3

Methodology

A national lifestyle survey among undergraduate full time students was

undertaken in 21 third level colleges in Ireland during the academic year

2002/2003.  The colleges included the seven universities, twelve institutes of

technology and two colleges of education, giving a total undergraduate

student population of over 100,000 (Annex 1).  A national sample size was

calculated using a 3% precision and a 95% degree of confidence, with a

breakdown for the colleges based on each college population. A detailed

protocol was provided.  Each college was required to generate a random

sample from the computerised enrolment list of their full time

undergraduate students, to distribute the questionnaire by mail and to

collect the completed questionnaires by either mail or drop off points on

campus.  A national prize draw was provided as an incentive to return

questionnaires. 

The questionnaire contained several sections including student living

conditions, general health, mental health, dietary habits, exercise habits,

accidents and injuries, sexual health and substance use   tobacco, illegal

drugs and alcohol. The questions were based on existing items used in

similar international, national and college surveys on lifestyle issues. The

national survey of SLAN was used for items on general health, dietary

habits, physical activity, tobacco and drug use and for some of the alcohol

items

2

. The items on accidents and injuries were from HBSC survey as were

a few of the sexual health questions

12,13

. The mental health section was from

two sources, a college and school survey

14,15

. The international ECAS survey

was used for items on drinking patterns

10

, the Trinity College survey for

sexual health

16

and the Higher Education Authority European Student Survey

for questions on student income and expenditure

1

. 

Data were analysed using the SPSS X statistical package. The analysis

examined differences between gender and across year in college. All of the

gender analyses were reported in this study and the statistically significant

(p<.01) variables across year in college were also reported.

17