During a survey conducted in 2023, approximately 54 percent of respondents in 17 Latin American countries claimed to be catholic. Meanwhile, 14.9 percent of the people participating in the survey said they did not profess any religion.In Mexico, more than 69 percent of respondents said they professed Catholicism. Particularly in Honduras, Colombia and Ecuador, most Christians think religion is very important in their lives.

The General Social Surveys (GSS) have been conducted by the National Opinion Research Center (NORC) annually since 1972, except for the years 1979, 1981, and 1992 (a supplement was added in 1992), and biennially beginning in 1994. The GSS are designed to be part of a program of social indicator research, replicating questionnaire items and wording in order to facilitate time-trend studies. This data file has all cases and variables asked on the 2022 GSS.

The 2022 cross-sectional General Social Survey has been updated to Release Version 3a as of May 2024. This Release includes the addition of an oversample of minorities (based on the AmeriSpeak® Panel), household composition and respondent selection data, and post-stratified weights for all years of the GSS.

To download syntax files for the GSS that reproduce well-known religious group recodes, including RELTRAD, please visit the "/research/syntax-repository-list" Target="_blank">ARDA's Syntax Repository.

In 2022, around 31.6 percent of the global population were identify as Christian. Around 25.8 percent of the global population identify as Muslims, followed by 15.1 percent of global populations as Hindu.

The World Religion Project (WRP) aims to provide detailed information about religious adherence worldwide since 1945. It contains data about the number of adherents by religion in each of the states in the international system. These numbers are given for every half-decade period (1945, 1950, etc., through 2010). Percentages of the states' populations that practice a given religion are also provided. (Note: These percentages are expressed as decimals, ranging from 0 to 1, where 0 indicates that 0 percent of the population practices a given religion and 1 indicates that 100 percent of the population practices that religion.) Some of the religions (as detailed below) are divided into religious families. To the extent data are available, the breakdown of adherents within a given religion into religious families is also provided.

The project was developed in three stages. The first stage consisted of the formation of a religion tree. A religion tree is a systematic classification of major religions and of religious families within those major religions. To develop the religion tree we prepared a comprehensive literature review, the aim of which was (i) to define a religion, (ii) to find tangible indicators of a given religion of religious families within a major religion, and (iii) to identify existing efforts at classifying world religions. (Please see the original survey instrument to view the structure of the religion tree.) The second stage consisted of the identification of major data sources of religious adherence and the collection of data from these sources according to the religion tree classification. This created a dataset that included multiple records for some states for a given point in time. It also contained multiple missing data for specific states, specific time periods and specific religions. The third stage consisted of cleaning the data, reconciling discrepancies of information from different sources and imputing data for the missing cases.

The Global Religion Dataset: This dataset uses a religion-by-five-year unit. It aggregates the number of adherents of a given religion and religious group globally by five-year periods.

Abstract

The Religion Battery is a consolidated list of items focused on religion in the United States. The dataset includes responses from 1999-2024.

Methodology

The Religion Battery leverages the same methodology as the Gallup Poll Social Series (GPSS).

Gallup interviews a minimum of 1,000 U.S. adults aged 18 and older living in all 50 states and the District of Columbia using a dual-frame design, which includes both landline and cellphone numbers. Gallup samples landline and cellphone numbers using random-digit-dial methods. Gallup purchases samples for this study from Survey Sampling International (SSI). Gallup chooses landline respondents at random within each household based on which member had the next birthday. Each sample of national adults includes a minimum quota of 70% cellphone respondents and 30% landline respondents, with additional minimum quotas by time zone within region. Gallup conducts interviews in Spanish for respondents who are primarily Spanish-speaking.

Gallup weights samples to correct for unequal selection probability, nonresponse, and double coverage of landline and cellphone users in the two sampling frames. Gallup also weights its final samples to match the U.S. population according to gender, age, race, Hispanic ethnicity, education, region, population density, and phone status (cellphone only, landline only, both, and cellphone mostly).

