BIBLIOGRAPHY
CALIAS, NALYN T., SILOY, MERVIN P. and WAKAT, SHARON A.
March 2009. A Loglinear Analysis on the Students’ Utilization of the Internet
Facilities at Benguet State University. Benguet State University, La Trinidad,
Benguet.
Adviser: DR. MARIA AZUCENA B. LUBRICA
ABSTRACT
The study aimed to determine the best-fitting model that approximately
summarizes the students’ frequency of utilizing the internet facilities of Benguet
State University and to determine the odds of the students’ frequency of utilizing
the internet facilities in relation to their gender, age, course, year level and
utilization outside Benguet State University.
A sample of 375 respondents was drawn from the 6118 college students of
Benguet State University using Stratified Random Sampling.
Results of the study revealed that gender and course affect the student’s
utilization of the BSU internet facilities. The frequency of utilization is independent
of the student’s age, year level and utilization outside BSU. Gender and course
revealed a very weak to weak association to the frequency of internet facilities
utilization of the students, while the age of the student showed a very weak
association to the frequency of utilization. Furthermore, year level showed a very
weak to moderate association to the student’s utilization. The findings also showed
that outside utilization revealed a weak to moderate association to the student’s
utilization of the BSU internet facilities. Moreover, the findings revealed that the
chances of often utilizing the BSU internet facilities by the students who are
female, aged 19 and above, second year, with only one computer subject, and who
have their own internet facilities is higher than those belonging to other groups.
A more thorough research on the internet usage is recommended using
either same technique or other techniques with wider scope, larger sample size and
more variables.
ii
TABLE OF CONTENTS
Page
Bibliography……………………………………………………………… i
Abstract…………………………………………………………………... i
Table of Contents………………………………………………………… iii
INTRODUCTION
Background of the Study…………………………………………. 1
Objectives of the Study…………………………………………… 3
Significance of the Study…………………………………………. 3
Scope and Delimitation…………………………………………… 4
REVIEW OF RELATED LITERATURE
Internet in Perspective……………………………………………. 5
Internet Information Services…………………………………….. 5
Purposes in Using the Internet……………………………………. 6
Importance of Internet…………………………………………….. 7
Problems Encountered in Using the Internet……………………… 8
Profile of Computer Users………………………………………… 8
Internet Use, Gender and Age……………………………………. 9
Studies on Loglinear Analysis……………………………………. 14
iii
THEORETICAL FRAMEWORK
Loglinear Analysis………………………………………………… 16
The Loglinear Model……………………………………………… 17
Saturated Model…………………………………………………… 17
Unsaturated Model………………………………………………… 18
Choosing a Model to Investigate………………………………….. 19
Fitting a Loglinear Model…………………………………………. 20
Parameter Estimates………………………………………………. 21
Testing Goodness of Fit of Loglinear Models……………………. 23
Loglinear Residuals……………………………………………….. 24
Definition of Terms………………………………………………... 25
METHODOLOGY
Respondents of the Study…………………………………………. 28
Data Analysis……………………………………………………… 29
RESULTS AND DISCUSSION
Frequency of Internet Facilities Utilization
in Relation to Gender………………………………………………. 31
Frequency of Internet Facilities Utilization
in Relation to Age Group………………………………………….... 32
Frequency of Internet Facilities Utilization
in Relation to Course……………………………………………….. 34
Frequency of Internet Facilities Utilization
in Relation to Year Level…………………………………………… 35
iv
Frequency of Internet Facilities Utilization
in Relation to Utilization Outside BSU ……………………………. 36
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
Summary………………………………………………………….... 39
Conclusions………………………………………………………… 40
Recommendations………………………………………………….. 40
LITERATURE CITED…………………………………………………….. 41
APPENDICES……………………………………………………………… 43
CURRICULUM VITAE…………………………………………………… 55
v
INTRODUCTION
Background of the Study
The “internet” has become a byword in this age of information technology.
To have an access on information, just simply click on the mouse. The internet is a
global communications network consisting of thousands of networks typically
interconnected by fiber optic cabling. The advances of technology now made it
possible to be connected via satellite so that internet access is possible even in
remote areas. The internet services include access to the worldwide web, e-mail, e-
chat and teleconferencing. The worldwide web links together relevant information
to the user’s fingertips (Kingat, 2003).
The vital role information plays in day-to-day lives makes the internet a
necessity not only for the students and teachers but also for the other people from
different walks of life. Since the internet provides students unlimited access to
knowledge and information available in the world, students now rely on the internet
for their researches and school needs.
The educational institutions’ one major function is to share information
with their client as part of their preparations for the employment market where stiff
competition awaits them. To be competitive, one needs to be equipped with
updated information and the best way of getting this is through internet access.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
2
Realizing the importance of the internet to education, Benguet State
University (BSU) got connected to cyberspace in 1997 through the partnership with
Saint Louis University. In 1999, the internet connection was enhanced with the
implementation of the Philippine Institutional University Cooperation (PIUC)
Program. Among the components of the program is an internet laboratory where
faculty members and students can visit for their information needs. The main
library of the university was the most accessible internet laboratory then.
Conceptualized as a tool for e-learning, three internet laboratories were
established by the administration. The first was established at the College of
Agriculture building, the second at the College of Nursing building and the third at
the College of Arts and Sciences building. The locations of these internet
laboratories were chosen based on whether or not the buildings had fiber optic
connection to the internet server and the buildings’ location in the university.
The internet laboratories provide internet services, software application and
Information Technology application. Bonafide teachers and students can use these
services. With the Php. 150.00 ICT Fund collected from each student for every
semester, each student may utilize the computers for a maximum of 15 hours per
semester. It is the privilege of every student to utilize these services.
This study then attempts to find out how well the internet laboratories are
carrying out its purpose in support of the curricular needs of the university. It also
attempts to determine how students utilize the services provided in the internet.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
3
Objectives of the Study
This study was primarily done to determine the students’ utilization of the
internet facilities at Benguet State University using Log linear Analysis.
Specifically, the study aimed to determine the best fitting model that approximately
summarizes the students’ frequency of utilizing the internet facilities of Benguet
State University and to determine the odds of the students’ frequency of utilizing
the BSU internet facilities in relation to their gender, age, course, year level and
utilization outside BSU.
Significance of the Study
The study could be useful to school officials in planning, improving and
enhancing the internet facilities and services of the university.
This study could also provide information on how students utilize these
facilities and avail of such services.
As a form of communication, the study will be used as a reference material
for students and interested individuals for a comprehensive and extensive research
on BSU’s internet laboratories.
Scope and Delimitation
The study was centered on the college students of the Benguet State
University (BSU)-Main Campus, school year 2008-2009. Survey questionnaires
were used to gather information from the students.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
4
The study was delimited to the gender, age, courses, year level of the
students, their frequency of utilizing the internet laboratories and utilization outside
BSU.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
REVIEW OF LITERATURE
This chapter presents related studies and literature in relation to the study.
These related literatures were presented along the following subject areas.
Internet in Perspective
Technological development involves complex system of computer networks
known as Internet. Claravall (2002) defined Internet as a collection of interlinked
computer networks or a network of networks which provide global connectivity.
Also, William (1999) defined internet as a large collection of networks that are tied
together so that many users can share their vast resources.
