Gambar halaman
PDF
ePub

PROVIDENCIA: A Case Study in Economic Censuses

calculating personnel requirements during the data collection phase, because there are marked differences in the time required to enumerate urban vs. rural areas. Since the interview time for the long forms is considerably longer than that for the short forms, separate estimates of the number of each type of form is also important for the calculation of personnel requirements.

2.4 Other factors

These are only a few of the data items needed to estimate personnel requirements accurately. Since transportation is an important item of cost, the kinds of roads and the availability and cost of various types of transportation (rail, bus, boat, private car, etc.) should be known for each province. In the Providencia Industrial Census, the questionnaires were left with the manager, owner, or other person in charge of the establishment in most cases, and the question arises as to whether the completed questionnaire should be mailed to the field office or to the enumerator directly, or held until picked up on another personal visit. Mail is, of course, less costly but may be considered unacceptable for other reasons.

Much of the relevant information may be brought out by the maps described in chapter II-3, but consideration should be given to the consolidation by province of the data needed in estimating personnel requirements. Such summary information is useful to management in trying to forecast the overall needs for manpower.

3. ESTIMATING MAN-WEEK REQUIREMENTS Exhibits II-5-5 to II-5-8 provide the means to estimate personnel requirements in terms of number of man-weeks and, in the case of exhibit II-5-7, the equivalent number of

full-time positions that are needed in each phase of the census.

3.1 Travel and completed interviews

In exhibit II-5-5, there are presented estimates of the number of kilometers of travel required in specific situations-mining forms, manufacturing forms, urban vs. rural, and household interviews. These were based on NSO's travel experience in the 1970 Census of Population and Housing. They are useful in this study because the number of kilometers travelled has an important bearing on the number of interviews expected to be completed per week. Note, for example, that in rural areas the long form manufacturing establishments are expected to be more scattered than are the mines or oil wells-about 33 kilometers of travel per completed mining form, as against 50 for the manufacturing long form. These figures may seem high but it should not be forgotten that they include allowances for check interviews, revisits to pick up the forms or to resolve editing problems, and any other personal contact with respondents. In urban areas, as expected, the travel requirements are considerably less.

Such travel estimates are useful not only in computing the man-week requirements, but in computing the cost of travel for the budget (chapter II-4).

3.2 Total man-weeks required

In the first two columns of exhibit II-5-5, there are shown the number of establishments to represent the physical workload and the estimated man-week requirements for each type of interview. In the case of the mining long form, for example, there are an estimated 1,200 forms to be collected at a rate of 12 per week, or a total 100 man-weeks. For

urban areas, there is no differentiation between manufacturing and mining forms, since there are relatively few mines in urban areas. In the case of households, the estimates are based entirely on the screening of interviews (during which eligibility for an in-scope questionnaire will be determined) because the number of "good" questionnaires is expected to be very small.

3.3 Operations calendar

Exhibit II-5-6 consists of a list of industrial census operations, consistent with that shown in exhibit II-4-2 where the NSO Division responsible for each operational step is given. It also shows the time period when the operational step was to be performed. Exhibit II-5-7 presents the personnel requirements in terms of man-weeks and number of equivalent full-time persons. In the data collection phase, the number of man-weeks is consistent with the estimates shown in exhibit II-5-5, while those under data processing and publication are new and are not presented elsewhere in the Providencia Case Study. They are consistent with the dollar budget estimates shown in exhibit II-4-3. Exhibit II-5-7 could be expanded to include two more columns showing the numbers of man-weeks and equivalent full-time employees available.

3.4 Man-weeks required, by quarter

Exhibit II-5-8 presents a summary of manweek requirements for each calendar quarter-from the advance planning phase to the publication of the final reports. It is a useful type of summary because it identifies possible problems; it was evident, for instance, that regular employees of NSO could be expected to perform all operational steps except possibly

those which took place during the first and second quarters of 1976. Requirements during these critical periods were analyzed, as described below.

