69805 v. 3 Technical Assistance on Poverty Analysis and Social Statistics Mission on the Census of Buildings and Dwellings March 21 – April 19, 2009 Questionnaire Design for the Census of Buildings and Dwellings - Draft - Consultant Report - Document III May 5, 2009 Contents Introduction 1. The option of obtaining socio-demographic data from the census of buildings and dwellings by addition of questions on buildings and/or dwelling units 2. Background on the questionnaire for the census of buildings and dwellings, based on the previous experiences in Lebanon 2.1 Lack of a national address system 2.2 Information on establishments for national accounts 2.3 Lack of legal basis for population censuses 3. Looking to the future 4. Implications for questionnaire development 5. Developing the questionnaire for the next census of buildings and dwellings 5.1 Questions and Definitions 5.2 Units of enumeration in population and housing censuses 5.3 Units of enumeration in the census of buildings and dwellings 6. Design issues for the questionnaire of the next census of buildings and dwellings 6.1 Observations on the Questionnaire of the 2004 census of buildings and dwellings1 6.2 Administrative Geography and issues raised by incomplete cadastral registration 6.3 Item by item commentary on the 2004 questionnaire 7. Developing the Housing content of the Census of Buildings and Dwellings Annex 1. Questionnaire of the Census of Buildings and Dwellings of 2004 Annex 2. Questionnaire of the Census of Buildings and Dwellings-Proposed Revisions Annex 3. Housing census topics by unit of enumeration 1 The following is based on the English translation of the questionnaire provided by CAS to the consultant on the first day of the mission. Questionnaire Design for the Census of Buildings and Dwellings 2 Introduction2. This note addresses development of the questionnaire for the census of buildings and dwellings. It identifies the recommendations related to the design of the questionnaire(s) that will be used to conduct the next census of buildings and dwellings. Several options were considered and discussed with relevant staff from the Central Administration for Statistics (CAS) for the next Census of Buildings and Dwellings, including: (i) census only with additional questions to provide socio-demographic data; (ii) census plus a concurrent (possibly lagged) large (10%) household survey with 10-20 population census-type questions to obtain key socio- demographic data and; (iii) census plus household survey plus minimal LSMS-type survey administered to a small (4,000-5,000) sub-sample of households. The second and third options would probably mean scheduling the census for 2011. The decision of which option will be taken needs to be made very soon, as the Director General of CAS will present the budget for the coming fiscal year (2010) to the Government by the end of May 2009. Depending on which option is adopted, there might be as many as three questionnaires involved: the census of buildings and dwellings questionnaire; the household survey questionnaire, and the LSMS-type questionnaire. Only the two first types of questionnaires are in scope for the time being and this note deals with the questionnaire on the Census. 1. The option of obtaining socio-demographic data from the census of buildings and dwellings by addition of questions on buildings and/or dwelling units The option of obtaining socio-demographic data by addition of questions to the dwelling unit form of the census of buildings and dwellings was examined closely and it was decided that this approach should not be pursued. The reasons for this conclusion are as follows: (i) the census questionnaire on dwelling units contains only three items: Type of residential unit (Principal, Secondary, Vacant, House for guard); Mode of use (Property, Renting, Illegal occupant, Vacant, Other); and Unit area (square meters), (ii) for units in buildings that contain more than one dwelling unit, which are common in urban areas and constitute a majority of all units in the country, it is accepted that the enumerator need not necessarily interview a resident of the unit to obtain this information, that it may be supplied, for example, by the building concierge or a resident of a neighboring unit, (iii) in the 2004 census, enumerators were paid (in Lebanese pounds) 500 for a enumerating a building, 1,000 for enumerating an establishment, and 250 for enumerating a dwelling unit. These figures might need further inquiries and development. Accordingly, there is a strong disincentive for the enumerator to take the time and trouble to interview residents of every unit, particularly in buildings with substantial numbers of units, (iv) the census questionnaire does not identify the respondent who supplied the information on each dwelling unit, so it is impossible to know in how many cases this information was obtained 2 The note is based on the assessments done by Griffith Feeney in the period March 21-April 4, 2009 and on brainstorming meetings held on Wednesday and Thursday, April 08 and 09, 2009 in the office of the Central Administration for Statistics in Beirut. The following persons attended to these meetings: Alissar Nasser (CAS), Ibtissam Jouni (CAS) Nader Keyrouz (CAS), Ziad Abdallah (CAS), Vicken Ashkarian, Lara Badre (CAS), Griffith Feeney (Consultant) and Chadi Bou Habib (World Bank). Other persons involved in the process are Najwa Yacoub, Marleine Bakhos and the Director General of CAS, Dr Maral Tutelian. Questionnaire Design for the Census of Buildings and Dwellings 3 from a resident of the unit, and in how many cases it was obtained from someone not a resident in the unit. (v) based on the facts noted in the preceding three paragraphs, it might be expected that in many cases, interviews were done with concierges and other respondents than residents in urban areas in which the majority of the population live in buildings containing many units. Of course these areas also contain a majority of the population, (vi) CAS staff expressed very strong reservations about inclusion of any questions on the dwelling unit component of the census questionnaire that would require enumerators to interview residents of dwelling units. The concerns were that this would impose a heavy additional burden on the work of the enumerators while at the same time be unlikely to provide information of acceptable quality. Because of the uncertainty over how frequently enumerators in the last census interviewed residents of dwelling units, it is strongly recommended that the next census include an item that identifies the source of the information on “residential units� as either a resident of the unit or some other person. Addition of this question will require very little space on the questionnaire, very little additional data processing work, and enumerators will certainly know the answer to the question. 2. Background on the questionnaire for the census of Buildings and Dwellings, based on the previous experiences in Lebanon The questionnaire for the census of buildings and dwellings is a complex, finely-wrought data collection instrument. The complexity reflects the following three principal influences. 2.1 Lack of a national address system The first influence is the lack of an effective national system of addressing dwelling units, business establishments, and other entities that operate on physical premises. Streets in Lebanon do not necessarily have generally recognized official names. In consequence, dwelling units, business establishments, and other entities do not necessarily have addresses that consist of the name of the street they are on, the number of the building on the street and, in the case of buildings containing multiple entities, an “apartment� or “suite� designation. Given this lack, the most effective way for census field workers to identify the entities that are the principal object of the enumeration is by first identifying and enumerating buildings, physical structures permanently secured to the ground. “Buildings� strictly speaking must be placed in quotes because the term encompasses, in the context of the census, nomad camps and open spaces used for economic activity as well as buildings proper. No effective national system for the addressing of buildings exists either, but buildings, to a greater extent than dwelling units or business establishments, are clearly identifiable by census field workers and can be “addressed� for the purpose of census field work by mapping them. The broad modus operandi of the census is therefore to first enumerate buildings and then to identify dwelling units, business establishments, and other entities of interest within buildings. Questionnaire Design for the Census of Buildings and Dwellings 4 The specifics of this procedure can be read from the sequence of questions on the questionnaire and will be described further below in that context. 