My 4 demands for scientific progress

Doubt serves as the foundation of good scientific practice. No body of scientific evidence is ever truly perfect or complete. The only way to obtain the most accurate picture possible is to remain skeptical, never letting go of our wits or our curiosity.

Even when the science is shown to be “wrong”, our wrongness exposes us to the gaps and the holes that allow us to fall into a better representation of the truth. It advances the study and reminds us that science is not absolute, that science allows us to test our logical assumptions and see how human logic holds up outside of personal experience and the confines of the human mind.

In our current political climate, it breaks my heart to see that this trend of doubt, which the general field of science so desperately relies on, is being replaced with an altogether denial of science on a national scale.

The outright denial of scientific proof is damaging.

When large numbers of people begin to believe that science is on the same level as witch craft, these people begin witch hunts against policy that is rooted in empirical evidence. This denial of science leads to the denial of mental health services which serve to protect our most vulnerable populations and the denial of policies which serve to protect our environment and the denial of policies which serve to advance human growth potential.

Returning national support to the belief of the efficacy of science relies on the implementation of four measures, which I consider my demands of the scientific and educational communities.

1.) A general increase in statistics and research literacy.

Do non-scientists need to know the difference between ANOVA and Hierarchical Regression?

Not necessarily.

However, they should, at the very least, be able to distinguish between exploratory research and replication. Not knowing this difference often leads to the irresponsible dissemination and absorption of new, and often poorly constructed, research (e.g. childhood vaccinations cause autism).

Some schools already attempt to increase science literacy with courses devoted to critical thinking.

I remember taking a class in high school where we were taught ten definitive ways to assess the reliability of a website (i.e. date, author, etc.), but this was just one small portion of one small class.

In an increasingly digital world, it is paramount that science literacy be implemented earlier and more regularly so it becomes a habit and not a choice.

2.) All studies should be triple-blind, if at all possible.

Surely, you have heard of double-blind research.

This type of research is when neither the researcher nor the participant know whether they are in the control group (usually a placebo) or the treatment group.

Double-blind research attempts to control for bias.

These biases include participants acting a certain way because they think they should or researchers assessing participants a certain way because they think they should.

There is at least one major limitation to double-blind research: the bias of the data analyst.

Different types of data analysis can produce different results depending on the assumptions an analyst makes prior to analysis. These assumptions can be subtly influenced by bias when an analyst knows who belongs to which group. This bias can also influence the way in which an analyst interprets the data.

Triple-blind research, as you may have guessed, accounts for the bias of participants, researchers, and analysts by preventing anyone from knowing much of anything about anyone else.

There are time and budget constraints associated with the implementation of triple-blind research, but if it is possible, it can and must be done for the sake of data integrity.

3.) Data integrity boards must be implemented nationwide.

Another way to ensure data integrity is to implement data integrity boards.

Data integrity boards are groups of highly trained researchers who validate every assumption and analysis implemented within a study. In other words, they are the people who make sure that what the researchers say their research did is what it actually did.

Considerations associated with this ask are the question of who is qualified to assess data integrity, who will fund these boards, and who will watch the watchmen.

This last point is especially important when considering the nepotism commonly associated with journal publications. Nepotism often leads to biases in favor of certain theories over others (e.g. Jabari and I go way back. Jabari works for X journal. Jabari high fives me and approves my flimsy research to be published.

But I am not here to offer answers to these questions— that would be a research project in itself.

4.) All data sets, especially those which are privately funded, should be made public.

I believe that everyone should be able to have access to all the world’s information.

Sure, if I come up with something brilliant, I want to receive credit for it; but that’s what creative commons copywriting is for.

Hiding brilliance behind paywalls does nothing but limit human understanding. If you want to profit from your research, then write books, create apps, make things. But don’t hide the data itself behind paywalls and copywrites. Such measure serve to increase distrust in science.

For an example of public dissemination of privately funded research, take a look at The Transparency Project, a data repository for addiction research.

In The Transparency Project’s own words,

Scientific information often is locked away in scientific journals, which tend to have limited accessibility. […] We believe that [public access and free summaries of research journals are] both publicly responsible and ethically desirable. Industry-funded scientific studies have often come under scrutiny by researchers, advocates, the media, and the public. The unfortunate history of tampering with science by scientific sponsors created this atmosphere of distrust. We seek to improve this complex situation by increasing accessibility to privately-funded data. It is our hope that this increased access will provide the impetus for the development of public-private research partnerships and simultaneously advance what we know about addictive behavior.”

Though these demands are by no means small tasks, if all four points are met, I believe that we’ll see a return of trust in scientific evidence and in the favor of public policy based on well-replicated, empirical evidence.

In terms of progress, we are long way off. Modern science is still relatively young. Psychology and neuroscience are even younger.

But youth is the best time for behavioral changes, right?