Development of early literacy skills assessment tool, validity and reliability study

This study aimed to develop an Early Literacy Skills Assessment Tool validity and reliability study to assess the early literacy skills of preschool children. The research sample is composed of 473 children of 48-77 months, 244 of whom are female and 229 male, studying at 23 schools in six districts of Ankara during the school year of 2011-2012. The research is designed in accordance with the survey model. Having reviewed the literature on the early literacy skills assessment, the sub-dimensions of planned assessment tool as well as an item pool were formed and content validity was conducted by consulting the expert opinions. The study was piloted first with 47, and then 150 children. As the purpose of the first piloting was to see whether the children could understand the instructions of the assessment tool or not, hence, no statistical analysis were performed at this stage. The second piloting aimed at analysing item discrimination and difficulty values of data collected excluding the sub-test of ?Story Comprehension?. For the Story Comprehension sub-test, the items to be included in the research were determined based on the responses from children and the expert opinions. After the item analysis, 116 of them remained in the five distinct sub-tests. The assessment tool items were evaluated by giving points of "1" for the correct responses and "0" for the incorrect ones. The applications were practiced with the kids on one-to-one basis. Exploratory, confirmatory factor analyses were performed and item discrimination values were examined within the scope of validity analyses of Early Literacy Skills Assessment Tool sub-tests. For the reliability, the sub-tests were investigated according to KR-20, test re-test and split-half reliability. The findings of the research revealed that according to total variance for the exploratory factor analysis of Phonological Awareness Skills Assessment sub-test was 59.73%. This sub-test is composed of five factors and 53 items. Print Awareness sub-test has three factors and 16 items with a total variance of 40.01%. The contribution of single factor structure in the Story Comprehension sub-test to the variance is 40.22%, in the Matching Images sub-test is 44.38% and Pre-Writing Skills sub-test is 56.70%. These three sub-tests are composed of 9 items each. In line with confirmatory factor analysis, x², Sd, RMSEA, GFI, CFI, AGFI and NNFI fit of the sub-sets were examined. The results of the confirmatory factor analysis displayed that Print Awareness and Pre-writing Skills Assessment sub-tests have a weak fit; while, other sub-tests have a fit value at the acceptable level. Item discrimination for the groups at the top and bottom 27% showed that all items are distinctive. Based on these results, Early Literacy Skills Assessment Tool has five sub-tests with a total of 96 items. The KR-20 reliability value for the sub-tests of Early Literacy Skills Assessment Tool ranges between 0.61 and 0.91. The KR-20 reliability value for the Phonological Awareness Skills sub-test is 0.91, its test re-test reliability is 0.92 and split-half reliability is 0.76. The KR-20 reliability value for the Awareness of Writing sub-test is 0.75; test re-test reliability is 0.72, while split-half reliability is 0.60. The KR-20 reliability value for the Story Comprehension sub-test is found as 0.61, test re-test reliability as 0.75 and split-half reliability as 0.64. The KR-20 reliability value for the Matching Images sub-test is 0.71, test re-test reliability is 0.64 and split-half reliability is 0.70. Finally, KR-20 reliability value for the Pre-writing Skills sub-test is 0.77, test re-test reliability of 0.86 and split-half reliability of 0.72.The scores children gained in the sub-tests were examined with respect to gender, age (48-53 months, 54-59 months, 60-65 months, 66-71 months, 72-77 months), and educational background of the parents. As for the analysis of scores from sub-tests according to gender, a significant difference is observed in the total score of Awareness of Writing and Pre-writing Skills sub-tests (p<0.05). In the Awareness of Writing subtest, the total average score of girls (X= 8.1) is significantly higher than that of boys (X= 7.4). In the Pre-writing Skills sub-test, girls? average score (X= 5.63) is higher than boys? average score (X =4.77). The scores from all the sub-tests increase significantly with the growing age. This result shows that the assessment tool developed could well reveal the developmental features of children. Furthermore, when the scores from the sub-tests are examined based on the educational background of families, it is seen that there is a significant difference between the Pre-writing Skills sub-test scores. When scores of children with mothers graduated from primary and high school or university graduates are compared, a significant difference is observed in favour of children with parents graduated from primary school (p<0.05). It can be explained by referring to the interaction of mothers with their children. The scores did not display any significant difference according to the educational backgrounds of fathers. This study concludes that sub-tests of Early Literacy Skills Assessment Tool developed for preschool children (48-77 months) are valid and reliable. Researchers and teachers can use subtests separately. The assessment tool is recommended to be worked on further for the norm studies. Keywords: Preschool Education, Early Literacy Skills, Assessment