National Information and Communication Technology Survey 2010 - Kenya

Abstract
---------------------------

In an effort to address the ICT data challenges, the Communications Commission of Kenya (CCK) partnered with Kenya National Bureau of Statistics (KNBS) to undertake a comprehensive National ICT Survey. This was planned and executed during the months of May and June 2010.

The main objective of the study was to collect, collate and analyse data relating to ICT access and usage by various categorizations in Kenya. The survey captured data and information on critical ICT indicators as defined by international bodies such as the International Telecommunications Union (ITU). These indicators focused on household and individuals; and the data was be disaggregated by age, gender, administrative regions, rural and urban locations.

The specific objectives of the study were to; Obtain social economic information with a view of understanding usage patterns of ICT services;
(a) Obtain social economic information with a view of understanding usage patterns of ICT services;
(b) Collect, collate and analyze ICT statistics in line with ICT indicators;
(c) Evaluate the factors that will have the greatest impact in ensuring access and usage of ICTs and;
(d) Develop a database on access and usage of ICT in Kenya

Geographic coverage
---------------------------

National coverage

Analysis unit
---------------------------

District, Household, Individual

Universe
---------------------------

Households from the sampled areas.

Kind of data
---------------------------

Sample survey data [ssd]

Sampling procedure
---------------------------

The National Sample Survey and Evaluation Programme (NASSEP IV) maintained by the Bureau was used as the sampling frame. The frame has 1,800 clusters spread all over the
country, and covers all socio-economic classes and hence able to get a suitable and representative sample of the population. The survey was distributed into four domains, namely:
1. National,
2. Major Urban areas,
3. Other Urban areas, and
4. Rural areas.

The major urban towns included Nairobi, Thika, Mombasa, Kisumu, Nakuru and Eldoret. All other areas defined as urban by KNBS but fall outside the major municipalities above were
categorized as 'other urban areas'. The rural domain was further sub-divided into their respective provinces, excluding Nairobi which is purely urban. For the 'rural' component, the districts that display identical socio-cultural and economic conditions have been pooled together to create strata from which a representative set of districts is selected to represent the group of such districts. A total of 42 such stratifications were done and one district in each categorization was selected. The major urban areas of the country namely Nairobi, Mombasa, Kisumu, Nakuru, Eldoret and Thika were all sub-stratified into five sub-strata based on perceived levels of income into the:
1. Upper income
2. Lower Upper
3. Middle
4. Lower Middle and
5. Lower.

In this survey, all the six 'major urban' are included while just a few of the 'other urban areas' are selected depending on their population (household) distribution.

Selection of the Clusters for the Survey
The selection of the sample clusters was done systematically using the Equal Probability Selection method (EPSEM). Since NASSEP IV was developed using Probability Proportional to
Size (PPS) method, the resulting sample retains its properties. The selection was done independently within the districts and the urban /rural sub-stratum.

Selection of the Households
From each selected cluster, an equal number of 15 households were selected systematically, with a random start. The systematic sampling method was adopted as it enables the distribution of the sample across the cluster evenly and yields good estimates for the population parameters. Selection of the households was done at the office and assigned to the Research Assistants, with strictly no allowance for replacement of non-responding households.

Sampling deviation
---------------------------

Owing to the some logistical challenges the following clusters were partially or not covered at all:
• One cluster in Tana River due to floods.
• Two clusters in Molo where households shifted to safer areas after the Post Election Violence (PEV). As a result, fewer than the expected households were covered.
• One cluster in Koibatek was covered halfway due to relocation of households to pave way for a large plantation.

Where there was no school found within the cluster, Research Assistant was allowed to sample an institution from a neighbouring cluster. In some districts, the schools were found to be very far from the cluster and therefore could not be covered. Where a cluster was to be covered over a weekend, it was often not possible to find a responsible person in institutions to respond to the questionnaire.

