The Household sample

The Household sample provides an example of hierarchical data. Unlike the Museum and Short Drinks samples, which are based on real surveys, the Household sample has been created artificially to demonstrate hierarchical data. The Household sample is a small data set that includes a very small number of cases at each level. This is to make it easy to understand what happens when you aggregate the data at the different levels.

▪UNICOM Intelligence Data File. This consists of the UNICOM Intelligence metadata model (MDM) file called household.mdd and case data in the form of a UNICOM Intelligence Data File called household.ddf. Like the XML version of the household sample, the .Ex variables from the original Quanvert sample have been removed from the metadata file, but are present in the case data file. By default, this data set is installed into the [INSTALL_FOLDER]\IBM\SPSS\DataCollection\7\DDL\Data\Data Collection File\ folder.

▪Quanvert database. This is a Quanvert levels project. By default, this is installed into the [INSTALL_FOLDER]\IBM\SPSS\DataCollection\7\DDL\Data\Quanvert\Household folder.

▪Surveycraft database. This consists of a HOUSHLD.QDT file and a HOUSHLD.VQ file which are installed by default into the [INSTALL_FOLDER]\IBM\SPSS\DataCollection\7\DDL\Data\Surveycraft folder.

▪XML data set. This consists of an .mdd file (household.mdd) and case data in the form of an XML file, in which the data is stored as HDATA hierarchical tables. This data set is based on the Quanvert database. However, the .mdd file has been modified slightly. For example, the person loop has been expanded and some of the Quanvert-specific information, such as the .Ex variables has been removed. These variables are present in the case data, but are invisible when you connect to the case data using the .mdd file. By default, this data set is installed into the [INSTALL_FOLDER]\IBM\SPSS\DataCollection\7\DDL\Data\XML folder.

▪Household questions. Respondents are first asked a number of questions about their household as a whole, such as the address, age of the building, and number of rooms.

▪Person questions. Respondents are then asked a number of questions about each person in the household, such as the person's name, age, gender, and occupation, and a grid question that asks the number of days he or she watches various TV channels.

▪Overseas trip questions. Respondents are also asked a number of questions about each overseas trip that each person in their household has taken in the previous year (if any), such as the purpose of the trip, number of days he or she was away from home, and countries that were visited.

▪Vehicle questions. Finally, respondents are asked a number of questions about each vehicle that belongs to their household, such as the vehicle's type, color, and annual mileage, and a grid question that asks the respondent to rate the vehicle's features.

Loops called person, trip, and vehicle are used to ask the person, overseas trip, and vehicle questions. The loops are iterated (and the questions are asked) as many times as necessary. For example, in a household of three people, the person loop is iterated three times; in a single-person household, it is iterated once. In a household that has no cars, bikes, or other vehicles, the vehicle questions are not asked at all and the vehicle loop has no iterations.

The structure of the tables in the case data in the hierarchical HDATA view of the data corresponds to the structure of the loops in the metadata. This means that because the trip loop is nested within the person loop, the trip table is a child of the person table. The two grids are also represented in the case data by hierarchical tables, each nested within its parent table.

This diagram shows the case data for two households (the first and third) represented in the HDATA hierarchical tables. (For simplicity, not all of the variables are shown.)

Name	Description	Type
Household	Household serial number	Long
Region	Region of main residence	Single response
Address	Address	Text
Tenure	Do you rent or own your home?	Single response
HouseType	Type of accommodation	Single response
AgeOfBuilding	Approximate age of the building	Single response
NumRooms	Number of rooms	Long
FloorArea	Approximate floor area in square meters	Double
Pets	Number of pets in household	Long
NumPersons	Number of persons in household	Long
NumVehicles	Number of vehicles owned by the people in the household	Long

Name	Description	Type
Person	Person serial number	Long
Age	Age last birthday	Long
Gender	Gender	Single response
Occupation	Occupational status	Single response
Name	Name	Text
Weight	Weight in pounds	Double
Languages	Languages spoken	Multiple response
Newspapers	Number of newspapers and magazines bought last week	Long
TVDays	Number of days on which TV channels were watched last week	Grid
NumTrips	Number of overseas trips	Long

Name	Description	Type
Trip	Trip serial number	Long
Country	Countries visited	Multiple response
DaysAway	The number of days spent away from home	Long
Satisfaction	Overall satisfaction with the trip on a scale of 1 to 10 (where 10 represents the highest satisfaction level).	Double
Modes	Number of different modes of transport used	Long
Purpose	Purpose of the trip	Multiple response

Name	Description	Type
Vehicle	Vehicle serial number	Long
VehicleType	Type of vehicle	Single response
VehicleAge	Age of vehicle	Single response
Maintenance	Approximate annual maintenance costs	Double
DaysUsed	Average number of days the vehicle is used in a week	Single response
Mileage	Approximate annual mileage	Long
YearsOwned	Number of years ownership	Double
Rating	Rating of features	Grid