Appendix A. Technical Description of the Survey

Transcription

1 Appendix A. Technical Description of the Survey The University of Pittsburgh, by demand of USAID (Colombia) performed a thorough research of Colombia s main public opinion and market research companies. Because of its 0-year experience as well as its leadership in complex, country-wide operations, the Centro Nacional de Consultoría (CNC) was chosen to develop the sampling process, field interviewing, data entry, verification and processing. Universe Sample Design The survey universe has National coverage of all adults living in the country s six regions: Bogotá, Atlantic, Pacific, Central, Oriental regions and the former Territorios Nacionales. The universe also comprehended all the adults in both rural and urban areas. The Universe was divided in two sectors: one with cities with more than inhabitants and another with cities with less than inhabitants. Population The sampling was circumscribed to all non institutional civil population: this means that people living in hospitals, orphanages, schools and military bases were excluded. Private households in these areas were contemplated. Final Selection Unit Since the questionnaire included questions not limited to the respondent but also to other household members, the observation statistical unit chosen was the household. The respondent had to live in only one household. Because every household belongs to a housing unit, sometimes shared by more than one household, with relative stability over time, each housing unit was chosen as the final selection unit. Sampling Method We chose the probabilistic, stratified, multistage method, with randomized selection of units in each stage. First, the sample is stratified by town size (cities with less and more than ), then by regions and areas.

2 It is multistage because within each urban area it starts with primary sampling units (sectors), followed by Secondary Units (sections), then Third Units (blocks) and Final Sampling Units (housing conglomerates) of to in urban areas and 0 to in rural areas. In each housing unit the interviewer selected only one household as Observation Unit. The respondent was chosen following gender and age quotas. Thus, in each block the interviewer had to include at least one man and one women from each of these age brackets: to years to 40 years More than 40 years Each interviewer was assigned one specific block. Once in the area, each interviewer listed the first 0 housing units he/she encountered. He/she was instructed to apply a maximum of interviews out of the 0 housing units per block, balancing out the age and gender quotas. The method selection was chosen according to the following considerations: a) We needed representative samples in the following levels: - Nationally First Stage Strata: Cities with more than inhabitants Cities with less than inhabitants Second Stage Strata: Bogota Atlantic Region Pacific Region Oriental Region Central Region Former Territorios Nacionales Third Stage Strata: Urban Area Rural Area Study Domains:

3 Cities with more than inhabitants (obliged selection) Cities with less than inhabitants b) For each stage, we calculated margins of error that corresponded to minimum quality standards c) We pursuit to facilitate operability of interviewing d) We looked for usage of the best and more updated sampling frame available for each municipality (population census, cartography, updated housing lists, among others) Sampling Frame The sampling frame is constituted by the updated cartographic inventory and housing lists obtained during the 99 Population Census. CNC obtained the 00 versions from the Colombian National Department of Statitics, DANE. Sample Calculations by Strata The sample is constituted by sampling points: 0 urban and 5 rural, distributed over 5 municipalities of out of the departments. Sample sizes, Confidence levels and Margins of Error

4 The confidence level anticipated for the national sample was 95% with a margin of error of.% assuming a 50/50 proportion for dichotomic variables. The margins of error for a 95% Confidence Level are: TAMAÑO DE MUESTRA Y MARGENES DE ERROR Nivel de Confianza del 95% Estratos Margen de error % Tamaño de Regiones M.A.S. M.P.C. Atlantica 5,45 5,5 Bogota,45,59 Central 5 5, 5,9 Oriental 4 5,9,05 Pacifica,04, Territorios Nacionales 54,4, Areas Urbana 0,9,0 Rural 9 4,9 5,0 Pais 50,5,5 Non Coverage Adjustment To ensure efficiency, sufficiency and coverage of the sample, we adopted a Non Coverage Adjustment system which guarantees the minimum confidence level and maximum margin of error. In addition, the system eliminates biases resulting from replacements of interviewed units. The system applies to the sample size (n) of each stratum a non coverage factor defined by: N*= (+t) n Where: T = non interview rate. N*= final sample size We chose a t average of 0.. Thus, N*= (+0.) 500 =.0 adults. By the end of the field interviewing we ended up with a real size of 49 adults. less than expected because of law and order issues (see Final Considerations about the interviewing process).

