Exploring the Conceptual and Psychometric Properties of Teachers’ Conception of Feedback

This chapter investigates the empirical evidence regarding the validity and reliability of an established questionnaire named TCoF. The Rasch analysis was conducted in six steps which are Rasch Fit statistics, item difficulty measurement, item polarity, uni-dimensionality, dispersion and reliability.

It is very important to check for validity and reliability of self-report scales or measures. The study is using survey design. The instrument consists of 37 items from 9 constructs namely irrelevance, improvement, accountability, encouragement and self-type, task, process, self-regulation, peer and self-assessment and timeliness. Respondents are chosen by random sampling. The questionnaire was filled out by 185 (N=185) undergraduate students who are also student teachers. Three measurement and evaluation specialists independently verified the information. The Rasch measurement model was used to do the data analysis. The analysis revealed that all items fit the model as their MNSQ values are between 0.50 and 1.50 unless for the three items. Two items showed a negative point measure correlation. Other items showed a value more than 0.20 so the item discrimination is good. Item reliability is 0.98 and teacher reliability is 0.89 meaning that they were both showed a very good reliability. The mean for person is 0.43 logits with a standard deviation of 0.70 logits, while the item mean is zero with a standard deviation of 1.05. The analyses yielded evidence that the instrument can be a useful scale to measure assessment concepts among teachers or student teachers. And, although this instrument originated from the western context, it is suitable to be used in Malaysian context. In total, four items were deleted altogether. The statistical analysis provides strong evidence to support the validity and reliability of the scale. The results provide an important foundation for an extension of subsequent research that makes use of this inventory. Although this inventory has been used for a while in industrialised nations, it still has to be confirmed to work in the educational setting of Malaysia. The teachers' notion, which is one of the most important aspects in determining how instructors use assessment for learning, may also be a major source of concern for the ministry.