GAP Toolkit 5: Types of Question and Types of Variable

GAP Toolkit 5
Training in basic drug abuse data management

and analysis
Training session 4
Types of question and types of

variable
Objectives
• Define a range of classifications for questions and

variables
• Discuss the use of levels of measurement in defining
variables in SPSS
Types of question
• Closed, open
• “Factual” and attitudinal
Closed questions
• The respondent selects from a list of mutually exclusive

and collectively exhaustive answers
• The answers are pre-coded
Example
• Has the patient been in treatment prior to this episode?

 Yes (1)
 No (0)
Example
• In the last 30 days, how many times (if any) have you
had 5 or more drinks in a row?
 None
1
2
 3-5
 6-9
 10 or more
“Other” Category
• An option on all but the simplest closed questions

• Ensures the list of options are exhaustive
• Allows flexibility in response
• Post-coded rather than pre-coded
Example
• Type of centre:
 Specialized treatment centre
 Therapeutic community
 General hospital
 Psychiatric hospital/unit
 Other (specify): …………………………..
Dichotomous questions
• A subset of closed questions

• There are only two possible answers
• The answers are mutually exclusive and collectively
exhaustive
Examples
1. Gender:
 Male
 Female
2. Has the patient been in treatment prior to this episode?
 Yes
 No
Multiple-response questions
• The question allows more than one response

• The categories are not mutually exclusive
• Frequently, a grouping of dichotomous closed questions
Example
• Mode of ingestion of primary substance

(X all that apply):
 Swallow
 Smoke
 Snort
 Inject
 Other (specify): ……………………………….
Likert Scales
• A type of closed question

• Designed to measure attitudes
Example
• Do you disapprove of people doing each of the
following:
– Trying marijuana once or twice
 Don’t approve
 Disapprove
 Strongly disapprove
 Don’t know
– Smoking marijuana occasionally
• (options repeated)
Open questions
• There are no constraints on the respondent’s answer

• The answers cannot be predicted before the
questionnaires are presented
• The answers must be coded after the questionnaires
are collected
Examples
1. Q30. Which new drugs or new patterns of use have

been reported?
2. Q13. Indicate primary substance of abuse, that is, the
most frequently used
3. Other (specify): ……………………..
Exercise: discussion
• Do Open or Closed questions appear more frequently in

the questionnaires used by your specific focal group?
Give reasons/possible explanations for these choices.
Response types
• Factual/attitudinal
• Direct/indirect
Types of variable
• Levels of measurement
• Types of variation
• Categorical vs. continuous
Levels of measurement
• Nominal
• Ordinal
• Interval
• Ratio
Nominal
• The data describe an attribute

• The set of possible values the variable can contain are
mutually exclusive and collectively exhaustive
categories
• The categories cannot be objectively measured
against each other
Examples: nominal data
• Gender: male and female

• Location: urban and rural
• Religion: Christian, Hindu, Muslim, Jew
• Race: white, black, coloured, mixed
• Referral source: self, employer, court
Ordinal
• The data are broken into categories that can be
ranked
• It is not possible to quantify the difference between the
categories
Example: ordinal
• Level of education:
 None
 Primary
 Secondary
 Tertiary
Interval
• The data are measured on a continuous scale, not
simply ranked
• The units of measurement are constant
• There is no absolute 0
Example: interval
• Temperature:
– Fahrenheit or Celsius
• Measured on a continuous scale
• No absolute 0
Ratio
• The data are measured on a continuous scale, not

simply ranked
• The units of measurement are constant
• There is an absolute 0
Examples: ratio
• Age
• Income
• Temperature on the Kelvin Scale
Types of variation
• Nominal: equal categories

• Ordinal: ordered categories
• Interval and ratio: a continuous scale
Types of variation
• Qualitative: nominal
• Quantitative: interval and ratio
• Quantitative and qualitative: ordinal
Exercise:
identify the levels of measurement
• Name of treatment centre • Highest level of education

• Referral source completed
• Gender • Employment status
• Age • Current marital status
• Home language • How old was the patient
• Region of permanent when they first began using
residence drugs regularly?
Level of measurement in SPSS
• Nominal
• Ordinal
• Scale
Exercise: measure
• Return to Ex1.sav and set the level of measurement for

the variables ID, DRUG, AGE and COND
• Save the file
Summary
• Question types: • Variable types:

– Closed/Open – Levels of measurement
– Factual/Attitudinal – Discrete
(categorical)/continuous
– Quantitative/qualitative

GAP Toolkit 5: Types of Question and Types of Variable

Uploaded by

Copyright:

Available Formats

GAP Toolkit 5: Types of Question and Types of Variable

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

GAP Toolkit 5: Types of Question and Types of Variable

Uploaded by

Copyright:

Available Formats

GAP Toolkit 5

Training in basic drug abuse data management

Types of question and types of

• Define a range of classifications for questions and

• The respondent selects from a list of mutually exclusive

• Has the patient been in treatment prior to this episode?

• An option on all but the simplest closed questions

• A subset of closed questions

• The question allows more than one response

• Mode of ingestion of primary substance

• A type of closed question

• There are no constraints on the respondent’s answer

1. Q30. Which new drugs or new patterns of use have

• Do Open or Closed questions appear more frequently in

• The data describe an attribute

• Gender: male and female

• The data are measured on a continuous scale, not

• Nominal: equal categories

• Name of treatment centre • Highest level of education

• Return to Ex1.sav and set the level of measurement for

• Question types: • Variable types:

You might also like