Science and later computers have fascinated me since I was a kid. When I decided to be a scientist in high school, I found science a flashlight that illuminated many biological and physical phenomena around me. I paid attention to many scientific facts in physics and biology. I had pure fun when I studied how scientists came up with laws of physics or equations. These laws explained an observed phenomenon. When I started studying science in college, I found science to crystallize my thinking, my drive, my life. I am always happy to read, watch, listen to scientific discoveries.
Scientific achievement starts with simple observations, followed by simple questions, followed by testing a hypothesis. Scientific researchers at the University of California, Berkeley defined science as: “In science, ideas can never be completely proved or completely disproved. Instead, science accepts or rejects ideas based on supporting and refuting evidence, and may revise those conclusions if warranted by new evidence or perspectives.” To prove a hypothesis in science, experiments with the right controls give excellent reproducible data. However, experiments have a tendency to disprove something based on science rather than proving it. To prove something in science, you need to design the right experiments and use the right statistical tools to analyze the data, with the right negative and positive controls. When a scientist wants to publish a manuscript, reviewers of the manuscript examine the controls first and don’t even bother reading the rest of the manuscript when controls are questionable. The proof of scientific observation planned to test one variable, only one, at a time creates a beginning. Studying other variables creates a story.
The shortcomings of science
A recent example of the shortcoming of scientific research is an article by a scientist Dr. Haruko Obokata who created stem cells quickly and cheaply from normal cells, published by Nature (one of the most prestigious scientific journals in the world). The whole research and outcome turned out to be falsified data. Not only could nobody repeat the experiments to generate stem cells from normal cell, but also Dr. Obokata herself could not generate any stem cells after an internal investigation was conducted by the university, where she worked. Dr. Obokata, including the coauthors, withdrew the manuscript leaving the journal to scramble. The editor-in-chief rushed to do some damage control to protect the reputation of the journal by withdrawing their support. Until now, nobody can understand how the reviewers did not spot the irregularities.
The evaluation of scientific research
In order for data to be valid, they should be reproducible. Several outspoken critics recently pointed out that up to 60% (it might be even higher; nobody knows for sure) of scientific research could not be reproduced. If this is true, it is a sad fact. Generally, to evaluate studies, we need to raise the following questions:
- Was the sample size big enough? (Although there is no set number, experts generally believe that the bigger the sample size, the better.)
- Was the experiment repeated several times to insure reproducibility (at least three times)?
- What are the opinions of the experts in this research?
Some strategies for understanding scientific research is reading newsletters specialized in different subjects from prominent scientific establishments , e.g., Harvard. These scientific newsletters don’t have ads, and the writing is short and sweet. Of course, we can read websites or Wikipedia for free on the Internet or read journals in university libraries, but the newsletter subscription is like a filter to weed out unrelated information, providing simple and clear language written by the experts in the field. You are buying your time. If you can afford these newsletters, buy them now, don’t wait.
This blog is published every week on Saturday before 10:00 pm. US Eastern time. Thank you for reading my blog. I would love to hear from you.