2.2 Information on establishments for national accounts The second influence on the design of the census of buildings and dwellings questionnaire was the urgent need perceived in the early 1990s for improving the then existing system of national accounts. It was decided that one element of this task should be, given the conditions existing in the country, to conduct a census of business establishments that would contribute to the construction and maintenance of a business register. This was an important, perhaps the fundamental driving force behind the first census of buildings and dwellings conducted in 1996. The literal translation from Arabic of the name of the census is “the census of buildings and establishments,� though this has tended to be rendered in English translation as “the census of buildings and dwellings.� While the census provides an initial basis for constructing a business register, maintaining the register requires updates from other sources, including, for example, the Ministry of Finance (MoF) tax register and Ministry of Justice (MoJ) business register. Updating faces numerous problems, including little homogeneity of definitions and legal restrictions on access to records. It has proven difficult to merge establishment information from the CAS census of buildings and dwellings with information from other sources. There are two primary problems with the MoF database: the activity classification used until recently did not conform to any international standard, and the coders were not well trained. CAS has been working with MoF for 2 ½ years to bring the MoF activity classification into line with international standards. CAS has trained approximately 100 MoF staff in coding using ISIC Rev. 3.1 (International Standard Industrial Classification of Economic Activities). Implementation of ISIC Rev. 4 does not seem practicable at this time. At the current rate of progress it is anticipated that 2-5 years of work will be required to finish the job. The MoJ database is problematic from the point of view of economic statistics because of the possibility of what might be described as purely “paper� entities that have legally recognized existence but lack significant economic assets and that may or may not be engaged in any economic activity. 2.3 Lack of legal basis for population censuses The third influence on the design of the questionnaire is that the 1979 statistics law and associated administrative rulings under which the Central Administration for Statistics operates do not give it authority to carry out a population census. It is appropriate to call attention to how unusual this situation is. Almost every country in the world has carried out at least one population census, and most countries have carried out several or many. The legal basis for the conduct of population censuses varies from country to country, but there are very few countries that do not authorize some government entity to carry out a population census at regular intervals. Lacking the authority the carry out a population census, CAS took the opportunity of the census to conduct an enumeration of dwelling units to provide an approximation of the kind of Questionnaire Design for the Census of Buildings and Dwellings 5 information provided by a population census. Counts of dwelling units are imperfect as proxies for counts of population, since the numbers of persons occupying a dwelling unit may vary widely, but they are useful. One of the most important uses of periodic population censuses is to provide a sampling frame for carrying out the many different kinds of household surveys that are an essential feature of every modern statistical system. The dwelling unit enumeration aspect of the census of buildings and dwellings provides this sampling frame in Lebanon. It may be noted that the last population census in Lebanon was conducted in 1932 under the French Mandate. Population censuses are typically carried out every ten years. A census taken more than 70 years in the past is useful only for historical studies. 3. Looking to the future Most or all of the information the census of buildings and dwellings provides on establishments and businesses would be more appropriately supplied by a business register. Practices and institutional arrangements vary from country to country, but the generally preferred approach to maintaining a business register is through the records of business entities maintained, usually, by the Ministry of Finance. A census of business establishments is a blunt instrument for this purpose because (i) it covers “establishments� (enumerated in the census) and not “enterprises�, and (ii) the information it provides rapidly becomes outdated, as CAS staff very well recognize. For CAS to maintain an up to date business register, however, it must have both the legal authority and appropriate institutional arrangements to access the relevant Ministry of Finance records. Under the 1979 statistics law, CAS lacks this authority. From this legal deficiency comes about the use of the census of buildings and dwellings to collect information on establishments. The preferred approach to maintaining an up to date business register in a cost effective manner is to provide CAS with legal authority to access the necessary Ministry of Finance records. This requires appropriate safeguards, of course, but as Lebanon’s national statistical office, CAS is legally responsible for and very well aware of the importance of confidentiality in the operations of official statistics. A new statistics law can provide CAS with the necessary legal authority to maintain a business register working with Ministry of Finance records (and perhaps other records, such as those maintained by the Ministry of Justice). A new statistics law can also address several other issues critical to the development of CAS’s capacity to fulfill its role as leader of Lebanon’s national statistical system. The Statistical Master Plan (SMP) recently prepared for Lebanon emphasizes the fundamental importance of adopting a new statistics law, and this is accordingly one of the priority issues identified by the cabinet paper now under review by the Prime Minister’s office. The SMP also calls for an enterprise survey. Therefore, the optimal orientation that CAS might adopt is to (a) defer the census of buildings and dwellings until a new legal framework has been established that allows the construction and maintenance of a business register using MoF and other administrative records, (b) reorient the census by removing most or all of the establishment questions from census, and (c) develop the census as a full-fledged housing census. Questionnaire Design for the Census of Buildings and Dwellings 6 4. Implications for Questionnaire Development The purpose of providing this background on the census of buildings and dwellings and the legal framework of Lebanon’s national statistical system is to ensure that discussion of questionnaire development for the next census does not loose sight of the forest for the trees. Ideally, the adoption of a new statistics law will obviate the necessity for collecting information (or at least so much information) on establishments in the census. This would make it possible to focus the census