A Likert scale is an evaluation scale that is often found on survey forms and measures how people feel about something. Technically, this is an ordinal scale with an interval response option. In the baseline study, there are 23 variables in 6 classifications. We have adopted the Likert scale to measure each variable. Options range from 1, which is totally the opposite, to 7. The reliability and validity of the questionnaire are analysed. As shown in Table 8.1, kmo values are greater than 0.7 for all variables and significant chi square values are less than 0.05. This indicates that the variables reach the generally acceptable level. The Kul-Bach reliability coefficients of the variables are higher than the acceptable level of 0.7 and the factor loads for each question are greater than 0.5. Therefore, the design of the questionnaire is appropriate and qualifies the requirements of the study. In light of the above discussions, it is clear that there is a need for increased research to identify acceptable and practical approaches to assessing IRA levels. Recommendations for best practices for the interval would include the disclosure of several IRA statistics, ideally from different families, to provide a balanced perspective on the IRA.

Practical levels of significance could be achieved from the outset with the proposals described above (z.B Burke and Dunlap, 2002; Brown and Hauenstein, 2005; LeBreton and Senter, 2008), to identify Cutoffs for decisions. Statistical significance would only be used if conclusions about the population were important and a performance analysis suggested sufficient effectiveness to reach agreement, although practical standards should also be taken into account, especially when performance is very high. For example, a researcher or practitioner may place little emphasis on statistical significance if he or she is not concerned about generalization to the population and, if there are very few judges, the researcher may be advised to consider a less severe level of significance (for example. B α – 0.10). It is important that, in a number of judges, the evaluation of the agreement generally does not focus on whether the sample came from a population with a systematic chance or concordance, but whether there is a degree of practically useful agreement.