What counts as a good justification for a claim?

The graph above shows the percentage of positive COVID-19 tests in Jacksonville, Florida in a one-month span from April to May 2020. The illustration was posted on Twitter by Lenny Curry, the mayor of Jacksonville. After seeing this seemingly hopeful graph in the news, I decided to do some further research. As COVID-19 tests were becoming more readily available in early April 2020, Jacksonville reduced the requirements to get tested. This led to more people testing, hence the decreased percentage of positive tests. In reality, COVID-19 numbers were sky-rocketing. This obvious manipulation of the public highlighted issues with the power structures in society.

Statistics are an adept way in which claims are justified in our daily lives. Although statistics may seem like good justification for a claim, they can easily be manipulated to reflect an ulterior motive or implicit bias. For the graph above, the motive was to reduce panic, although this may be an admirable cause, the true nature of the COVID-19 situation in Florida was not reflected. It is almost impossible to fact-check statistics given in the news, thus we rely on a variety of sources. But what is the ‘correct’ source? In a world where algorithms determine articles we do and don’t see, ‘justification’ for a claim becomes relative. So how can a claim ever be truly justified?

A study by two Ph.D. students concluded that there are three criteria in order for a claim to be justified: empirical consistency, plausibility of claims, and observation reliability (Lin, Jer-Yann and Guo, Ding-Ying). This states that evidence should be coherent with claims and previous knowledge, and be repeatable. For the last few thousand years, mathematics has been a universal justification for everyday claims. Unlike statistics, mathematics can be replicated and proven time and time again. Mathematics is based on fundamental truths from real life, one plus one gives two. In conclusion, in the context of our society, mathematics satisfies the criterion outlined by the referenced study, thus is good justification for a claim. 

The object above is a digital motion sensor I used to conduct a practice IA in Physics. I used it to gather data about the distance of an oscillating object. Upon collecting the data, some anomalies were found with random spikes in the measurement of distance. Since the investigation was done using the scientific method, an iterative cycle of questioning, experimenting, and concluding, the anomalies were dismissed.

The scientific method is a controllable, replicable, and justifiable method to prove a claim (Castillo, M.). The modern scientific method has been used since 1930 and has been repeated countless times. The scientific method is based on representative evidence. Representative evidence is accepting the common case over the exceptional case as correct. The exceptional case comes in the form of anomalies. We tend to ignore anomalies as it does not fit the expectations that we have for results. This method has limited our perspective of knowledge throughout history. As of this point, we have no way of guaranteeing that an anomaly is really an anomaly. There is the possibility that anomalies don’t exist and the anomalies we see is the correct data. One example of this is the electron orbital theory. The theory states that electrons are highly likely to be found in a mathematically defined region in space around the nucleus of an atom. However, part of that theory states that there is a non-zero probability that an electron that is supposed to be in the orbital is anywhere in space. A seemingly anomalous result may in fact be the truth.

In summary, even in the scientific method, humans experience confirmation bias. We look for data that is in line with our hypothesis and discard all ‘anomalies’. All the scientific theories we have now seem like the best justification for claims. Scientists have proved the unimaginable. However, 150 years ago, people believed that electrons orbited nuclei spherically, perhaps in 100 years, people will dismiss our theories of today as we dismiss the theories of the past.

This is a diagram I came across while studying IGCSE Economics. The diagram shows wage, plotted on the horizontal axis, against the number of hours worked, plotted on the vertical axis. The diagram shows that people are willing to work more hours if the wage they are given is higher until a point, where they are no longer willing to work. This was memorable as it was the first time I had seen a non-linear curve in an economics diagram and the most behavioural economics that I have learnt.

Human behaviour forms the basic understanding of economics and most human sciences. One can argue that humans are predictable. The basic laws of supply and demand are based on human behaviour. When the price increases, people are less willing to buy. But, like anything, there are exceptions. The theory above can be tested by taking samples of people's wages and the number of hours they are willing to work. However, even if one million people follow the trend, the claim stated above cannot be fully justified. There are situations in which people are willing to keep working more hours as their wage increases. A 2010 study concluded that human behaviour is 93% predictable (Northeastern University). Although this is a high percentage, a human-based claim will always be subject to a pre-existing, implicit, or explicit bias causing unpredictabliity and thus can never be justified. A good example of this is cryptocurrencies: fluctuations in cryptocurrencies are due to human behaviour, it is well known that predicting these fluctuations is almost impossible.

As much as we can predict human behaviour, one could argue that a claim based on human behaviour can never be fully justified with one-hundred percent certainty.

Economicshelp, www.economicshelp.org/wp-content/uploads/2013/09/backward-bending-supply-curve.jpg. Accessed 13 Mar. 2022.

Castillo, M. "The Scientific Method: A Need for Something Better?" American Journal of Neuroradiology, vol. 34, no. 9, Sept. 2013, pp. 1669-1671, AJNR. www.ajnr.org/content/34/9/1669.

Curry, Lenny. “Tweet image.” Twitter, May 24, 2020, https://twitter.com/lennycurry/status/1264577936079958022/photo/1.

Lin, Jer-Yann, and Ding-Ying Guo. "Undergraduates' Criteria to Justify Claims Proposed after Laboratory Experiments." Education Research International, vol. 2011, 2011, p. 1, Hindawi. www.hindawi.com/journals/edri/2011/612109/#references. Accessed 19 May 2022.

Northeastern University. "Human Behavior is 93% Predictable, Research Shows." 19 Feb. 2010, cos.northeastern.edu/news. Accessed 19 May 2022.

Vernier, www.vernier.com/product/gomotion/. Accessed 10 Mar. 2022.