5

6 Staff involved with the Survey CNC involved its 5 different branches (Bogota, Cali, Medellín, Barranquilla and Bucaramanga) in order to ensured a very high quality survey gathered in a very short time. Because of the country s current situation we were advised to stay in most of the areas very short periods of time, which implied complicated operations. Because of the complexity of the questionnaire we used our most experienced interviewers, many of them with more than 5 years of field experience. CNC involved a total of 0 staff members distributed as follows: ASSIGNMENT TOTAL STAFF Field Coordinators 5 Field Supervisors 9 Field Interviewers 5 Field Quality Supervisors 0 Codifiers Data Entry Staff Data Validation Staff for Field Interviewing and Data Entry 9 Directorial and Professional Personnel 5 Administrative Staff Human Resources 0 0. Universe, Population By Region and Urban/Rural Areas COLOMBIA: POBLACION TOTAL, PROYECCION AÑO 004 Region Area Urbano Rural Atlantica Bogota Central Oriental Pacifica Territorios Nacionales Distribucion Porcentual (%) Region Area Urbano Rural Atlantica 9, 00 Bogota 00,0 0,0 00 Central,, 00 Oriental,,4 00 Pacifica,, 00 Territorios Nacionales 9, 0,4 00,, 00

7 Size and Sample Distribution by Strata TAMAÑOS Y DISTRIBUCION DE LA MUESTRA POR ESTRATOS RESUMEN GENERAL Urbano Rural Atlantica + de 00,000 habitantes - de 00,000 habitantes 4 9 Atlantica 9 4 Bogota + de 00,000 habitantes Bogota Central + de 00,000 habitantes de 00,000 habitantes 9 0 Central 9 5 Oriental + de 00,000 habitantes - de 00,000 habitantes 9 0 Oriental 9 4 Pacifica + de 00,000 habitantes de 00,000 habitantes 4 Pacifica 9 4 Territorios Nacionales - de 00,000 habitantes 54 Territorios Nacionales RESUMEN GENERAL MUESTRA UPM PUNTOS DE MUESTREO Regiones Atlantica 0 40 Bogota Central 5 4 Oriental Pacifica 4 Territorios Nacionales

