Introduction to Laboratory Research

Essential questions 

 Research methodology is used to establish causal relationship between actions and effects that they cause. The essential questions are:

• What is the strength of evidence provided by various research methods?
• What research method is most suitable for a particular research purpose?

Content

Historically, the earliest scientific research method was descriptive qualitative (also known as ethnological, case studies). It is based on the detailed description of events, situations, documents, factors, and outcomes as they are. While it is good for record keeping, this method has the lowest evidence of effectiveness for the accurate causation and prognosis based on the evidence [1].

The next step in gaining evidence of effectiveness was a descriptive quantitative method. It is based on measuring effects of intervention on the subject and recording outcomes as they are (usually one time), instead of providing just their non-quantitative description. 

Correlational or regression analysis method allows to establish a quantitative relationship between two or more variables and has greater evidence of effectiveness. Correlation does not imply causation but establishes how the two variables vary together [2]. 

Quasi-experimental research is based on the comparison of outcomes in two groups of subjects in empirical study: one with a specific intervention and another without those interventions; however, random subject assignments to these two groups is not used. It can be affected by confounding variables that cause a spurious association [3].

A marked step up from it is an experimental research method in which the subjects are randomly assigned to the two groups: one that will receive an intervention (experimental), and the one that will not (control). This allows to exclude the effect of non-random factors on the research outcomes and provides a more accurate causation and prognosis [4]. In true experimental research all experimental conditions are controlled, and one parameter is varied. The outcome of the research experiment is based on the effect of the variable parameter on the response of the experimental subject.

Meta-analysis is based on the compilation and synthesis of the results of multiple research studies. It aims to establish the average impact of the specific or similar interventions. It has the greatest evidence of effectiveness and predictive value.   

Teaching science laboratories were in existence for a shorter period of time than the scientific, clinical, and industrial research laboratories. Teaching laboratories are designed to expose students to the learning of properties of natural object through manipulations, hands on interaction, and by application of methodology and theoretic basis of contemporary science. The National Science Education Standards require more emphasis on student learning “in the context of inquiry, technology, science in personal and social perspectives, and history and nature of science” [6]. Teaching labs are essential for the completion of these goals [7].

Suggested readings

 Savin-Baden M., Major C. (2013). Qualitative Research: The Essential Guide to Theory and Practice. London: Routledge.

Dietrich C. F. (1991) Uncertainty, Calibration and Probability: The Statistics of Scientific and Industrial Measurement (2nd ed.), Higler Series on Science and Technology. CRC Press, Boca Raton, FL.

Rossi P. H., Lipsey M.W., Freeman H. E. (2004). Evaluation: A Systematic Approach (7th ed.). SAGE Publications.

Hinkelmann, K., Kempthorne, O. (2008). Design and Analysis of Experiments, Volume I: Introduction to Experimental Design (Second ed.). Wiley.

Haidich A.B. (2010) Meta-analysis in medical research. Hippokratia. 14(Suppl 1):29-37.

National Science Education Standards. (1996) https://www.nap.edu/read/4962/chapter/8

National Science Teachers Association (NSTA). 2007. NSTA Position Statement: The Integral Role of Laboratory Investigations in Science Instruction.