effort on collecting maximum relevant socio-demographic data. At the very least, it should be expected that the upcoming census will be the last to collect substantial information on business establishments, that its role in this regard will be overtaken by a new arrangements enabled by a new statistics law. 5. Developing the questionnaire for the next census of buildings and dwellings 5.1 Questions and definitions Given the background of the census of building and dwellings, it is recommended that development of the questionnaire for the next census not go beyond incremental changes to the questionnaire for the last census. Indeed, questionnaire issues involve not only the questions posed, but the precise, statistical definitions of the entities to which they refer. The definitions used in the 2004 census follow international standards and do not require revision unless international standards have changed, but there may be considerable need for “fine-tuning� the application of these definitions to the practicalities encountered in the field. The experience of the last census revealed some weaknesses, which CAS staff clearly understand, and these should be addressed. Moreover, as in every such very large scale data processing operation, it is important to be ruthless in eliminating detail that is not manifestly needed by users. CAS needs to carefully research users needs, ascertain what information is essential, and limit the questionnaire accordingly. 5.2 Units of enumeration in population and housing censuses A fundamental starting point for any census is the identification of “units of enumeration,� statistically defined entities about which information will be collected, together with a precise delineation of which units are to be included in the enumeration and which units are to be excluded. In censuses of population and housing, the most important units of enumeration are: (a) persons, (b) households, (c) housing (“dwelling�) units, and (d) collective living quarters. (a) “Persons� requires no explanation. Questionnaire Design for the Census of Buildings and Dwellings 7 (b) Households are socially defined groups of persons who “make common provision for food and other essentials for living;� understood to include “single person� households as a special case. (c) Housing units are physical entities that may be occupied by households. Housing units may be vacant or occupied by more than one household. (d) Collective living quarters include, for example, hotels, institutions such as hospitals and student dormitories, and military camps. Paragraph 2.415 of the UN Principles and Recommendations provides a three digit coding of different types of collective living quarters. The purpose of referring here to units of enumeration in population and housing censuses is to call attention to the very detailed discussion of the issues involved in the UN Principles and Recommendations for Population and Housing Censuses Revision 2 (2008), which are relevant, if not necessarily directly applicable, to the census of buildings and dwellings. 5.3 Units of enumeration in the census of buildings and dwellings The definitions of units of enumeration for the census of buildings and dwellings are available in Arabic only, but the units may be inferred in general terms from the questions on the questionnaire. The units of enumeration for the census of buildings and dwellings are: (1) buildings; (2) “units� in buildings, meaning physical subdivisions of buildings; (2a) “residential units,� a category of “units;� (2b) “non-residential units,� also a category of “units;� and (3) establishments. The census of buildings and dwellings is not a population and housing census, and is not even primarily a housing census, so the enumerated entities require explicit identification and definition. A fundamentally important component of the training of field workers for any census consists of teaching precise conceptual and operations definition of the several units of enumeration used. It is estimated that development of enumerator training materials, which should include photographs of actual field situations, would require the full time effort of three people for 12-16 weeks. CAS staff noted that some of the enumerators recruited for the 2004 census enumeration were engineering students who were highly knowledgeable in construction materials, and that this caused problems in enumerator training because they argued with the trainers about proper terminology. 6. Design issues for the questionnaire of the next census of buildings and dwellings 6.1 Observations on the Questionnaire of the 2004 census of buildings and dwellings3 Two recommendations apply throughout the questionnaire. The first is to number all questionnaire items in a suitable manner. It is convenient to be able identify the unit of enumeration to which each question applies from the item number. This may be accomplished by combining numbering with alphabetic does representing different parts of the questionnaire, e.g., 3 The following is based on the English translation of the questionnaire provided by CAS to the consultant on the first day of the mission. Questionnaire Design for the Census of Buildings and Dwellings 8 “B1� for the first question on buildings. This scheme is used below to number all items on the 2004 census questionnaire, which are not. The second general recommendation is to add additional explicit skip instructions to enumerators to clarify which questions are to be addressed to which units of enumeration and to minimize duplication of categories in different questions. 6.2 Administrative Geography and issues raised by incomplete cadastral registration Approximately 65 percent of the national territory of Lebanon is covered by cadastral registration, meaning (roughly) that ownership of land is documented in cadastral registers that identify property boundaries identified by land survey, current owners of the properties, past transfers of ownership, and so on. For the remainder of the country, property boundaries and even in some cases ownership and related issues may be more or less ambiguous. This is relevant to census work because a fundamental principle of enumeration for population and housing censuses, of which the census of buildings and dwellings is an instance, is that the national territory to be censuses is partitioned into clearly delineated enumeration areas (“EAs�) that can be assigned to an enumerator for enumeration. EAs are generally formed as subdivisions of the lowest level administrative units (if the units are dense) or as combinations of lowest level administrative units (if the units are sparse). In this case, EAs are assigned a unique place in the geographic hierarchy of administrative units because each EA is a subdivision of an administrative unit. This correspondence between EAs and administrative units is critical to the coding and processing of the data collected in the census. Areas with incomplete cadastral registration can pose problems for this assignment. Boundaries of enumeration areas may be well defined by the CAS cartographic unit based on topographic maps and remote sensing imagery (satellite images and aerial photography). However, if boundaries of administrative units are not clearly defined, questions may arise regarding the proper classification in the administrative unit hierarchy. 