Influence of Duration for Composition and Letter Writing Examination on Performance: An Experimental Design Approach

Abstract : Construct irrelevant variance is one of the major threats to generation of valid and reliable scores. Construct irrelevant variance occurs when performance of candidates in an examination is not explained or accounted for by the subject matter assessed but the variance observed in the scores is influenced by other intervening factors. The duration of the examination is one critical element that can have a negative impact of validity of test scores due to the fact that an inappropriately timed examination can severely limit a candidate’s ability to demonstrate what he or she knows understands and can do. Primary School Leaving Examination requires candidates to write composition and letter pieces within a period of 60 minutes. The relevance and appropriateness of the 60 minutes duration has not been determined.  There is need therefore to collect empirical evidence to support or challenge the relevance of the 60 minute duration. The current study uses an experimental design to compare performance of learners required to write composition and letter under different experimental conditions. The establishment of an appropriate duration would significantly enhance the construct validity of the examination and subsequent inferences made on the basis of the generated scores.