Author: Professor Dr. Syed Arif Kamal  Previous Article  Next Article  Printable Version PDF

Observations and Experiments in Science

The scientific method is closely related to the natural and the observable world. Questions and hypotheses come from observations and are tested by fresh observations. Observations are available to us in nature, e. g., volcano eruption, generation of heat in the Sun. In general, we do not have much control on these observations. However, planned observations of the heavenly bodies (e. g., motions of planets) are the essence of the observational astronomy. On the other hand, experiments are planned and controlled. These are, mostly, performed in laboratories. The purpose of an experiment is to better understand the natural world and not to merely understand the experimental data. If one constructs a triangle with a model, an experiment and a simulation (a computerized experiment) lying on its vertices nature lies at the center with all the four interconnected (Fig. 1). Recall that all theories and laws are valid within certain limits. An experiment can only produce an estimate of the true state of nature.

Fig. 1. Model, simulation and experiment

Following are the steps needed to solve a problem using scientific method (Fig. 2):


Observing a natural phenomenon.


Developing a question about cause and effect.


Formulating a tentative answer to the question (hypothesis) through inductive generalization from the observations.


Working out the consequences (predictions) of the formulated hypothesis.


Testing the hypothesis by taking further observations/conducting an experiment.

Take the famous example of the Newton's apple.


Apple falls towards the earth.


Why does the apple not go in other direction?

Tentative Answer:

The earth attracts everything towards itself?


A ball, which is left without support, shall fall towards the earth.


Ball eventually falls towards the earth.

Fig. 2. Steps in problem solving

The observations on which a scientific theory is based must be reproducible by anyone with the proper training and the facilities. If one takes observations at different places and at different times, one must obtain the same results. Such observations are called stable. Two observers conducting the same experiment under identical conditions must get identical results. Such observations are termed as objective. Not all the observations are identical at different places. If an observer measures atmospheric pressure at Karachi and the other measures at Nathiagali the values shall be different. Similarly, all observations are not identical at different times. One observes stars during the night and sun during the day. Are the laws based on these observations valid at different places and at different times? Science is based on the notion that the physical laws are globally valid. In fact it does not make any sense to do science, spend a lot of money and time, only, to find laws, which are not valid at different places and, during different periods of times. If one is not careful one may try to formulate a general law, which does not satisfy these conditions. For example, it never snows in Karachi or in Lahore. If one makes a law that it never snows in Pakistan, it would be incorrect because it does snow in Murree (law not valid at different places). Else, from the general observation that it is never dark during the day, a law formulated that it is never dark during the day time is wrong because it is dark during a total solar eclipse (law not valid at different times).

The great scientific experiments served to elaborate the formal aspects of method (decide between rival hypotheses, find the form of a law inductively, explore the characteristics of a naturally occurring process, use models to simulate processes, exploit an accident, interpret null results), develop the content of a theory (find hidden mechanism of a known effect, provide existence evidence, decompose a simple phenomenon), establish techniques (accuracy and care in manipulation, power and versatility of apparatus).

There are good experiments and there are bad experiments. Many are either poor experiments, which can result in wrong decisions, or inefficient experiments, which result in excessive cost or time delay in reaching a decision. The main reasons are lack of training in the strategy of experimenting, incorrect decisions based on insufficient data, bad engineering decisions by setting vague objectives for projects, inappropriate statistics courses taught in the colleges and in the universities.

A scientific approach to every problem whether it is in science, engineering, economics or finance is sure to bring results. To do so one must develop a keen observation of nature asking the questions how and why. A scientist must always follow the truth. Nobody is scared in science. Nothing is accepted by faith. All statements must be tested by observations and experiments.

Appeared in AGAHI, magazine of the Shaheed--Millat Government Degree College for Women, Azizabad, Karachi, 2007, pp 1-3 (Intellectual Preview)

Updated: May 1, 2016 (0000h UTC)  Previous Article  Next Article  Printable Version PDF

Back to Articles Page

Debates Proposal Writing Problem Solving PDF Early Talent Research Participation Program Life History

Best viewed with 800 x 600 screen resolution using Microsoft Internet Explorer. Copyright 1998-2016. Professor Dr. Syed Arif Kamal