6.3 Item by item commentary on the 2004 questionnaire In this section the items and questions on the 2004 questionnaire are listed in sequence with commentary, questions and recommendations following each item. For some items and questions, no commentary is called for. Cover page (CP): Location of EA and metadata CP1 Geocode data CP2 Numbers of pages added for buildings/units/establishments/centers and subsidiaries CP3 Site processing metadata CP4 Central office processing metadata Buildings (B) B1 Number of building in EA B2 Name of building or name of building owner B3 Name of street or name of nearby landmark Questionnaire Design for the Census of Buildings and Dwellings 9 B4 Number of building on the street (if any) These four questions aim to identify the enumerated building unambiguously in the absence of a uniform national system for addressing buildings, dwelling units, establishments, and other entities. B5 Building Status (enter code) 1 – Construction complete, usable 2 – Building unusable due to war damage 3 – Building unusable for other reasons 4 – Construction complete, only partially usable due to war damage 5 – Construction complete, only partially usable for other reasons 6 - Under construction 7 – Construction stopped 8 – Improvised structure 9 – Open space used for economic activity 10 – Nomad camp These building status codes are extremely confusing, as CAS staff well understand. Part of the problem is the sheer number of dimensions packed into the responses, including: usable versus partially usable versus unusable; standard building versus improvised structure; building versus nomad camp versus open space; unusable due to war damage versus unusable for other reasons (not necessarily damage). The sequencing of the categories defies ready comprehension, and there are no skip instructions. The following categorization is recommended as a minimal alternative. It can of course be elaborated if any of the information omitted is deemed essential to meet user needs. “Continue� in the following means “continue to the following question on the questionnaire.� B5 Building Status (enter code) 1 - Construction completed, in use 2 - Construction completed, not in use (vacant) 3 - Improvised building (skip to B12) 4 - Nomad camp (skip to B12) 5 – Open space used for economic activity (skip to B12) This simplified coding aims to identify information that is essential for enumerators to identify “buildings� (including improvised buildings, nomad camps and open spaces used for economic activity) that contain dwelling units and/or establishments that they need to enumerate. Note the use of explicit skip instructions that recognize that questions B6 to B11 are not applicable to improvised buildings, nomad camps, and open spaces used for economic activity. B6 External building material 1 – Rock 2 – Concrete or hollow block covered with glass or aluminum 3 – Hollow block covered with marble, rock, or ceramic 4 – Patched hollow block with paint or other 5 – Un-patched hollow block or reinforced concrete 6 – Other, specify Questionnaire Design for the Census of Buildings and Dwellings 10 This question was included to give some indication of living conditions. It was less than wholly satisfactory for this purpose, and it is unclear whether it should be included in the questionnaire for the next census. CAS staff had expressed some skepticism about the value of this question. CAS needs to review user need for this question and decide on this basis whether to keep it or discard it. If the next census involves a concurrent household survey, the question on external building material should be dropped from the census questionnaire and might be included (perhaps along with other dwelling unit questions) on the survey questionnaire. B7 Building use 1 – Residential only 2 – Non-residential only 3 – Mixed 4 – Under construction 5 – Vacant 6 – Closed (if unable to be determined) The primary objective of the census of buildings and dwellings is to provide sampling frames for various kinds of surveys, including household surveys and surveys of business establishments. Buildings are comprised of one or more “units,� which may be dwelling units, places of business, or dwelling units and places of business. The later are identified operationally by the presence of a sign on the unit advertising the presence of some kind of business establishment. This question about building use is a “filter� question to identify which buildings contain dwelling units (Residential only), establishments (Non-residential only), or units that are both dwelling units and establishments (Mixed). From there, the usefulness of categories 4 to 6 seems unclear. “Under construction� and “vacant� have just been ascertained in question B5. Is it really necessary to repeat them two items further on in the questionnaire? Is there some reason enumerators cannot be instructed to skip this question for buildings to which it does not apply? Also, the precise meaning of ““if unable to be determined�� category is not clear. Indeed, what exactly does “Closed� mean? “Available and suitable for residential or non-residential use but not being so used at the time of enumeration� seems a reasonable interpretation. If so, then in this case the parenthetical “if unable to be determined� does not make sense. B8 Year construction was completed (enter code or year) 0 – Unknown 1 – 1950 or before 2 – 1951-1960 3 – 1961-1970 4 – 1971-1975 5 – 1976-1980 6 – 1981-1985 7 - 1986-1990 8 – 1991-1993 9 - 1994 and later (enter year construction ended) CAS has been using calendar year groupings so far and “year construction ended� for the last 10 years prior to the census. It is recommended that CAS consider simplifying the response to “year construction ended� for all units. Of course respondents will not be able to answer accurately in Questionnaire Design for the Census of Buildings and Dwellings 11 all cases, but the same is true of age of persons in a population census. If the respondent does know the exact year, the enumerator asks him or her to provide a “best guess.� The results are certain to provide more useful information than the coded year groupings, and of course years may be grouped in tabulation. This change should of course be field tested. B9. Connections and services (mark 1 for “yes� or 2 = “no� for each) 1 – Sewage system 2 – Water system 3 – Well 4 – Generator 5 – Guard 6 – Parking area 7 – Lift For this question (“building equipment�) it might be possible to identify additional characteristics and add them (connections to municipal services, electricity, telephones, for example). However, it is important to know what are the other sources of data on such characteristics, and to know for what purpose(s) precisely the information is being collected. To make the sample frame derived from the census useful for a greater variety of surveys, for example? To provide information on coverage of municipal/community services? B10 Number of floors (enter number) Below ground Above ground B11 Number of units by use Residential (enter number) Non-residential (enter number) Residential and working place (enter number) Undetermined (enter number) Total number of units (sum numbers above) The questions on numbers of floors in the building can be filled in the questionnaire sequence indicated, but the numbers of units (residential, non-residential, residential and working place, undetermined) and number of establishments are filled in only after the enumerator has completed the “List of units in the building (Residential or Non residential). This is not clear from the questionnaire. Some amendment to the questionnaire to make this clearer without reference to the training manuals might be appropriate. In theory, enumerators have the training manual with them at all times and consult it as needed; practice may fall short of this ideal. One slight subtlety in these topics is that a “non-residential� unit is not necessarily a business establishment. It might be, for example, a store room of some kind. Indeed, “Residential� and “non-residential� are apparently mutually exclusive (a unit cannot be both) and exhaustive (every unit falls into one of these categories. On this understanding, the third category, “Residential and working place,� is therefore ambiguous. Evidently “residential� is intended to be understood as “residential only,� meaning used as a residence, but not as a place of work. If this is the case, the following more explicit designations and ordering are recommended. ALTERNATIVE VERSION OF B11 B11 Number of units by use Residential, not used for work (enter number) Residential and working place (enter number) Non-residential (enter number) Questionnaire Design for the Census of Buildings and Dwellings 12 Total number of units (sum numbers above) Placing