Kwazulu-Natal Income Dynamics Study 1993-1998, Waves 1-2 - South Africa

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

The 1993 Project for Statistics on Living Standards and Development was an integrated household survey similar in design to a World Bank Living Standards Measurement Survey. The survey collected data on the socio-economic condition of households. Households in Kwazulu-Natal province were re-surveyed from March to June 1998 for the Kwazulu-Natal Income Dynamics Study. Combining these two survey datasets has yielded a panel (or longitudinal) dataset in which the same individuals and households have been interviewed at two points in time, 1993 and 1998. These are the first two waves of the KIDS panel study.

The institutions collaborating in the KIDS study include the University of KwaZulu-Natal (UKZN), the University of Wisconsin-Madison and the International Food Policy Research Institute (IFPRI).

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

The survey covered households in the KwaZulu-Natal Province, on the east coast of South Africa.

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

Households and individuals

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

The Kwazulu Natal Income Dynamics Study 1993-1998 covered all household members.

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

Sample survey data

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

The 1993 sample was selected using a two-stage self-weighting design. In the first stage, clusters were chosen with probability proportional to size from census enumerator subdistricts (ESD) or approximate equivalents where an ESD was not available. In the second stage, all households in each chosen cluster were enumerated and a random sample of them selected. (See PSLSD, 1994, for further details.)
In 1993, the KwaZulu-Natal portion of the PSLSD sample was designed to be representative at the provincial level, conditional on the accuracy of the 1991 census and other information used for the sampling frame, and contained households of all races. Due to the geographic concentration of African and Indian households, KIDS-unlike the PSLSD-limits its scope to African and Indian households. In the KwaZulu-Natal province, Africans represent 85 percent of the population and Indians represent 12 percent. Compared with their representation nationally, White and Coloured people are underrepresented in KwaZulu-Natal. Effectively, the numbers of White and Coloureds in the KwaZulu-Natal sample are too small, and too geographically concentrated in a few clusters, to permit meaningful inference. The KIDS study has thus been limited to the first two population groups.

PSLSD was a survey of households. However, households are a complicated object to define, particularly in longitudinal studies. To transform KIDS from a single-round household survey into a longitudinal household panel study required a redefinition of the sampling unit. In 1998, a decision was made to follow the core household members with the intention of capturing the major decision makers within the household. A household member is a core person if he/she satisfied any of the following criteria (the self-declared head of household from the 1993 survey):

- Spouse/partner of the self-declared head of household (from the 1993 survey)
- Lives in a three generation household and all of the following are true:
- Child of the self-declared household head, son/daughter-in-law of the household head, or niece/nephew of self-declared head
- At least 30 years old
- Has at least one child living in household
- Spouse/partner of person satisfying criteria

Thus all heads of households and spouses of heads are automatically classified as core and in some three-generation households, adult children are also included in this cateogry. In this way, we can see the 1993 survey as the baseline information for a random sample of dynasties. The efforts of the 1998 and 2004 surveyors to find the location of the 1993 core members can then be seen as a way to keep track of the 1993 dynasties.

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

Face-to-face [f2f]

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

KIDS re-interviews the KwaZulu-Natal (KZN) sample of the 1993 nationwide survey known as the Project for Statistics on Living Standards and Development (PSLSD.) The original project was financed by the World Bank and had the characteristics of the Living Standard Measurement Surveys. Reflecting their origin, all three waves of fieldwork for KIDS-1993, 1998, and 2004-collected information on household composition, expenditure on food and on other durable and non-durable goods, education, health, agricultural production, employment, and additional sources of labor and non-labor income. To ensure comparability, the 1998 and 2004 questionnaires largely followed the 1993 version of the questionnaire, however, a few modules have been added and removed. For example, the 1998 survey added sections on assets to marriage, economic shocks, and social capital and trust.

