suggested that a properly conducted interview
was required in order to make meaningful suggestions for
improvements on equipment or working methods. The questionnaire
was constructed in three parts; Part A on personal particulars,
Part B on training and work experience and Part C on issues
consisting of the five clusters of factors and the 30 elements of
the clusters as shown in Table 2. Further refinements to reduce
errors and ambiguity were made after the questionnaires were
tried on a sample of ten undergraduate volunteers. It was accepted
that the main constraint to the conduct of the study was the time
availability of the workers. Questionnaire forms were sent by post
and potential respondents were initially given 2 weeks to respond.
Verification interviews were then arranged and conducted during
breaks and after shift periods. Part A and Part B are straightforward.
For Part C respondents were requested to rate each of
the 28 questions on the safety elements on the five-point Likert
scale, varying from ‘‘not important’’ (1) to ‘‘extremely important’’
(5). The distribution of the Likert-point score from the respondents
for each safety element was computed to determine the mode for
each factor. The mode for each element was further strengthened
by examining the maximum, the minimum, the mean and standard
deviation as the value for the strongest probable Lickert score
value for the element. The most influential factor could then be
determined by examining and comparing the overall scores for
each factor. A similar analysis was conducted for the sub-factors
of the most influential factor to determine the most influential
sub-factor among them.
3. Results and discussion
The total number of respondents was 275. Based on this sample
size together with a confidence level of 95% the margin of error is
5.89%. Table 3 shows that the construction industry engages 92%
male workers. Employers have the perception that male workers
had better abilities and endurance than their female counterparts