The Development of the Text Evaluation Scale for Child Rights: A Study of Validity and Reliability

Abstract

This study was conducted with the aim of developing The Text Evaluation Scale for Child Rights. There are four
different sample chosen for the face validity, content validity and construct validity (for pilot scheme and main study)
of the study. For face validity, a sample group of 3 experts chosen with the method of purposeful sampling including
the researcher was formed. For content validity, snowball method was determined and studied with 12 experts. For
the study of construct validity, random sampling method was performed for the sample selection in pilot study
involving 120 people and in main study involving 510 people.
The theoretical framework of the scale was determined by means of the attempts, conventions, studies regarding the
subject basing upon the United Nations Organization Child Rights Convention and additional protocols to this
convention. One could get minimum 40 and maximum 200 points from the five-point Likert scale which consists of
28 positive and 12 negative (total 40) items. The scale has two sub-dimensions which are content and author. The
fact that Cronbach Alpha reliability coefficient is high regarding the sub-dimensions of the scale (author
sub-dimension= 0,822 content sub-dimension= 0,834) shows that the items in the sub-dimensions are consistent with
one another. Cronbach Alpha value for the whole of the scale was determined as 0,90 which means that the scale is
highly reliable. Besides, in the scale, there are items of which factor loading value is higher than 0,45. When validity
and reliability results are examined, it can be seen that the scale could be utilized to evaluate the text in terms of
being suitable for the child rights.