In the

Each operational step in exhibit II-5-6 and exhibit II-5-7 was re-examined in an effort to determine whether permanent employees of NSO would be able to fulfill the indicated task, or whether other individuals would be needed. In the case of data collection, which represented the largest item of cost, the county agents could perform most of the work. The item "FOD regular employee" would, however, require 1,000 man-weeks. assumed 10-week collection period, this meant that at least 100 individuals would be needed. Of these, the provincial field offices could supply the services of 20 full-time equivalent persons (roughly two-thirds of the time of the permanent field staff). Therefore, a recruitment and training program would be necessary, with training for the county agents as well as for FOD employees.

The two items under "Data Processing and Publications" which appeared to require special attention were the editing of long forms (250 man-weeks) and editing of short forms (150 manweeks). Other requirements, it was found, could be met by transferring employees, either full-time or part-time, from other work. In the operation of long form editing, six of the statisticians required were found within MMD, and another six were transferred from the Trade and Services Division. The editing of short forms was taken care of by the permanent staff of DPD (five persons) and by transfer from FOD (four persons). Thus the only recruitment and training required was in the data collection phase.

Chapter 11-6. ADMINISTRATIVE CONTROL SYSTEM

1. INTRODUCTION

Control over expenditures of human energy, as well as money, is of profound importance in any rational undertaking. Although courses of action are sometimes dictated by emotional or other irrational decisions, the existence of a comprehensive and well-thought-out plan may tip the scales in favor of sounder approaches to problems. The NSO had such a plan for Providencia's 1975 Industrial Census, which contributed greatly to its success.

1.1 NSO control system

In order to maintain progress and meet deadlines each operational step must be supervised closely, both by the individuals directly responsible for the operation, and by top management. Detailed records are necessary, and frequent progress reports must be circulated among the supervisory personnel. The maintenance of such records, the preparation and distribution of such progress reports, and the actions taken by supervisors as a result, constitute the NSO administrative control system described in this chapter.

This system was not developed exclusively for the industrial census. It is a generalpurpose system which has been employed by NSO for several years, although the details are usually adapted to specific projects rather than used without change. At the heart of the system are detailed records reflecting the location and movement of census materials, and the expenditure of funds. In general, records are kept for each operational step named in Form 120, which is entitled Time and Cost Schedule (exhibit II-4-1). The individuals in charge of

maintaining the detailed records prepare progress reports, which are forwarded to the Control Staff of the Administrative Services Division (ADMS), with copies to supervisory personnel as needed. The ADMS Control Staff does not actually exercise control but provides the data needed by management personnel to supervise and control the progress and flow of work, to control expenditures, and to keep all features of the project consistent with its time and cost schedule. Consolidated progress reports are prepared by the ADMS Control Staff (daily during crucial periods, weekly or monthly at other times) which are circulated among members of the Executive Committee and others concerned with the status of the project.

The system reports on the status and movement of census materials and the expenditure of funds at all stages. It permits management personnel to determine the location of materials, to oversee the progress of the work, to make decisions regarding priorities and the flow of materials to each operation, to identify trouble spots where special attention may be required, and to observe and regulate the outflow of expenditures in comparison with the budget and with the available funds.

2. APPLICATION OF THE SYSTEM TO THE
INDUSTRIAL CENSUS

The Providencia Industrial Census has been conducted at intervals of at least 10 years, and is consequently regarded as the equivalent of a one-time operation. For such operations, control systems are not as elaborate as those required for monthly surveys, where materials are constantly flowing to and from the field, and where

detailed accounts of enumerator performance are used to build up and maintain a more-orless permanent field staff. Furthermore, the Providencia Industrial Census, although more complex in its data content than a population census, is a much smaller operation requiring fewer controls. The Census of Population and Housing for 1970 employed 15,000 individuals at the peak, and the 1975 Industrial Census only 300. Fifteen million individuals and three million households were enumerated in 1970, compared with 70,000 reporting units in the 1975 project, with less than 15,000 of them on the relatively difficult long form.

The top management of NSO is aware from experience that great restraint is necessary in adapting the control system to individual projects. Otherwise, over-enthusiasm about the importance of keeping up the flow of information may lead to the maintenance of unnecessary records, and supervisors may spend so much time on records and progress reports that the flow of work suffers. To eliminate unnecessary record-keeping and to adapt the system to the industrial census, the Executive Committee provided guidelines to each NSO Division to indicate which records were to be kept.