Moreover, Claravall (2002) enumerated the features of Internet. These are
as follows: it is global; it is not controlled by individuals; it can be used to transmit
all kinds of data in digital form; it can be accessed easily given the appropriate
equipment; and it has a significant impact on the way people live, work and
communicate.
Internet Information Services
According to O’Brien (1999), the most popular uses of the internet are as
World Wide Web (www); e-mail; use net; internet; relay chat; file transfer protocol
(FTP); telnet.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
6
As pointed by Ince (2001), internet provides the user with some space on a
web server for their own websites; e-mail addresses, web access, printing,
downloading, filtering assistance and training, chat line games, e-commerce, library
website and domain names, acts as intermediaries in e-commerce and offer other
technical assistance.
Purposes in Using the Internet
The most common reasons why higher education students access the
Internet are to identify and retrieve information relevant to research, to send
messages, or to collect data (Robinson, 1994).
Furthermore, Wells (1996) also reported that the use of Internet in higher
education focuses in the information pathway as both are means to an end unto
itself.
Also, Tolhurst and Blancton (2003) noted the reasons of students in using
Internet tools. They can exchange information quickly and conveniently; access
experts in different fields; receive regular updates on topic interested in; gain wide
area access to data; gain access to archived information; translate and transfer data
between machines; have fun and be entertained.
Likewise, Maughan (1999) added the reasons why computer users liked
best the Internet. The reasons were: opportunity to improve skill and experience on
the Internet; to explore real life issues in depth and in real time; opportunity to
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
7
work or study independently at one’s own place; uniqueness of the information
retrieved; more freedom to select themes and interpret information on an individual
basis; and discovery of topics associated with communication and information
which were new to them.
According to Ang and Loh (1996), communication which is 34.4% is the
main reason in using Internet. This was followed by access to databases and
research.
Importance of Internet
Vice President Al Gore said that the challenge to the nation’s communities
is the use of the new technologies to improve educational opportunities motivates
students and help tap their natural curiosity (WHPR, 1995). The new generation of
students has grown up in a world of computers. Wherever one looks, there are
children playing video games in an arcade or a handheld game between classes but
there are no kids with books or asking their friends for tips to do better in school. It
is generally believed that the “nation at risk” and that teachers are doing a poor job
teaching the youth of today. Some educators feel that the use of technology in
schools will allow teachers to do a better job in today’s challenging environment by
motivating students in new ways (U.S. Congress, 1995). The uses of the World
Wide Web excites and motivates students and are used in some schools as early as
kindergarten (Anonymous).
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
8
Yu and Huang (1996) reported that about 73 percent of on-line users use
only email services in the network. They also stated that most information from the
internet is useful for research.
Problems Encountered in using the Internet
As pointed out by Maughan (1998), problems in the use of Internet are
difficulty in accessing certain sites; difficulty in owning a computer.
Lehnus (1997) find out that the problems in using internet are they do not
know how to use the internet, no interest in using the internet, no unit available, the
unit does not provide floppy disk drive for using data and the internet is not always
available.
Profile of Computer Users
Majority of females (56%) learned how to use computers in school and
minority of males (35%) learned computers in school (Clarke, 1990). This is
because males had a greater access to computers outside the school than females.
Clarke (1989) added that the use of computers at home was significantly lower for
females than males. Moreover, Fyer (1994) revealed that males typically work with
computers. Studies reported that brothers and fathers not sister and mothers use the
computer at home.
A case study (Mayer, 1999) found that students who had extension
experience in using educational computing software at an after-school computer
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
9
club scored higher in comprehending word problems than did equivalent students
who had less or no exposure at all.
Internet Use, Gender and Age
Another study was done at Rutgers University in California which studied
college students and the Internet in a different prospective. It examined the amount
of usage and negative consequences that Internet overuse has on college students.
The main issues treated in this research are: What contributes to Internet overuse?;
how does gender, age, ethnicity social position of the student in the society affects
this?; what are the consequences and the site effects?
The survey consisted of 43 multiple choices on Internet usage, study habits,
academic performance and personality measures. Data were grouped to form
Internet dependent groups and the nondependent groups. It was correlated with
these factors such as guilt, not having control, using the Internet less if responded
by his friends, staying up late as well as the academic impairment and missed
classes. Recreational Internet was broken down into formats like: length of time
using the Internet, Web browsing, e-mails, chat rooms, newsgroup, online shopping
and MUDs. Then, students were asked about their feelings and attitudes towards
Internet and what effects, positive or negative this has on their lives where 572
students responded to survey. Two-thirds of the students were females, 90% were
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
10
in the first three years of college. Most of these students were enrolled in
journalism, communication, media studies courses.
When it came to results, only 9% of the sample agreed or strongly agreed
that they might have become psychologically dependent on the Internet but the
overall hours spent online for the whole sample were considerable, average of three
hours per day. Within the dependent group Internet, use was more than double of
that of the total sample. Students that spend too much time on the Internet stated:
“Sometimes I think it would be better if I spent less time on the Internet;” “Some
people have suggested to me that I spent too much time on the internet;” “I feel that
I do not always have really good control over my Internet use;” “Sometimes I feel
guilty about the amount of time I spend on the Internet”.
Researchers found out that Internet-dependent students spend nearly three
times online more than nondependent. Males, 33% of the total sample, comprised
of the self-reported Internet dependent group. In response to the question, about
“How often has your schoolwork been hurt because of the time you spend on the
Internet?” (p.373), about 14% of the students reported that their schoolwork has
been hurt occasionally, frequently or very frequently. Of the students in the
academically impaired group, 40% reported that their Internet use has kept them up
late at night frequently. Taking into consideration students social life aspects, this
study concluded that Internet dependency stem partly from lonely students
communicating with their friends and family because they feel alone in everyday
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
11
life. According to the result of this study the frequency of chat/IRC was higher than
any other online activity (p.372). The study showed that Internet can be also
harmful if students don’t manage the time spent online. This study pointed out that
Internet overuse can have negative consequences on academic work as well as
social life. It concluded, Internet dependents students, were students that didn’t
have an enhanced social life. This study also showed that there are gender
differences when it comes to Internet usage with males being more dependent.
A similar study was conducted at seven different colleges in the U.S. as
well as one school in Europe, for a total of eight institutions: a mid-sized private,
technical/engineering school, two small private liberal arts colleges, and two large
public state universities (all in the northeastern U.S.) and a non-technical college in
Ireland. There were 1,302 useable surveys with 649 men and 647 women
responding.
The study was designed to be a preliminary investigation into various
aspects of Internet use among college students. It also found out that “While the
typical Internet using students use the Internet for 100 minutes per day, there is a
small group of students that use the Internet to the degree that it interferes with
other aspects of their lives” (Anderson). Approximately 10% of Internet-using
students have used the Internet to the point that their usage meets criteria that are
parallel to those of other forms of dependence. The study focused on how excessive
Internet use results in academic, social or lifestyle difficulties. The overall time
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
12
spent online by the students was 100 minutes per day including online activities
such as WWW, e-mails, games, chatting and MUDs. The researchers divided the
students in to two groups, “the high use group” and “the low use group”. As
compared to the nondependent, the dependents were significantly more likely to
indicate that their on-line use negatively affected their academics, meeting new
people and their sleep patterns. In addition, the dependents were significantly more
likely to report; spending more than 3 consecutive hours on-line twice in the
previous week, have gotten less than 4 hours of sleep more than once due to on-line
activity, looking for an alternative way to go on-line when not in school, and to use
on-line activity to feel better when feeling down (Anderson). Like the previous
research, this study shows how Internet overuse influences negatively the students’
academic work and social life.
