In late September 2014 I was chatting with a former colleague, WC who is a young chap working on his MBA project. Our Facebook conversation revolved around WC’s studies which led to his asking me to help him with the statistical analysis of his MBA project which I consented.
One thing I knew I would not be any better than WC. That is the use of the statistical package SPSS to do the number crunching. I have never used SPSS in my life. When I was a doctoral student back in the late 1980s, we had a rudimentary statistical analysis software that ran from the PC but for “serious” number crunching, we would need to key in the raw data into the minicomputer, the VAX.
It was the early days of the PC era and we just saw these funny looking “typewriters” in the university library. It was the IBM XT personal computer, with 5 and a quarter inch magnetic disk drive. Ever the inquisitive soul that I was, through trials and many errors (including many attempts to learn how to save my work!) I managed to learn enough to use the PC for word processing and later data capturing using Lotus 1-2-3. By then more PC would be made available including one in my department. With some reading from a guidebook (that was the period before the era of Internet) I managed to learn how to export data from Lotus 1-2-3 delimited by commas which the VAX could understand. I no longer had to spend hours at the VAX terminal (and having to queue for it) but to do my raw data transcribing using the PC’s spreadsheet programme. When it came to analyzing these data, it would be just a matter of minutes for me to load in the pre-crunched raw data and obtain my statistical analysis within a few minutes. In the world of the “PC semi-literates” I was the “king” of the faculty. Even the computer technicians did not know how to export Lotus 1-2-3 data in comma-delimited text file and referred many fellow researchers to me.
I was a sought after person for a while! Analysis of variance, T-test, Chi-square etc which I learned as an undergraduate in 1982 was something I understood well throughout my Master’s and doctoral studies. I even helped many fellow postgraduate students in analysing their research data. I also shared my knowledge with some of these guys who in turn were able to pass this skill along to others.
Fast forward 25 years, when I thought of helping WC, I realised that learning to use PC-based statistical analytical software like SPSS would not be difficult. In fact I found an Open Source version called PSPP that works very well under my Linux-Ubuntu operated 7-years-old hand-me-down laptop from my son.
I used about 2 days to learn how to use PSPP and by the third day, I had familiarized myself enough of the basics to be able to import raw data from WC. I was also able to carry out most of the required statistical analysis using PSPP. However when I tried to make sense of the analysis churned out by PSPP, I suddenly realised that, after over 20 years of not using my core statistical analysis knowledge, I had to relearn how to interpret these results. My previous career as a commercial research scientist in Singapore in the early to mid 1990s did not necessitate the use of much in terms of statistical analysis. We were interested in finding out if a particular treatment work in multiplying good quality plants inside our test-tubes and moved on without much resources to carry out statistically verifiable experiments on these studies. We were worried about bottomline and I was not allowed to publish any proprietary research work which had commercial implications. Thus for 20 odd years, I did not have the chance to use my knowledge of statistical analysis, hence my “losing” much of my former flair in this area!
Luckily, together with WC (plus the help of Google Search), I was able to regain the bulk of my knowledge in this area. WC was happy with the discussion we had which I hope will lead to his getting a good grade for his MBA project.
There are two key take home lessons from this story:
- No matter how much of an expert you are in an area, if you do not make use of the knowledge and skills acquired for a prolong period of time, sooner or later, as I had discovered about my knowledge of interpreting statistical analysis, you will find that you have lost these skills.
- “Old dog” (a.k.a. yours truly) can still learn new tricks. In my case, I learned as much from WC as he did from me during our long discussion on how he should be analyzing his MBA project data and how he could interpret the statistical analysis.
The best way to re-gain your knowledge is in relearning and using this knowledge. You may, as I did, not only regain your old skills but pick up new skills (in using PSPP) as well!