3. ADVANTAGES AND DISADVANTAGES OF THE ESSAY METHOD

Interviews with children are not easy, because the pupil at an ordinary elementary school looks at any adult asking questions in the school environment as some person of authority; the child therefore tries to give the answers which he thinks are expected, rather than his own views. Interviews at home are out of the question, as it would be impossible in the majority of cases to meet the child alone.

If the child writes his opinions in an essay, this effect of shyness can be avoided. Children also do not see anything extraordinary in an essay, as it is part of their curriculum. It may therefore be assumed that they will express a fairly frank opinion. To judge from the large amount of criticisms of school meals which was found, in the essays, this assumption was correct.

Another advantages of this method is that children are trained to give the pros and cons for any judgment they make in their essays. In an investigation like this, where the purpose is not only to find out momentary attitudes, but also what opposing motivations lead to an attitude, this attempt by the children to see and express both sides of the picture is most useful.

On the whole, the results which were obtained by the analysis of the essays were interesting, and will answer a number of the questions on which the Department wanted information. However, the method suffered from one great handicap; it was impossible to make a statistical analysis showing the significance of individual attitudes. A number of teachers found it necessary to give their children sub-headings to work by. As can well be imagined, these headings varied to a certain extent from teacher to teacher, and therefore in certain essays groups of problems are mentioned which do not appear in others.

The statistical results of this inquiry therefore, must not be taken as final. It cannot show the numerical strength of the different motives and attitudes.

Because of the difficulty of statistical analysis, it was extremely hard to compare Mr. LeGros Clark’s results with our own. Qualitatively, both investigations give similar results, but no quantitative comparison is possible.

At the end of the investigation we came to the conclusion that essays can be a most useful tool in social investigations if they are conducted in the following way:

the teacher must give the children the title, together with a number of sub-headings; these sub-headings must be the same for the whole sample, and must be introduced by all the teachers with exactly the same words. It must also be impressed on the teacher that under no circumstances must there be any discussion on the subject beforehand If these two simple rules are observed, statistical, analysis will become possible.