A study revised Internet usage more specifically taking into consideration
the university or college size. A sample of 349 were taken undergraduate students
from a medium size Midwestern university and 184 undergraduate students from a
small, private liberal arts university. Each student completed a sheet requesting
demographic, information about Internet access, amount of time spent on-line
weekly, and the types of Internet applications visited. The students were also asked
if time online interfered with work, school or interpersonal relations, and to explain
the nature of the interference. Both results from medium sized and small university
students reported having high Internet access. In the medium size university men
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
13
(M=6.89hr/week) spent more time online than women (W=4.66hr/w) but on the
contrary, also in the small university men (M=2.90hr/w) spent more time online
than women (W=2.19hr/w). The most important findings were that the small
private university uses Internet less than the public one. One of the reasons was that
the public university students are less likely to sacrifice their education for non-
academic Internet use.
An internet survey about gender and the internet was conducted with a
sample of 630 Anglo American graduates. The survey consisted of questions e-mail
and Web use. It specifically tried to point out the gender differences in internet
usage. Path Analysis to identify mediators of gender differences in internet use
revealed that computer self-efficacy, loneliness, and depression accounted in part
for gender difference, but that gender continued to have a direct effect on use often
these factors were considered (p. 363). The survey consisted of three sections: 1)
multiple aspects of e-mail, 2) web use, 3) a subset of motivational and cognitive
factors. The results of these study showed that females used e-mails more than the
males did, males used Web more than females, and females reported more
computer anxiety and less stereotypic computer attitudes (p.370). Some of the
factors that link to this are thought to be: computer self-efficacy, loneliness and
depression, which had to deal directly with the gender factor. In conclusion, the
factors that contribute to internet use and overuse are gender, ethnicity, possibility
of internet access including here students that own a computer and the other factor
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
14
is the size and potential of the university or college to provide their students with
internet access.
Studies on Loglinear Analysis
The analysis of cross-classified data changed quite dramatically with the
publication of a series of papers on loglinear models in 1970’s (Goodman, 1970).
Loglinear analysis can also be applied in medicine. In the study of patients
with community acquired pneumonia (CAP), presence of diabetes mellitus,
neoplastic diseases and neurologic diseases with significantly associated with
mortality with a significant odds ratio as perpendicular of mortality (Task Force on
CAP, 1998 as cited by Jularbal et. al, 2008).
Chayao et. al. (2005) employed loglinear analysis on the occurrence of
crimes in La Trinidad using socio-demographic and socio-economic profile of the
respondents.
An analysis of the socio-demographic and economic characteristics by
women in La Trinidad using loglinear was made by Wakat (2002). She used odds
ratio and its function Yule’s Q to get the proportion and to determine the
relationship of socio-demographic characteristics such as age, sex, civil status, and
kind of disabilities of citizens.
Add to this, Febronia (2004) applied loglinear analysis of some factors
affecting income of disabled persons in La Trinidad, Benguet. Results of the study
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
15
revealed that gender, age, civil status, kinds of disabilities and educational
attainment were related to the income of the respondents with different level of
association. The odds ratio and Yule’s Q was used to determine such relationship.
The independent model fits data on the cross tabulation of income and each of the
socio-demographic characteristics. Analysis revealed that gender, age, civil status
and kind of disabilities did not affect the disabled respondent’s income.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
THEORETICAL FRAMEWORK
Log-linear Analysis
Data that consist of count of people, places or things arise naturally when
summarizing surveys. Individuals in these surveys were classified using certain
variables as criteria. As the classifications were used simultaneously, the data that
were usually presented in tabular form were referred to as categorical data.
Loglinear techniques, propagated among social scientists like Goodman, were
excellently suited to analyze this type of data.
To discuss the analysis, a two-dimensional table was considered. Let
n be
ij
the observed frequencies in the
th
i row and
th
j columns. The data format is
presented below.
Table 1. Observed frequency distribution in an r x c table.
Variable Y
(columns)
1 2 3 .... j
Totals
1
n n n .…
n
n
11
12
13
1
j
.
1
2
n n n ….
n
n
21
22
23
2
j
.
2
Variable X
. . . …. .
.
(rows)
.
. . . …. .
.
.
.
. . . …. .
.
i
n n n … .
n
n
1
i
i2
i3
ij
r.
r
n n n ….
n
n
1
.
2
.
3
.
.
j
..
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
17
where:
n is the sample joint frequency distribution;
ij
I
n
the column marginal total
2
= ∑
n =
ij
j
j=1
I
J
n =
total number of observation
..
∑∑
n =
ij
i=1
j=1
J
n
the row marginal total
2 = ∑
n =
ij
i
i=1
The Loglinear Model
The loglinear model is one of the specified cases of generalized linear
model for Poisson-distributed data. It is an extension of the two-way contingency
table where the conditional relationship between two or more discrete categorical
variables is analyzed by taking the natural logarithm of the cell frequencies within
the contingency table.
The following model refers to the traditional chi-square test where two
variables, each with two levels (2x2 tables), are evaluated to see if an association
exists between the variables.
Saturated Model
A
B
AB
ln(
F ) = μ + λ + λ + λ
ij
i
j
ij
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
18
where:
ln(
F ) = is the log of the expected cell frequency of the cases for cell
ij in
ij
the contingency table
μ = is the overall mean of the natural log of the expected frequencies
A
λ = the main effect for variable A
i
B
λ = the main effect for variable B
j
AB
λ = the interaction effect for variables A and B
ij
λ = represents an “effect” which the variables have on the cell frequencies
Given that the saturated model has the same amount of cells in the
contingency table as it does effect, the expected cell frequencies will always
exactly match the observed frequencies, with no degrees of freedom remaining
(Knoke and Burke, 1980). In order to find a more parsimonious model, set some of
the effect parameters to zero. For instance, if the effect parameters
AB
λ are set to
ij
zero then the remaining terms is the unsaturated model.
Unsaturated Model
The unsaturated model contains both row and column effects as well as the
grand mean. Mathematically, the unsaturated model is written as
A
B
ln(
F ) = μ + λ + λ
ij
i
j
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
19
where:
ln(
F ) = is the log of the expected cell frequency of the cases for cell
ij in
ij
the contingency table
μ = is the overall mean of the natural log of the expected frequencies
A
λ = the main effect for variable A
i
B
λ = the main effect for variable B
j
λ = represents an “effect” which the variables have on the cell frequencies
The unsaturated model lacks an interaction effect parameter between A and
B. Implicitly, this model hold that the variables are unassociated. Note that the
unsaturated model’s most important assertion is that Y and X are not associated
which is analogous to the chi-square test for independence.
Choosing a Model to Investigate
Typically, either theory or previous empirical findings should guide this
process. However if a priori hypothesis does not exist, there are two approaches
that one could tale.
1. Start with the saturated model and begin to delete higher order interaction
terms until the fit of the model to the data becomes unacceptable based on the
probability standards adopted by the investigator.
2. Start with the simplest model (independence model) and add more complex
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
20
interaction terms until an acceptable fit is obtained which cannot be significantly
improved by adding further terms.
Fitting Loglinear Models
Once a model has been chosen for investigation the expected frequencies
need to be tabulated. For two variable models, the following formula can be used to
compute the direct estimates for non-saturated models.
