Paper	Individual Factors Affecting Online Health Information Perception	Methods of the Analysis	Limitations
[21]	Race/ethnicity	Health Information National Trends Survey	The self-reported data are subject to both validity and bias issues. Collected data do not allow for longitudinal analysis of respondents. The sample does not help to capture nationally representative and generalizable trends. The model specification strategy has limited sample sizes.
[24]	Health literacy	Online survey	The number of samples is relatively small and only consists of people in China. The study failed to gain deep insights into how patient compliance changed with internet health information-seeking behavior. The used scales are more inclined to measure one’s trust in physicians.
[23]	Health literacy	Nationally representative middle-aged and older internet users	The theorized direction of relationships as portrayed can only be confirmed through the use of a longitudinal research design. Individuals who never used the Internet to locate health information were excluded from the analysis.
[22]	Health literacy	Online survey	The cross-sectional design of the study restricts the ability to infer causal relationships between health literacy and health information source preferences. The quality of the actual health information sources used by participants in the study is not been evaluated.
[25]	Health literacy, gender, and age	Online survey	The generalizability of results to various racial/ethnic minority groups is not possible.
[20]	Age, education, gender, and health status	A regional and population health survey	While the samples are of acceptable size, the cross-sectional design effectively prohibits a causal analysis of the relationships. The re-categorization of some of the dichotomous variables reduces the power of the estimates. Although this approach ensures a sufficient number of individuals in the final sample, it may also subject the survey to the element of self-selection, given that participation is voluntary.