Author: Nazia Bashir
Supervisor: Dr. Saima Dawood
Degree: MSc
Year: 2008-2010
University: Institute of Applied Psychology, Punjab University, Lahore, Pakistan
Abstract
The present study was conducted to develop an Achievement Test of Urdu for the children studying in Class I, II & III of 6-8 years. The main purpose of the study was to develop a test to judge or evaluate the performance of a child in his /her mother language (Urdu) by using standardized test of his / her age group. Moreover, previously no comprehensive test was available for this age range of children. Therefore, efforts were made to develop such test in Urdu language that could cover both elements of the Urdu subject: Reading and Writing.The present test with 389 items consisted of seven subtests of Urdu Reading which include Identification of Words (63 items); Nonsense Words (14 items); Reading Comprehension, Part a (9 items) & Part b (12 items); Receptive Coding, Part a (14 items), Part b (15 items) & Part c (13 items); Syllable (13 items); Phonemes, Part a (10 items) & Part b (12 items); Quick Reading, Part a (1 passage of 52 letters), Part b (1 passage of 40 words) & Part c (1 passage of 36 words) and three subtests of Urdu Writing which include Writing Alphabets (38 items); Dictation (26 items)and Proofing, Part a (12 items) & Part b (10 items). Scoring was based on 1 and 0 point. For the development of test items, an item pool was generated from the curriculum of Class I, II & III. Urdu books for Class I, II & III by Punjab Text Book Board were used to construct the items of each subtest. There was total 434 items. Later, the test was sent for its review to 15 senior subject teachers, having experience of at least 3 years in teaching Urdu as a subject and was working in different Government Schools. No significant error or revision was reported. Only 8 items were discarded as having Low mean value of 5. Total 426 items were administered on sample. Initially, newly developed test was administered on 2 students of each class (1 boy & 1 girl) to see any other ambiguity according to child understanding. No significant error or problem was reported by the children. Therefore, no changes were made in the try out version of the test.Data was collected through convenient sampling. Govt. Junior Model Girls High School, Wahdat Road, Lahore was selected to collect data. From each class 60 students (30 boys & 30 girls) were selected to administer the Urdu Achievement Test individually. After administration, item analysis was conducted manually to determined Difficulty and Discriminating level of each item. Then 389 items out of total 426 items (91 of total items) were selected for final version of the test where both levels: Difficulty and Discrimination were within desired range (0.8 – 0.2) and (0.2 – 0.8) respectively. Later, the items of each subtest were arranged in order of increasing difficultly.The reliability of the test was determined by two methods: Split Half Reliability which was (r =.97, p<.01) and Kuder Richardson Coefficient which was r =.98). The validity of the test was established by threeprocedures: Content Validity, Construct Validity and Concurrent Validity. Content Validity was determined by designing the test content on the basis of curriculum taught in specific classes. After developing the test, it was sent for its review to 15 experienced teachers who- had been working in school as senior subject (Urdu) teachers from at least 3 years. Only those items were selected that obtained mean value above or at 5. Thus, 426 out of 434 items (98 of total items) were retained. To determine Construct Validity, the test along with each item’s Discrimination and Difficulty indices was sent to expert who had fair knowledge about test development as well as Urdu language at the end. Internal consistency method was also used to establish reliability and strengthen the construct validity. Correlation matrix revealed significantly high correlation between scores of subtests and total test. Concurrent Validity was established by correlating marks obtained by the students in class exams in the subject of Urdu with their marks obtained on the test (r = .83, .90 & M,p<.01) for Class I, II & III, respectively. In order to determine time limit, time was recorded for each administration. The average time taken was calculated to have an exact idea about the time limits for each subtest as well as for complete test. The average time required to complete the test was 45 minutes.
Keywords: Achievement Test, Reliability, Validity, Children.
Visit University to get the full thesis.
Corresponding Address: Department of Applied Psychology, University of the Punjab, Lahore, Pakistan. Email: chairperson@appsy.pu.edu.pk, Phone: 92-42-9231245