∑
In n
ij ∑
J
ij
i=1
j−
E(
x =
1
)
(1)
∑∑
I
J nij
i=1
j=1
For large tables, an interactive proportion fitting algorithm (Deming-
Stephen algorithm) is used to generate expected frequencies. This procedure uses
marginal tables fitted by the model to ensure that the frequencies sum across the
other variables to equal the corresponding observed marginal tables (Knoke and
Burke, 1980).
The interactive proportional fitting process generates maximum likelihood
estimates of the expected cell frequencies for a hierarchical model. In short,
preliminary estimates of the expected cell frequencies are successfully adjusted to
fit each of the marginal sub-tables specified in the model. For example, in the
model AB, BC, ABC, the initial estimates are adjusted to fit AB then BC and
finally to equal the ABC observed frequencies. The previous adjustments become
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
21
distorted with each new fit, so the process starts over again with the most recent
cell estimate. This process continues until an arbitrarily small difference exists
between the current and previous estimates (Christensen, 1997).
Parameters Estimates (Odds ratio)
Once estimates of the expected frequencies for the given model are
obtained, these numbers are entered into appropriate formulas to produce the effect
parameter estimates (λ'
s) for the variables and their interactions. The effect
parameter estimates are related to odds and odds ratios. Odds are described as the
ratio between the frequency of being in one category and the frequency of not being
in the category. In symbol,
Ω
Odds
ratio
1
=
(2)
Ω2
where:
n
11
Ω =
1
nij
n
21
Ω =
2
n2
j
An odds ratio above 1 indicates a positive association among variables,
while odds ratios smaller than one indicate a negative association. An odds ratio
equal to 1 demonstrates that the variables have no association (Knoke and Burke,
1980). It was noted that odds and odds ratio are highly dependent on a particular
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
22
model. Thus, the associations illustrated by evaluating the odds ratios of a given
model are informative only to the extent that the model fits well.
Odds ratio ranges from 0 to infinity, with one indicating statistical
independence. Values less than 1.00 implies a “negative association” while values
greater than 1.00 means a “positive association”. The greater the departure of the
ratio from 1.00, in either direction, it implies stronger the relationship. The measure
of association, Yule’s Q, is a simplified function of odds ratio:
OR −1
(
n )(
n ) − (
n )(
n )
'
11
22
12
21
Yule sQ =
=
(3)
OR +1
(
n )(
n ) + (
n )(
n )
11
22
12
21
While the Yule’s Q ranges from -1 to +1, with zero indicating no
relationship, odds ratio takes only positive values, have no upper limit and is one
when no relationship exists (i.e., the two conditional odds are equal). Odds ratios
larger than one indicate direct covariation between variables, while odds ratios
smaller than one indicate an inverse relationship.
Adopting the interpretation of Yule’s Q by Landicho (1998), the
associations could be,
0 <
Q < 2
.
0 = very weak association
2
.
0 ≤
Q < 4
.
0 = weak association
4
.
0 ≤
Q < 6
.
0 = moderate association
6
.
0 ≤
Q < 8
.
0 = strong association
8
.
0 ≤
Q < 0
.
1 = very strong association
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
23
Testing Goodness of Fit of Loglinear Model
Once the model has been fitted, it is necessary to decide which model
provides the best fit. Comparing the expected frequencies for each model assesses
the overall goodness-of-fit of a model. The Pearson Chi-Square or likelihood ratio
L² can be used to test a model’s fit. However, the L² is more commonly used
because it is statistic that is minimized in maximum likelihood estimation and can
be partitioned uniquely for more powerful test of conditional independence in
multiway tables. The formula for the L² statistic is as follows:
⎛
f ⎞
L2 = 2∑
f ln⎜
ij ⎟
(4)
ij
⎜
⎟
⎝
Fij ⎠
The likelihood ratio L² follows a chi-square ( 2
χ ) distribution with the
degrees of freedom (df) equal to the number of lambda terms set equal to zero.
Thus, L² statistic tests the residual frequency that is not accounted the effects in the
model (the λ parameters set equal to zero). The larger the L² relative to the
available degrees of freedom, the more the expected frequencies depart from the
actual cell entries. Therefore, large L² indicates that the model does not fit the data.
Hence, the model should be rejected (Tabachnick and Fidell, 1996). The likelihood
ratio can be used to compare an overall model within a smaller, nested model. The
equation is as follows:
2
2
2
L comparison=
L mod 1
el −
L mod
el2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
24
Model 1 is the model nested within model 2 the degrees of freedom (df) are
calculated by subtracting the df of model 2 from the df of model 1 if the L²
comparison statistics is not significant, then the nested model (1) is not
significantly worse than the saturated model (2). Therefore, choose the more
parsimonious (nested) model.
Loglinear Residuals
In order to investigate the quality of fit of a model, one should evaluate the
individual model cell residuals. Residual frequencies can show why a model fits
poorly or can point out the cells that display a lack of fit in a generally good-fitting
model. The process involves standardizing the residuals of each cell by dividing the
difference between frequencies and frequencies expected the square root of the
frequencies expected. In symbol,
⎛
F −
F ⎞
⎜
obs
exp ⎟
Re
siduals = ⎜
⎟
(5)
⎝
Fexp
⎠
The cell with largest residuals shows where the model is not appropriate.
Therefore, if the model is appropriate for the data, the residual frequencies should
consist of both negative and positive values of approximately the same magnitude
that are distributed evenly across the cell of the table.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
25
Definition of terms
Algorithm. This is a set of procedures that, when followed in a step-by-step
manner, will provide an optical solution to a problem.
Categorical data. It refers to data that consist of count of people, place as
things grouped in any system of classification.
Contingency table. It present a multinomial count data classified in two
scales or dimension of classification.
E-chat. This is much like text messaging. It is conversing on the web by
typing the message and receiving reactions immediately.
E-mail. It is electronic mail. Writing correspondence sent through the
internet much like conventional postal mail faster and cheaper.
Dependent Variable. This refers to variable that depends on one or more
explanatory variables.
File
transfer protocol (FTP). They are downloaded data files, programs,
report, articles, magazines, books, pictures, sounds and other type of files from
thousands of source to your computer system.
Gender. This refers to either male or female.
Group A. they are courses with two or more computer subjects: B.S.
Information Technology, B.S. Applied Statistics, B.S. Development
Communication and B.S Library Science.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
26
Group B. They are courses with one computer subject: B. Elementary
Education, B. Secondary Education, B. S. Forestry, B. S. Entrepreneurial
Technology, B. S. Home Economics, B. S. Nutrition and Dietetics, B. S. Nursing,
B. S. Agriculture, B. S. Agribusiness, B. S. Environmental Science, B. S.
Agricultural Engineering, and Doctor of Veterinary Medicine
IAC. It means Internet Access Data.
Independent Variable. This is the variable used to predict values of the
dependent variable in regression analysis.
Internet access. It refers to ability to use internet.
Keyword. This is a word to type in a search engine to help look for websites
related to your topics of interest.
Loglinear Analysis. This is used to analyze the relationship of association of
cross-tabulated nominal data.
Model. It is an abstract symbolic representation of a problem, it depicts the
functional relationship among the variables.
Nominal data. This is a discrete observation that can be sorted into
categories.
Parameter. It is any characteristic of the population in the study.
Parsimonious model. The probability distribution that describes the number
of random occurrences in a given time period.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
27
Respondent. This refers to the student who will furnish the information or
answer the questionnaire.