8 Sample Selection by Strata, Areas and Cities

9 44, 5 5 9, 9, 44, TOTAL, 9 9, 9,, ,, 9, 5, CALI 4 4,, PASTO NARIÑO 4,4 0,,9 4, ,,0 99,0, BUCARAMANGA SANTANDER, 4 4, 9,, CUCUTA NORTE DE SANTANDER 0,,4, 0, VILLAVICENCIO META 0,, 9, 0, SOACHA CUNDINAMARCA 0, , 5, 9, ,0 94,0, IBAGUE TOLIMA, 5, 4,, PEREIRA RISARALDA 0,, 9, 0, ARMENIA QUINDIO 0,, 9, 0, NEIVA HUILA 4 4, 9, 0, MANIZALES CALDAS 0,, 9, BELLO 4, ,5 95,5 4, MEDELLIN 5,4 0, 99, 5, ,4 0, 99, 5, ,4 9, 9,, , 9,, SANTA MARTA MAGDALENA 0,,9, 0, MONTERIA CORDOBA 0, 9, 0, 0, DUPAR CESAR, 5 5 5, 9,4, CARTAGENA BOLIVAR 0, 0, 99,4 0, SOLEDAD ATLANTICO, 4 4 0, 99,, BARRANQUILLA ATLANTICO % de la Rural Urbana % Rural del Municipo % Urbano del Municipo % Poblacion Rural Urbana Tamaño Punto de Poblacion año 004 Municipio Departamento Region Distribucion de la ciudades con 00,000 o mas habitantes 44, 5 5 9, 9, 44, TOTAL, 9 9, 9,, ,, 9, 5, CALI 4 4,, PASTO NARIÑO 4,4 0,,9 4, ,,0 99,0, BUCARAMANGA SANTANDER, 4 4, 9,, CUCUTA NORTE DE SANTANDER 0,,4, 0, VILLAVICENCIO META 0,, 9, 0, SOACHA CUNDINAMARCA 0, , 5, 9, ,0 94,0, IBAGUE TOLIMA, 5, 4,, PEREIRA RISARALDA 0,, 9, 0, ARMENIA QUINDIO 0,, 9, 0, NEIVA HUILA 4 4, 9, 0, MANIZALES CALDAS 0,, 9, BELLO 4, ,5 95,5 4, MEDELLIN 5,4 0, 99, 5, ,4 0, 99, 5, ,4 9, 9,, , 9,, SANTA MARTA MAGDALENA 0,,9, 0, MONTERIA CORDOBA 0, 9, 0, 0, DUPAR CESAR, 5 5 5, 9,4, CARTAGENA BOLIVAR 0, 0, 99,4 0, SOLEDAD ATLANTICO, 4 4 0, 99,, BARRANQUILLA ATLANTICO % de la Rural Urbana % Rural del Municipo % Urbano del Municipo % Poblacion Rural Urbana Tamaño Punto de Poblacion año 004 Municipio Departamento Region Distribucion de la ciudades con 00,000 o mas habitantes 00, ,, 00, GRAN TOTAL 55, , 5, 55, TOTAL, 54 0,4 9,, ,,5,5, VILLA FATIMA* VAUPES, 5, 4,, PUERTO ASIS PUTUMAYO, 4,9 5,, FLORENCIA CAQUETA TERRITORIO S NACIONALES, , 50,4, , 4,4,, RIOFRIO 0, 4 4,0,0, FLORIDA 9, 4,5,5, BUGA, 4 0,4 9,, SANDONA NARIÑO, 4,9,, CUMBITARA NARIÑO, 4 9,, PADILLA CAUCA 5, 4 55,0 45,0, POPAYAN CAUCA 4, , 5, 4, , 4,,, RIONEGRO SANTANDER, 4 40, 59,9, CONFINES SANTANDER, 4 50, 49,, SUTAMARCHAN BOYACA, 4 5, 4,, GUAMAL META 0, 4 49, 50,, MONTERREY CASANARE 9, 4 9,, FACATATIVA CUNDINAMARCA, 4 5,4 4,, TUNJA BOYACA, 4 59, 40,, MUZO BOYACA, , 54,, , 4 5, 4,4, QUINCHIA RISARALDA 5, 4 4,9 5,, SANTUARIO RISARALDA 4, 4 49,0 5,0, COYAIMA TOLIMA, 4 5, 4,9, MANZANARES CALDAS, 4 5, 4,, ICONONZO TOLIMA, 4 9,4 0,, MARINILLA 0, 4,,4, ABEJORRAL 9, 4 4,0 5,0, CALDAS, ,0 5,0, ,9 0, 9,, CHALAN SUCRE,9 4, 5,, MAJAGUAL SUCRE,9 4, 5,, CERRO SAN ANTONIO MAGDALENA 5,9 5, 4,, PELAYA CESAR 4,9 4,9 5,, AYAPEL CORDOBA,9 5,0 44,0, SAN MARTIN CESAR,9,,, SAN JUAN NEPOMUCENO BOLIVAR % de la muestra Rural Urbana % Rural % Urbano % Poblacion Nacional Poblacion Rural del segmento Poblacion Urbana del segmento Poblacion del segmento Tamaño por punto de Rural (0 a ) Tamaño por punto de Urbano ( a ) Rurales Urbanos Poblacion año 004 por punto de Municipio seleccionado del segmento Departamento Segmentos Region Distribucion de la ciudades con 00,000 o menos habitantes 00, ,, 00, GRAN TOTAL 55, , 5, 55, TOTAL, 54 0,4 9,, ,,5,5, VILLA FATIMA* VAUPES, 5, 4,, PUERTO ASIS PUTUMAYO, 4,9 5,, FLORENCIA CAQUETA TERRITORIO S NACIONALES, , 50,4, , 4,4,, RIOFRIO 0, 4 4,0,0, FLORIDA 9, 4,5,5, BUGA, 4 0,4 9,, SANDONA NARIÑO, 4,9,, CUMBITARA NARIÑO, 4 9,, PADILLA CAUCA 5, 4 55,0 45,0, POPAYAN CAUCA 4, , 5, 4, , 4,,, RIONEGRO SANTANDER, 4 40, 59,9, CONFINES SANTANDER, 4 50, 49,, SUTAMARCHAN BOYACA, 4 5, 4,, GUAMAL META 0, 4 49, 50,, MONTERREY CASANARE 9, 4 9,, FACATATIVA CUNDINAMARCA, 4 5,4 4,, TUNJA BOYACA, 4 59, 40,, MUZO BOYACA, , 54,, , 4 5, 4,4, QUINCHIA RISARALDA 5, 4 4,9 5,, SANTUARIO RISARALDA 4, 4 49,0 5,0, COYAIMA TOLIMA, 4 5, 4,9, MANZANARES CALDAS, 4 5, 4,, ICONONZO TOLIMA, 4 9,4 0,, MARINILLA 0, 4,,4, ABEJORRAL 9, 4 4,0 5,0, CALDAS, ,0 5,0, ,9 0, 9,, CHALAN SUCRE,9 4, 5,, MAJAGUAL SUCRE,9 4, 5,, CERRO SAN ANTONIO MAGDALENA 5,9 5, 4,, PELAYA CESAR 4,9 4,9 5,, AYAPEL CORDOBA,9 5,0 44,0, SAN MARTIN CESAR,9,,, SAN JUAN NEPOMUCENO BOLIVAR % de la muestra Rural Urbana % Rural % Urbano % Poblacion Nacional Poblacion Rural del segmento Poblacion Urbana del segmento Poblacion del segmento Tamaño por punto de Rural (0 a ) Tamaño por punto de Urbano ( a ) Rurales Urbanos Poblacion año 004 por punto de Municipio seleccionado del segmento Departamento Segmentos Region Distribucion de la ciudades con 00,000 o menos habitantes

