Bio 131: Secondary Sources for Student Critiques (Barton)

Secondary sources generally provide interpretation and analysis of primary sources. In the sciences, popular secondary sources aim to summarize, explain, or simplify scientific information for the general public. Examples of popular secondary sources include news articles, podcasts, blogs, or websites.

While popular secondary sources are important for the public to stay informed, they can often over-simplify or sensationalize scientific research to the point of no longer providing accurate information. Being able to critically read these sources is an important skill to have.

There are many types of experimental evidence. Most secondary sources will not outright say the type of experimental evidence, but the original research should. Important factors in how strong evidence is: mitigating bias, sample size, time period, and reproducibility. 

• Randomized Controlled Trials: Studies which randomly place participants in groups to test the effect of a new treatment versus a placebo or standard treatment. Conclusions are generally considered strong because of mitigated bias, but a small sample size can be a limitation.
  • Blind vs. Double-Blind Trials: The purpose of blinding trials is to try and minimize any potential bias of the participants or researchers on the outcome of the study. Blind Trials refer to trials in which study participants do not know what group they are in, in order to minimize potential bias. Double-Blind Trials refer to trials in which study participants and researchers do not know which group is which. Note that Blind and Double-Blind Trials can only be performed on humans. 
  • Ex.: A study follows two groups of cancer patients, one group receiving traditional treatment and the other receiving a new treatment, to determine whether the new treatment is more effective.
     
• Animal and Cell Studies: Conducting research on animals can be useful in predicting effects in humans. However, some observed effects can differ, so subsequent human trials are required before a particular effect can be said to be seen in humans. Tests on isolated cells can also produce different results to those in the human body. 
  • Ex: Researchers study Simian Immunodeficiency Virus in chimpanzees to better understand how Human Immunodeficiency Virus functions in humans.
     
• Computer Simulations: Broadly, computer programs are used  to emulate a physical system in order to make heuristic inferences, predictions of data we do not or cannot have, or for better understanding data that we do have. Results need to be interpreted by experts in the field in order to be determined as reliable. 
  • Ex.: A climate scientist simulates how high sea levels will rise if glacial melt continues at its current rate.
     
• Correlational Studies: Correlational Studies observe the relationship between two variables without interfering with them. The purpose is to determine whether there is a positive correlation, negative correlation, or no correlation.
  • Ex.: A study looks at the difference in caffeine consumption between a group of coffee drinkers and a group of non-coffee drinkers to determine whether there is a correlation between high caffeine consumption and high blood pressure. 
     
• Cohort Studies: A cohort study observes a large group of individuals over time to find correlation between individuals. They are generally considered to be strong observational studies given population size and time period, but interpretations of the data can be inaccurate.
  • Prospective vs. Retrospective Studies: Prospective studies collect data in-real-time to find correlation between subjects. Retrospective studies look at pre-existing data over a set amount of time to find correlation between subjects.
  • Ex.: A prospective study of female college students, some who vape and some who don't, to determine possible long-term effects of vaping.

• Case-Controlled Studies: Similar to Retrospective Cohort Studies, Case-Controlled Studies look back at data from two groups of subjects, one with a particular condition and one without, to determine a factor that caused the condition. They can help to inform further studies, but cannot determine causality.
  • Ex.: Researches look at two groups of men age 50-60, one group with heart disease and the other without, to determine whether there is an association between lifelong sleep habits and heart disease. 
     
• Cross-Sectional Studies: Cross-Sectional Studies examine the relationship between a population and a particular condition in order to gain insights into the population. They cannot determine causality or effect, and instead are used for statistics gathering and hypothesis creating.
  • Ex.: A sample of rural Americans have their prevalence of nutrient deficiency assessed.
     
• Case Reports: Case Reports are written records of a researcher's observation of the effect of a particular treatment on a small group of individuals. Their aim is to inform whether further research should be done on a larger population. 
  • Ex.: A doctor records the results of an experimental treatment on two individuals with Covid-19.
     
• Anecdotal and Expert Opinions: One person presents their opinion on a particular topic. Strong anecdotal and expert opinions will come from experts in the field, contain references, be backed by credible science, and point out any limitations of the research. 
  • Ex.: A magazine article quotes an expert in the field of chronobiology saying that following daylight saving time is more in-tune with our bodies' biological clocks than not following it. 

Questions for determining whether a secondary source is reliable:

• Who is the author of the work? What have they published before?
• Does the author mention other reliable sources? 
• Is the author's tone professional and not sensationalist?
• Was the work published by a reputable newspaper or website?
• How up-to-date is the information? 

Questions for determining whether research is reliable:

• Do the researchers have institutional or educational backgrounds that lend credibility to what they are stating?
• Do the researchers describe their methods in-depth, and list any possible limitations or conflicts of interest?
• Is the research being cited by other researchers? 
• Is the research published in a reputable journal or pre-print server? 
• Where did funding for the research come from?