research era. I have a friend whose supervisor refused to sign his project in a University in the North because he used advanced statistical analysis. The Oga (in our term) insist descriptive statistics is still the best in their area of study. So, I ask myself, how do we move research forward with such supervisors still in the system.
Our Analyst: We still need them in the system. Old soldiers never worn out completely. We need their experience. The key thing is that they need to have mindset shift. Let me share a story. One of my academic papers was rejected recently by a journal in Nigeria. I told my co-author that this work will be published in an international journal. We sent it out. The reviewer said the value addition of the paper is the statistical approach we used. Mindset shift and upskilling remains the best strategies.
Mustapha Adeitan: I agree we need them in the system as experience cannot be replaced with anything. However, early career researchers and we on the advance study level should see this as a challenge and face it seriously. No doubt, the elders will leave the stage one day for those behind them to also perform on the stage. How well are we preparing for our performance to meet the world standard research approaches?
Rasheed Adebiyi: There is a need to introduce statistical courses especially at the higher levels – Masters and PhD’s.
Our Analyst: We need statistical communication as a course. It is a big issue when postgraduate students do not know how to use Microsoft excel for simple data analysis and visualisation. How do we get our thoughts to the beneficiaries of researches?
Mustapha Adeitan: Statistical communication should be a big issue we should look at especially in our department. In the current world of Journalism, Data Journalism has come to stay and it’s gaining more attention by day. If a graduate cannot do simple analysis with Microsoft Excel, then, how would such graduate venture into Data Journalism. It will be a minus.