Ferðavenjur Íslendinga - 01.02.1998, Page 31

Ferðavenjur íslendinga 1996
29
N
ika
n
ika
+ n
*
ika
N = mean population aged 0-74 years,
n = number of respondents,
n* = number of sampled individuals with domicile in
' Iceland but residence abroad,
k = sex, and
a = age groups {0-5, 6-10, 11-15,..., 65-69, 70-74}.
As described above the question on income in the year
1995 in the first stage of the tourism survey was not fully
comparable to the questions on the same subject in subse-
quent surveys. The non-response of these particular ques-
tions was somewhat greater than for other questions, but this
non-response was not equally distributed between all re-
spondent groups. In order to correct this inconsistency and
counter the effects of non-response on the frnal results, all
answers in the first survey given by respondents who were
neither married nor cohabiting were omitted and so were
answers from respondents living in households with two or
more grown-ups. The next step was to calculate the probable
value for income in 1995 by means of regression analysis
based on valid answers in all three surveys and an examina-
tion of the relationships between annual income and the
variables sex, age, socio-economic group, education, marital
status, number of children in household, and household size.