Stratified Simple Random Sampling. This is the procedure of dividing the
population into a number of internally homogenous, non-over lapping strata.
Telnet. This is to log on to and use thousands of internet computer system
around the world.
Usenet. It refers to the collective term given to newsgroups which are
accessible via internet.
Variable. This is a characteristic of interest that is measurable and
observable in every aspect in the study.
Website. It is used for a set of link themed pages which are stored on a web
server.
World Wide Web (www). A web for short generally refers to the internet. It
is the compilation of web pages that are connected together through link.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
METHODOLOGY
Respondents of the Study
The respondents of the study were the college students of Benguet State
University-Main Campus. Out of the 6118 college students of Benguet State
University, a sample of 375 respondents was chosen as the sample of the study
using Stratified Random Sampling. The population was subdivided into two
subpopulation called strata. Stratum I included students who are taking up B.S.
Information Technology, B.S. Applied Statistics, B.S. Development
Communication, and B.S Library Science. The students from these courses are
classified as Group A. These are courses with two or more computer subjects. The
B. Elementary Education, B. Secondary Education, B. S. Forestry, B. S.
Entrepreneurial Technology, B. S. Home Economics, B. S. Nutrition and Dietetics,
B. S. Nursing, B. S. Agriculture, B. S. Agribusiness, B. S. Environmental Science,
B. S. Agricultural Engineering, and Doctor of Veterinary Medicine students
constituted stratum II. They are the Group B having only one computer subject.
The number of students in each stratum was determined using the proportional
allocation with the given formula:
Stratified Random Sampling for Proportional Allocation:
nN
n
i
=
(6)
N
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
29
N
where:
n = sample size =
, Slovin’s Formula:
2
1+
Ne
where:
N = the population size
N = subpopulation size of the
th
i stratum
i
e = margin of error
Stratum I consisted of 2,495 students which is 40.78% of the total
population. And stratum II with 3,623 students constituted 59.22% of the total
population. Thus the numbers of respondents to be taken were 153 and 222 from
stratum I and II, respectively. The sample of each subpopulation was drawn using
simple random sampling.
Survey questionnaire consisting of structured questions were administered
to the students. The respondents were asked to fill up the questionnaires by
checking their preferences from the choices given. Their gender, age, courses and
year level were asked . Their frequency of utilizing the internet laboratories and
their utilization outside Benguet State University were asked, also.
Data Analysis
The data were summarized, tabulated, analyzed and interpreted according to
the objectives of the study. Computations of summary statistics tests were
facilitated with the Statistical Packages for Social Sciences (SPSS).
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
30
Table 2. Distribution of the student respondents
VARIABLES/VARIABLE CODE
FREQUENCY
PERCENT
Gender
1-Male 152
40.5
2-Female 223
49.5
Age
1-15-18 years
188
50.1
2-19-and above
187
49.9
Course
1-Group A
153
40.8
2-Group B
222
59.2
Year
1-First Year
85
22.7
2-Second Year
101
26.9
3-Third Year
112
29.9
4-Fourth Year
77
20.5
Frequency of Utilization
1-Never 57
15.2
2-Occasionally 240
64.0
3-Always 78
20.8
Outside Utilization
1-Own internet facility
40
8.8
2-Computer Shops
335
89.3
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
RESULTS AND DISCUSSIONS
Frequency of Internet Facilities Utilization
in Relation to Gender
Table 3 shows the distribution of the Benguet State University (BSU)
students according to gender and the frequency of internet facilities utilization. The
odds ratio 1.86 means that the tendency of male to never utilize the facilities than
their tendency to always utilize the facilities is greater compared to the tendency of
female to never utilize it. The odds ratio 0.84 means that the chance of male to
utilize the internet facilities occasionally is lesser than their chance to always
utilizing it, compared to the females. A Yule’s Q value of 0.30 indicates a weak
association between male and never utilizing the internet facilities while a Yule’s Q
with -0.09 indicates a negative very weak association between being male and
occasionally utilizing BSU internet facilities.
The model that best fits the observed frequencies is the independence
model. An L² of 7.060, significant at 5% level indicates that the students’
frequency of utilizing the internet facilities is affected by their gender.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
32
Table 3. Relationship between frequency of internet facilities utilization and
gender of the students in BSU
FREQUENCY
GENDER
Never Occasionally Always
TOTAL
Male 32
89
31 152
Female 25
153 45 223
TOTAL 57 242 76
375
Odds ratio
(Yule's Q)
Male 1.86
(0.30)
0.84
(-0.09)
-
Female
-
-
-
MODEL
MODEL
DESCRIPTION
df
L² SIGNIFICANCE
Independence
μ+λ freq
gender
i
+λj
2 7.060*
0.0293
Legend: * - Significant at 5% level of significance
Yule’s Q in bold face
Interpretation:
0<|Q|<0.2 = very weak association
0.2≤|Q|<0.4= weak association
0.4≤|Q|<0.6= moderate association
0.6≤|Q|<0.8= strong association
0.8≤|Q|<1.0= very strong association
Frequency of Internet Facilities Utilization
in Relation to Age Group
An odds ratio of 1.05 in Table 4 indicates that the tendency of younger
students (15-18 years old) not to utilize the internet facilities is almost the same
with the older students (19 years old and above) who tend to do the same. The
Yule’s Q 0.02 implies a very weak association between age group and frequency of
utilization of the internet facilities. Likewise, the chance of the younger students
who opted to utilize the internet facilities occasionally than always is almost the
same with the older students, who opted to do the same. A Yule’s Q value of -0.04
indicates a negative very weak association between age group from 15-18 to 19
years old and above and frequency of utilization from occasionally to always.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
33
The best model that shows the best fit for the data is the independence
model. This is evident in the presence of low L² relative to the available degrees of
freedom. Likewise, the level of significance equal to 0.8719 implies that the
hypothesis of goodness of fit cannot be rejected since it is greater than 0.05. Hence,
age group does not affect the student’s frequency of utilizing the internet facilities.
Table 4. Relationship between frequency of internet facilities utilization and age
group of the students in BSU
FREQUENCY
AGE
Never Occasionally Always
TOTAL
15-18 30
119 39
188
19 and above
27
123
37
187
TOTAL 57 242 76 375
Odds ratio
(Yule's Q*)
15-18 1.05
(0.02)
0.92
(-0.04)
-
19 and above
-
-
-
MODEL
MODEL
DESCRIPTION
df
L² SIGNIFICANCE
Frequency
Effect
μ+λ freq
i
1
154.590**
0.0000
Age Effect
μ+λ age
j
2
0.003ns
0.9588
Independence
μ+λ freq
age
i
+λj
2 0.274ns
0.8719
Legend: ** - Highly significant at 1% level of significance
ns – Not significant
Yule’s Q in bold face
Interpretation:
0<|Q|<0.2 = very weak association
0.2≤|Q|<0.4= weak association
0.4≤|Q|<0.6= moderate association
0.6≤|Q|<0.8= strong association
0.8≤|Q|<1.0= very strong association
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
34
Frequency of Internet Facilities Utilization
in Relation to Course
As shown in table 5, the model that provides the best fit is the independence
model. The model has an L² of 7.9431 with a significance of 0.02. This means the
model is significant at 5% level of significance. This result implies that the
students’ frequency of utilizing the internet facilities could be attributed to their
course.
