Comparative Study

En Español  In Finnish   Contents

The comparative method is often used in the early stages of the development of a branch of science. It can help the researcher to ascend from the initial level of exploratory case studies to a more advanced level of general theoretical models, invariances, such as causality or evolution.

Comparative StudyThe design of comparative research is simple. Your objects are specimens or cases which are similar in some respects (otherwise, it would not be meaningful to compare them) but they differ in some respects. These differences become the focus of examination. The goal is to find out why the cases are different: to reveal the general underlying structure which generates or allows such a variation.

Comparation is one of the most efficient methods for explicating or utilizing tacit knowledge or tacit attitudes. This can be done, for example, by showing in parallel two slides of two slightly different objects or situations and by asking people to explain verbally their differences.

The method is also versatile: you can use it in detail work as a complement to other methods, or the entire structure of a research project can consist of the comparison of just a few cases.

  Observed
state of things
Case 1 Case 2
Aspect A A1 A2
Aspect B B1 B2
Aspect C C1 C2
In comparative study, you are examining two (or more) cases, specimens or events, often in the form of a table such as can be seen on the right where a column is reserved for each case, here called "Case 1" and "Case 2". On the basis of the target of your study you have to decide which are the interesting aspects, properties or attributes that you will have to note and record for each of the cases. In the table on the right, these aspects are called A, B and C. During the process of analysis, you then can add new aspects or drop out fruitless ones. Those aspects that are similar in both the cases need not be recorded, because here you are not making two case studies but only a comparison of the cases.

The final goal of research is usually to reveal the systematic structure, invariance, that is true not only for the cases that were studied, but for the entire group (population) where the cases came from. In other words, the goal is to generalize the findings. Of course, it would be foolhardy to assert anything about a larger group, if your study consisted of just two cases. The plausibility of your generalisation will increase, if you have instead of "Case 1", several cases from the same group, let us call it "Group 1", and similarly several cases from "Group 2". If all or the majority of these pairs show the same invariance, its credibility will quickly rise. There are statistical methods to calculate the credibility, or statistical significance of the findings. The question whether the found invariance then is true even outside the population, is something that the researcher normally leaves to be speculated by the readers of his report.

In the case that you wish to compare more than two groups, or the number of cases is large, the study begins to approach classification, a method that is discussed on another page.

In comparative like in most other studies there are two different styles, both of which will be discussed below:

Descriptive Comparison

In descriptive study of products there are many situations where comparison is an adequate method. You could, for example, study comparable products which have been designed by different designers or made by different producers. Or you can study the same type of products as they are used in the same circumstances but in different countries.

Steam irons An example of comparison can be found in the study Products as Representations, by Susann Vihma. She examined metaphors of domestic equipment. Among her study objects were twelve steam irons, five of which are presented in the photo above. She found out that when studying each specimen in separation it was not easy to grasp its symbolic message; it became easier when the object was studied together with other similar objects or when two objects could be compared to each other.

Comparison may be useful even when the researcher is not interested in differences but in a single case. If the interesting object belongs to the researcher's own cultural environment, it is not always easy to perceive its special characteristics. The case may appear too obvious and non-problematic. "A fish cannot see that it is living in water." One method to reveal the specific nature of a too well known object is to compare it to other cases or specimens from another context.

In exploratory study it often happens that you need gradually add new aspects of comparison, or have to redefine them when your knowledge of the object increases. It is also common that in the initial phases of the study you only can reach descriptive answers to the question what the object is and what it is like, Another, more difficult task then is to explain or answer the question why the object is as it is.

  Case 1 Case 2
Potential reason + --
Potential effect + --
In comparative analysis you can apply all the usual types of explanation: by earlier events, by later events, and contextual explanation. It can be useful to make a table, like the one on the right, of potential reasons and potential effects. If there is conformity between likely reason and likely effect (i.e. there is effect only when the reason is present) it augments the plausibility of the hypothetical explanation. However, a mere correlation between two variables does not yet definitively confirm a hypothetical explanation, because the correlation can be due to other reasons which have not been registered, see a list of possible explanations.

It can be difficult to discover all potential causal influences in empirical study only, therefore it is usually advisable to start by doing a thorough study of literature for finding theory and data of comparable cases.

Field work tends to entail, in spite of its usually good validity, often mediocre reliability of the findings because of disturbances that obstruct discovering those relationships that the researcher would want to study. If such is the case, you should consider complementing the comparison with other methods like interview (if people are mixed up in the activity to be studied) or an experiment with appropriate shielding to keep out any disturbing influences.

Another usual technique for reducing not desirable influences on the object of study is to select the cases to be compared so that they are as similar as possible. For example, if you want to compare a case in your home town to another similar case, you should select the latter from another nearby town of the same size.

Normative Comparison

The difference between descriptive and normative styles of comparison is that in normative analysis one of the principal criteria is evaluative like "satisfaction", "usefulness" etc., and the aim of the study is to point out the best (in this respect) among the alternatives that are being studied. The final aim perhaps is not only to find the best, but also to improve it or similar objects later on. The principles and methods of normative comparison are explained on a separate page.

En Español  In Finnish   Contents

August 3, 2007.
Comments to the author:

Original location: http://www2.uiah.fi/projects/metodi