THE SCIENCE PROCESSES
Illinois State University
The following is a list of the thirteen science processes advocated by the American Association for the Advancement of Science (AAAS). These are best thought of as a set of intellectual skills that are associated with acquiring reliable information about nature. Each process is defined. In addition, comment about the inherent nature of each of the skills is provided. The first eight processes are called "basic processes" and are appropriate for children in the primary grades. The last five are called "integrated processes" and are more appropriate for children at grades four and above.
This is the most fundamental of all of the processes. Observation may be defined as the gathering of information through the use of any one, or combination of the five basic senses; sight, hearing, touch, taste, and smell.
The term observation may also be used to express the result of observing. In other words one might observe and, as a result, gather observations. These observations can also be called data or facts.
Observation should suggest objectivity as opposed to the expression of opinion. For example, "John is a bad boy" is not an observation. On the other hand, "John exhibits behavior that we characterize as bad" is an observation. "John is throwing Mary out of the window" is also an observation.
Skilled observers seem to proceed from general perceptions of a system to more specific ones so the nature of skilled observing can be thought of as analytical. Systems are first observed as a whole then analyzed for subsystem information. Subsequently, subsystems can be treated as a whole and subjected to further analysis in an ever tightening spiral. Technology can be used to amplify the senses, which provides for even more analysis. A microscope, for example, is a technology that allows us to see things that are too small to be seen with the unaided eye.
In summary, observation is an objective process of gathering data through the use of one's senses applied in an analytical way.
Measurement is an observation made more specific by comparing some attribute of a system to a standard of reference. An example is when the length of an object is expressed in terms of the length of a meter or when the mass of an object is expressed by referring to a standard such as a gram. Measurement and observation are the only process skills that are actually two forms of the same thing.
There are many standards that can be employed to make observations more precise. For instance, academic scholarship can be expressed as a grade. When one receives an "A" or a "C" in a course one's performance has been measured relative to a standard. In a similar fashion, a four star restaurant is a measure of quality.
As one can see from these examples, a measurement can range from highly concrete and universal to rather conditional. Observing that a stick is 27 centimeters long requires little interpretation. The meaning is rigid and understood by anyone, anywhere who is familiar with the metric system. On the other hand, being an "A" student may require considerable interpretation with meaning highly dependent upon circumstance. And, of course, with respect to restaurants, "Charlie's Four Star Chili Dog Heaven" may be just that to some.
The nature of this process entails the description of some system attribute by comparison to a standard of reference.
Classification is the process of grouping objects on the basis of observable traits. Objects that share a given characteristic can be said to belong to the same set. The process is somewhat arbitrary depending upon the identifying trait selected.
This is an important process to science because of an underlying assumption that kinship in one regard may entail kinship in others. Science assumes that to a large degree the universe is consistent with it's laws holding true everywhere. Therefore, if a set of objects share one thing in common they may well share other attributes.
Also there is the notion of realness or depth. This means that the more characteristic a trait is of a particular system the closer the kinship of those sharing the trait. For example, consider the idea of a marble. What makes a marble a marble? Is color a fundamental component of being a marble? We could, of course, classify objects on the basis of color but is that a deep characteristic? Because some marbles are red does it follow that all red objects are marbles? The issue here is that some traits are more expressive of the essence of the system than are other shared traits. In most instances we should seek to classify on the basis of traits that are essential to the idea of the set.
The nature of the skill of classification is two fold. First, one must be able to identify traits and, second, one must select traits that express the deeper essence of the system.
Quantification refers to the process of using numbers to express observations rather than relying only on qualitative descriptions.
The process has two major values. First, by expressing something in numerical terms the need for translation of verbal meaning is reduced. Second, the use of numbers allows mathematical logic to be applied to attempts to explore, describe and understand nature.
For example, consider a situation where one might try to describe the various hair colors of students in a classroom. Try making an accurate and complete description using only qualitative terms. At best we might develop groupings based on generic names such as brunette and blonde (I am sure you will recognize these as an example of classification, as described above). The problem we must deal with is that terms such as brunette and blonde are not absolute. Some brunettes are obviously darker than others and some blondes are clearly lighter than others and we need a scheme that will allow us to express such variation. Numbers will allow us to do that. For example, suppose Sally's hair is the darkest and Jeff's is the lightest. If we assigned a number such as 10 to Sally and 1 to Jeff a range has been developed within which all other shades must fall. Incidentally, the range could be reversed with Sally being assigned the 1 and Jeff the 10. It really doesn't matter and the scheme would work just as well. Either way, by defining color as a number the arithmetic logic of sequencing can be applied to the problem. In so doing, we find that all observers of hair color are playing by the same rules. Everyone is accepting the quantitative logic so that there is no question that haircolor #7 must fall somewhere, probably midway, between #6 and #8. This leaves a lot of room for describing very subtle differences. For instance, we can have some idea of the color difference between a 6.9 and a 7.2 but try describing that difference in qualitative terms.
Consequently the nature of the skill of quantification is one of application where one seeks precision of expression by transferring the logic of mathematics to qualitative problems.
Inferring is an inventive process in which an assumption of cause is generated to explain an observed event. This is a very common function and is influenced by culture and personal theories of nature.
Inferences can also influence actions. For example, suppose two students receive a poor grade on some project. One student observes the poor grade and infers that the reason he received it was because the teacher does not like him. The second student infers that he did not spend enough time on the project. Would you expect these two students to respond to the poor grade in the same way? In both cases the event was the same but different inferences about the cause of the event would likely lead to very different responses.
The nature of this process is inventive within the parameters of cosmology and culture.
