This study has some data limitations. First, usage is measured by users' retrospective memory, rather than via actual data; consequently, reliability of the data needs to be strengthened using a secondary data source. This study measured the usage and frequency variables by an ordinal manner to be convenient for respondents; however, it would be better if a research measures those variables in a continuous variable. For future study, it is recommended that research include the explicit questions of why respondents seek out different sources so that the finding can be informative to information managers and centers