Examples on categorical variables
The data used in the following
examples are part of those collected in a questionnaire study of neighbourly
help (see Webley &
Lea 1993, Human Relations 46, 65-76). On the next page you will
find an extract from the questionnaire. People living in different districts
were sent different coloured questionnaires, so we knew when the forms
came back where they had come from. The corresponding data (including a
variable giving the district the respondent lived in) are stored in the
Singer file /singer1/eps/psybin/stats/neighbor.MTW
The data are coded as follows. District names are replaced by numbers,
and these occupy the last column of the worksheet. The remainder of the
columns each correspond to one question. For each question, the first answer
category was given code 1, the second code 2, and so on.
The four districts can be briefly described as follows. District 1 was
a long-established residential area near the city centre, with housing
dating from the late nineteenth century. Originally working class, it now
has a considerable middle class population with some student and other
temporary accommodation. District 2 was a working-class housing estate
dating from the 1930s, with mainly rented accommodation but some owner
occupation. District 3 was the oldest part of a more recently developed,
mainly middle-class, almost exclusively owner-occupied estate, dating from
the 1960s. District 4 was the most recently developed part of a more sought-after
middle-class residential area, with smaller but almost entirely owner-occupied
properties dating from the 1970s and 1980s.
- Retrieve this worksheet into Minitab. Use INFO to examine what variables
it contains, and DESCRIBE or HIST to get an idea of the distribution of
responses between the categories.
- All the variables in this study are categorical, but some of them consist
of ordered categories, and some of them of unordered categories that is,
some of them are measured at the ordinal level and some of them at the
nominal level. Which are which?
- Use multiple regression, including dummy variables where appropriate,
to find out how people's ratings of the neighbourliness of the area where
they live are related to the answers they gave to the other questions and
to the district where they now live.
- With all the other variables taken into account, are the following
variables significantly associated with rated neighbourliness?
- age group
- number of people known by name
- the district where people now live
- Can you relate the rated neighbourliness of a district to its social/demographic
characteristics?
NEIGHBOURLINESS SURVEY
(Extract)
About how long have you lived where you do now?
Less than 6 months
6-12 months
1-3 years
3-10 years
Over 10 years
Where were you living before you moved to your present house?
In the same neighbourhood?
Elsewhere in Exeter
Elsewhere in Devon
Elsewhere in Britain
Abroad
How neighbourly do you think the area where you now live is?:
Very unfriendly
Not very friendly
About average
Fairly friendly
Very friendly
Roughly how many people in your street, or in the streets just
near you, do you know the names of?
None
1-5
6-20
More than 20
How many of those people (not counting children) would you call by
their first names?
None
1-5
6-20
More than 20
Your sex:
Male
Female
Your age:
Under 18
18-30
31-50
51-65
Over 65
Stephen Lea
University of Exeter
Department of Psychology
Washington Singer Laboratories
Exeter EX4 4QG
United Kingdom
Tel +44 1392 264626
Fax +44 1392 264623
Send questions and comments to the
departmental administrator or to the
author of this page
Goto Home page for this
course | handout
for this topic | dummy
test paper
Goto home page for: University of
Exeter | Department of
Psychology | Staff
| Students |
Research | Teaching
| Miscellaneous
(access count since 1st January 1997).
Document revised 7th February 1997