A Systematic PLS-SEM Approach to an Assessment of Unemployment among University Graduates in Madagascar

Bibliographic Data

ID22010563
AuthorsMili Mili (0000-0002-2596-9534, Central South University), Mili, F O Ursule (Central South University), Holihasinirina Judith (Central South University), Natolonjanahary Tantely Clara (University of Toliara)
Year2022
Volume12
Issue09
Pages423-438
Publication date2022-01-01
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueAdvances in Applied Sociology (JOURNAL)
Journal identifiersISSN: 2165-4328 • E-ISSN: 2165-4336
PublisherScientific Research Publishing, Inc. (PUBLISHER)
DOI10.4236/aasoci.2022.129034
OpenAlexW4296053331
LanguageEN
Citations received1
References cited54

The number of university students has markedly increased over the last half-century along with the remarkable expansion of universities and their campuses, reflecting the high demand. The objectives of this study were an empirical analysis of unemployment among university graduates in Madagascar, the PLS-SEM approach to some attributes, and its relationship with unemployment and education, age, and gender may have impacts on DV. This study was based on an assessment of unemployment among university graduates in Madagascar. We tried to check how their attribute can influence job seeking. A semi-structured questionnaire was adopted from existing literature and modified as per the study area. The main indicators of research were: Lack of Awareness (LOA); Lack of Opportunity (LOO); Lack of Proper Skills (LOPS); Over Qualified (OQ); Unemployment (U); Weak GDP (W-GDP). In this study, 450 completed and filled questionnaires were collected from the study area and a pre-test was carried out to check the reliability of the data and to avoid any discrepancies. PLS-SEM is used to analyze data to check relationships among variables, since it is the most often recommended way for predicting and assessing explained variables to account for the largest potential variance, and PLS-SEM is one of the most effective approaches for predicting outcomes. We carried out a bootstrapping test using 5000 replicates to evaluate the degree to which our findings were consistent with the hypothesis. PLS-SEM direct path analysis revealed that LOA -> U (β = 0.051; p 0.444); LOO -> U (β = -0.164; p 0.014); LOPS -> U (β = 0.358; p 0.000), OQ -> U (β =0.215; p 0.000); W-GDP -> U (β = 0.415; p 0.000) are significant/insignificant values. These results offer support to hypotheses H1 rejected and H2-H5 accepted. As per the results of this model, we can say Lack of Opportunity (LOO); Lack of Proper Skills (LOPS); Over Qualified (OQ); Unemployment (U); Weak GDP (W-GDP). As per research results, we can say that some reasons for unemployment are Lack of Opportunity (LOO); Lack of Proper Skills (LOPS); Over Qualified (OQ); Unemployment (U); Weak GDP (W-GDP). Age, gender, and education are considered as control variables under research, as per the results of the direct relationship among control variables and dependent variables: age (β = -0.111; p 0.004), gender (β = -0.025; p 0.526), and education (β = -0.056; p 0.126). As per values of direct relationships among control variables and DV, age has a significant relationship with DV while gender and education do not show any direct relationship

Biology · Econometrics · Economic growth · Economics · Statistics · Unemployment · Unemployment rate · COVID-19 Pandemic Impacts · Economic Growth and Development · Mathematics · Psychology · Taxation and Compliance Studies · Accounting

Tools

Open DOIOpen Access