Demographic weighting targets are based on the most recent Current Population Survey figures for the aged 18 and older U.S. population. Phone status targets are based on the most recent National Health Interview Survey. Population density targets are based on the most recent U.S. Census.

Usage

Previous versions of the Religion Battery have more rows and columns than the current version (v. 3.0). This is because the previous data releases contained fields unrelated to religion. The current release was cleaned/streamlined to reflect the topic of interest and isolate the surveys related to that topic.

For more information about included variables, please see

Supporting Files.

Bulk Data Access

Data access is required to view this section.

Roman Catholic is the most common religion affiliation in Peru. In a survey carried out in 2023, almost 64.5 percent of Peruvian respondents claimed to be of catholic faith, whereas the second most chosen religion was Evangelism, with around 19 percent of the people interviewed.

In 2023, North Korea was the country with the strongest suppression of Christians with an index value of 97.8. Somalia followed behind with 91.6, with Yemen in third.

The World Watch List The index ranges from 0 (complete freedom) to 100 (total suppression). The survey for the World Watch List included various aspects of religious freedom: the legal and official status of Christians, the actual situation of Christians living in the country, regulations from the state as well as factors that can undermine the freedom of religion in a country. Christianity worldwide Even though Christianity is the largest religion worldwide, there are many areas of the world where Christians are persecuted. As the list shows, this is especially the case in countries with larger Muslim populations. Moreover, these countries are characterized by their authoritarian nature. The global distribution of religions varies strongly; whereas almost 100 percent of Hindus and Buddhists are found in the Asia-Pacific, Christians are spread around most world regions except the Middle East and North Africa.

According to a survey conducted in South Korea in 2023, over 60 percent of respondents reported no religious affiliation, while approximately 20 percent identified as Christians and 16 percent as Buddhists. Religious population South Korea is a multi-religious society where Christianity, Buddhism, and various other religions coexist with shamanism. According to a previous study, the domestic religious population appeared to decline over time after reaching its peak in 2005, at nearly 25 million people. In contrast, the share of people who are religiously unaffiliated has increased in recent years. Within the last two decades, the religiously unaffiliated population has increased from about 40 percent to more than 60 percent. Shamanism Shamanism has continued to significantly influence the daily lives of many South Koreans. According to a survey conducted in 2023, about 40 percent of respondents reported having consulted a fortune-teller within the past year. Roughly 80 percent of those respondents were already affiliated with a religion.

The survey contained questions about church and religion. The respondents were asked whether they were members of a church or other religious community, and whether they were members of the Evangelical Lutheran Church of Finland. They were asked to estimate how significant the various reasons for being a member of the church were to them, and whether they had ever thought about resigning from it. Those who were not members of the Evangelical Lutheran Church were asked whether they had ever thought about joining it. The respondents were also asked to present their views on the Evangelical Lutheran Church with the help of a number of opposite adjective pairs (e.g. honest - dishonest, rich - poor, reliable - unreliable, tolerant - intolerant, efficient - inefficient). Furthermore, the respondents were asked how often they participated in various congregational activities, such as church services, communion, church concerts or other religious meetings, and how often they prayed, read the Bible or the New Testament, watched religious programmes on television, or listened to them in the radio. Views were also probed on the afterlife, the existence of God, the creation of the world, and Jesus' status as the Son of God. The respondents were presented with a list of duties belonging to the Finnish Evangelical Lutheran Church and its congregations, and asked to estimate how well they had fulfilled these obligations. Opinions were also charted on different religions and religious denominations, religious upbringing, Jesus, the Bible, the ordination of women, and the liturgical reform in Finland. Background variables included respondent's age, gender, education, occupational group, income group, province of residence, and type of municipality.