Table 5. Relationship between frequency of internet facilities utilization and
course of the students in BSU
FREQUENCY
COURSE
Never Occasionally Always
TOTAL
Group A
16
111
26
153
Group B
41
131
50
222
TOTAL 57
242
76 375
Odds ratio
(Yule's Q*)
Extensive 0.75
(-0.14)
1.63
(0.24)
-
Non-extensive
-
-
-
MODEL MODEL
DESCRIPTION
df
L² SIGNIFICANCE
Independence
μ+λ freq
course
i
+λj
2 7.9431*
0.02
Legend: * - Significant at 5% level of significance
Group A - courses with 2 or more computer subjects (BSIT, BSAS, BSDC and BLIS)
Group B - courses with only 1 computer subject (BEE, BSE, BSF, BSET, BSHE, BSND, BSN, BSA, BSES, BSAEng and DVM)
Yule’s Q in bold face
Interpretation:
0<|Q|<0.2 = very weak association
0.2≤|Q|<0.4= weak association
0.4≤|Q|<0.6= moderate association
0.6≤|Q|<0.8= strong association
0.8≤|Q|<1.0= very strong association
An odds ratio of 0.75 implies that the tendency of the Group A students to
never utilize the facilities over those who always utilize the facilities was found to
be lower than the tendency of Group B students. The association between groups
and frequency of utilization of internet facilities is a negative very weak association
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
35
indicated by the Yule’s Q value -0.14. The odds ratio of Group A in occasionally
utilizing the net is 1.63 times greater than in always utilizing it over the Group B.
The Yule’s Q 0.24 implied a very weak association between Group A and
occasionally utilizing the facilities.
Frequency of Internet Facilities Utilization
in Relation to Year Level
Table 6 shows the relationship between year level and frequency of internet
facilities utilization. The odds of first years from never (0.91) or occasionally (0.48)
in utilizing the Benguet State University (BSU) internet facilities to always
utilizing is lower than the fourth year students. The association between being a
first year and in never utilizing the facilities is a negative very weak association
(-0.05), and in occasionally utilizing the facilities is a negative weak association
(-0.35). The odds ratios of 1.13 and 1.04 imply that the tendency of second years to
either never or occasionally utilize the internet facilities is almost the same with
their tendency to always utilize it as compared to the fourth years. There is both a
very weak association between being a second year and either utilizing the facilities
never or occasionally as indicated by the Yule’s Q values 0.06 and 0.02,
respectively. The junior’s tendency to never utilize the internet is 0.32 times lesser,
and to occasionally utilize the internet is 0.61 times lesser than their tendency to
always utilize it over the seniors. The Yule’s Q value -0.52 and -0.24 indicate that
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
36
the association between being a junior and to never utilize the net is a negative
moderate association, and to occasionally utilize it is a negative weak association.
This is further shown in the model that gives the best fit for the observed
frequencies. The saturated model with an L² of 14.113 and significance of 0.0284
fits the data. The significance value lesser than 0.05 means rejecting the model
hence, the students’ year level does not affect their frequency of utilization on BSU
internet facilities.
Table 6. Relationship between frequency of internet facilities utilization and year
level of the students in BSU
FREQUENCY
YEAR LEVEL
Never Occasionally Always
TOTAL
I 19
45
21
85
II 17
69
15
101
III 9
75
28
112
IV 12
53
12
77
TOTAL 57 242 76 375
Odds ratio
(Yule's Q*)
I 0.91
(-0.05)
0.48
(-0.35)
-
II 1.13
(0.06)
1.04
(0.02)
-
III 0.32
(-0.52)
0.61
(-0.24)
-
IV
-
-
-
MODEL MODEL
DESCRIPTION df
L² SIGNIFICANCE
Saturated
μ+λ freq*year
ij
2 14.113*
0.0284
Legend: * - Significant at 5% level of significance
Yule’s Q in bold face
Interpretation:
0<|Q|<0.2 = very weak association
0.2≤|Q|<0.4= weak association
0.4≤|Q|<0.6= moderate association
0.6≤|Q|<0.8= strong association
0.8≤|Q|<1.0= very strong association
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
37
Frequency of Internet Facilities Utilization
in Relation to Utilization Outside BSU
The parsimonious model that best-fit the data is the independence model
(Table 7). The model’s L² = 2.913 is not significant at 5% level of significance.
This indicates that utilization of internet facilities outside Benguet State University
(BSU) does not affect the student’s frequency of BSU internet facilities utilization.
The two main effects do not interact with each other but they have significant effect
on the data.
The odds of the students whose outside utilization is their own facilities, in
never utilizing the internet is 2.66 times greater than in always utilizing it as
compared to the students whose outside utilization is in computer shops. The odds
ratio 1.71 implied that the tendency of students having their own internet in
utilizing the BSU facilities occasionally is greater than their tendency in utilizing it
always over the students not having their own net facilities. The Yule’s Q values
0.45 indicated that the association between the students having own internet and
never utilizing the BSU internet is moderate and 0.26 indicated a weak association
between students having own internet and occasionally utilizing the BSU facilities.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
38
Table 7. Relationship between frequency of internet facilities utilization and
utilization outside BSU
UTILIZATION
FREQUENCY
OUTSIDE
Never Occasionally Always
TOTAL
Own Internet
9
26
5
40
Computer Shops
48
216
71
335
TOTAL 57
242
76
375
Odds ratio
(Yule's Q*)
Own Facilities
2.66
(0.45)
1.17
(0.26)
-
Internet Shops
-
-
-
MODEL
MODEL
DESCRIPTION df L² SIGNIFICANCE
Frequency Effect
μ+λ freq
i
2 154.591**
0.0000
Outside Utilization
Effect
μ+λ out
j
1 265.244**
0.0000
Independence
μ+λ freq
out
i
+λj 2
2.913ns
0.2330
Legend: ** - Highly significant at 1% level of significance
ns – Not significant
Yule’s Q in bold face
Interpretation:
0<|Q|<0.2 = very weak association
0.2≤|Q|<0.4= weak association
0.4≤|Q|<0.6= moderate association
0.6≤|Q|<0.8= strong association
0.8≤|Q|<1.0= very strong association
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
Summary
For the data of the cross-classified variables such as (frequency of internet
facilities utilization) by (gender/age/course/utilization outside BSU), the best fitting
model that accounted for the variation of the cell frequencies is the independence
model. The saturated model best fits the data of the cross-classified variables:
(frequency of internet facilities utilization) by (year level). Gender and course
revealed a very weak to weak association to frequency of internet facilities
utilization of the students, the age of the student showed a very weak association to
the frequency of utilization. Furthermore, year level showed a very weak to
moderate association to the student’s utilization. The findings also showed that
outside utilization revealed a weak to moderate association to the student’s
utilization of the BSU internet facilities. On one hand, students, aged 15-18 years,
belonging to the Group B, who are second years, and have their own internet
facilities were more likely to never utilize the BSU internet facilities than to always
utilize it. On the other, the female students, aged 19 and above, who are second
years belonging to Group A, who have their own internet are more likely to
occasionally utilize the facilities than to always to utilize it.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
40
Conclusion
Based on the result of the study, the following conclusions were drawn: The
gender and course are the variables that affect the student’s utilization of the BSU
internet facilities. The frequency of utilization is independent from the students’
age, year level and utilization outside BSU.