the first two categories together clarifies for the enumerator the intended meaning of the categories. It is recommended that the explicit code for “Undetermined� be deleted from the questionnaire, with enumerators instructed to use a standard mark (e.g., “0�) when it is impossible to assign one of the three explicitly indicated codes. The intention is to discourage enumerators from too easily allowing a “could not be determined� entry on the questionnaire. In the event, it is clearly advantageous to use a single code for “could not be determined� throughout the questionnaire. B12 Number of establishments B13 Number of parcel CAS staff have indicated that enumerators should make this entry only after canvassing the establishments in the building. It is recommended that CAS consider (a) including instructions on the questionnaire that indicate this explicitly and/or (b) repositioning this item so that it occurs after the listing of establishments, so that the correspondence between order on the questionnaire and order in filling in the questionnaire is preserved. B14 Nature of building use by public administration, … This item needs considerable work, in particular the development of a suitable classification and coding scheme. There are also some questions about what entities are supposed to be enumerated. Units (U) U1 Number of building in enumeration area (control) U2 Serial number of unit in the building U3 Floor the unit is on U4 Number of unit on the floor U5 Use of the unit (enter code) 1 – Residential 2 – Residential and work 3 – Non-residential 4 – Under construction [unit or building?] 5 – Vacant 6 – Closed [what does this mean? Not functional?] There appears to be an ambiguity between the units of enumeration: “building,� “unit,� and “establishment.� Establishments might comprise many units in a building; depending on how precisely “unit� is defined. If an establishment occupies multiple floors of a building, for example, are these floors collectively a “unit.?� FOR RESIDENTIAL UNITS Questionnaire Design for the Census of Buildings and Dwellings 13 It is recommended that the following question be added for residential units: Was the respondent who provided the information for this unit a resident of the unit? (mark 1 for “yes� or 2 = “no� for each) U6 Type of residential unit (enter code) 1 – Principal residence 2 - Secondary residence 3 – Vacant 4 - House for guard Are these categories mutually exclusive and exhaustive? “Principal residence� and “secondary residence� are mutually exclusive. The categorization suggests that a “House for [a] guard� cannot be a principle residence? Why not? And if it cannot be a principal residence, is it not a secondary residence? Should “Secondary residence� be “Secondary residence, other than house for guard?� What does “Vacant� mean? At first the answer seems obvious, but a secondary residence is by definition unused for more or less extended periods of time, during which it is “vacant.� And when a secondary residence is in use, the corresponding primary residence may be “vacant.� Do these categories need rethinking, or merely more precise articulation? On the types of residential units, CAS staff suggested that a new question on type of dwelling / residential unit might be incorporated, identifying, for example, “Villa,� “single family dwelling,� etc. However, there are “type� questions in the Living Conditions surveys that can be reworked for this purpose, while keeping questions in the census of buildings and dwellings as they are. From the interaction with CAS staff, it appeared that it might be desirable to identify dwelling units in which informal sector work is taking place. However, obtaining accurate information on this on the census form might prove difficult or impossible. U7 Mode of use of residential unit (enter code) 1 – Owned by occupants 2 – Rented to occupants 3 – Illegally occupied 4 – Vacant 5 – Other What does “Other� mean? Why are the first four categories not exhaustive? Why is “vacant� repeated here when it just appeared in the previous question? U8 Unit area of residential unit (enter area in square meters) U9 Type of non-residential unit activity (enter code) 1 – Permanent production establishment 2 – Temporary or seasonal production establishment 3 – Public administration 4 – Auxiliary activity unit 5 – Vacant 6 – Closed 7 – Other (specify) Questionnaire Design for the Census of Buildings and Dwellings 14 What exactly does “vacant� mean here? Suitable for some non-residential use, but not used at the time of enumeration? What does “closed� mean? What is the difference between “closed� and “vacant�? Why do these categories precede, rather than follow, the “Vacant� and “Closed� responses? I.e., why not group together responses that correspond to the unit being in use as opposed to not in use? CAS needs to think through the intended meanings of these categories and clarify and/or recast them. U10 Name of the establishment or name of the unit occupant (write name) U11 Number of floors related to the unit, if on more than one floor (enter number) By examining U10 and U11, two questions arise. First, how precisely is the unit of enumeration “unit� (note the two different meanings of the word “unit� here, potentially confusing, but seemingly unavoidable) is defined. Second, what is the relationship between “units� of buildings and “establishments?� Questions on “units� of buildings precede questions on “establishments� on the questionnaire, and the enumeration of units is apparently the basis for the enumeration of establishments. “Units� are apparently physical entities, however, whereas establishments are conceptual entities with multiple possible relationships with physical units. How exactly are enumerators instructed to identify establishments for enumeration? The structure of the questionnaire suggests that the enumerator first identifies “units� in a building and then ascertains which of these units are occupied by establishments. But a “floor� would presumably be the largest possible identifiable physical unit in a multi-story building, and there is no obvious general method by which an enumerator could look at a building and ascertain that some collection of floors is occupied by an establishment. This suggests that for multi-story buildings enumerators identify establishments directly, without the intermediary step of dividing the building into “units.� If this is the case, the meaning of “unit,� and therefore the count of units, becomes ambiguous. This in turn suggests that the understanding of “unit� is flexible, referring either to a physical unit, which may be residential or non-residential, or to a “unit� that is actually an establishment rather than a physical unit. CAS needs to clarify these issues. Establishments (E) Question E8 below indicates, or at least suggests, that the establishments enumerated here includes not-for-profit as well as for-profit establishments. E1. Serial number of establishment in building (enter number) E2. Number of building in enumeration area (copy number from Buildings form) E3. Serial number of the unit in the list of units E4. Official name of the establishment (enter name) How important is it for the enumerator to enter the full official name of the establishment? Is this is well defined in every case? Does the enumerator record the respondent’s first response? Is the enumerator instructed to probe for a fuller name if the name first given sounds truncated? Does the enumerator examine a document of some kind to confirm the name? Are names an issue in matching census records with records of establishment maintained elsewhere in the national Questionnaire Design for the Census of Buildings and Dwellings 15 statistical system, e.g., the Ministry of Finance? Is sufficient emphasis is placed on recoding full and accurate names in field worker training? E5. Name of owner or director (enter name) Which name should the enumerator enter under which circumstances? Does it matter? Why or why not? E6. Principal activity (enter description for ISIC Rev 3.1 coding) ISIC (International