About the statistics

1. Administrative information

1.1. Name

Smoking habits

1.2. Subject group

03.01 - Health conditions

1.3. Frequency and timeliness

Data is collected quarterly and published annually at the beginning of each year.

1.4. Regional level

County

1.5. Responsible division

330 - Division for Health Statistics

1.6. Legal authority

The Statistics Act §2-1 (voluntary)

1.7. Legal document(EU)

Not relevant.

1.8. International reporting

Statistics Norway reports annual figures to the OECD.

2. Background and purpose

2.1. Purpose and history

The purpose of the survey is to obtain information about smoking habits in Norway. Statistics Norway has collected data on smoking habits since 1973 and on snuffing habits since 2008. The data is now collected on commission from the Directorate on Health.

2.2. Users and applications

The statistics are primarily used by the Directorate for Health as well as research institutions and the media.

3. Statistics production

3.1. Population

The statistics cover the Norwegian population at age 16–74 years, and their smoking and snuffing habits.

3.2. Data sources

The questions about smoking habits are asked as a part of the quarterly Holiday and Travel Survey. The Statistics Norway’s population database is used to define the populations and draw the samples. In addition, information about the level of education is linked from register.

3.3. Sampling

The survey is based on a representative sample of 2000 persons in the age of 16-74 years each quarter, in total 8000 persons per year.

3.4. Collection of data

The data collection is primarily done by telephone interviews as a part of the Holiday and Travel Survey.

 

3.5. Control and revision

The data is collected by computer-assisted interviews. There are several procedures for electronically controlling the registration of answer in the questionnaire. In some cases the interviewer gets a “warning” when recording an answer, in other cases only values within certain pre-specified limits are accepted. In addition, there are checks that ensure that only valid codes for the answer categories are accepted.

3.6. Estimation

Not relevant.

3.7. Confidentiality

Not relevant.

4. Concepts, variables and classifications

4.1. Definitions of the main concepts and variables

The share of smokers is measured by two questions: "Do you sometimes smoke?" and "Do you smoke daily or occasionally?".

The share of snuff users is measured by one question: ”Do you use snuff daily, occasionally, or never? ”.

4.2. Standard classifications

Not relevant.

5. Sources of error and uncertainty

In the course of a year, four surveys with a total net sample of 5000 are carried out.

There are possible sources of error in any interview survey. Additionally, even in a sample as large as 5000, error margins of more than one percentage point in either direction may exist. To reduce the effect of random fluctuations in the data, particularly in presentations with a line diagram, a three-year moving average is normally applied. This is particularly important in periods where the sample has been smaller than it is today, and also when the results are broken down, by age group for example. In practice, a three-year moving average means that the average of the results for three consecutive years is calculated, and that this average represents the second year. From 2009 Statistics Norway has decided to use the figures for each year.

5.1. Measurement and processing errors

Measurement errors are errors that occur during data collection and are caused by the survey instrument: the form, questionnaire or measuring device used for data collection may lead to the recording of wrong values. The respondent may also consciously or unconsciously, give erroneous data or interviewers may influence the answers given by respondents. 

Processing errors are errors introduced during data processing in Statistics Norway, such as coding, data entry, data editing, imputation etc. Efforts have been made to minimise both measurement and processing errors by developing a control system in the questionnaire.

5.2 Non-response errors

Non response errors are errors by unit non-response, i.e. that the individual has failed to respond, or item non response, i.e. that the respondent has failed to respond to some but not all the questions in the survey. 

The unit non-response in this survey is between 35 and 45 per cent. For more information on biases caused by non-response, please see the publications for each Holiday and Travel survey.

5.3. Sampling errors

Sampling errors arise from the fact that the estimates are based on a sample and not a census of the entire population. The sampling error is measured by the standard error which estimates the expected deviation between the survey estimate and the estimate that would have been obtained if a complete enumeration had been carried out. 

The exact standard deviation is not calculated for this survey. However, by using the figure below one can find the expected standard deviation. The number of observations in this survey is about 5000. 

Figure 1. Expected standard deviation for observed percentages by different sample sizes 

Number of observations

5/95

10/90

15/85

20/80

25/75

30/70

35/65

40/60

50/50

25

5,4

7,5

8,9

10,0

10,8

11,5

11,9

12,2

12,5

50

3,8

5,2

6,2

7,0

7,6

8,0

8,3

8,6

8,7

100

2,7

3,7

4,4

4,9

5,3

5,6

5,9

6,0

6,2

500

1,2

1,6

2,0

2,2

2,4

2,5

2,6

2,7

2,7

1000

0,8

1,2

1,4

1,5

1,7

1,8

1,8

1,9

1,9

1500

0,7

0,9

1,1

1,3

1,4

1,4

1,5

1,5

1,6

2000

0,6

0,8

1,0

1,1

1,2

1,3

1,3

1,3

1,4

3000

0,5

0,7

0,8

0,9

1,0

1,0

1,1

1,1

1,1

4000

0,4

0,6

0,7

0,8

0,8

0,9

0,9

0,9

1,0

5000

0,4

0,5

0,6

0,7

0,7

0,8

0,8

0,8

0,9

5.4. Other sources of error

Not relevant.

6. Comparability and coherence

6.1. Comparability over time and space

Statistics Norway has collected data on smoking habits in Norway since 1973.

6.2. Coherence with other statistics

Not relevant.

7. Availability

7.1. Publications and other links

The statistics are published on http://www.ssb.no/english/subjects/03/01/royk_en/. In the left column there is a link to the StatBank where it is possible to define ad-hoc tables.

The Directorate of Health also publishes statistics based on the data, along with other information about tobacco use, sale and legislation.

 

7.2. Microdata

The final data are sent to Norwegian Social Science Data Services (NSD).


2011 © Statistics Norway