The purpose of this study is to describe the underlying topics and the topic evolution in the 50-year history of educational leadership research literature. Method: We used automated text data mining with probabilistic latent topic models to examine the full text of the entire publication history of all 1,539 articles published in Educational Administration Quarterly (EAQ) from 1965 to 2014. Given the computationally intensive data analysis required by probabilistic topic models, relying on high-performance computing, we used a 10-fold crossvalidation to estimate the model in which we categorized each article in each year into one of 19 latent topics and illustrated the rise and fall of topics over the EAQ's 50-year history. Findings: Our model identified a total of 19 topics from the 1965 to 2014 EAQ corpus. Among them, five topics-inequity and social justice, female leadership, school leadership preparation and development, trust, and teaching and instructional leadership-gained research attention over the 50-year time period, whereas the research interest appears to have declined

The purpose of this study is to examine the public opinion on the Common Core State Standards (CCSS) on Twitter. Using Twitter API, we collected the tweets containing the hashtags #CommonCore and #CCSS for 12 months from 2014 to 2015. A Common Core corpus was created by compiling all the collected 660,051 tweets. The results of sentiment analysis suggest Twitter users expressed overwhelmingly negative sentiment towards the CCSS in all 50 states.Five topic clusters were detected by cluster analysis of the hashtag co-occurrence network. We also found that most of the opinion leaders were those who expressed negative sentiment towards the CCSS on Twitter. This study for the first time demonstrates how text mining techniques can be applied to education policy research, laying the foundation for real-time analytics of public opinion on education policies, thereby informing policymaking and implementation. , and #gagop) co-occurred in a tweet "Jeremy Spencer talking about the coming storm Nathan Deal will face with #CommonCore #gagov #StopCommonCore #gagop http://t.co/oyM0WIBBsw", then there were six co-occurrence ties connecting the four hashtags in the hashtag co-occurrence network: (1) CommonCore-gagov, (2) CommonCore-StopCommonCore, (3) CommonCore-gagop, (4) gagov-StopCommonCore; (5) gagovgagop; and (6) StopCommonCore-gagop. We wrote R code to repeat this procedure for all 660,051 tweets in the Common Core corpus to build the hashtag co-occurrence network. We then ran the faction algorithm-one of the network clustering algorithms-to partition the network (de Amorim, Barthélemy, & Ribeiro, 1992;Glover, 1989Glover, , 1990, thereby detecting the clusters of hashtags in the network. According to network science (Borgatti et al., 2013), the cooccurrence relationships between hashtags in the same cluster are closer than the ones in different clusters. Thus, the clusters of hashtags manifest the frequently co-occurred topics and their interconnections in the Common Core discourse on Twitter. Further, to ensure the robustness of network partitions, following the recommendations for cluster analysis, we ran the faction algorithm multiple times with different initial partitions by using different random number seeds (Borgatti et al., 2013). If the same subgroups always emerged, then the network partition is considered robust. Therefore, in this study we examined the subgroups that are consistently detected by using the faction algorithm. Communication Network AnalysisTo identify the opinion leaders in the Common Core discourse on Twitter, five centralities-Indegree, Outdegree, In-Bonacich Power, Out-Bonacich Power, and betweenness degree-were calculated as the indicators of each Twitter user's influence in the Twitter communication network. Opinion leaders, according to Rogers (2003), are those who occupy the PUBLIC SENTIMENT AND OPINION ON THE COMMON CORE 20 central structural locations in the communication network. In this study, the opinion leaders were those who have high centrality, calculated by performing social netwo...