The General Social Surveys (GSS) have been conducted by the National Opinion Research Center annually since 1972, except for the years 1979, 1981, and 1992 (a supplement was added in 1992), and biennially beginning in 1994. The GSS are designed as part of a program of social indicator research, replicating questionnaire items and wording in order to facilitate time-trend studies. Items on religion in the 1980 GSS include religious preference, church attendance, beliefs about life after death, and attitudes toward organized religion. In addition, the 1980 GSS contains a module on voluntary organization membership, including a measure of membership in church-affiliated groups.

To download syntax files for the GSS that reproduce well-known religious group recodes, including RELTRAD, please visit the "/research/syntax-repository-list" Target="_blank">ARDA's Syntax Repository.

This data set contains the raw data used for the analysis of the research article “Religion, social desirability bias and financial inclusion: Evidence from a list experiment on Islamic (micro-)finance”. In particular, it comprises information regarding the attitudes of the Muslim poor towards the use of non-Islamic financial products and services, based on survey data and list experiment data of 2,145 borrowers of an Islamic microfinance institution (Akhuwat) from the city of Multan, Punjab Province, Pakistan, collected in 2017. Data collection took place in Urdu.

This study is Pew Research Center's most comprehensive, in-depth exploration of India to date. For this report, Pew surveyed 29,999 Indian adults (including 22,975 who identify as Hindu, 3,336 who identify as Muslim, 1,782 who identify as Sikh, 1,011 who identify as Christian, 719 who identify as Buddhist, 109 who identify as Jain and 67 who identify as belonging to another religion or as religiously unaffiliated). Interviews for this nationally representative survey were conducted face-to-face under the direction of RTI International from November 17, 2019, to March 23, 2020. Respondents were surveyed about religious beliefs and practices, religious identity, nationalism, and tolerance in Indian society.

The majority of Japanese adhere to Shintoism, a traditional Japanese religion focusing on rituals and worship at shrines. In 2021, around 48.6 percent of the total population of Japan participated in Shinto practices. Closely behind is Buddhism, with more than 46 percent of the population adhering to its practices. Most Japanese thus practice both religions. The original ShintoShintoism adherents worship spirits or gods at shrines, often publicly, through rituals and traditions. Shintoism is deeply ingrained in the Japanese culture and way of life, but the number of followers has been decreasing over the past years, as has the number of Japanese who are adhering to any kind of religion. Tradition seems to have to make way for modern attitudes and choices – still the number of people who call themselves Shinto adherents is quite high, even if they do not actively practice it. Everything zen in BuddhismWhile Shintoism seems to become a thing of the past, the number of Buddhists, on the other hand, has been stable over the last decade, and if anything, has only slightly decreased. Japan is, in fact, among the countries with the largest Buddhist communities, right after China, of course, and Thailand. This might be due to Buddhism being able to adapt much easier to modern times and its adherents’ everyday lives, as well as a better PR machine – Buddhism is, after all, also quite popular in the Western world.

The work plan activities in Kiribati related to the updating of the listing of all households and institutions in Kiribati is to produce a sex and age disaggregated population count that forms the basis for a sampling frame for the upcoming Social Indicator Survey (SIS) and Household Income and Expenditure Survey (HIES). It also serves the purpose of digitalising and harmonising enumeration areas (EAs) to facilitate random sampling and census planning. To achieve this, SPC was engaged to conduct the following activities:

1. Planning and budgeting: prepare a comprehensive plan and budget for the household listing.
2. Mapping: prepare field maps to be used in the listing; digitalise EA boundaries and harmonisation of new EA framework; training and capacity building of the Ministry of Environment, Lands and Agricultural Development; prepare maps for the selected EAs in the SIS.
3. Listing questionnaire design, enumerator training and technology: develop a tablet-based household listing questionnaire and associated training resources, and set up of technology (e.g., server, tablet interviewer application, backup protocols); support Kiribati's National Statistics Office (KI-NSO) to conduct training of enumerators in all aspects of the collection; and administer South-South support to Kiribati for the duration of the listing.
4. Sample design: design the sample and field plan for the SIS; and build capacity of KI-NSO in sample design and field work planning.

