Wednesday, August 15. 2007
(Fifth and last of the Series)
By ATTY. MAQTAHAR L. MANULON
To ensure good quality census results, a number of strategies are enforced before, during and after enumeration. Training of census workers, i.e. enumerators and census area supervisors and team supervisors is among these strategies that give confidence to our workers because they are taught the various concepts and procedures of conducting the census.Â
During enumeration, close supervision of the ENâ€™s work and regular review of the accomplished questionnaires are basic strategies that assure good quality data. Other than the designated supervisors from among the ranks of the census workers there are also regular NSO personnel from provincial, regional and central offices who do spotchecking and field supervision. These supervisors were given additional training owing to the nature of their duties and responsibilities of handling people and preparing reports. They are required to observe the ENs during interview, reinterview households already interviewed and to correct errors there and then. There are claims in the past although not proven that census enumerators doctored their data under the mango tree. The NSO condemns this act of enumerator and is asking the public to report if there isÂ any to NSO office or itsÂ personnel. In Region 9, the NSO has deployed about 1800 enumerators and supervisors to cover the entire area for the whole month of August.
After enumeration, processing of the questionnaires comes in. This is composed of manual and machine processing at provincial, regional and central offices of NSO. A number of processors will be hired to do the work. They will also be given training to ensure correct editing and consistency checking of the various questionnaires. The manual processors check for correct geographic identification, review of entries for completeness of questionnaires, consistency and acceptability of responses, coding of selected items, document preparation, and bundling/packaging of forms for transmission. In some instances field validation is done to verify unacceptable count or incomplete data.
Machine Processing covers (a) Automated receipt and control at various data processing centers; (b) Data encoding of CP Forms 5 and 7 at PPS; (c) Scanning of accomplished questionnaires at RDPC 2007; (d) Processing of image files at CDPC 2007 that includes interpretation, validation, verification, completeness checking of data items, file preparation, and back up data files; (e) Post data capture processing at the CO that includes consistency checking, imputation, tabulation of results, and creation of Public Use File (PUF).
At regional office, all CP Forms 1, 2, 4, 5, 7 and barangay/EA/block maps shall be scanned. This process captures images of all documents into digital form. All image files of CP Forms 2 and 4 shall be subjected to qualify assurance, stored in optical media, and transmitted to CDPC 2007 for further processing. Scanning of CP Forms 4 will be done only after all CP Forms 2 have been processed. Moreover, CP Forms 1, 5, 7, and barangay/EA/block maps shall also be scanned after all CP Forms 2 and 4 have been processed for archiving purposes.
Documents will be scanned using mid-volume scanner producing image files named after their corresponding serial numbers.
Some data items in the images of CP Forms 2 and 4 shall be subjected to interpretation using ICR software to determine if they can be processed at CDPC 2007. Questionnaires that failed ICR interpretation shall be re-scanned until the images are identified by ICR.
Image File will be copied into CD/DVD (Compact Disc) for transmittal and back-up purposes. One copy of CD/DVD shall be sent to CDPC 2007 for further processing and another as a back-up copy at the RDPC 2007.
At Central Data Processing Center in Manila, when scanning of CP Forms 2 for a city/municipality and CP Forms 4 for a province has been completed, the Regional Data Processing Center shall send the image files to the CDPC 2007. To facilitate the management and control of all image files, TRACS (CO version) shall be used. This version of TRACS shall automatically record the number of CDs received from RDPC 2007, the assignment of EAs to data encoders, and perform other transactions. The system is also capable of generating the required operational and management reports.
Data capture using ICR software will be done in two passes. First Pass â€“ Interpretation of all Optical Mark Recognition (OMR) data items with 100 percent verification for geographic and household ID. Verification means that data items will be encoded and the keyed data will be compared to the corresponding interpreted figure. If a mismatched occurred, data encoder would be prompted to re-type the correct entry. The output data files in this phase will be used in evaluating the final counts, i.e., population counts by sex by barangay. Second Pass â€“ all data items will be interpreted with 100 percent verification for write-in entries except for geographic and household IDs which have already been captured in the first pass.
Resulting data files shall be subjected to key sample verification using key-form-image program. This is to check the correctness and completeness of the work of the data encoders. Once the data file for a province passed the editing phase, that is, the data are free of errors; detailed characteristics of the population shall be generated and evaluated at HSD. These tables shall then be transmitted to the ROs and POs for further evaluation before IRD generates the final tables for publication.
During machine processing, the NSO uses the image reader (scanner) stationed in each of its regional offices nationwide. While the cost of the machine is high, it shortens the processing period and make the release of the results much earlier.
The time table to release the population count down to barangay level is February 2008. This is done after the proclamation has been made by the President of the Philippines. â€” Atty. Manulon is the regional director of NSO-9