In general, those operational steps involving few questionnaires or other physical units, and those requiring only a short time for completion, need only a few simple records. In the industrial census, the entire canvass of multi-unit enterprises, single-unit establishments with more than 100 persons engaged, government owned-and-operated plants, and electricity and gas plants, was conducted by the experts in the Manufacturing and Mining Division. There were less than 500 reporting units in total and, except for a simple checkin file to control receipts, the only written records were the regular progress reports to the Control Staff of ADMS.

It should be borne in mind that, in terms of economic importance and consequent effect on the published results of the census, these establishments constituted by far the most important group included in the census. They accounted for about one-third of the total industrial output (value added).

2.1

Determination of current status of work

An important use of the control system was to enable management personnel to determine the current status of work in any operation at any time. Records of work that were moving in and out of an operation (such as records of long forms passing through the editing process) were so maintained that reports on status could be readily compiled when they were needed. For the long form editing operation, which took 5 months altogether, the regular progress reports were prepared weekly. Other summary reports, such as the reports of questionnaires received and edited by province, were prepared only once a month. Special reports related to quality of returns, answers to specific inquiries, etc., were prepared occasionally on request from the Director or from the head of the Manufacturing and Mining Division.

During most of 1976, which was a critical year for collection, processing, and publication, the industrial census was an item on the agenda for the regular weekly meetings of the Executive Committee. Summary progress reports, as up to date as possible, were circulated in advance to those expected to attend the meeting. The heads of the divisions which were directly concerned with the census not only answered questions from other staff members, but reported to the group orally on the current status of the project. the end of July, as the bulk of the processing neared completion, discussions of the industrial census became shorter and less frequent.

Toward

2.2 Measurement of costs; comparison with budget

Important features of the summary progress reports were the current and cumulative expenditures for the census, together with similar figures projected from the budget. A frequent occasion for investigation by management personnel was when any of the expenditure figures differed from the budget by excessive amounts. During the course of conducting the census, it became evident that there were some poor estimates in the budget, and these were amended as necessary. Since it was not possible at this stage to change the total allocation of funds for the census, it was necessary to "transfer" funds from one budget account to another in order to revise the budget. Until it was exhausted, the "not specified" fund was the source of such transfers; afterward, transfers were made from other NSO accounts.

Payroll charges and other payments for personal services assignable to the industrial census were reported by the Personnel Division to the ADMS Control Staff for inclusion in the progress reports. Special reports from the Ministry of Agriculture accounted for a proportion of the salaries of the county agents during the time when they were active. ADMS Control Staff also received special reports on the proportion of time spent on the industrial census by the few employees who worked on it part-time. The salaries of heads of divisions and other top management personnel were not allocated but were included in the progress reports in the form of an overhead charge.

Non-payroll costs, such as the cost of paper, printing, and supplies, were named specifically in the progress reports where they could be directly allocated to the industrial census. Other costs of this kind were presumed to be included in the overhead charge.

2.3 Preservation of records

Many of the basic records, and all of the summary progress reports, were preserved for their historical value. These documents will provide very useful data for preparing the budget for the next industrial census, and will assist in planning that census. Since these materials consist of a running record of the census, and reflect its failures as well as its successes, they will provide an excellent starting point for a discussion of improvements.

2.4 Control of printed material

One of the first tasks of the ADMS Control Staff in the fall of 1975 and early in 1976 was the preparation of progress reports showing the flow of printed material. These reports showed, for example, the transfer of material from PUB to FOD, from FOD to the provincial field offices, etc. FOD was responsible for seeing that supplies of printed material, consisting of maps, training material, pocket manuals, blank questionnaires, progress report forms, enumerator's records, etc., reached the provincial field offices well in advance of the beginning of enumeration on 1 February 1976. From the field offices, the materials were further distributed to enumerators, either directly or through the county agents. Except for Estrella city, all operations were conducted directly or through the county agents, who acted not only as enumerators but occasionally as supervisors. In Estrella, all operations were conducted through the provincial field offices and the field supervisors who were in charge of the enumerators.

The distribution of all of the material was under the direction of one person in FOD, who maintained simple records showing the date of shipment, kind of material, number of copies, methods of transportation, and destination. similar record was maintained in each field

A

« SebelumnyaLanjutkan »