Version 01: Cleaned, labelled and anonymized version of the Master file.

-HOUSEHOLDS/INSTITUTIONS: Household ID and building type.
-INDIVIDUALS: ID, age, sex, religion.

• Collection start: 2018
• Collection end: 2018

Christianity was the largest religion in a high number of the countries included in the survey. Of the countries, Peru, South Africa, and Poland had the highest share of Christians at around 75 percent. Moreover, around 90 percent in India and Thailand stated that they believed in another religion, with Hinduism and Buddhism being the major religion in the two countries respectively. Sweden and South Korea were the only two countries where 50 percent or more of the respondents stated that they did not have any religious beliefs.

DRAKO is a Mobile Location Audience Targeting provider with a programmatic trading desk specialising in geolocation analytics and programmatic advertising. Through our customised approach, we offer business and consumer insights as well as addressable audiences for advertising.

Mobile Location Data can be meaningfully transformed into Audience Targeting when used in conjunction with other dataset. Our expansive POI Data allows us to segment users by visitation to major brands and retailers as well as categorizes them into syndicated segments. Beyond POI visits, our proprietary Home Location Model determines residents of geographic areas such as Designated Market Areas, Counties, or States. Relatedly, our Home Location Model also fuels our Geodemographic Census Data segments as we are able to determine residents of the smallest census units. Additionally, we also have audiences of: ticketed event and venue visitors; survey data; and retail data.

All of our Audience Targeting is 100% deterministic in that it only includes high-quality, real visits to locations as defined by a POIs satellite imagery buildings contour. We never use a radius when building an audience unless requested. We have a horizontal accuracy of 5m.

Additionally, we can always cross reference your audience targeting with our syndicated segments:

Overview of our Syndicated Audience Data Segments: - Brand/POI segments (specific named stores and locations) - Categories (behavioural segments - revealed habits) - Census demographic segments (HH income, race, religion, age, family structure, language, etc.,) - Events segments (ticketed live events, conferences, and seminars) - Resident segments (State/province, CMAs, DMAs, city, county, sub-county) - Political segments (Canadian Federal and Provincial, US Congressional Upper and Lower House, US States, City elections, etc.,) - Survey Data (Psychosocial/Demographic survey data) - Retail Data (Receipt/transaction data)

All of our syndicated segments are customizable. That means you can limit them to people within a certain geography, remove employees, include only the most frequent visitors, define your own custom lookback, or extend our audiences using our Home, Work, and Social Extensions.

In addition to our syndicated segments, we’re also able to run custom queries return to you all the Mobile Ad IDs (MAIDs) seen at in a specific location (address; latitude and longitude; or WKT84 Polygon) or in your defined geographic area of interest (political districts, DMAs, Zip Codes, etc.,)

Beyond just returning all the MAIDs seen within a geofence, we are also able to offer additional customizable advantages: - Average precision between 5 and 15 meters - CRM list activation + extension - Extend beyond Mobile Location Data (MAIDs) with our device graph - Filter by frequency of visitations - Home and Work targeting (retrieve only employees or residents of an address) - Home extensions (devices that reside in the same dwelling from your seed geofence) - Rooftop level address geofencing precision (no radius used EVER unless user specified) - Social extensions (devices in the same social circle as users in your seed geofence) - Turn analytics into addressable audiences - Work extensions (coworkers of users in your seed geofence)

Data Compliance: All of our Audience Targeting Data is fully CCPA compliant and 100% sourced from SDKs (Software Development Kits), the most reliable and consistent mobile data stream with end user consent available with only a 4-5 day delay. This means that our location and device ID data comes from partnerships with over 1,500+ mobile apps. This data comes with an associated location which is how we are able to segment using geofences.

Data Quality: In addition to partnering with trusted SDKs, DRAKO has additional screening methods to ensure that our mobile location data is consistent and reliable. This includes data harmonization and quality scoring from all of our partners in order to disregard MAIDs with a low quality score.

