Background: Phonological short-term memory is a part of working memory, as working memory consists of four components: the central executive, the phonological loop, the visuo-spatial sketchpad and the episodic buffer. Aim of the work: To construct a valid and reliable test for assessment of phonological short-term memory, which can be applied in language child disorders as on specific language impaired children. This might help to clarify more points in their diagnosis and therefore to guide a better management. Subjects and methods: The subjects of the study were 2 groups: The first group consisted of 102 normal children between the ages of 4 to 8 years old, with average number of 25 children in each with 1 year age range. The second group included 31 specific language impaired children that were selected to match the control group as regard age and IQ level. The constructed test items were 17 in number. The test included the following items: digit span, syllable repetition ( for 1,2,3,4 syllables ), non-sense word repetition (NSWR) of (1,2,3 and 4 syllables),dissimilar word set recall (short word set recall of 1 and 2 syllables, long word set recall of 3 and 4 syllables) and lastly similar word set recall (of 1,2,3 and 4 syllables). Results: Reliability of the test was measured by test re-test, alpha Cronbach and split half method. Validity of the test was also measured using internal consistency, contrasted group validity and factorial validity. Conclusion: The constructed test for measuring phonological short-term memory (PSTM) showed high reliability and validity and could be used on Arabic speaking children from 4 years to 8 years of age.