10 Geographical Location of the Sample TERRITORIOS NACIONALES

11 Cuadro No Distribucion de la ciudades con 00,000 o mas habitantes Region Departamento Municipio ATLANTICO BARRANQUILLA 4 ATLANTICO SOLEDAD BOLIVAR CARTAGENA 5 CESAR DUPAR CORDOBA MONTERIA MAGDALENA SANTA MARTA 4 MEDELLIN 0 BELLO CALDAS MANIZALES 4 HUILA NEIVA QUINDIO ARMENIA RISARALDA PEREIRA TOLIMA IBAGUE 4 CUNDINAMARCA SOACHA META VILLAVICENCIO NORTE DE SANTANDER CUCUTA 4 SANTANDER BUCARAMANGA 54 NARIÑO PASTO 4 CALI 9 TOTAL 5 Distribucion de la ciudades con menos de 00,000 habitantes Region Departamento Municipios MUESTRA BOLIVAR SAN JUAN NEPOMUCENO CESAR SAN MARTIN CORDOBA AYAPEL CESAR PELAYA MAGDALENA CERRO SAN ANTONIO SUCRE MAJAGUAL SUCRE CHALAN 9 CALDAS ABEJORRAL MARINILLA TOLIMA ICONONZO CALDAS MANZANARES TOLIMA COYAIMA RISARALDA SANTUARIO RISARALDA QUINCHIA 0 BOYACA MUZO BOYACA TUNJA CUNDINAMARCA FACATATIVA CASANARE MONTERREY META GUAMAL BOYACA SUTAMARCHAN SANTANDER CONFINES SANTANDER RIONEGRO 0 4 CAUCA POPAYAN CAUCA PADILLA 4 NARIÑO CUMBITARA NARIÑO SANDONA 4 BUGA FLORIDA 4 RIOFRIO 4 CAQUETA FLORENCIA TERRITORIOS PUTUMAYO PUERTO ASIS NACIONALES VAUPES VILLA FATIMA* 54 TOTAL GRAN TOTAL.50 TERRITORIOS NACIONALES

12 Final Considerations about the Questionnaire and Field Interviewing About the questionnaire: the questionnaire was long, 45 minutes in average, but in general respondents were willing to answer and we had very few interviews left incomplete. Because of Colombia s internal conflict, some questions were especially sensitive. Some interviewers mentioned that the series on page of the questionnaire generated some discomfort among some respondents. But surprisingly we had very few refusals to answer. About field interviewing: by demand of local authorities, we were not able to apply the survey in the rural areas of Cumbitara, Nariño and Chalán, Sucre, because of the fragile security situation in these towns. In the rest of the country, some of our interviewers were stopped and questioned by illegal groups who finally allowed us to perform the interviewing process. Although many of the respondents were located in areas with heavy presence of illegal groups, there were no reports of pressure over respondents to induce any type of answer. To the contrary, field interviewers emphasized the free will of those who agreed to be part of the study. CNC would like to extend its gratitude to every staff member involved in this study, specially to the brave men and women who defied security warnings and took the risk to perform a very well done job.