The American Religious Identification Survey (ARIS) 2008 is the third in a landmark time series of large, nationally representative surveys that track changes in the religious loyalties of the U.S. adult population within the 28 contiguous states from 1990 to 2008. The "/data-archive?fid=ARIS2001" Target="_blank">2001 and 2008 surveys are replicas of the "/data-archive?fid=NSRI1990" Target="_blank">1990 survey, and are led by the same academic research team using an identical methodology of random-digit-dialed telephone interviews (RDD) and the same unprompted, open-ended key question "What is your religion, if any?" Interviewers did not prompt or offer a suggested list of potential answers. Moreover, the self-description of respondents was not based on whether established religious bodies or institutions considered them to be members. To the contrary, the surveys sought to determine whether the respondents regarded themselves as adherents of a religious community. The surveys tap subjective, rather than objective, standards of religious identification. The value of this unique series of national surveys, which allows scientific monitoring of change over time, has been recognized by the U.S. Bureau of the Census The Bureau itself is constitutionally precluded from such an inquiry into religion, and so has incorporated NSRI/ARIS findings into its official publication the Statistical Abstract of the United States since 2003.

Abstract

The 1993 Indonesia Family Life Survey (IFLS) provides data at the individual and family level on fertility, health, education, migration, and employment. Extensive community and facility data accompany the household data. The survey was a collaborative effort of Lembaga Demografi of the University of Indonesia and RAND, with support from the National Institute of Child Health and Human Development, USAID, Ford Foundation, and the World Health Organization. In Indonesia, the 1993 IFLS is also referred to as SAKERTI 93 (Survai Aspek Kehidupan Rumah Tangga Indonesia). The IFLS covers a sample of 7,224 households spread across 13 provinces on the islands of Java, Sumatra, Bali, West Nusa Tenggara, Kalimantan, and Sulawesi. Together these provinces encompass approximately 83 percent of the Indonesian population and much of its heterogeneity. The survey brings an interdisciplinary perspective to four broad topic areas:

• fertility, family planning, and contraception • infant and child health and survival • education, migration and employment • the social, economic, and health status of adults, young and old

Additionally, extensive community and facility data accompany the household data. Village leaders and heads of the village women's group provided information in each of the 321 enumeration areas from which households were drawn, and data were collected from 6,385 schools and health facilities serving community residents.

Geographic coverage

National coverage

Analysis unit

• Communities
• Facilities
• Households
• Individuals

Universe

Household Survey data were collected for household members through direct interviews (for adults) and proxy interviews (for children, infants and temporarily absent household members). The IFLS-1 conducted detailed interviews with the following household members: - The household head and their spouse - Two randomly selected children of the head and spouse aged 0 to 14 (interviewed by proxy) - An individual age 50 and above and their spouse, randomly selected from remaining members - For a randomly selected 25 percent of the households, an individual age 15 to 49 and their spouse, randomly selected from remaining members.

The Community and Facility Survey collected data from a variety of respondents including: the village leader and his staff and the leader of the village women's group; Ministry of Health clinics and subclinics; private practices of doctors, midwives, nurses, and paramedics; community-based health posts and contraceptive distribution centers; public, private, and religious elementary schools; public, private, and religious junior high schools; public, private, and religious senior high schools. Unlike many other surveys, the sample frame for the survey of facilities was drawn from the list of facilities used by household survey respondents in the area.

Kind of data

Sample survey data [ssd]

Sampling procedure

The Household Survey Sampling Procedure

The household survey component of the 1993 IFLS was designed to collect contemporaneous and retrospective information on a wide array of family life topics for a representative sample of the Indonesian population. In IFLS1 it was determined to be too costly to interview all household members, so a sampling scheme was used to randomly select several members within a household to provide detailed individual information. IFLS1 conducted detailed interviews with the following household members: - the household head and his/her spouse - two randomly selected children of the head and spouse age 0 to 14 - an individual age 50 or older and his/her spouse, randomly selected from remaining members, and - for a randomly selected 25% of the households, an individual age 15 to 49 and his/her spouse, randomly selected from remaining members.