Standard Industrial Classification; detailed metadata is available online in the Classifications Registry of the UN Statistics Division’s website, unstats.un.org). On the basis of the experience with the last census, was the training of field workers sufficient to enable them to record responses that could be reliably coded by central office coders? Should more training time be allocated to this? Should training materials be revised? Should the ISIC revision used be changed for the next census? E7. Is the establishment registered in the Business Record? (enter “1� for yes or “2� for “no�) Clarify how the enumerator ascertains the answer to this question. On the basis of the respondent’s answer only, or by reference to document of some kind? E8. Legal status of the establishment (enter code according to registration status) [CODES FOR ESTABLISHMENTS REGISTERED IN THE BUSINESS RECORD] 1 – Individual establishment 2 – Joint company 3 – Partnership 4 - Stock partnership company 5 – Limited liability company 6 – Joint stock company 7 – Cooperative company 8 – Holding company 9 – Offshore company 10 – Franchise [CODES FOR ESTABLISHMENTS NOT REGISTERED IN THE BUSINESS RECORD] 1 – Individual establishment 2 – Joint company 3 – Particular partnership 4 – Society or cooperative or committees of union or party 5 – Religious organization Does this question presume that respondent’s are able to identify their establishment as being one of the coded categories? Does the enumerator examine any document to confirm the respondent’s reply? Does the enumerator ask the question, wait for the respondent to reply, and then decide what code is appropriate, perhaps with additional probing questions? Does the Questionnaire Design for the Census of Buildings and Dwellings 16 enumerator read the categories off to the respondent and ask the respondent to choose among them? How thoroughly are field workers trained in the definitions of these designations? It is recommended that CAS field tests focus on these questions and produce a report on the results of the test. E9. Does the establishment operate on a commercial basis (for profit)? (enter “1� for yes or “2� for “no�) Are all categories in the classifications in question E7 applicable to not-for-profit establishments? If not, should this question precede question E7 and include appropriate skip instructions to enumerators? What is the relation, if any between operating on a commercial basis and registered or not registered in the business record? This illustrates a question to which a diagnostic cross-tabulation of establishments by these dimensions may be relevant. E10. Registered capital (for registered establishments only) E11. Year establishment began operation (enter year) The “year of operation� question is problematic because establishments that have been doing business for many years without having registered will, when registering, report this variable inaccurately, giving a year much later than the year of initial operation. This question asks the enumerator to enter the year in all cases, without resort to any coding of groups of years (e.g., “Before 1950�), in contrast with the question on the year construction of buildings was complete. What is the rationale for this different treatment of the responses to the two questions? Is it expected that the year an establishment began will be more accurately remembered than the year construction of a building was completed? Has any analysis of previous census results for “heaping� on, for example, years ending in “0� or “5�? E12. Does the establishment possess an accounting system? (enter “1� for yes or “2� for “no�) CAS tested questions on financial turnover for 2004 but found very high non-response and decided against including a turnover question. Establishments in Lebanon have since moved toward using accounting services for their declarations to the Ministry of Finance. This is likely to make responses to the question “Does the establishment possess an accounting system?� more accurate than in the past. CAS proposes to ask a question on annual turnover for establishments with an accounting system. Will enumerators be instructed to confirm the answer by reference to a document of some kind? Pre-testing of the results of this question will be most useful if appropriate tabulations are made and analyzed before decisions on the final questionnaire are made. E13. Economic organization (enter code) 1 – Single establishment Questionnaire Design for the Census of Buildings and Dwellings 17 2 - Headquarters having subsidiaries 3 – Subsidiary with accounting system 4 – Subsidiary without accounting system The “subsidiaries� question may be problematic. Family businesses in particular may involve two or more places of business in such a way that the distinctions between “head office� and “subsidiary� are difficult or impossible to draw. E14. If question E12 is code 2, number of subsidiaries (enter number) E15. Is the work of the establishment permanent or seasonal? (enter code) 1 – Permanent 2 – Seasonal E16. If answer to E14 is “Seasonal,� specify what months of the year the establishment is active (circle numbers) 1 2 3 4 5 6 7 8 9 10 11 12 E17. Number of workers (enter number) This is the only socio-demographic item in this section of the questionnaire. Non-response was substantial and the quality of the responses is questionable. CAS staff noted that this information is not available from the Ministry of Labor. There are also issues related to lack of registration of and documentation on foreign laborers. E18. Telephone number E19. Cell phone number E20. Email address E21. Is the establishment registered for the VAT? (enter “1� for Yes or “2� for “No�) E22. If answer to E20 is “Yes,� enter VAT registration number E23. Finance number within Ministry of Finance (enter number if available) Notes For questions E22 and E23, are enumerators instructed to confirm answers by reference to documents? Enumerator comments Notes The rationale for the distinction between comments “concerning a building� and “general comments� is unclear. It is recommended that it be clarified or, perhaps, replaced with a more detailed breakdown. It is also recommended that the purpose of this section be explicitly identified as being for the enumerator to record questions and issues that arise during the enumeration for discussion with the group leader and/or the field controller. Entries on this section of the questionnaire may then serve as partial basis for controller review of the enumerator’s work and the periodic discussions between the enumerator and the controller. Training of enumerators, group leaders and field controllers would then address explicitly and specifically how this section is to be used to improve the quality of the enumeration. Questionnaire Design for the Census of Buildings and Dwellings 18 7. Developing the Housing content of the Census of Buildings and Dwellings The Census of Buildings, Dwellings and Establishments name in Arabic (Census of Buildings and Establishments) is more accurate to describe the operation. Indeed, of the 48 items on the questionnaire, only 3 refer to housing units (under the title “For residential Units�) of which two, questions U6 and U7 (see Annex 1) are part of the 13 core topics on housing units identified in Annex 3. Of course, many questions under “Buildings�, mainly B6 and B9, cover some of the 13 core topics, but it is not clear if the coverage goes to the level of the housing units. Also, it happens that residents of dwellings are not contacted by enumerators, especially in multiple dwellings units’ buildings. To have the CBD becoming a real Housing Census, enumerators would have to visit every housing unit and obtain responses from residents on various characteristics of the dwelling. Moving toward having more housing characteristics covered in the questionnaire is not a major challenge. Indeed, for the matter of re-organizing and develop the questionnaire, CAS staff has all the skills and expertise to do the job without any major technical assistance. The UN Statistics