Evolutionary Path From Ornithomimus Dinosaurs to Birds: Image of Dinosaur With Feathers

A paleontologist's discovery of the fossil of an ostrich-like dinosaur with tail feathers and skin in the legs is stirring up excitement in the scientific world, in case you hadn't noticed. Other discoveries pointing toward that evolutionary path have been made before -- in China, for example, as I remember. However, this new one adds importantly to the accumulating mass of related finds, from which it also benefits, I think, in terms of the internet chatter we're witnessing.

READ/SEE: Dinosaur With Feathers

Next time you eat chicken -- or ingokho to many Kenyans -- just remember it's a living dinosaur. Give it the respect it deserves.

Control: A Key Standard of Scientific Method

Nine key standards, or principles, of scientific research can be gleaned from texts on research method. As I see them, these are:

1. Universality

2. Replicability

3. Control
4. Measurability
5. Validity
6. Reliability (Pending)

7. Objectivity (Pending)

8. Ethics 

9. Representativeness.

I say something about each of the nine standards in other posts. What I want to briefly talk about here is the control standard. 

Control

By control we mean establishing or maintaining the integrity or identity or boundaries of the sample/specimen/subject/site against all disturbance, unwanted/unintended influence or “contamination” by external/extraneous factors or even by the researcher himself or herself. In effect, control means quality control. 

In experimental design, Yu (n.d.) points out, the kind of control called for is, specifically, variance control  -- which is indeed a form of quality control. According to Leedy, control suggests establishing parameters: “Parameters are important. All research is conducted within an area sealed off by given parametric limitations. By such control, we isolate those factors which are critical to the research. Control is important for replication. An experiment should be repeated under the identical conditions and in the identical way in which it was first carried out. It is also important for consistency with the research design” (Leedy, 1980: 46).

Giorgio (n.d.) notes that, in the laboratory setting, the control function tends to be too narrowly carried out in practice, in a way that even conceptually distorts the meaning of control. Thus:

"...control practically comes to mean manipulate. That is because the laboratory setting is itself the creation of the scientist, and it is deliberately created in order for the researcher to be able to dictate to the situation. Any thing that is left to chance is seen as a demerit. The dictionary con firms these ideas by defining control as exercising “authority or dominating influence over; to hold in restraint or check; to direct or regulate.” Thus a strong sense of control means being able to do things at will, to be able to have desires transformed into realities in as brief a time as it takes to accomplish the task."

But, he argues, "control does not have to be defined so narrowly." Thus, for him, control:

"...could mean to be respectfully present in a situation so that one could respond knowledgeably and appropriately with readiness to the perceived spontaneous happenings of a situation. In this sense, control refers to a certain kind of “intentional presence,” a certain way of looking, an informed being present that at the same time does not actively intervene in the unfolding of the event to be observed... [The] knowledge that the researcher has guides his expectations so that he knows how to interpret or evaluate his ongoing perceptions. The researcher can do this without active intervention, but it is still a form of control because directedness and restraint are used by the researcher in perceiving the situation. Thus control implies observation, that is, it is an educated looking."

Admittedly, in the social sciences, in contrast to the natural sciences, control is more varied and more nuanced, and sometimes more problematic, but not impossible. For example, determining the parameters of, or controlling for variance in, a social research includes such considerations as scope and study limitations. Moreover, one can ensure control over the following, among others: [1] the setting of the interview, or the conditions under which the respondent is interviewed, [2] the accuracy/precision of the questions asked and the responses given. [3] the language of the interview. [4] the quality of training given to interviewers, other research assistants and supervisors. [5] the actual supervision process. [6] the parameters of both the experimental and the control group.

Archaeologists and paleontologists, who do not represent the image of the typical natural scientist, and who routinely work away from the public view, offer outstanding templates for quality control in their persistent quest for new discoveries. They are quite adept at protecting the sites of their ongoing work against contamination and interference, in time and space. In important ways, the work of the crime (or accident) investigator, often partially done in full view of the media and the public, at its best obeys the strictures of the control standard,

Clearly, controls are crucial not just in experimental designs and field observations, but also, and perhaps more dramatically so, in all situations in which the conscious and pre-emptive gathering and establishing of incontrovertible evidence is called for; for example, in the prosecutor's mandate, or in the context of a claim to/of the discovery of a consequential artifact. Both of which are typically subjected to strict and detailed scrutiny, skepticism and even attack.

REFERENCES:

Leedy, Paul D. (1980) Practical Research: Planning and Design. Second Edition. New York:  Macmillan Publishing Co.,

Giorgi, Amedeo (n.d.) "The Role of Observation and Control in Laboratory and Field Research Settings"

Jaikumar, Maheswari (2014) "Experimental Research Design" Slideshare.net

Mann, CJ (2003) "Observational Research Methods. Research Design II: Cohort, Cross Sectional and Case-Control Studies"

Yu, Chong-Ho (n.d.) "Experimental Design as Variance Control"