Moreover, the findings revealed that the chances of often utilizing the BSU
internet facilities by the students who are females, aged 19 and above, second
years, with only one computer subject, and who have their own internet facilities is
higher than those belonging to other groups.
Recommendation
Based on the conclusions, the following are suggested: To be able to use the
multi-way log linear analysis, a large sample size should be obtained. Data from the
logbook of internet laboratories may also be used for the research. Further
investigation is recommended similar to this study with wider scope research area
including more variables specifically, those related to internet usage like family
background, psychological, social and physical environment using either similar
technique or other techniques.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
LITERATURE CITED
ANG, R. H. and LOH, C. M. 1996. Internet Developing in Asia.
http://www.isoc.org/inet96/proceedings/hl/hl-1.htm
BAGTA, F. N. et al. 2004. Log-linear Analysis of Some Factors Affecting Income
of Disabled Persons in La Trinidad, Benguet. Undergraduate Thesis.
BSU. La Trinidad, Benguet.
CHAYAO, B. et al. “Log-linear Analysis of Occurrences of Crimes in La Trinidad,
Benguet”. Undergraduate Thesis, Benguet State University. La Trinidad,
Benguet. 2005. Pp. 13-17.
CLARAVALL, N. J. 2002. Basic Definition of Terms. Teaching e-information
Skills: A Source Guide. BSU-PIUC, La Trinidad, Benguet. P.4.
DAPLIAN, E. 2005. Virtual Library Usage among Students of Benguet State
University. Undergraduate Thesis. Benguet State University. La Trinidad,
Benguet.
DOBHERTY, A. 2000. The Internet: Destined to become a Passive Surfing
Technology. Journal of Educational Technology. Englewood Cliffs, New
Jersey. Pp. 61; 62-63.
GUNN, H. 2002. Virtual Library Supporting Student Learning. Retrieve.
September 5, 2004. http://www.slaic.ubc.ca/course/lib500/00.01-
wt2/www/5Gannon/import.3htm
IDIE, N. 1996. Interview with Sony Corporation Manager. World Executive’s
Digest. Japan. Pp. 57-58.
INCE, D. 2001. Dictionary of the Internet. P. 64.
JULARBAR, M. et al. 2008. Log-linear on the Nutritional Status of Underfive
Children in La Trinidad, Beguet. Pp. 6-10.
KINGAT, JOAN. S. 2003. ”Internet Use for Instruction in Benguet State
University”. B.S. Thesis, Benguet State University. 2002. Pp. 1-2; 11-13.
MAUGHAN, G. 1998. Using the Internet to Study Communication Technology
Issues. Journal of the International Technology Education Association. P. 9.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
42
MAYER, R. E., SCHUSTACK, M. W. and BLANCTON, W. E. 1999. What do
Children Learn from Using Computers in an Informal, Collaborative
Setting?. Journal of Educational Technology. Englewood Cliffs, New
Jersey. Pp. 27-28.
O’BRIEN, J. 1999. Management Information System: Managing Information
System Technology in the Internet Worked Enterprise. Pp. 200-201.
ROBINSON, M. 1994. Improving Science: Teaching with e-mail. Computers in the
School. Pp. 11; 95-107.
TOLHURST, W. A., M.A. and BLANCTON, K. A. 2003. Using the Internet.
India polis, IN, QUE Corporation. P. 22.
WAKAT, J. 2002. Log-linear Analysis of Socio-demographic and Socio-economic
Characteristics Affecting the Income of Women in La Trinidad, Benguet. .
Undergraduate Thesis. BSU. La Trinidad, Benguet. P.11-15.
WELLS, J. G. 1996. Search Strategies for the World Wide Web. The Technology
Teacher. Pp. 55; 34-36.
WILLIAM, B. 1999. Internet for Teachers. California: IDG Books Worldwide.
P.9.
YU, C. and HUANG, N. 1996. China: Access to the Internet Growing. Newsbytes
XN96010810@clari.et.
http://www.iteawww.org. Pp. 15-17.
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
APPENDIX A
Data Set
Respondents Gender
Age
Course
Year
Frequency
Outside
1 2
1
1
2
2
2
2 2
1
1
2
3
2
3 2
1
1
3
2
2
4 2
2
1
2
2
1
5 2
1
1
2
2
2
6 2
1
1
1
2
2
7 1
2
1
3
1
2
8 2
1
1
2
3
2
9 1
2
1
4
2
2
10 2
2
1
3
2
2
11 2
2
1
3
2
2
12 2
2
1
4
3
2
13 1
2
1
4
2
2
14 2
2
1
3
3
1
15 1
2
1
3
2
2
16 1
2
1
3
2
2
17 1
1
1
3
2
2
18 1
1
1
1
1
2
19 2
2
1
4
2
2
20 2
2
1
3
2
2
21 2
1
1
3
2
2
22 2
1
1
1
3
2
23 2
1
1
3
3
2
24 1
2
1
4
2
2
25 1
1
1
2
2
2
26 1
2
1
4
2
2
27 2
1
1
1
2
2
28 2
2
1
4
1
2
29 2
2
1
4
2
2
30 2
2
1
4
1
2
31 1
2
1
4
1
2
32 1
1
1
1
1
1
33 2
2
1
4
2
2
34 1
2
1
4
2
2
35 1
2
1
3
1
2
36 2
1
1
1
2
2
37 1
2
1
3
2
2
38 2
1
1
2
2
2
39 1
2
1
3
2
2
40 2
2
1
2
2
2
41 1
2
1
4
2
2
42 1
2
1
2
3
2
43 1
1
1
2
2
2
44 1
1
1
1
3
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
44