Division has a collection of population and housing census schedules online4. The "Census Questionnaires" link in this page contains an alphabetical list of countries with questionnaires in original and translation. Sometimes population and housing are together, sometimes housing is separate. CAS staff is aware of this resource and have downloaded numerous questionnaires. The UN Principles and Recommendations for Population and Housing Censuses Revision 2 (2008) has recommended a list that is reproduced below5 in Annex 3. If CAS judges user demand for housing information sufficient to justify the inclusion of additional housing questions on the census questionnaire, this list provides a starting point for deciding which questions to consider. Note however that many of these questions would require enumerators to interview residents of households. As discussed in Section 1 of Document III, it cannot be assumed that enumerators in 2004 obtained information on dwelling units by interviewing residents of these units. The questionnaire did not identify respondents as residents or non-residents of the dwelling unit they were providing information on, so statistical evidence on the question is not available, but it is possible that information on a large number of housing units was not provided by residents of these units. Indeed, beyond developing the questionnaire, the problem is the implications of changes in the questionnaire for field operations. Including questions that would require enumerators to interview residents of every dwelling unit would have major implications for field operations. The 2004 census counted roughly 1.5 million "units", of which roughly 400,000 were establishments and 1.1 million were dwelling units. Less than 250,000 dwelling units were in buildings containing a single unit; the rest were in buildings containing multiple units; units in which enumerators did not necessarily obtain information from residents of the dwelling and had a strong incentive (the low piece rate for dwellings) not to do so. Indeed, it is possible that even for single unit buildings enumerators obtained information from neighbors or passersby if a 4 World Population and Housing Census Program 2010 page http://unstats.un.org/unsd/demographic/sources/census/2010_PHC/default.htm. 5 Source: UN Principles and Recommendations for Population and Housing Censuses Revision 2 (2008), Chapter VII, ", page 188 Table 2 "Housing census topics by unit of enumeration. Questionnaire Design for the Census of Buildings and Dwellings 19 resident was not readily available. The additional cost and manpower that would be required to have an exhaustive Housing Census may be gauged by considering that for an enumerator to (i) visit a dwelling unit, interview a resident and obtain housing information would require roughly the same time and effort as (ii) visiting and obtaining information for a single establishment. This suggests that CAS would need to do for about 75% of 1.1 million dwelling units, or about 800,000 units, what they did, in terms of effort, for 400,000 establishments. This implies that the total labor of field work done in 2004 would have to be multiplied by 3, with implications in terms of budget, field staff, and various support operations, etc. Of course this is a very rough estimate. It is possible that costs, manpower, etc. might have to be multiplied only by 2, or by more than 3. The point is that adding questions that would require enumerators to interview residents of every household would greatly increase the cost of the census. The magnitude of such an operation is far beyond what CAS can do, taken into consideration the available budget and staffing. The Statistical Master Plan lays out in detail precisely what steps need to be taken to ensure the adequate staffing needed to have the national statistical office functioning and developing efficiently. Other initiatives can at best bring marginal improvements on the fringes of some activities, but this should not distract attention from what is the most important issue for CAS activities and development at this point in time: adoption of the Statistical Master Plan. Questionnaire Design for the Census of Buildings and Dwellings 20 Annex 1. Questionnaire of the Census of Buildings and Dwellings of 2004 Cover page: Location of EA and metadata CP1 Geocode data CP2 Numbers of pages added for buildings/units/establishments/centers and subsidiaries CP3 Site processing metadata CP4 Central office processing metadata Buildings B1 Number of building in EA B2 Name of building or name of building owner B3 Name of street or name of nearby landmark B4 Number of building on the street (if any) B5 Building Status (enter code) 1 – Construction complete, usable 2 – Building unusable due to war damage 3 – Building unusable for other reasons 4 – Construction complete, only partially usable due to war damage 5 – Construction complete, only partially usable for other reasons 6 - Under construction 7 – Construction stopped 8 – Improvised structure 9 – Open space used for economic activity 10 – Nomad camp B5 Building Status (enter code) 1 - Construction completed, in use 2 - Construction completed, not in use (vacant) 3 - Improvised building 4 - Nomad camp 5 – Open space used for economic activity B6 External building material 1 – Rock 2 – Concrete or hollow block covered with glass or aluminum 3 – Hollow block covered with marble, rock, or ceramic 4 – Patched hollow block with paint or other 5 – Un-patched hollow block or reinforced concrete 6 – Other, specify B7 Building use 1 – Residential only 2 – Non-residential only 3 – Mixed 4 – Under construction 5 – Vacant 6 – Closed (if unable to be determined) B8 Year construction was completed (enter code or year) 0 – Unknown 1 – 1950 or before 2 – 1951-1960 3 – 1961-1970 4 – 1971-1975 5 – 1976-1980 6 – 1981-1985 7 - 1986-1990 8 – 1991-1993 9 - 1994 and later (enter year construction ended) B9. Connections and services (mark 1 for “yes� or 2 = “no� for each) 1 – Sewage system 2 – Water system 3 – Well 4 – Generator 5 – Guard 6 – Parking area Questionnaire Design for the Census of Buildings and Dwellings 21 7 – Lift B10 Number of floors (enter number) Below ground Above ground B11 Number of units by use Residential (enter number) Non-residential (enter number) Residential and working place (enter number) Undetermined (enter number) Total number of units (sum numbers above) B11 Number of units by use Residential, not used for work (enter number) Residential and working place (enter number) Non-residential (enter number) Total number of units (sum numbers above) B12 Number of establishments B13 Number of parcel B14 Nature of building use by public administration, … Units U1 Number of building in enumeration area (control) U2 Serial number of unit in the building U3 Floor the unit is on U4 Number of unit on the floor U5 Use of the unit (enter code) 1 – Residential 2 – Residential and work 3 – Non-residential 4 – Under construction 5 – Vacant 6 – Closed [FOR RESIDENTIAL UNITS] U6 Type of residential unit (enter code) 1 – Principal residence 2 - Secondary residence 3 – Vacant 4 - House for guard U7 Mode of use of residential unit (enter code) 1 – Owned by occupants 2 – Rented to occupants 3 – Illegally occupied 4 – Vacant 5 – Other U8 Unit area of residential unit (enter area in square meters) [FOR NON-RESIDENTIAL UNITS] U9 Type of non-residential unit activity (enter code) 1 – Permanent production establishment 2 – Temporary or seasonal production establishment 3 – Public administration 4 – Auxiliary activity unit 5 – Vacant 6 – Closed 7 – Other (specify) U10 Name of the establishment or name of the unit occupant (write name) U11 Number of floors related to the unit, if on more than one floor (enter number) Establishments E1 Serial number of establishment in building (enter number) E2 Number of building in enumeration area (copy number from Buildings form) E3 Serial number of the unit in the list of units E4 Official name of the establishment (enter name) E5 Name of owner or director (enter name) Questionnaire Design for