Page 30 - Šolsko polje, XXVII, 2016, no. 3-4: IEA ICILS in druge sodobne teme, ur. Eva Klemenčič
P. 30
šolsko polje, letnik xxvii, številka 3–4
with differing levels of SES: those from lower SES schools tend to use
computers more often for “drill and practice” activities, solving problems
and analysing data, while the students studying at higher SES schools use
computers for written expression, they tend to use computers and the
internet at home more often and in general use the technology in high
er-order thinking activities (Hohlfeld et al., 2008). The differences at the
school level can be explained with the school-level SES, which is related to
different other student characteristics. Lower-SES schools differ in both
the methods and technologies used for instruction (Hohlfeld et al., 2013).
Method
As stated earlier, this study tests the differences in CIL of 8th grade stu
dents based on their SES. The analysis uses stepwise multiple linear regres
sion models, adding control variables to the regression models in order to
partial out their effect from the relationship. This way the characteristics
bridging the digital divide in CIL are identified.
Data
The data stem from IEA’s International Computer and Information Lit
eracy Study (ICILS) 2013, which evaluates CIL achievement and collects
background information from students at grade 8 and their schools and
teachers across 21 countries. The target population of the students was
defined as those enrolled in the grade representing eight years of school
ing or its equivalent within the country. In most countries this defini
tion matched students in grade 8. The only exception is Norway, where
students in grade 9 and their teachers were sampled. The minimum re
quirement of the number of sampled schools is 150 within each country.
The sample of schools was drawn using Probability Proportional to Size
(PPS). That is, the more students in the target grade a school has, the more
likely it is to be sampled. Within each school 20 students were sampled
across all its target-grade classes, disregarding the number of classes with
in the school. At least 15 teachers within schools were sampled regardless
of the subject they teach in the target grade (Meinck, 2015). However, the
sampled teachers are not necessarily the teachers of the sampled students
which means that link between students cannot be linked to their teach
ers. Therefore teacher characteristics cannot be used as the explanatory
variables of student achievement or as the control variables of the relation
ship between student achievement and their SES. The list of 21 countries
participating in ICILS 2013, their sample sizes and population estimates
are presented in Table 1. As the table shows, four countries did not meet
the sampling requirements of sampling at least 150 schools within the
28
with differing levels of SES: those from lower SES schools tend to use
computers more often for “drill and practice” activities, solving problems
and analysing data, while the students studying at higher SES schools use
computers for written expression, they tend to use computers and the
internet at home more often and in general use the technology in high
er-order thinking activities (Hohlfeld et al., 2008). The differences at the
school level can be explained with the school-level SES, which is related to
different other student characteristics. Lower-SES schools differ in both
the methods and technologies used for instruction (Hohlfeld et al., 2013).
Method
As stated earlier, this study tests the differences in CIL of 8th grade stu
dents based on their SES. The analysis uses stepwise multiple linear regres
sion models, adding control variables to the regression models in order to
partial out their effect from the relationship. This way the characteristics
bridging the digital divide in CIL are identified.
Data
The data stem from IEA’s International Computer and Information Lit
eracy Study (ICILS) 2013, which evaluates CIL achievement and collects
background information from students at grade 8 and their schools and
teachers across 21 countries. The target population of the students was
defined as those enrolled in the grade representing eight years of school
ing or its equivalent within the country. In most countries this defini
tion matched students in grade 8. The only exception is Norway, where
students in grade 9 and their teachers were sampled. The minimum re
quirement of the number of sampled schools is 150 within each country.
The sample of schools was drawn using Probability Proportional to Size
(PPS). That is, the more students in the target grade a school has, the more
likely it is to be sampled. Within each school 20 students were sampled
across all its target-grade classes, disregarding the number of classes with
in the school. At least 15 teachers within schools were sampled regardless
of the subject they teach in the target grade (Meinck, 2015). However, the
sampled teachers are not necessarily the teachers of the sampled students
which means that link between students cannot be linked to their teach
ers. Therefore teacher characteristics cannot be used as the explanatory
variables of student achievement or as the control variables of the relation
ship between student achievement and their SES. The list of 21 countries
participating in ICILS 2013, their sample sizes and population estimates
are presented in Table 1. As the table shows, four countries did not meet
the sampling requirements of sampling at least 150 schools within the
28