摘要
---------------------------
1993年生活水平与发展统计项目是一项综合的家庭调查，其设计与世界银行生活水平测量调查相似。该调查收集了关于家庭社会经济状况的数据。1998年3月至6月，夸祖鲁纳塔尔省的家庭为夸祖鲁纳塔尔收入动态研究进行了重新调查。将这两个调查数据集相结合，产生了一个面板（或纵向）数据集，其中相同的个人和家庭在两个时间点接受了访谈，即1993年和1998年。这些是KIDS面板研究的头两个波次。

KIDS研究中的合作机构包括夸祖鲁纳塔尔大学（UKZN）、威斯康星大学麦迪逊分校和国际食物政策研究所（IFPRI）。

地理覆盖范围
---------------------------
该调查涵盖了南非东海岸夸祖鲁纳塔尔省的家庭。

分析单位
---------------------------
家庭和个人

总体
---------------------------
夸祖鲁纳塔尔收入动态研究1993-1998覆盖了所有家庭成员。

数据类型
---------------------------
样本调查数据

抽样程序
---------------------------
1993年的样本采用了两阶段自加权设计。在第一阶段，从人口普查户籍员子区（ESD）或近似等效区域中选择簇，这些区域没有ESD。在第二阶段，对每个选择的簇中的所有家庭进行编号，并从中随机选择样本。（参见PSLSD，1994年，以获取更多详细信息。）
1993年，PSLSD样本的夸祖鲁纳塔尔部分旨在在省级水平上具有代表性，条件是1991年人口普查和其他用于抽样框架的信息的准确性，并且包含所有种族的家庭。由于非洲和印度家庭在地理上的集中，KIDS与PSLSD不同，其范围仅限于非洲和印度家庭。在夸祖鲁纳塔尔省，非洲人占85％，印度人占12％。与他们在全国的比例相比，白人和有色人在夸祖鲁纳塔尔省代表性不足。实际上，白人和有色人在夸祖鲁纳塔尔样本中的数量太少，并且在几个簇中地理上过于集中，无法进行有意义的推断。因此，KIDS研究已被限制在头两个种群。
PSLSD是一项家庭调查。然而，家庭是一个复杂的定义对象，特别是在纵向研究中。将KIDS从单轮家庭调查转变为纵向家庭面板研究需要重新定义抽样单位。1998年，做出了一项决定，跟随核心家庭成员，目的是捕捉家庭中的主要决策者。如果一个人满足以下任何一项标准（1993年调查中自我声明的家庭负责人）就被认为是核心成员：
- 自我声明的家庭负责人的配偶/伴侣（1993年调查中）
- 生活在一个三代家庭中，且以下所有条件都成立：
- 自我声明的家庭负责人的子女，家庭负责人的儿媳/女婿，或自我声明的负责人的侄子/侄女
- 至少30岁
- 至少有一个孩子生活在家庭中
- 满足上述标准的个人的配偶/伴侣

因此，所有家庭负责人和配偶都被自动归类为核心成员，在某些三代家庭中，成年子女也被包括在这个类别中。通过这种方式，我们可以将1993年的调查视为一个随机宗族样本的基线信息。1998年和2004年调查员寻找1993年核心成员位置的举措，可以看作是追踪1993年宗族的途径。

数据收集方式
---------------------------
面对面 [f2f]

研究工具
---------------------------
KIDS重新访谈了1993年全国调查的夸祖鲁纳塔尔（KZN）样本，该调查被称为生活水平与发展统计项目（PSLSD）。原始项目由世界银行资助，具有生活水平测量调查的特征。反映其起源，KIDS-1993、1998年和2004年的三个波次现场工作均收集了有关家庭构成、食品和其他耐用及非耐用商品支出、教育、健康、农业生产、就业以及额外劳动和非劳动收入来源的信息。为了确保可比性，1998年和2004年的问卷在很大程度上遵循了1993年问卷的版本，但是增加和删除了一些模块。例如，1998年调查增加了关于资产、婚姻、经济冲击和社会资本及信任的章节。