This process deals with projecting events based upon a body of information. One might project in a future tense, a sort of trend analysis, or one might look for an historical precedent to a current circumstance. In either case, the prediction emerges for a data base rather than being just a guess. A guess is not a prediction. By definition, predictions must also be testable. This means that predictions are accepted or rejected based upon observed criteria. If they are not testable they are not predictions.
It is not unusual to find that a data base is not available for a particular system. In such cases predictions about that system are not possible. The first step in understanding such a mystery system would be to observe it as objectively as possible with the goal being to acquire the data base necessary to develop predictions.
The nature of the skill of predicting is to be able to identify a trend in a body of data and then to project that trend in a way that can be tested.
The process skill of relationships deals with the interaction of variables. This interaction can be thought of as a kind of influence--counter influence occurring among a system's variables.
Relationships can occur in multiple or single dimensions. An example of a multiple dimension relationship is speed with distance and time representing the two dimensions. Single dimension relationships can only be expressed relative to something else as in the location in space of some object. It's location can only be expressed with relative terms such as over, under, near, far, etc.
Of course the notion of relationships can be extended into more abstract areas such as values, friendships, marriage, love, and growth, for examples.
The inherent nature of this skill is that it requires analytical thought in which one seeks to dissect cause from effect. The causal elements are the system's variables and the effect is the resulting interaction.
This process actually refers to a group of skills, all of which represent some form of systematic reporting of data. The most common examples include data display tables, charts and graphs. The process is conceptually fairly simple and is frequently based upon some type of two or three dimensional matrix with the axes representing the system variables and the cells of the matrix representing the interactions.
The purpose of the communication skills is to represent information in such a way that the maximum amount of data can be reviewed with an eye toward discovering inherent patterns of association.
The inherent nature of this process skill involves the ability to see and, consequently, represent information as the interplay among influencing variables.
9. Interpreting data
This process refers to the intrinsic ability to recognize patterns and associations within bodies of data. Obviously there is a direct contribution of the previous process, communication, to interpreting data. The better the data is represented the more likely one will detect associations within the data.
Interpretation probably requires creative thinking that results in the invention of conceptual umbrellas that can encompass the data.
10. Controlling variables
This process is also a kind of group process because one may engage in several different behaviors in an attempt to control variables. In general, this skill is any attempt to isolate a single influent of a system so that it's role can be inferred. The process is an attempt to achieve a circumstance or condition in which the impact of one variable is clearly exposed. The use of experimental and control circumstances, standardizing procedures and repeated measures are only a few of the ways in which variables might be controlled.
Understanding the nature of the skill requires analytical thinking in which the system under study can be reduced to a set of interacting components. The next step is to establish some circumstance that allows the scientist to observe one component in isolation.
11. Operational definitions
An operational definition is one that is made in measurable, or observable terms. An operational definition should not require interpretation of meaning nor is it relative. The meaning of the defined term must be explicit and limited to the parameters established for the definition.
An operational definition is primarily a research tool and related to the concern for controlling variables. The major function of operational definitions is to establish the parameters of an investigation or conclusion in an attempt to gain a higher degree of objectivity.
Consider this example. An investigator suggests that by applying some treatment a class of students will become more intelligent. The problem here lies with the word intelligent. What does it mean? And, more to the point, what does the investigator mean with the word? In order to evaluate the treatment intelligence must be defined in a very clear way. Perhaps, in this case, defining intelligence as a score on an IQ test makes sense. Such a definition (intelligence = IQ score) would be an excellent example of an operational definition.
In terms of the nature of the skill, we are again dealing with analytical issues. An individual who is skillful a making operational definitions is one who can engage in reductionistic thinking that defines phenomena as a collection of components which interact.
Hypothesizing is, again, an intrinsic and creative mental process rather than a more straight forward and obvious behavior. Consequently, developing this ability is probably less a product of linear training but more a function of intuitive thinking that emerges from experience.
Defined, an hypothesis in a response, or potential solution, to a specific research question, or problem. For our purposes we will insist upon a rather rigid use of the term and will restrict it to the second step in the classical scientific algorithm as outlined in the next process.
The kind of hypothesis one produces is also heavily dependent upon one's world view. For instance consider the individual whose world view is based upon anthropomorphic and supernatural beliefs. This person is likely to develop anthropomorphic and supernatural hypotheses in response to questions so disasters become a function of angered gods and good times result from happier gods. A result of western science has been to replace the supernatural worldview with one steeped in the physics of Newton and the philosophy of Descartes. This has lead to an industrial age cosmology characterized by cause and effect and the separateness of the observed from the observer. Therefore current explanations (or hypotheses) are more likely to take the form of a causal chain forged link by link by observations which seem to lead inevitably to a conclusion.
The nature of the skill is to recognize that objectively gathered observations are justified into an explanation as a result of having an operational cosmology, or worldview. Secondly, a good hypothesizer recognizes that explanations are inventions rather than discoveries and subject to rejection based upon facts. Beyond this no one is really sure how hypotheses are actually generated. No one really knows what goes on in the mind that results in the hypothesis but it seems reasonable to suspect that information, perceptions, and ideas are being combined and recombined until a particular combination seems to make sense.
This process is a systematic approach to solving a problem. Usually experimenting is synonymous with the algorithm called scientific method which follows these five basic steps:
PROBLEM---->HYPOTHESIS---->PREDICTIONS---->TEST OF PREDICTIONS---->EVALUATION OF HYPOTHESIS
In experimentation each step emerges from the previous one. The purpose of the process is to judge the extent to which an hypothesis might be true and to set a standard whereby that judgement is made. Consequently, scientists tend to think in terms of probabilities of truth rather than absolute correctness.
As a term, experimenting is frequently used in a much broader way than described here. It is not unusual to hear teachers applying the term to any activity or demonstration but, strictly speaking, experimentation should be reserved for the process of systematically evaluating hypotheses.
© Copyright 1997 Jerry Jinks