Household Selection The IFLS sampling scheme stratified on provinces, then randomly sampled within provinces. Provinces were selected to maximize representation of the population, capture the cultural and socioeconomic diversity of Indonesia, and be cost effective given the size and terrain of the country. The far eastern provinces of East Nusa Tenggara, East Timor, Maluku and Irian Jaya were readily excluded due to the high costs of preparing for and conducting fieldwork in these more remote provinces. Aceh, Sumatra's most northern province, was deleted out of concern for the area's political violence and the potential risk to interviewers. Finally, due to their relatively higher survey costs, we omitted three provinces on each of the major islands of Sumatra (Riau, Jambi, and Bengkulu), Kalimantan (West, Central, East), and Sulawesi (North, Central, Southeast). The resulting sample consists of 13 of Indonesia's 27 provinces: four on Sumatra (North Sumatra, West Sumatra, South Sumatra, and Lampung), all five of the Javanese provinces (DKI Jakarta, West Java, Central Java, DI Yogyakarta, and East Java), and four provinces covering the remaining major island groups (Bali, West Nusa Tenggara, South Kalimantan, and South Sulawesi). The resulting sample represents 83 percent of the Indonesian population. (see Figure 1.1 of the Overview and Field Report in External Documents). Table 2.1 of the same document shows the distribution of Indonesia's population across the 27 provinces, highlighting the 13 provinces included in the IFLS sample.

The IFLS randomly selected enumeration areas (EAs) within each of the 13 provinces. The EAs were chosen from a nationally representative sample frame used in the 1993 SUSENAS, a socioeconomic survey of about 60,000 households.The SUSENAS frame, designed by the Indonesian Central Bureau of Statistics (BPS), is based on the 1990 census.The IFLS was based on the SUSENAS sample because the BPS had recently listed and mapped each of the SUSENAS EAs (saving us time and money) and because supplementary EA-level information from the resulting 1993 SUSENAS sample could be matched to the IFLS-1 sample areas.Table 2.1 summarizes the distribution of the approximately 9,000 SUSENAS EAs included in the 13 provinces covered by the IFLS. The SUSENAS EAs each contain some 200 to 300 hundred households, although only a smaller area of about 60 to 70 households was listed by the BPS for purposes of the annual survey. Using the SUSENAS frame, the IFLS randomly selected 321 enumeration areas in the 13 provinces, over-sampling urban EAs and EAs in smaller provinces to facilitate urbanrural and Javanese-non-Javanese comparisons. A straight proportional sample would likely be dominated by Javanese, who comprise more than 50 percent of the population. A total of 7,730 households were sampled to obtain a final sample size goal of 7,000 completed households. Table 2.1 shows the sampling rates that applied to each province and the resulting distribution of EAs in total, and separately by urban and rural status. Within a selected EA, households were randomly selected by field teams based upon the 1993 SUSENAS listings obtained from regional offices of the BPS. A household was defined as a group of people whose members reside in the same dwelling and share food from the same cooking pot (the standard BPS definition). Twenty households were selected from each urban EA, while thirty households were selected from each rural EA. This strategy minimizes expensive travel between rural EAs and reduces intra-cluster correlation across urban households, which tend to be more similar to one another than do rural households. Table 2.2 (Overview and Field Report) shows the resulting sample of IFLS households by province, separately by completion status.

Selection of Respondents within Households For each household selected, a representative member provided household-level demographic and economic information. In addition, several household members were randomly selected and asked to provide detailed individual information.

The Community Survey Sampling Procedure

The goal of the CFS was to collect information about the communities of respondents to the household questionnaire. The information was solicited in two ways. First, the village leader of each community was interviewed about a variety of aspects of village life (the content of this questionnaire is described in the next section). Information from the village leader was supplemented by interviewing the head of the village women's group, who was asked questions regarding the availability of health facilities and schools in the area, as well as more general questions about family health in the community. In addition to the information on community characteristics provided by the two representatives of the village leadership, we visited a sample of schools and health facilities, in which we conducted detailed interviews regarding the institution's activities.

