Today in Ethiopia, there is a widely heard academic commonrntalk, especially in the tertiary level of learning, about a greatrnvariation in the English language proficiency level of governmenurnand mission secondary school graduatese The degree of thisrndiffierence has not so far been assessed.rnThe reserch desccibed in this paper attempted to examinernwhether or not there was a significant difference in the Englishrnlanguage proficiency level of government and mission second~yrnschool leavers, usin~ two very well reputed instruments. Theserninstruments are the cloze procedure and the dictation, perhapsrnthe most qualified integrative tests of language proficiency. Sincernconstructing discrete-point proficiency tests is extremely difficultrnfor ordinary classroom teachers, like the writer of this paper,rnunskilled in the technicalities of dev8loping such tests, theserntwo integrative tests, which are considered to be perhaps equallyrnvalid and reliable to measure the language proficiency level of thernlearner, have been employed as best alternative devices for thernpurpose.rn11.11 the possible necessary cautions essential in developingrnsuch tests, have been taken in order to determine the gra~3 levelrnof the cloze and the dictation passages, the Fog Index, one of thernformulas currently used for the purpose has been used and accordingrnto this formula all the cloze and the dictation passages have a difficultyrnlevel of grade 12. In addition to this the personalrnjudgement of English language teachers in the Addis Ababa University,rnabout their fitness, has been considered.rnThey were validated against the Michign Test of EnglishrnLanguage Proficiency Form-Bo The results showed that the cloze andrnthe dictation tests used in this paper, correlated at .97. And inrnorder to examine their relinbility the Kuder Richardson formularn(Kr-21) has been adopted. This revealed a reliability coefficientrnof .95 for the cloze ~nd .96 for the dictation tests.rnThe t-test technique was used to evaluate the differences inrnthe Mean scores of each test, so as to assess the extent of thernvariations in the performance level of English between these tworngroups of student. The findings showed that there is a singificantrndifference between government and mission secondary school leaversrnat p .001 level of confidence.rnOn the basis of these findings, it was recommended that selectionrnfor admission of students into the colleges and the universitiesrnhas to be made on the basis of an English language proficiency testrnor College Entrance English Language ~xamination, on the basis ofrnwhich placement of students into an appropriate program of Englishrnlanguage learning could be mnde so that each group can learn at arnpace commensurate nith its ability. An inquiry to make a thoroughrnrevien of the English curriculum of the lower levels has also beenrnmade. Further research on the subject has also been invited, sornthat a reliable solution to the problem can be found.