Protocol (science)

From Wikipedia, the free encyclopedia
Jump to: navigation, search

In the natural sciences a protocol is a predefined written procedural method in the design and implementation of experiments. Protocols are written whenever it is desirable to standardize a laboratory method to ensure successful replication of results by others in the same laboratory or by other laboratories. Detailed protocols also facilitate the assessment of results through peer review. In addition to detailed procedures and lists of required equipment and instruments, protocols often include information on safety precautions, the calculation of results and reporting standards, including statistical analysis and rules for predefining and documenting excluded data to avoid bias. Protocols are employed in a wide range of experimental fields, from social science to quantum mechanics. Written protocols are also employed in manufacturing to ensure consistent quality.


Formal protocol is the general rule in fields of applied science, such as environmental and medical studies that require the coordinated, standardized work of many participants. Such predefined protocols are an essential component of Good Laboratory Practice (GLP)[1] and Good Clinical Practice (GCP)[2] regulations. Protocols written for use by a specific laboratory may incorporate or reference standard operating procedures (SOP) governing general practices required by the laboratory. A protocol may also reference applicable laws and regulations that are applicable to the procedures described. Formal protocols typically require approval by a laboratory official before they are implemented for general use.

Manufacturing protocols are required by current good manufacturing practice (cGMP) for manufacture of foods, pharmaceuticals, and medical devices.

In a clinical trial, the protocol is carefully designed to safeguard the health of the participants as well as answer specific research questions. A protocol describes what types of people may participate in the trial; the schedule of tests, procedures, medications, and dosages; and the length of the study. While in a clinical trial, participants following a protocol are seen regularly by research staff to monitor their health and to determine the safety and effectiveness of their treatment. Since 1996, clinical trials conducted are widely expected to conform to and report the information called for in the CONSORT Statement, which provides a framework for designing and reporting protocols. Though tailored to health and medicine, ideas in the CONSORT statement are broadly applicable to other fields where experimental research is used. Clearly defined protocols are also required by research funded by the National Institutes of Health.[3]


Safety precautions are a valuable addition to a protocol, and can range from requiring goggles to provisions for containment of microbes, environmental hazards, toxic substances, and volatile solvents. Procedural contingencies in the event of an accident may be included in a protocol or in a referenced SOP.


Procedural information may include not only safety procedures but also procedures for avoiding contamination, calibration of equipment, equipment testing, documentation, and all other relevant issues. These procedural protocols can be used by skeptics to invalidate any claimed results if flaws are found.


Equipment testing and documentation includes all necessary specifications, calibrations, operating ranges, etc. Environmental factors such as temperature, humidity, barometric pressure, and other factors can often have effects on results. Documenting these factors should be a part of any good procedure.

Calculations, statistics and bias[edit]

Protocols for methods that produce numerical results generally include detailed formulae for calculation of results. Formula may also be included for preparation of reagents and other solutions required for the work. Methods of statistical analysis may be included to guide interpretation of the data.

Many protocols include provisions for avoiding bias in the interpretation of results. Approximation error is common to all measurements. These errors can be absolute errors from limitations of the equipment or propagation errors from approximate numbers used in calculations. Sample bias is the most common and sometimes the hardest bias to quantify. Statisticians often go to great lengths to ensure that the sample used is representative. For instance political polls are best when restricted to likely voters and this is one of the reasons why web polls cannot be considered scientific. The sample size is another important concept and can lead to biased data simply due to an unlikely event. A sample size of 10, i.e. polling 10 people, will seldom give valid polling results. Standard deviation and variance are concepts used to quantify the likely relevance of a given sample size. The mass media and the public often use average and mean values interchangeably, which can lead to dubious and even misleading arguments. The placebo effect and observer bias often require an experiment to use a double blind protocol and a control group.

Blinded protocols[edit]

A protocol may require blinding to avoid bias.

A single blind protocol requires that the experimenter does not know the identity of samples or animals during the testing and calculations. It is appropriate when no human subjects are involved.
A double blind protocol comes into play when human subjects are tested and requires insuring neither the experimenter nor experimental subjects have knowledge of the identity of the treatments or the results until after the experiment is complete.
A triple blind protocol is used when it is required that neither the experimenter, experimental subject(s), nor statisticians have knowledge of the identity of the treatments or the results until after the experiment is complete.

An experimenter may have latitude defining procedures for blinding and controls but may be required to justify those choices if the results are published or submitted to a regulatory agency. When it is known during the experiment which data was negative there are often reasons to rationalize why that data shouldn't be included. Positive data are rarely rationalized the same way.


A protocol may specify reporting requirements. Reporting requirements would include all elements of the experiments design and protocols and any environmental factors or mechanical limitations that might affect the validity of the results.

See also[edit]


  1. ^
  2. ^ "Guideline for Good Clinical Practice E6 R1" (PDF). International Conference on Harmonisation of Technical Requirements for the Registration of Pharmaceuticals for Human Use. 1996-06-10. Retrieved 2009-07-10. 
  3. ^ "Understanding Clinical Trials". National Institute of Health. Retrieved 2012-04-21.