the Census of Buildings and Dwellings 22 E6 Principal activity (enter description for ISIC Rev 3.1 coding) E7 Is the establishment registered in the Business Record? (enter “1� for yes or “2� for “no�) E8 Legal status of the establishment (enter code according to registration status) [CODES FOR ESTABLISHMENTS REGISTERED IN THE BUSINESS RECORD] 1 – Individual establishment 2 – Joint company 3 – Partnership 4 - Stock partnership company 5 – Limited liability company 6 – Joint stock company 7 – Cooperative company 8 – Holding company 9 – Offshore company 10 – Franchise [CODES FOR ESTABLISHMENTS NOT REGISTERED IN THE BUSINESS RECORD] 1 – Individual establishment 2 – Joint company 3 – Particular partnership 4 – Society or cooperative or committees of union or party 5 – Religious organization E9 Does the establishment operate on a commercial basis (for profit)? (enter “1� for yes or “2� for “no�) E10 Registered capital (for registered establishments only) E11 Year establishment began operation (enter year) E12 Does the establishment possess an accounting system? (enter “1� for yes or “2� for “no�) E13 Economic organization (enter code) 1 – Single establishment 2 - Headquarters having subsidiaries 3 – Subsidiary with accounting system 4 – Subsidiary without accounting system E14 If question E12 is code 2, number of subsidiaries (enter number) E15 Is the work of the establishment permanent or seasonal? (enter code) 1 – Permanent 2 – Seasonal E16 If answer to E14 is “Seasonal,� specify what months of the year the establishment is active (circle numbers) 1 2 3 4 5 6 7 8 9 10 11 12 E17 Number of workers (enter number) E18 Telephone number E19 Cell phone number E20 Email address E21 Is the establishment registered for the VAT? (enter “1� for Yes or “2� for “No�) E22 If answer to E20 is “Yes,� enter VAT registration number E23 Finance number within Ministry of Finance (enter number if available) Last page: Enumerator comments Comments concerning a building Building # Comments General Comments Questionnaire Design for the Census of Buildings and Dwellings 23 Annex 2. Questionnaire of the Census of Buildings and Dwellings-Proposed Revisions Note: See text of section 6.3 for commentary on questions for which revisions may be needed but for which revision requires further research by CAS. These are not shown below. Cover page: Location of EA and metadata CP1 Geocode data CP2 Numbers of pages added for buildings/units/establishments/centers and subsidiaries CP3 Site processing metadata CP4 Central office processing metadata Buildings B1 Number of building in EA B2 Name of building or name of building owner B3 Name of street or name of nearby landmark B4 Number of building on the street (if any) B5 Building Status (enter code) 1 – Construction complete, usable 2 – Construction complete, not in use (vacant) 3 – Under construction (skip to B12) 4 – Construction stopped (skip to B12) 5 – Improvised building (skip to B12) 6 – Nomad camp (skip to B12) 7 – Open space used for economic activity (skip to B12) B6 External building material 1 – Rock 2 – Concrete or hollow block covered with glass or aluminum 3 – Hollow block covered with marble, rock, or ceramic 4 – Patched hollow block with paint or other 5 – Un-patched hollow block or reinforced concrete 6 – Other, specify B7 Building use 1 – Residential only 2 – Non-residential only 3 – Mixed 4 – Under construction 5 – Vacant 6 – Closed (if unable to be determined) B8 Year construction was completed (enter code or year) [coding of year groups deleted] B9. Connections and services (mark 1 for “yes� or 2 = “no� for each) 1 – Sewage system 2 – Water system 3 – Well 4 – Generator 5 – Guard 6 – Parking area 7 – Lift B10 Number of floors (enter number) Below ground Above ground B11 Number of units by use Residential (enter number) Non-residential (enter number) Residential and working place (enter number) Undetermined (enter number) Total number of units (sum numbers above) B11 Number of units by use Residential, not used for work (enter number) Residential and working place (enter number) Non-residential (enter number) Questionnaire Design for the Census of Buildings and Dwellings 24 Total number of units (sum numbers above) B12 Number of establishments B13 Number of parcel B14 Nature of building use by public administration, … Units U1 Number of building in enumeration area (control) U2 Serial number of unit in the building U3 Floor the unit is on U4 Number of unit on the floor U5 Use of the unit (enter code) 1 – Residential 2 – Residential and work 3 – Non-residential 4 – Under construction 5 – Vacant 6 – Closed [FOR RESIDENTIAL UNITS] [Note inserted question U5A below] U5A Was the respondent who provided the information on this unit a resident of the unit? (mark “1� for “yes� and “2� for “no�) U6 Type of residential unit (enter code) 1 – Principal residence 2 - Secondary residence 3 – Vacant 4 - House for guard U7 Mode of use of residential unit (enter code) 1 – Owned by occupants 2 – Rented to occupants 3 – Illegally occupied 4 – Vacant 5 – Other U8 Unit area of residential unit (enter area in square meters) [FOR NON-RESIDENTIAL UNITS] U9 Type of non-residential unit activity (enter code) 1 – Permanent production establishment 2 – Temporary or seasonal production establishment 3 – Public administration 4 – Auxiliary activity unit 5 – Vacant 6 – Closed 7 – Other (specify) U10 Name of the establishment or name of the unit occupant (write name) U11 Number of floors related to the unit, if on more than one floor (enter number) Establishments E1 Serial number of establishment in building (enter number) E2 Number of building in enumeration area (copy number from Buildings form) E3 Serial number of the unit in the list of units E4 Official name of the establishment (enter name) E5 Name of owner or director (enter name) E6 Principal activity (enter description for ISIC Rev 3.1 coding) E7 Is the establishment registered in the Business Record? (enter “1� for yes or “2� for “no�) E8 Legal status of the establishment (enter code according to registration status) [CODES FOR ESTABLISHMENTS REGISTERED IN THE BUSINESS RECORD] 1 – Individual establishment 2 – Joint company 3 – Partnership 4 - Stock partnership company 5 – Limited liability company 6 – Joint stock company 7 – Cooperative company 8 – Holding company Questionnaire Design for the Census of Buildings and Dwellings 25 9 – Offshore company 10 – Franchise [CODES FOR ESTABLISHMENTS NOT REGISTERED IN THE BUSINESS RECORD] 1 – Individual establishment 2 – Joint company 3 – Particular partnership 4 – Society or cooperative or committees of union or party 5 – Religious organization E9 Does the establishment operate on a commercial basis (for profit)? (enter “1� for yes or “2� for “no�) E10 Registered capital (for registered establishments only) E11 Year establishment began operation (enter year) E12 Does the establishment possess an accounting system? (enter “1� for yes or “2� for “no�) E13 Economic organization (enter code) 1 – Single establishment 2 - Headquarters having subsidiaries 3 – Subsidiary with accounting system 4 – Subsidiary without accounting system E14 If question E12 is code 2, number of subsidiaries (enter number) E15 Is the work of the establishment permanent or seasonal? (enter code) 1 – Permanent 2 – Seasonal E16 If answer to E14 is “Seasonal,� specify what months of the year the establishment is active (circle numbers) 1 2 3 4 5 6 7 8 9 10 11 12 E17 Number of workers (enter number) E18 Telephone number E19 Cell phone number E20 Email address E21 Is the establishment registered for the VAT? (enter “1� for Yes or “2� for “No�) E22 If answer to E20 is “Yes,� enter VAT registration number E23 Finance number within Ministry of Finance (enter number if available) Last page: Enumerator comments Comments concerning a building Building # Comments General Comments Questionnaire Design for the Census of Buildings and Dwellings 26 Annex 3. Housing census topics by unit of enumeration Questionnaire Design for the Census of Buildings and Dwellings 27 Questionnaire Design for the Census of Buildings and Dwellings 28