Penelitian ini bertujuan untuk menghasilkan sebuah alat ukur (tes) berpikir kritis yang valid dan reliabel untuk digunakan, baik dalam lingkup pendidikan maupun kerja di Indonesia. Tahapan penelitian dilakukan berdasarkan tahap pengembangan tes menurut Hambleton dan Jones (1993). Kisi-kisi dan pembuatan butir didasarkan pada konsep dalam tes Watson-Glaser Critical Thinking Appraisal (WGCTA). Pada WGCTA, berpikir kritis terdiri dari lima dimensi yaitu Inference, Recognition Assumption, Deduction, Interpretation dan Evaluation of arguments. Uji coba tes dilakukan pada 1.453 peserta tes seleksi karyawan di Surabaya, Gresik, Tuban, Bojonegoro, Rembang. Data dikotomi dianalisis dengan menggunakan model IRT dengan dua parameter yaitu daya beda dan tingkat kesulitan butir. Analisis dilakukan dengan menggunakan program statistik Mplus versi 6.11 Sebelum melakukan analisis dengan IRT, dilakukan pengujian asumsi yaitu uji unidimensionalitas, independensi lokal dan Item Characteristic Curve (ICC). Hasil analisis terhadap 68 butir menghasilkan 15 butir dengan daya beda yang cukup baik dan tingkat kesulitan butir yang berkisar antara –4 sampai dengan 2.448. Sedikitnya jumlah butir yang berkualitas baik disebabkan oleh kelemahan dalam menentukan subject matter experts di bidang berpikir kritis dan pemilihan metode skoring.Kata kunci: Pengembangan tes, berpikir kritis, item response theory DEVELOPING CRITICAL THINKING TEST UTILISING ITEM RESPONSE THEORYAbstractThe present study was aimed to develop a valid and reliable instrument in assesing critical thinking which can be implemented both in educational and work settings in Indonesia. Following the Hambleton and Jones’s (1993) procedures on test development, the study developed the instrument by employing the concept of critical thinking from Watson-Glaser Critical Thinking Appraisal (WGCTA). The study included five dimensions of critical thinking as adopted from the WGCTA: Inference, Recognition Assumption, Deduction, Interpretation dan Evaluation of arguments. 1453 respondents from Surabaya, Gresik, Tuban, Bojonegoro and Rembang were used for trailing the test. The dichotomous data were analized using the Item Response Theory with two parameter logistic model using statistical program Mplus ver. 6.11. Several assumptions were tested prior the IRT analysis; the test of unidimensionality, local independency and Item Characteristic Curve (ICC). Amongst 68 items only 15 items had good discrimination parameter. Difficulty item level ranged from – 4.95 to 2.448. The study was limited in producing high number of qualified items due to its failure in finding subject matter experts in critical thinking area and inadequate choice in scoring method.Keywords: test development, critical thinking, Item response theory

Self-esteem is increasing in the United States according to temporal meta-analyses of the Rosenberg Self-Esteem Scale. However, it remains unclear whether this trend reflects broad social ecological shifts toward urban, affluent, and technologically advanced or a unique cultural history. A temporal meta-analysis of self-esteem was conducted in Australia. Australia shares social ecological and cultural similarities with the United States. On the other hand, Australian culture is horizontally individualistic and places a stronger emphasis on self-other equality compared to American culture. For this reason, the strengthening norm of positive self-esteem found in the United States may not be evident in Australia. Consistent with this possibility, the findings indicated that self-esteem among Australian high school students, university students, and community participants did not change between 1978 and 2014.