Abstract—Identification of employability skills expected by
employers in engineering disciplines is vital for fresh
engineering graduates who enter to Sri Lankan work force.
Factor Analysis was found to be one of the best methods for
identifying these skills, and use of proper sample sizes in such
an analysis is essential for reliable results. From this view point,
this study establishes a method for investigation of applicable
sample sizes in extraction of employability skills of civil
engineering graduates. For this purpose, a questionnaire survey
was carried out among civil engineering organizations to
measure employer expectation on thirteen technical and fifteen
nontechnical specific attributes identified as important. These
attributes were measured using five point Likert scale based on
their importance levels. In the factor analysis process, the
communality values of each attribute that represent the degree
of participation in the analysis was considered as the key
indicators to examine their consistency with effective sample
sizes. This study used fifty random samples generated for
sample sizes ranging from 25 to 100 with 25 increments and the
total sample size 122. From the results, it was found that the
communality values showed consistent trend above the sample
size 75. Moreover, it should be noted that questionnaire should
be properly designed and also respondents should be correctly
chosen when this sample size is selected for questionnaire
surveys. Finally the finding of this study will help in planning
and conducting questionnaire surveys effectively and data
analysis with a proper sample sizes in order to make reliable
predictions in related studies.
Index Terms—Employability skills, factor analysis, random samples, sample size.
M. M. G. V. Shyamalee is with the Industrial Training and Career Guidance Unit, Faculty of Engineering, University of Peradeniya, Sri Lanka (e-mail: firstname.lastname@example.org).
K. S. Walgama is with the Department of Engineering Mathematics, Faculty of Engineering, University of Peradeniya, Sri Lanka (e-mail: email@example.com).
W. M. V. S. K. Wickramasinghe is with the Department of Civil Engineering, Faculty of Engineering, University of Peradeniya, Sri Lanka (e-mail: firstname.lastname@example.org).
S. Dissanayake is with the Department of Civil Engineering, Kansas State University, 2118 Fiedler Hall, Manhattan, KS, 66506, USA (e-mail: email@example.com).
Cite: M. M. G. V. Shyamalee, K. S. Walgama, W. M. V. S. K. Wickramasinghe, and S. Dissanayake, "Influence of Sample Sizes in Extraction of Employability Skills of Engineering Graduates Using Factor Analysis," International Journal of Information and Education Technology vol. 7, no. 11, pp. 830-836, 2017.