Print ISSN: 2204-1990

Online ISSN: 1323-6903

Keywords : reliability


Fazli Wadood, Fazal Akbar Imran Ullah

Journal of Contemporary Issues in Business and Government, 2021, Volume 27, Issue 5, Pages 2419-2431

Pilot testing is the most appropriate tool and play indispensable role while conducting large scale survey to increase the reliability, validity and practicability of the questionnaire, especially in management, social sciences and education studies. Although there are several widely adopted rules of thumb regarding the respondents of pilot testing in respective circumstances like ‘How to conduct and whom to include in the sample?’ is one of the most frequently asked questions in survey research. The prime aim of this study is to provide clear guidelines for future researchers to increase their research reliability. In the first phase we have reviewed and discussed different pilot studies. Secondly, we reviewed the suggested and adopted rules related to pilot testing. In the last, we present the findings/guidelines based on analysis of the different studies. This step to by step guidelines does not provide only researchers the basic understanding of how to conduct pilot testing but also enhances their understanding regarding the importance of the validity and reliability of their studies.

A Step by Step of Content ValidityAssessment Process on Approach Towards Safety Knowledge, Safety Climate and SMEs Safety Performance Framework

Nor Azma binti Rahlin; Ayu Suriawaty @ Siti Aisyah Bahkia; Zainudin Awang; Roslida Abdul Razak; Sidah Idris

Journal of Contemporary Issues in Business and Government, 2021, Volume 27, Issue 2, Pages 6253-6271
DOI: 10.47750/cibg.2021.27.02.615

This article examines the issues and the conceptual basis in the operationalized definition of safety knowledge. The importance of safety knowledgeemphasizes the mediation role of safety knowledge in the indirect link between safety climate and safety performance assessment in small and medium enterprises and the functional nature of content validity by applying a step-bystepcontent validity assessment. A content validation assessment process aims to increase representativeness, relevancy, clarity, comprehensiveness of the measurement's validitybesides
to ensure the instrument answers the research question. In business, social science, and management science, many variables of interest and research outcomes are abstract concepts and  its function by robust theoretical and conceptual explanation. Thus, a step-by-stepcontent validity assessment offered a bridge to achieve a valid and reliable measurement, and it was an essential element of quality research.

A General Linear Model Approach: Development Of Psychological Well-Being, Remote Working, Employee Engagement, Job Satisfaction, Scales, Data Analysis And Reporting Concerning To Information Technology Sector

Dr. KDV Prasad; Dr. Mruthyanjaya Rao Mangipudi

Journal of Contemporary Issues in Business and Government, 2021, Volume 27, Issue 1, Pages 1006-1035

In this article the author elaborates on the construction and development of Remote
Working, Employee Engagement, Job Satisfaction, Psychological well-being Scales, and
the appropriated statistical methods needed to analyze the respondent’s data, reporting the
results, using secondary data, with psychological well-being as a dependent variable. The
measurement scales constructed are remote working 18 items, with 4 sub-scales; job
engagement 20 items with 5 sub-subscales, work-life balance 18 items are the independent
factors, which have a causal effect on a dependent factor psychological well-being of an
employee which has 8 items with 6 sub-scales. The reliability of the survey instrument
assessed measuring reliability statistic Cronbach alpha, construct validity, and content
validity are tested using face validity and convergent validity taking expert opinion and
factor analysis respectively. As the study is measuring six dependent variables related to the
psychological well-being of an employee, the author has used the statistical procedure
generalized linear model a multivariate multiple regression model, which allows more than
one dependent variable in the system and the results were inferences and presented. The
measured Cronbach’s alpha value indicates that the overall survey instrument was reliable
and maintained internal consistency. The results of GLM indicate statistically gender and
age differences that are influencing the psychological well-being of an employee. A notable
observation was Self-Acceptance was statistically significant and influenced by almost all
the predictor variables. The results were presented in detail and inferences provided
wherever appropriate

Construct Validity and Reliability of Creativity and Innovation in Public Sector: A Rasch Measurement Model Approach for Pilot Study

Saharah Sapiyi; NurRiza Mohd Suradi; Zainol Mustafa

Journal of Contemporary Issues in Business and Government, 2021, Volume 27, Issue 1, Pages 2779-2792

A pilot study is imperative to test the questionnaire items, confirm and check the research instrument’s reliability to obtain the best items. This pilot research aimed to test the reliability of the developed public sector instrument and identify its weakness. The instrument was purposefully designed to recognise the factors that affect creativity and innovation in the Malaysian public sector and their influence on government agencies’ performance. The instrument included 90 items and was distributed to 120 civil servants from several ministries mostly located in Putrajaya and Kuala Lumpur, Malaysia. The goal of this instrument was to measure five study constructs: individual creativity (IC), team creativity (TC), organisational innovation (OI), innovation processes (IP), and organisational performance (OP). The method employed to analyse the validity and reliability of the items and respondents in this study was derived from the Rasch Measurement Model Approach, which is far more valid and well-grounded than only relying on Cronbach’s Alpha output produced. The Winsteps version 3.73 was used to verify the items’ functionality in aspects such as the item’s reliability and separation of the item-respondent, polarity item, item fit in measuring constructs,item difficulty level, the respondent’s ability, and the standardised residual correlations. It also enabled the removal of items based on polarity items’ statistics and the item’s suitability. At the end of the analysis, it was established that there was a total of 9 items that were discarded because they did not meet the inspection criteria specified per the Rasch Model. A total of 81 items were recorded on the final instrument that could be used to measure the five constructs.