Descriptive Analysis of the Vacancy Database - A Web-Based Approach to Measure Skill Mismatches and Skills Profiles for a Developing Country - Libros y Revistas - VLEX 879445079

Descriptive Analysis of the Vacancy Database

AutorJeisson Cárdenas Rubio
Páginas150-190
150 C
A WEB -BASE D APPRO ACH TO M E A SU R E S KILL MISM ATCHE S AND S KILL S PROFI LES FO R A DEV ELOPI NG COUN TRY
7. Descriptive Analysis of
the Vacancy Database
151
c
7.1. Introduction
From a theoretica l point of view, Chapter 2 discu ssed what can be understoo d
as skill m ismatches and how this phenom enon might arise in a par ticular econ-
omy (e.g. due to imperfect in formation). Chapter 3 showed t hat this problem
has specif‌ic re levance in countr ies like Colombia. Inde ed, evidence from t he
labour marke t in this count ry suggests t hat skill m ismatches might ex plain a
substantia l portion of high u nemployment and in formalit y rates. One factor
that hinder s the design of well- orientated public pol icies to tackle sk ill mis-
matches is the abse nce or scarcity of detailed la bour market informa tion. More
specif‌ical ly, given the h igh cost of collecting infor mation about labour demand
for skills t hrough surveys , the composition and dyna mics of Colombian labour
demand are re latively unk nown.
However, information rega rding unmet l abour demand ca n be collected
from job porta ls with the implementat ion of relatively novel data min ing tech-
niques. These on line sources might provide va luable information in real t ime
and at a low cost for the ana lysis of labour dem and and, thus, for t he early
identif‌ication of lab our demand for ski lls as well as pos sible skill shor tages.
A better unders tanding of this infor mation can provide proper infor mation to
traini ng providers and policym akers and, in th is way, might improve education
and public policy des igns to tackle is sues of unemployment a nd informality.
This chapte r describes the mai n characterist ics of vacancy data collected a nd
organised in C hapters 5 and 6. Sec tion 7.2 shows the number of vacancie s
and job positions dem anded on job portals. T hen Section 7.3 displays the geo-
graphical cover age of the Colombian vacancy d atabase. The fou rth section
provides a descr iptive analys is of the labour dema nd for skills in Colom bia,
and analys es labour demand composition by educ ation, occupation (at a four-
digit level), new job title s, skills, a nd experience re quirements. Se ction 7.5
shows labour dema nd organised by sec tors. The sixt h section ana lyses the
C c
152 C
A WEB -BASE D APPRO ACH TO M E A SU R E S KILL MISM ATCHE S AND S KILL S PROFI LES FO R A DEV ELOPI NG COUN TRY
most notable trends i n the Colombian labour de mand by occupation: occ u-
pations with h igher demand, occ upations with a con siderable increase, a nd
occupations for whic h demand has decreased over t ime. Section 7.7 describes
the distr ibution of wages offered by employers, a nd the last section describes
other (secondar y) characteristics of the vac ancy database, such as t he type of
contract and the du ration of vacancies.
7.2. Vacancy database composition
Chapters 5 and 6 h ave described the met hods and chal lenges involved in
obtaining a nd organising vacancy in formation from job portal s. As a result of
those methods, a Colom bian vacancy database h as been generated to be tested
and analy sed for public policy recom mendations. The sa mple period ru ns
from Januar y 1, 2016 to December 31, 2018. Each observation in the dat abase
is a vacancy. By the def‌in ition that has been ap plied in this book , a vacancy
can requir e one or more people (the total numb er of jobs or job placements
available) (see Chapter 5). Followi ng the above def‌inition, the total nu mber
of
observation s (vacancies) in the database ar e 2,247,959, while the nu mbers
of jobs
are 5,720,513 (Table 7.1). Consequentl y, a vacancy advertiseme nt on average
contains 2.5 job pl acements.
As shown in Table 7.1, by volume, most of the vacancies (55.7%) and jobs
advertised (tota l vacancies) come from Jobportal _a, followed by Jobportal_ b
(33.4%) and Jobportal_c (10.8%). Likewise, 65.2 % of the total number of job s
originate f rom Jobportal_a, fol lowed by Jobportal_b (23.7%) and Jobporta l_c
(10.9 %) 86 (Section 7.4 will discu ss the types of job tit les posted on each job porta l).
86 This res ult reaff‌i rms that webs ites such as Jobpor tal_c do not nec essarily c ontain the major -
ity of job adver tisements, eve n when the website sta ted that it had 263,621 job vaca ncies on
October 30, 2017 (see Chapte r 5). As mentioned in C hapter 5, when clic king on some vaca ncy
announce ments on Jobport al_c, a new window re directed t he search by opening anot her
website where t he vacancy was ori ginally posted (e.g. Jo bportal_b).

Para continuar leyendo

Solicita tu prueba

VLEX utiliza cookies de inicio de sesión para aportarte una mejor experiencia de navegación. Si haces click en 'Aceptar' o continúas navegando por esta web consideramos que aceptas nuestra política de cookies. ACEPTAR