Household youth and middle-aged unemployment in Kwara state, Nigeria.
This data is a questionnaire survey for analysing youth and middle-aged unemployment in Kwara state, Nigeria. This is a coded dataset obtained from Ilorin, Offa, and Lafiagi, representing the three senatorial districts in Kwara state. This research has been approved by the ethics committee of the Universiti Sains Malaysia (JEPEM), and all the conditions have been duly followed while on the field. This data was obtained from 1120 households from six local governments within the three senatorial districts. The dataset, a table describing the data, and the questionnaire with the informed consent form are hereby attached.
Steps to reproduce
The study gathered information through questionnaires and utilized a random and systematic probability sampling technique. The households were randomly selected to give everyone an equal opportunity to participate, as Agresti and Finlay (2009) suggested. Meanwhile, the communities were divided into areas for enumeration, and a specified number of questionnaires were given out in each area to prevent spatial dependence and ensure adequate representation. The reason for using both random and systematic sampling techniques is that while random sampling ensures representation, it does not address spatial dependence. In contrast, systematic sampling helps reduce spatial dependence but needs to be more representative. By using both techniques, the study aims to achieve a good sample, as noted by Cheng and Masser (2003). Questionnaire Design This research aims to design a questionnaire and model specification that aligns with the study's objectives and theory, ensuring no important information is left out during data analysis. The questionnaire covers various topics, including household head, individual and demographic characteristics, community features, education, employment status, living standards, and dependent nature. To achieve this, the questionnaire contains both open-ended and close-ended questions. Measurement of variable for analysis After data collection, some variables were recoded to fit the specified models for each study objective. For instance, certain categorical variables, such as gender and vocational training, were changed to a binary format of 0 and 1 where applicable, while ordered variables were coded as necessary.