Penggunaan Setara Linguistik (Linguistic Equidistant) dalam Skala Likert : Kajian Maklum Balas Kepuasan Pelanggan Institut Pendidikan Guru Malaysia

Use of Linguistic Equidistant in Likert Scale: A Customer Satisfaction Feedback Study of The Malaysian Institute of Teacher Education

Abstract

Skala Likert merupakan instrumen yang meluas digunakan dalam penyelidikan pendidikan bagi mengukur sikap, persepsi dan tahap kepuasan. Walau bagaimanapun, sifatnya yang berskala ordinal menjadikan penggunaan kaedah analisis tertentu, khususnya penggunaan skala, sering dipertikaikan. Isu kesetaraan linguistik dalam label respons antara kategori positif dan negatif, boleh menjejaskan kesahan dan kebolehpercayaan dapatan. Kajian ini dijalankan bagi menilai kesan reka bentuk skala dan kesetaraan linguistik terhadap pengukuran kepuasan pentadbir sekolah terhadap guru baharu lulusan Program Ijazah Sarjana Muda Perguruan di Institut Pendidikan Guru Malaysia. Tujuannya adalah untuk mengenal pasti sama ada penambahan kategori penengah dan perbezaan label respons memberi kesan terhadap skor purata dan tafsiran prestasi. Kajian ini menggunakan kaedah deskriptif secara retrospektif dengan menganalisis data maklum balas pentadbir sekolah bagi tiga kohort guru baharu. Instrumen penilaian berbeza mengikut kohort, iaitu skala 4 mata dan skala 5 mata. Analisis melibatkan pengiraan perbandingan peratusan antara reka bentuk skala yang setara dan tidak setara. Dapatan menunjukkan peralihan daripada skala 4 mata kepada 5 mata dengan kategori penengah meningkatkan skor purata mencapai 99.8%. Walau bagaimanapun, apabila kategori “Agak Memuaskan” dikeluarkan, skor keseluruhan menurun kepada 94.98%, dengan “Pembelajaran Sepanjang Hayat” mencatatkan nilai terendah. Perbezaan ini membuktikan bahawa reka bentuk skala dan label kategori mempengaruhi dapatan secara signifikan. Kesimpulannya, reka bentuk skala yang tidak seimbang dan label yang tidak setara boleh mewujudkan bias positif dan mengaburkan kelemahan prestasi sebenar. Kajian mencadangkan supaya skala disusun secara seimbang dengan label yang setara dari segi intensiti makna, atau memasukkan kategori negatif yang sepadan, bagi meningkatkan ketepatan pengukuran dan kebolehpercayaan penilaian prestasi guru.

The Likert scale is a widely used instrument in educational research to measure attitudes, perceptions and satisfaction levels. However, its ordinal nature makes the use of certain analytical methods, particularly the use of scales, often controversial. The issue of linguistic equivalence in response labels between positive and negative categories, can affect the validity and reliability of findings. This study was conducted to assess the impact of scale design and linguistic equivalence on the measurement of school administrators' satisfaction with new teachers who graduated from the Bachelor of Teaching Program at the Malaysian Institute of Teacher Education. The purpose was to identify whether the addition of an intermediate category and the difference in response labels had an impact on the average score and interpretation of performance. This study used a retrospective descriptive method by analyzing school administrators' feedback data for three cohorts of new teachers. The assessment instruments differed according to the cohort, namely a 4-point scale and a 5-point scale. The analysis involved calculating the percentage comparison between the equivalent and unequal scale designs. The findings showed that the shift from a 4-point to a 5-point scale with the intermediate category increased the average score to 99.8%. However, when the “Somewhat Satisfactory” category was removed, the overall score decreased to 94.98%, with “Lifelong Learning” recording the lowest value. This difference proves that the scale design and category labels significantly influenced the findings. In conclusion, unbalanced scale designs and unequal labels can create positive bias and obscure actual performance weaknesses. The study suggests that scales be structured in a balanced manner with labels that are equivalent in terms of meaning intensity, or include corresponding negative categories, to improve measurement accuracy and reliability of teacher performance assessments.