Page 179 - Contributed Paper Session (CPS) - Volume 7

CPS2056 Nurul Hafizah Azizan et al.
are subjective in nature, which is also referred to as latent constructs.
Therefore, extra caution should be given when constructing a questionnaire
with these types of variables. This is to ensure that the questionnaire clearly
depicts the actual meaning of the variables to be investigated.
In practice, these types of constructs will be measured indirectly using
several indicators or multiple reflective items with appropriate types of rating
scale to represent these components. This enables the respondents to provide
their responses based on the scales corresponding to the items. Based on the
responses given by respondents, the composite scores of these multiple items
will be computed as a single meaningful factor(s) where it can be used in
further analyses. Since these latent constructs are subjectively measured, it is
crucial for the researchers to ensure that the measurement instrument used
for the survey meets the psychometrics requirements of both validity and
reliability. It is acknowledged that the development of questionnaire as a
measurement instrument for a survey that fulfils requirement of the
psychometric properties is not an easy task. It requires several important steps
to be followed which takes considerable time and effort from the researchers.
The utility of the measurement instrument will depend on its quality, as it
can be useful if certain psychometric and practical requirements are met
(Shultz, Whitney, & Zickar, 2013). The quality of measurement instrument can
be affected by many factors and this requires careful attention at the
beginning of the questionnaire development. As Radhakrishna (2007)
remarked, a systematic development of a valid and reliable questionnaire is a
must to reduce measurement errors. Designing a questionnaire that produces
usable data is not as easy as it might seem. In fact, the development of
inappropriate instrument will lead to poor quality of data, misleading
conclusions and affect the recommendation to be given (Boynton, 2004),
consequently leading to ineffective corrective action taken to deal with the
problems in hand. A well-developed questionnaire will produce precise
findings. This is because it could avoid or reduce potential errors and biases,
which will lead to better quality data. Only with appropriate and well-
developed questionnaire, a relevant information can be obtained and better
decision can be made.
Therefore, before the questionnaire can be distributed to the respondents,
it is advisable to follow three main guidelines to minimise bias in a survey
research as highlighted by Sekaran and Bougie (2016). These include principles
of wording, principles of measurement and general appearance of the
questionnaire. In constructing a questionnaire, any researcher should take
particular concern on the measurement to be used. This is because
measurement is fundamental in any scientific research. Sekaran and Bougie
(2016) mentioned that measurement of the variables has become an integral
part of research, where the variables of interest should be measured correctly

166 | I S I W S C 2 0 1 9