|dc.description.abstract||The goal of this research was to find out whether a statistical signal of a steward’s purpose could be ascertained through viewing the length of the value sets or the lengths of the intersections of pairs of value sets, with the same domain intension.
We gathered 239 value sets from 22 stewards on the National Library of Medicine’s Value Set Authority Center (VSAC) and 8 domain categories (e.g. Cardiovascular Disease, Infectious Diseases, Mental Health, etc.). We used an interview with an expert to find out the purpose/role of stewards. And we conducted a second interview to decide which purpose/role fit each steward. Using the results, we conducted statistical analysis (Univariate, ANOVA, Linear Regression) to determine variability of the data and the correlation between the purpose of each steward and the value set size.
The results suggested that there is no statistical significance in variability of the value set size by purpose. However, the results of the regression analysis indicated that the purpose of an organization is statistically significant and has a positive correlation with the value set size.
Unlike the conventional methods for checking the sensitivity and specificity of a value set, such as face validity, our method is a quick way of classifying value. However, we were limited in our data size stemming solely from VSAC. An expanded data set encompassing value sets from other vendors could yield different results.
The results of this research, from the data we gathered, supports our hypothesis that the purpose of an organization has a correlation with the value set size. However, further study still needs to be done as the availability of data for our research was limited to VSAC.||