Learning problem-solving skills emphasizes reasoning abilities to determine conclusions based on-premises, determining alternative solutions in decision making, thinking creatively in building alternatives to get keys, and critical thinking to evaluate the best solutions in answering problems. This study aims to identify biased publications using the trim and fill method in determining the effectiveness of learning on the problem-solving abilities of junior high school students. This research is a quantitative study with a meta-analysis approach. Data collection based on quantifiable variable numerical information analysis results from mathematical problem-solving abilities in two different learning groups. The data analysis technique uses bias publication analysis with the trim and fills method with the analysis procedure of calculating the effect size, heterogeneity test, calculating the summary effect using the random-effect model, and forest-plot analysis and biased publication analysis. The results showed no publication bias. The validity of differences in innovative and conventional learning effectiveness was valid on junior high school students' mathematics problem-solving skills. This shows that innovative learning effectively improves the mathematical problem-solving ability of junior high school students.

Improving the quality of human resources is the main supporting factor in increasing national productivity in various fields and development sectors. The government's productive investment activities that spur the nation's competitiveness in the global era prioritize Indonesia's education development. This study aims to cluster provinces in Indonesia based on educational indicators using the Agglomerative method consisting of the Average Linkage and Ward methods. Data collection is based on documentation techniques obtained from Statistics Indonesia in 2018. Data analysis used hierarchical cluster analysis consisting of data standardization, determining the size of the similarity or dissimilarity between data, the clustering process with a distance matrix, and seeing the characteristics of the cluster results formed. The second clustering method is by doing the initial grouping and determining the excellent cluster based on the average standard deviation ratio to the standard deviation between groups. Clustering results show the Ward method with the number of collections as many as 4 clusters and produces a ratio with a value of 0.01 smaller than the Average Linkage method. It shows that the cluster analysis method using the Ward method has better group accuracy quality than the Average Linkage method.

This meta-analysis intends to determine the impact of cooperative learning model research findings on the mathematics learning outcomes of Indonesian vocational high school students. Based on moderator variables, such as grade level and student sample size, identify efficacy and publication bias measures. Data from descriptive analysis, which included the mean, standard deviation, and sample size, were collected from 21 research projects based on applying the cooperative learning model in the experimental group and direct learning in the control class. Forest plot analysis was the data analysis method used. The results of the measure of the effectiveness of the cooperative learning model on mathematics learning outcomes in grades 10 and 11 were 0.87 and 0.92, with each effect size category being medium. While the effectiveness of the learning model is measured using a sample size of 1 to 30 students, and more than 30 students were 0.94 and 0.83, respectively, each has a medium effect size category. The results of other analyses show no publication bias. The findings of this study provided teachers with information on how to apply effective cooperative learning models to mathematical learning outcomes in 10th and 11th grade and the efficiency of learning with large class size.

An alternative for determining an accurate major for prospective high school students is not only based on academic scores but also on the results of the scholastic aptitude test (SAT). Verbal ability is an SAT subtest that assesses language management, vocabulary, and problem-solving abilities through a complete language study. This study developed a verbal ability test instrument for junior high school students consisting of the ability of synonyms, antonyms, and analogies. The data was collected from 300 junior high school students in grade nine who took a test with dichotomous data. The data analysis approach used one-order confirmatory factor analysis (CFA) with correlation factors. The results showed that CFA with correlation factors indicated the construct validity of the instrument was valid with the index criteria value =446.80, <em>df</em>=389, <em>p</em>-value=0.02267, root mean squared error of approximation (RMSEA)=0.022, goodness of fit index (GFI)=0.91, adjusted goodness-of-fit (AGFI)=0.89, and comparative fit index (CFI)=0.98. Then, construct reliability has good reliability with coefficient values for each dimension of 0.93, 0.95, and 0.84. As for the composite reliability of 0.88. It shows that using the verbal ability test instrument is feasible and has a reliable scale to measure the ability of junior high school students.