The most important considerations in gathering data are completeness and accuracy. Incomplete ness of data may be in the kinds of facts gathered or in the number of cases considered. That is, certain facts which would help the concern form its conclusions are not considered, or a generalization is made before sufficient cases have been considered to make sure of the application of the law of averages. Inaccuracies may occur in observation of the facts, in their tabula tion, or in their classification. The receiver of the report will find it valuable to possess a knowledge of the sources of information on which reports are based, and of tests that may be used to check the accuracy of that information. Such information of the sources of data and tests for accuracy is often included in the report, usually in the introduction, or in the form of foot-notes in connection with each unit of data the accuracy of which might be questioned.
3. Preliminary outline.—Careful analysis of the problem involved in the report is best made in the form of a logical outline. This outline virtually con stitutes a table of contents of the report. The outline form is often drawn up before the information neces sary to fill it in is obtained. It is the bird's-eye-view of the report, which indicates clearly the broadest divisions into which the report itself is divided. The subdivision of the report into all its details requires; of course, keen appreciation of the purpose of the re port.
If an official of a company were asked to prepare for the president Or a board of directors in any busi ness a report designed to throw as much light as pos sible on the problem of improving correspondence thruout the organization, his preliminary outline might contain the following broad divisions: I. Present Conditions II. Possible Improvements III. Best Means of Obtaining Improvements.
What are present conditions? is, logically, the first question to be answered. Even tho the reporter knows a great deal about present conditions, he must make sure that he knows them well enough to criti cize them constructively. Thoro acquaintance with present conditions is the first step, and it is a very important one. General impressions concerning con ditions are not sufficient; they must be verified. The reporter must do as little guessing as possible about actual conditions. He must look for excellences as well as faults in the present system. When making an investigation he usually finds what he looks for. Unless lie looks for good points as well as bad points, and for the causes of both good and bad conditions, a one-sided view of the case results and improve ments cannot be so well planned. Improvements must be based upon whatever is good as well as bad in the system that is being investigated.
The questionnaire method is a good means of get ting information about present conditions. In this case one sets down on paper a set of pertinent ques tions to which written answers are to be given by members of the organization who are best qualified to answer or he can have personal conferences with these members. In general, the less the report-maker is dependent on others for his information, the better, because "others" are seldom free from a desire to make a good showing, and because these same "others" are often the originators of faults in the present sys tem.
Given the broad, general divisions of a report, as above suggested, the construction of the permanent outline then becomes a process of logical analysis or classification. It is very important that the report be unified and that the proper proportion be observed —that the less important parts be properly sub ordinated and that the more important parts be given due prominence.
4. Analyzing the as the purpose of the entire report serves to suggest the broad divisions, so the purpose of each main division suggests its sub divisions. Present conditions must be thoroly known in order to make any changes for the better. In the case under discussion, improvement lies in two general directions : lowering the cost of handling the corre spondence, and increasing the effectiveness of the letters. Immediately certain pertinent questions about present conditions are suggested: What is the cost of handling the correspondence? How does this cost compare with that in other businesses? How does the cost of correspondence in one department of this business compare with the cost in other depart ments? What are the main items of cost in the han dling of the correspondence? How do these items of cost compare with the cost of similar items in other departments and in other business houses? And thus, by the question method, the problem of cost is an alyzed with a view to finding out whether it may be lowered, and if so, where and how. Then questions of a similar nature are asked concerning the effective ness of the correspondence: Are the letters effec tive or not? Why, or why not? On what tests does the answer rest? What are the results of the various types of letters written in the different departments? How do these results compare with those of similar letters written by other concerns? If there is any difference, what is the explanation? What, in gen eral, constitutes an effective letter? What are the most important considerations in effective letter-writ ing? What is the correspondence department's weakness and what is its strength? Such questions would be the basis of the investiga tion of conditions, and these questions suggest the main subdivisions of the report, which might be some what like the following: I. Present conditions A. General statement of the correspondence methods B. General statement of cost C. Effectiveness analyzed 1. Favorable criticisms a. General b. Types of letters c. Departments 2. Unfavorable criticisms a. General b. Types of letters c. Departments D. Cost analyzed I. Where the cost is low a. General b. Types of letters c. Departments 2. Where the cost is high a. General b. Types of letters c. Departments II. Improvements suggested A. General improvements B. Departments, or types of letters III. "Best Means of Obtaining Improvements" would then consist of definite recommendations of the action considered necessary to effect the improvements sug gested in Part II.