Quality of proxy reports on labor force survey : case study : Tulkarm governorate

العناوين الأخرى

جودة بيانات الوكيل في مسوح القوى العاملة : دراسة حالة : محافظة طولكرم

مقدم أطروحة جامعية

Imarah, Muhammad Ahmad

مشرف أطروحة جامعية

Abu Hasan, Hasan

أعضاء اللجنة

Sadiq, Tariq
Barakat, Ali

الجامعة

جامعة بيرزيت

الكلية

كلية الدراسات العليا

القسم الأكاديمي

الإحصاء التطبيقي

دولة الجامعة

فلسطين (الضفة الغربية)

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2013

الملخص الإنجليزي

The purpose of this study is to evaluate the quality and reliability of proxy reports based on the self reports.

620 individuals representing 310 household from Tulkarm Governorate participated in this study by indirect interview.

The questionnaire included important items which were derived from Palestinian Labor Force Survey (PLFS) questionnaire.

These items about the employment status, working hours in the reference period, readiness to work, number of workdays and wages.

Intraclass correlation and Kappa Coefficient were used to assess the agreement between proxy reports and self reports.

Stuart-Maxwell test was used to determine the statistical significance for the qualitative measure of the percentage of bias.

The paired t test was used to determine the statistical significance of bias for the quantitative measure.

The agreement values were wobbling.

In some items, the agreement values were over 0.9 whereas in other items they were very low.

Our analysis confirm that proxy reports about quantitative questions tended to be biased downward in most quantitative questions.

We classified proxy reporters according to relationship with self-reporters.

We found that the results vary from one question to another, for example, the spouse answers were the best in some questions and the worst in the others.

We also classified proxy reporters according to their group age.

We found that the proxies who were aged over 30 years were significantly more likely to give consistent answer than other proxies.

Moreover, we classified proxy reporters according to their confidence rating about a given question.

We found that the proxies who rated their confidence as a high were significantly more likely to give consistent answer than other proxies.

The results of the study indicates that there are a hidden factors which they affect on the proxy reports, but this study did not take them into consideration.

The existence of statistical significance at aggregate level confirms the validity of this claim.

التخصصات الرئيسية

الرياضيات

عدد الصفحات

130

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-707000