A priori we wanted data on the major sources of outpatient health care, public and private, and on elementary, junior secondary, and senior secondary schools. We defined eight strata of facilities/institutions from which we wanted data. Different types of health providers make up five of the strata, while schools account for the other three. The five strata of health care providers are: government health centers and subcenters (puskesmas, puskesmas pembantu); private doctors and clinics (praktek umum/klinik); the private practices of midwives, nurses, and paramedics (perawats, bidans, paramedis, mantri); traditional practitioners (dukun, sinshe, tabib, orang pintar); and community health posts (posyandu, PPKBD).The three strata of schools are elementary, junior secondary, and senior secondary. Private, public, religious, vocational, and general schools are all eligible as long as they provide schooling at one of the three levels.

Our protocol for selecting specific

Abstract

The EPC-Customer Satisfaction Survey 2014 collected information to obtain and establish a baseline for customer's satisfaction on the EPC services and to identify the areas of the corporation's services that need improvement. The CSS results are planned to provide updated information to design new strategies for improving the services of the corporation. The overall outcome of the CSS 2014 is to assist and recommend relevant strategies to improve and upgrade the service of the EPC to its clients.The Customer Satisfaction Survey 2014 was conducted on the domestic or household level as well as all the other types of customers registered with the corporation.

Geographic coverage

National Regional

Analysis unit

EPC customers such as domestic, commercial, school, religion, government, industrial, hotels

Universe

EPC users or customers

Kind of data

Sample survey data [ssd]

Sampling procedure

There were seven types of customers, namely: domestic, commercial, religion, school, government, hotel and industrial in the EPC frame or their list of population which was given to SBS for sampling selection. It took several months for both parties to sort the list of registered customers with the corporation especially the domestic clients, so that they can be easily searched and identified during the field work or data collection period, therefore the SBS offered it list of households as part of domestic customers for the EPC to avoid the delay with the survey timeframe.

The total number of households with SBS was 26,205 which were counted from the latest census of population and housing 2011. Out of that total households with SBS, 25,262 or 96percent of households were with electricity. The total number of customers proposed by the corporation was about 200 in which 100 from the domestic and 80 from the other types of customers, however to accommodate the non response cases, the SBS increased the sample size to 250 in which 150 were from household or domestic customers and 100 from other types of customers.

Household/domestic sample

The sample of domestic customers for the CSS 2014 was drawn from the master sample frame of the list of occupied households compiled in the most recent Population and Housing Census 2011. The sample size was based on a 95 percent confidence interval of ± 5 percent margin of error. This means that if the survey found that 50 percent of respondents were satisfied with induction meter services of EPC, we could be 95 percent sure of getting the same result had we interviewed everyone in the population give or take 5 percent. An 80 percent response rate and a design-effect of 1.2 was used to allow for clustering of the complex design. After taking into account all those features, it resulted in the required sample size of 150 selected households.

In national statistical surveys, the region of Apia Urban Area (AUA) represented the urban population while the regions of North West Upolu (NWU), Rest of Upolu (ROU) and Savaii represented the rural population. Therefore in order to achieve the sample size of 150 for the domestic customers, a representative probability sample of households was selected in two stages.

The first stage involved the selection of clusters or enumeration area (EAs) from the master sample frame using stratified systematic sampling with probability proportional to size. A total of 30 primary sampling units or clusters were selected in which 6 clusters were from the urban areas and 24 clusters were selected from the rural areas. The design did not allow for replacement of clusters or households.

In the second stage, a total of 5 households were selected from each cluster using systematic equal probability selection. Normally an updated household listing from selected clusters could have been done to select 5 households. However, due to the delay in sorting of customers list and it was towards the end of the year, and the fact that the census 2011 was just completed in the previous three years, it was seen not necessary to conduct a fresh household listing which would have taken SBS another two months to carry out causing delay to the survey.

Other Types of Customers

The sample for the CSS other types of customers such as commercial, religion, school, government, hotel and industrial was drawn from the master sample frame of the list of all the 3767 customers registered with the EPC . The commercial type has 2587 customers, religion with 751customers, school with 229 customers, governments with 118 customers, hotels with 75 customers and industries with 47 customers .The sample size was based on a 95 percent confidence interval of ±5 percent margin of error, assuming an 80 percent response rate. To achieve a representative probability sample, the systematic method was used to select the 100 customers of other 6 types apart from the domestic customers.

Mode of data collection

Face-to-face [f2f]

Research instrument

A structured English questionnaire was prepared by the EPC team to collect the feedback from the corporation's customers. However, SBS made some improvements in terms of instructions between questionnaire sections in order to make the interviewing flow properly from beginning to end. The questionnaire was also translated into the Samoan language to complement the English questionnaire so that the interpretation of questions by the field enumerators was consistent in the field. A cover page of the questionnaire was also developed so that selected customer's identifications were clearly noted. The options for the survey status were also listed to account for non-coverage of EPC customers during the fieldwork. The Survey Questionnaire consists of four sections with a cover page in the beginning for the Identification of selected households and other types of customers. Section A has seven questions about the type of meters the customers used and the service provided by EPC to pay bills and buy cash power units, and open questions to state some reasons why the customers were not satisfied with the service given by the EPC, areas of paying of electricity bills and selling cash power units. Section B contains five questions on the management of complaints lodged to the corporation and the satisfaction of service provided. Section C asked two types of questions in which one was a rating question on the perception of the customers of the EPC service, and the second was a ranking question of the mediums that the public used to get EPC public awareness. Section D was open for the customers to list any of their comments about the service of EPC for improvement.

Cleaning operations

After coding, the computer data program was created using CSPro 5 software for data entry. After testing the program, the data entry was conducted in one week (March 24th-28th). The data editing, cleaning and weighting of the data took another two weeks (April 1st-11th) to complete, leaving three weeks (April 14th - May 2nd) to analyse and write the analysis report to meet the deadline.

Data editing was done using writing option in CSPro 5.0.

Response rate

A total of 150 households were selected to represent the domestic customers and 139 households were occupied during the field work period. Of the occupied households only 133 were successfully interviewed resulting in a household response rate of 95.7 percent. The 6 households which were selected but not able to answer the questionnaire because of they had no access to electricity during the survey period; most of them were in the island of Savaii.

For other types of EPC customers in which 100 were selected, only 97 customers were found during the survey period. From these customers, 94 were able to complete the survey while the others were no longer operating due to the following reasons: one was destroyed by tsunami, another changed its customer type, and the last was not in the location previously identified in the list of sample respondents.

This is explained in the final analytical report.

Sampling error estimates

Any survey will be affected by sampling errors and non-sampling errors. The latter is difficult to measure but can be greatly reduced by the application of high quality survey management, efficient field supervisions, skilful enumerators, good control of data coding and data processing, sufficient resources, etc. Sampling errors are usually calculated using relevant sampling estimation formulae and computer programs. For the CSS 2014, the variance formula for complex design was used to calculate sampling errors. Dr Ren Ruilin of ICF Macro developed specific sampling error estimation templates in Excel for use by developing countries like Samoa where expensive computer programs like SAS could not be purchased. The Excel templates used the Taylor linearization method of variance estimation for survey estimates like means and proportions. The design effect (DEFT) for each estimate was also calculated whereby a DEFT value of 1.0 indicates that the complex design used was just as efficient as the simple random sampling and a value more than 1 indicates an increase in sampling error due to the design and vice versa. In addition, the confidence limits of 95 percent can also be estimated for each variable which provides the range of values for which the true value falls.

Details of sampling errors are presented in the sampling errors appendix of the report.