Weighting

• Pages: 31-31

Weighting

Benefit s of gender equality th rough infrastruct ure provision: an EU-wid e survey 31

7. Weighting

In this survey-based research, the selected sam-

ple size was proportional to the size of the adult

population in ever y participating Member State.

In particular, in each Member State, the sample

size for the survey was 0.0001 % of the popula-

tion size. The sample size of each Member State

thus acts as an implicit weighting factor in the

analyses and it is not necessary to use addi-

tional weights in the microdata to ensure that

the sample is representative of the EU.

The sample was selected proportionally to age

(18-39, 40- 64 and ≥ 65) and gender. As it was

a random selection of cases in each Member

State, some discrepancies may have occurred

between the age and gender distribution of the

sample and that of the population. In order to

determine the reach of these discrepancies by

Member State, the percentage distribution by

gender and age of the sample was calculated and

compared to their distribution in the population.

Table 37 (Annex 3) shows the percentage dis-

tributions, where the unweighted percentage

(u (%) ) refers to the propor tion of the sample

that belonged to that category, whereas the

weighted percentage (w (%) ) is the proportion

of the population belonging to that category.

For instance, the percentage of men aged 18-39

was 20.24 % and 17.38 % in the Austrian sample

and population, respectively.

Columns u (%) and w (%) reveal the ex tent of

the discrepancy/similarity between sample and

population percentages, by Member State.

It is wor th noting the lack of 18-39 male respond-

ents in Luxembourg and Malta, even though

the percentage of this par t of the population is

19.60 % and 19.74 %, respectively.

Table 38 (Annex 3) shows a percentage distribu-

tion similar to that in Table 1, but with data dis-

aggregated by sex and labour status (employed

vs unemployed). It was a prerequisite that 30 %

of respondents should be employed. Table 38

reveals the extent of compliance with the crite-

rion in every MS, by gender.

Finally, Table 39 (Annex 3) includes the per-

centage distribution of rural, intermediate and

urban areas in the sample and the population

of each Member State. It was a prerequisite that

30 % of respondents should live in rural areas.

Table 39 reveals the extent of compliance with

the criterion in every MS, by gender.