October 18, 2016: Recent reports in the national media have suggested that four officers of the Central Statistical Office (CSO) were somehow able to manipulate the Labour Force statistics which were disseminated to the public of Trinidad and Tobago. In accordance with Principle four of the United Nations Fundamental Principles of Official Statistics (UNFPOS), it has become necessary for the Acting Director of Statistics to publicly clarify this troublesome matter.
Firstly I wish to unreservedly assure all stakeholders that Official Statistics disseminated by the CSO are collected and processed in accordance with strict international statistical standards which include rigorous and robust quality assurance procedures. In light of the CSO’s adherence to the UNFPOS in concert with constant peer review from our international development partners, users of Official Statistics are guaranteed a high level of accuracy.
The management of the CSO has determined that they were derelict in their duties. The four officers are Temporary Assistant Field Interviewers and are accused of not visiting enumeration districts and households at the time periods prescribed by the sampling design of the Continuous Sample Survey of Population (CSSP). It is alleged that the officers rather used information from past records to complete questionnaires and listing records instead of conducting contemporaneous interviews.
National Statistical Offices (NSOs) and other research organisations all over the world encounter problems where field officers do not actually visit the target households of the respective surveys, but fill out questionnaires based on old data (previous interviews). This is why practitioners of national statistical reporting globally, employ comprehensive quality assurance procedures to discover and correct errant data. At the CSO, when completed questionnaires are returned to the office by field officers, these documents are carefully checked by trained Statistical Editors so as to identify errors of consistency. Responses from households that are not pre-coded are then coded using International or National Classifications by specially, highly trained staff.
Edited and Coded questionnaires are then sent to the CSO’s Computer Division for electronic data capture. Electronic data are then edited and verified by specialised data validation software. Data are then returned to the CSSP section of the CSO where the data are subject to further error checking. When a ‘clean’ (error-free) Data file is established, the information is then inserted into formatted tables which are either used in preparing the Annual Labour Force Report or the Quarterly Labour Force Bulletin.
During the course of these routine operations the suspicious work of the four Temporary Assistant Field Officers were discovered. Senior field staff were then dispatched to the relevant enumeration districts and the households where the updated labour data were collected. The offending officers were reported to the Permanent Secretary of the Ministry of Planning and Development and officials in the ministry are now investigating the matter for necessary action. The accuracy of the published data was therefore in no way affected, however sadly the timeliness of the release of such data was negatively impacted. It must be emphasised that four junior officers such as Temporary Assistant Field Interviewers cannot influence data on a national scale.
From an International perspective, data are now being captured on the field via tablets equipped with GPS and audio visual recording to verify that field officers conduct interviews where and when they are supposed to. It should be noted that the CSO field staff are now being trained to use such devices which have already been purchased. When these tablets come into use, the progress of each field officer will be monitored by the respective supervisors remotely. The implementation of this measure is at present being supervised by experienced research consultants; KAIRI.
Moving forward, the management and staff of the CSO have long recognised the need for substantial improvements in the efficacy of national statistical reporting in Trinidad and Tobago. The CSO has acknowledged that a number of its data sets are not provided in a timely fashion and institutional restructuring is needed. Further, the National Statistical System is in need of greater coordination and strengthening.
By way of example the CSO is currently obligated to provide estimates of the Gross Domestic Product (GDP) to the Ministry of Finance in early August of each calendar year. These estimates are based on data from the first six months of the respective years and are therefore axiomatically subject to substantial review when actual (full year) data are subsequently collected. These provisional data therefore are not robust enough to meet the standards of Official Statistics. Accordingly the IMF has recommended that the CSO discontinue this practice and instead provide robust, quarterly GDP figures and leave annual GDP estimates to the policy makers. However this can only be achieved if the CSO is allowed to use data from the Vat Office as control data, at present legislative barriers prevent this.
The Cabinet appointed Task Force for the Implementation of the National Statistical Institute (NSI) is tasked with the organisational transformation of the CSO into an independent, transparent and well governed institute. Beyond the institutional changes the Task Force is also responsible for redesigning the National Statistical System (NSS) by inter alia drafting modern statistical legislation to empower the new NSI to provide improved statistical products which the CSO did not have the authority to deliver. The Management of the CSO has been actively supporting the efforts of the Task Force and looks forward to the new and improved NSI and NSS in 2017.
Sean O’Brien
Director of Statistics (Ag.)