سبتمبر
09
2011
0
مدى اختلاف الخصائص السيكومترية لأداة القياس في ضوء تغايير عدد بدائل الاستجابة والمرحلة الدراسية
كتبه المشرف
رسالة ماجستير
Abstract in English:
The objective of the current study is to recognize the extent of effect of the multi
alternatives response of measurement tools and school stages difference on reliability
and validity coefficients. Also it tried to acknowledge the extent of difference of styles or
types of the individual’s response in the light of change of the number of measurement
alternatives and school stages.
For data collection we used scale tension list measurement for Charles Splirger & others
prepared and translated to Arabic by Dr. Nabeel Al-Zahar and Dr. Danis Hosfer. Three
additional forms are made of this measurement differs only in the number of response
alternatives. Hence we now have four forms of the measurement (two alternatives form,
three alternatives form, four alternatives and the five alternatives). The tension feature
measurement for Charles Spliger & others, prepared & translated to Arabic by Dr. Al-
Behairi as external examiner for measurement of concurrence validity.
All these measurements were applied on a sample consist of 451 students from different
stages of general education at Jeddah city. They are used in a multi-stage punch style
as follows: 172 from class (6) of elementary school, 136 student from 3rd class of
intermediate school and 143 students from 3rd class of secondary school, scientific
section. We took in consideration while presenting the measurement to the students the
following ranking: two alternatives measurement, three alternatives measurement, five
alternative measurement and at last the four alternatives (original) measurement.
The time interval between each application is 14 days simultaneously at each of the
application school.
For reply of the study inquiries a group of statistical analysis are made (Z. V. and M) to
know the difference between each of the reliability & validity coefficients as used by the
statistician Ka in order to recognize the difference occur at response style of each
individual according to difference of alternatives numbers and age stage.
From the discussion of the study result we can conclude the following:
1) The measurement tools of the psychometrical characteristics are in general affected
by the change of the number of response alternatives, and hence it is not
independent of it. If the number of response alternatives increase the values of
reliability and validity coefficients increases, and with statistical function difference at
level of 0.01 and 0.05.
2) The difference in school stage affects the validity co-efficient of measurement tools
more than affection on reliability co-efficient. As it is noticed that the validity coefficient
of the measurement tool increases with the increase in the mental and
knowledge growth of the students.
3) The five alternatives measurement showed high reliability degree compared to other
measurements. On the other hand the four alternatives measurement has higher
validity co-efficient value than other measurements.
4) The values of the reliability & validity coefficients increases when the alternative
tools are many, particularly when the respondents are of high caliber and have high
incentive for participation supported with clear perception to the measured featured
and the importance of its measurement.
5) The response style of individuals differs according to number of alternatives and
difference of school stage as it becomes of more spread and diversified as the
mental and knowledge growth increases at the individual. Because availability of
multi-alternatives grants the respondent more allowance to express himself with
more precision reflecting what actually in his mind without falsification or tendency to
one style of response.
In the light of results reached by this study and the outcome of the previous studies
results. A number of recommendations are suggested which concentrates on the
importance of the number of measurement tools and that they should be scientifically
specified instead of being randomly selected. I.e. the measurement shall include 5
alternatives for higher knowledge stages and less for lower knowledge stages. Also we
ensure on sensitivity of the tool and its distinguishing capability, taking care of its validity
when applied on samples of high difference of ages and knowledge level even if it
showed high reliability.
الحقل:
المؤلفون:
اللغة: