The observed variables are composed of a single item, mean score, or total score of multiple items to increase the accuracy of the estimation in the use of structural model. This study tried to verity the effect of item parceling through analyzing empirical data as well as theoretical overview of item parceling. Item parceling has the advantage of improving the accuracy of the estimation and model fit. On the other hand, it has the risk of being misused as expedient to report the favorable goodness of fit. Therefore, it is necessary to understand the proper procedure of using item parceling. This study reviewed theoretical and empirical studies about item parceling and tried to explore the effect of item parceling and its procedure using empirical data. Also, a detailed comparison is made of the effect between item parceling according to the exploratory factor analysis and the domain-representative parceling approach. The result of the study suggested that the use of item parceling has a good influence on stable factor loading and model fit. However it was confirmed that that the original meaning of the scale may be distorted if the parceling is not followed by the appropriate procedure. Thus the study is proposing that it is necessary to examine the factor structure of the scale before to use the appropriate item parceling method.