Mode of data collection
---------------------------

Face-to-face [f2f]

Research instrument
---------------------------

Household questionnaire: This will be used to collect background information pertaining to the members of the household and businesses operated by household members. It will collect information about each person in the household such as name, sex, age, education, and relationship to household head etcetera. This information is vital for calculating certain socio-demographic characteristics of the household. The Business module in the household questionnaire will be used to collect information pertaining to usage of ICT in businesses identified in the household. To estimate the magnitude, levels and distribution of ICT usage in the country, all the selected respondents 15 years and above will be subjected to business questionnaire.

Institutional Questionnaire: This will collect information pertaining to institutions providing ICT related programmes in the country. This information will be analyzed to identify gaps and other issues of concern, which need to be addressed in the promotion ICT provision in the country.

Cleaning operations
---------------------------

As a matter of procedure initial manual editing was done in the field by the RAs. The supervisors further checked the questionnaires and validated the data in the field by randomly sampling 20 per cent of the filled questionnaires. After the questionnaires were received from the field, an office editing team was constituted to do office editing.

Data was captured using Census and Survey Processing System (CSPRO) version 4.0 through a data entry screen specially created with checks to ensure accuracy during data entry. All questionnaires were double entered to ensure data quality. Erroneous entries and potential outliers were then verified and corrected appropriately. A total of 20 data entry personnel were engaged during the exercise.

The captured data were exported to Statistical Package for Social Sciences (SPSS) for cleaning and analysis. The cleaned data was weighted before final analysis. The weighting of the data involved application of inflation factors derived from the selection probabilities of the EAs and households detailed in section 2.2.7, on weighting the Sample Data.

Response rate
---------------------------

The overall response rate stood at 85.9 per cent. Nairobi had the lowest response rate at 69.4 per cent while the highest (94.6 per cent) was realized in North Eastern. More than 95.5 per cent of all the sampled households were occupied out of which 85.9 per cent were interviewed.

为应对信息通信技术（ICT）数据挑战，肯尼亚通信委员会（CCK）与肯尼亚国家统计局（KNBS）合作，开展了全面的全国信息通信技术调查。此次调查于2010年5月至6月期间规划并实施。

研究的主要目标是收集、整理和分析与肯尼亚各类群体ICT接入和使用相关的数据。调查涵盖了国际组织如国际电信联盟（ITU）定义的关键ICT指标。这些指标主要集中在家庭和个人层面，并按年龄、性别、行政区域、城乡位置进行数据细分。

具体研究目标包括：
(a) 收集社会经济信息，以了解ICT服务使用模式；
(b) 收集、整理和分析与ICT指标相一致的ICT统计数据；
(c) 评估确保ICT接入和使用的最重要因素；
(d) 建立肯尼亚ICT接入和使用数据库。

地理覆盖范围：全国。
分析单元：区、家庭、个人。
总体范围：样本区域的住户。
数据类型：样本调查数据 [ssd]。
抽样程序：使用国家样本调查与评估计划（NASSEP IV）作为抽样框架。该框架在全国范围内拥有1,800个集群，覆盖所有社会经济阶层，因此能够获取一个合适且具有代表性的样本。调查分为四个领域：
1. 国家；
2. 主要城市地区；
3. 其他城市地区；
4. 乡村地区。
...
数据收集方式：面对面 [f2f]。
研究工具：家庭问卷：用于收集与家庭成员和家庭成员经营的业务相关的背景信息。将收集有关家庭中每个人的信息，如姓名、性别、年龄、教育程度和与户主的关系等。这些信息对于计算家庭的社会人口统计特征至关重要。家庭问卷中的业务模块将用于收集有关家庭中确定的业务ICT使用情况的信息。为了估计国家ICT使用的规模、水平和分布，所有15岁及以上的选定受访者都将接受业务问卷。
...
数据清理操作：...
响应率：总体响应率为85.9%。内罗毕的响应率最低，为69.4%，而最高的是东北部，达到94.6%。所有样本住户中有超过95.5%被占用，其中85.9%接受了访谈。