Statistical forecasting is futile beyond a point

EVERY year, on March 8, International Women’s Day, the world routinely recalls severe gender disparities in societies. The prevailing gender gap is also under scrutiny when the World Economic Forum (WEF) releases its annual report on the Global Gender Gap Index. A few days ago, the 2024 index was revealed.

The WEF calculates the index using predetermined procedures pertaining to several facets of lifestyle as well as a few preset measures. In 2006, the Gender Gap Index encompassed 115 countries, representing over 90 per cent of the world’s population. By 2024, the index would have expanded to cover 146 countries.

How is such an index value read? For every given index, a value of ‘1’ denotes the absence of any gender difference in that regard, whereas a value of ‘0’ denotes the most extreme possible disparity. A higher value corresponds to a smaller disparity. Notably, these indices highlight the disparities in the achievements of men and women in a given society, regardless of the opportunities that society may present, rather than the general level of economy, education, health or political stability of the society. It is not always true that indices representing highly developed nations have larger values. For instance, the US ranks 43rd, with an overall 74.7 per cent score of gender difference. It is far lower than that of countries like Namibia (80.5 per cent) and Nicaragua (81.1 per cent).

In which direction is the gender gap moving worldwide? Of the sample of 146 countries included in the 2024 report, 50.1 per cent reported overall score rises, 6.1 per cent indicated no change in score and the others reported score decreases. Seven European economies are among the top 10, along with New Zealand (ranked fourth), Nicaragua (ranked sixth), and Namibia (eighth). Importantly, no nation has yet attained complete gender parity. Iceland has closed 93.4 per cent of the gender gap, making it the country closest to achieving gender parity. It’s the only country to achieve a score above 90 per cent. Eight more nations have closed their gender gap by 80 per cent or more.

How about India? Well, we are approximately two-thirds of the way towards closing the gender gap, with a 64.1 per cent score in 2024. India is ranked 129th overall, slightly behind where it was a year ago (127th).

Four pillars or dimensions are used to generate the Global Gender Gap Index. They are: economic participation and opportunity, educational attainment, health and survival and political empowerment. India’s scores according to the other three indices are not great, even if the country does pretty well in the political empowerment criterion (65th rank), according to the WEF report.

But how efficient are those kinds of specific measures? Well, any indicator pertaining to inequality, hunger, democracy, etc. may inevitably give rise to disputes and the accuracy of its particular conclusions is frequently called into question. Undoubtedly, there are a number of significant inherent problems with most of these indices, including the validity of the methodology and the nature and accuracy of the data.

For example, the Global Gender Gap Report finds that 96 per cent of the gender gap in global health and survival, 94.9 per cent of the gap in educational attainment, 60.5 per cent of the gap in economic participation and 22.5 per cent of the gender gap in political empowerment have been closed. The overall index is the weighted average of these four with equal weights. These four pillars, together with their sub-divisions, are undoubtedly not the optimal choices. There may be other important aspects of lifestyle that represent the gender gap, one may argue. And why are the weights of these four pillars precisely equal, which seems rather arbitrary?

And what about the quality of the data, which is gathered from many sources across nations? How reliable and accurate is the data as it is provided “as is, as available” and “without warranty of any kind, either express or implied, including, without limitation, warranties of merchantability, fitness for a particular purpose and non-infringement,” as acknowledged by the WEF?

The WEF report has a humorous element, in my opinion, as well. The WEF typically projects when it will be possible to eliminate the gender gap worldwide. For instance, based on the current progress rates, the estimated timeframe for achieving full gender parity is 134 years, which translates to around five generations beyond the 2030 Sustainable Development Goal (SDG) objective.

Well, how good is that prediction? I believe it to be, at the very least, a misinterpretation of statistics. It should be almost impossible to predict what will happen in 134 years based on data from 2006-2024! In 2018, the WEF report predicted that the world would take another 108 years to close the overall gender gap across the 106 countries covered since 2006. We observe that the anticipated time interval has grown by an additional 26 years after six years!

Keep in mind that such statistical forecasting makes sense within a reasonable limit beyond the data range. Take an example from a different scenario. As demonstrated by the 1986 catastrophe of the US space shuttle Challenger, which broke apart 73 seconds into its launch and killed all seven crew members aboard, statistical forecasting is useless beyond a certain point. The failure of the primary and secondary redundant O-ring seals in a joint in the shuttle’s right solid rocket booster was determined by retrospective investigations to be the reason for the accident. The Rubber O-rings’ capacity to seal joints was diminished due to stiffness caused by the record-low temperature on the morning of the launch, which was 31°F outside. For any of the 23 previous launches (prior to the Challenger accident), 53°F was the lowest temperature recorded. Thus, the Challenger was launched at a temperature that was much outside of the data range, a fully uncharted territory where predictions could occasionally become meaningless.

In actuality, even for fiction writers, it is exceedingly challenging to forecast social shifts and technological advancements in the moderate-to-distant future. Furthermore, we must never forget that decision-makers, extraordinary social reformers and other stakeholders can always expedite this process in order to attain gender equality ahead of schedule. None can predict when a society will get a figure like Raja Rammohan Roy or Ishwar Chandra Vidyasagar, Kemal Ataturk, Begum Rokeya or Jyotirao Phule, despite the fact that the two halves of the sky differ greatly in brightness and in nature across nations.