Appendix A continued…
45 2
1
1
2
2
2
46 2
1
1
2
2
2
47 1
2
1
3
3
2
48 1
2
1
3
2
2
49 1
1
1
2
2
1
50 2
1
1
1
3
2
51 2
1
1
2
2
2
52 2
1
1
2
3
2
53 2
1
1
2
2
2
54 2
1
1
1
2
2
55 1
1
1
3
2
2
56 2
1
1
1
2
2
57 2
1
1
1
2
2
58 2
1
1
3
3
2
59 2
1
1
1
2
2
60 2
1
1
1
2
2
61 2
1
1
1
2
2
62 2
2
1
4
2
2
63 2
1
1
1
2
2
64 2
1
1
2
2
1
65 1
2
1
4
3
2
66 2
2
1
4
2
2
67 1
2
1
4
2
2
68 2
1
1
1
3
2
69 2
1
1
1
2
2
70 2
1
1
1
2
2
71 2
1
1
1
3
2
72 1
1
1
1
3
2
73 1
2
1
1
2
2
74 1
1
1
1
2
2
75 2
1
1
2
2
2
76 1
2
1
4
2
2
77 2
1
1
2
2
2
78 2
1
1
2
2
2
79 2
1
1
2
2
2
80 2
1
1
2
2
2
81 2
1
1
2
2
2
82 2
1
1
2
2
2
83 2
1
1
2
2
2
84 2
1
1
2
2
2
85 2
1
1
1
2
2
86 2
1
1
3
2
2
87 1
2
1
4
2
2
88 1
1
1
3
3
2
89 2
2
1
4
2
2
90 2
2
1
4
2
2
91 2
1
1
1
2
2
92 2
2
1
4
2
2
93 1
2
1
4
3
2
94 1
2
1
4
2
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
45
Appendix A continued…
95 2
2
1
2
2
2
96 2
2
1
3
2
2
97 1
2
1
3
3
2
98 2
1
1
1
2
2
99 2
1
1
1
2
2
100 1
1
1
1
2
2
101 1
1
1
1
2
2
102 1
1
1
3
2
1
103 1
1
1
1
2
2
104 1
2
1
3
2
2
105 2
1
1
2
2
2
106 1
2
1
1
2
2
107 1
2
1
3
1
1
108 2
1
1
2
1
1
109 1
1
1
1
1
1
110 1
1
1
3
2
1
111 1
2
1
3
2
1
112 1
2
1
3
3
2
113 1
2
1
3
2
1
114 2
1
1
2
2
2
115 1
1
1
3
2
2
116 1
1
1
1
2
2
117 2
1
1
3
3
2
118 1
1
1
1
2
2
119 2
2
1
3
2
2
120 2
1
1
1
2
1
121 2
2
1
3
2
2
122 1
2
1
4
2
1
123 2
2
1
4
2
1
124 1
2
1
3
1
2
125 2
2
1
3
2
2
126 2
2
1
4
2
2
127 2
1
1
3
2
2
128 2
1
1
2
3
2
129 2
2
1
4
2
2
130 1
2
1
3
1
2
131 2
1
1
2
2
2
132 1
1
1
1
2
2
133 2
2
1
4
2
1
134 1
1
1
1
3
2
135 1
2
1
4
2
1
136 2
2
1
4
2
2
137 2
2
1
4
1
2
138 2
2
1
2
2
2
139 2
2
1
3
2
1
140 2
1
1
1
3
2
141 2
1
1
2
2
2
142 2
1
1
2
2
2
143 2
2
1
2
2
2
144 2
2
1
4
1
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
46
Appendix A continued…
145 2
2
1
4
2
2
146 1
2
1
4
2
2
147 2
2
1
3
1
1
148 1
1
1
1
1
2
149 2
2
1
2
2
1
150 1
1
1
1
2
1
151 1
2
1
2
2
1
152 2
2
1
4
3
2
153 1
2
1
3
3
1
154 2
1
2
3
2
2
155 2
2
2
3
2
2
156 1
2
2
4
1
2
157 2
1
2
2
2
2
158 2
1
2
2
2
2
159 2
2
2
3
3
2
160 1
2
2
4
2
2
161 2
2
2
3
3
2
162 2
2
2
3
3
2
163 1
1
2
2
1
2
164 2
2
2
3
3
2
165 2
2
2
3
2
2
166 1
1
2
2
1
2
167 2
1
2
3
2
2
168 2
2
2
3
2
2
169 2
2
2
3
2
2
170 2
2
2
3
2
2
171 1
2
2
2
2
2
172 2
2
2
4
3
2
173 1
1
2
2
2
2
174 2
1
2
2
3
2
175 1
1
2
2
2
1
176 2
1
2
1
1
2
177 2
1
2
1
2
2
178 2
2
2
3
2
2
179 2
1
2
2
3
2
180 2
2
2
4
2
2
181 2
2
2
4
1
2
182 2
1
2
1
1
2
183 1
1
2
2
2
2
184 1
1
2
2
2
2
185 2
2
2
4
2
2
186 2
2
2
4
2
2
187 2
1
2
3
3
2
188 2
2
2
4
3
2
189 2
2
2
3
3
2
190 2
2
2
3
2
1
191 2
1
2
3
2
2
192 2
1
2
3
2
2
193 2
1
2
3
2
2
194 2
1
2
1
2
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
47
Appendix A continued…
195 2
1
2
1
2
2
196 2
1
2
3
2
2
197 1
2
2
2
3
2
198 1
1
2
1
3
1
199 1
1
2
1
2
2
200 2
2
2
3
3
2
201 1
2
2
3
3
2
202 1
1
2
1
3
2
203 1
1
2
1
2
2
204 1
2
2
4
2
2
205 1
2
2
4
1
2
206 1
2
2
4
1
2
207 1
2
2
4
1
2
208 1
2
2
3
2
2
209 1
2
2
4
2
2
210 1
1
2
1
2
2
211 2
2
2
4
2
2
212 2
1
2
2
2
2
213 2
1
2
2
2
2
214 1
1
2
2
2
2
215 2
1
2
1
2
2
216 1
2
2
3
3
2
217 1
2
2
3
2
2
218 1
2
2
3
2
2
219 2
1
2
2
3
2
220 2
1
2
2
2
2
221 2
1
2
2
2
2
222 2
1
2
2
2
1
223 1
2
2
3
3
2
224 1
2
2
3
2
2
225 2
2
2
4
2
2
226 1
1
2
2
2
2
227 1
2
2
3
3
2
228 2
2
2
4
2
2
229 2
2
2
3
2
2
230 1
2
2
4
1
2
231 1
2
2
2
1
2
232 1
1
2
2
1
2
233 1
2
2
4
1
2
234 1
2
2
3
1
1
235 2
1
2
2
2
2
236 1
2
2
2
1
2
237 2
1
2
2
3
2
238 1
2
2
3
2
2
239 2
1
2
2
3
2
240 1
1
2
2
2
2
241 2
1
2
2
2
2
242 1
1
2
2
2
2
243 2
2
2
4
2
2
244 2
2
2
2
2
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
48
Appendix A continued…
245 2
1
2
3
2
2
246 2
1
2
2
2
2
247 2
1
2
2
2
2
248 2
1
2
2
2
2
249 2
2
2
3
2
2
250 1
1
2
3
2
2
251 2
1
2
1
2
2
252 1
1
2
2
2
2
253 1
1
2
1
2
2
254 2
2
2
4
2
1
255 2
1
2
2
2
2
256 1
2
2
3
2
2
257 1
2
2
4
2
2
258 2
2
2
4
2
2
259 1
2
2
3
2
1
260 2
2
2
3
2
2
261 2
2
2
4
2
2
262 2
2
2
3
2
2
263 2
2
2
3
2
2
264 2
1
2
1
2
2
265 2
1
2
1
1
2
266 2
1
2
1
3
2
267 2
1
2
2
3
2
268 2
2
2
3
2
2
269 2
1
2
1
1
2
270 2
1
2
1
1
2
271 2
1
2
1
1
2
272 2
1
2
2
1
2
273 1
2
2
2
1
2
274 2
1
2
1
2
2
275 1
2
2
3
3
2
276 2
2
2
1
2
2
277 1
1
2
2
2
2
278 2
1
2
2
2
2
279 1
2
2
2
2
2
280 2
2
2
3
2
2
281 2
1
2
3
2
2
282 1
2
2
3
2
2
283 2
2
2
3
2
2
284 2
2
2
2
3
2
285 2
1
2
1
2
2
286 2
1
2
1
2
2
287 2
2
2
4
2
2
288 1
2
2
4
2
2
289 2
2
2
4
3
2
290 1
1
2
1
3
2
291 2
1
2
1
3
2
292 1
1
2
3
2
2
293 2
2
2
3
2
2
294 1
2
2
3
2
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
49
Appendix A continued…
295 1
1
2
1
3
1
296 2
1
2
3
3
2
297 2
1
2
3
2
2
298 1
1
2
2
2
2
299 2
2
2
3
2
1
300 1
2
2
3
3
2
301 1
2
2
3
2
2
302 1
2
2
3
2
2
303 2
1
2
2
3
2
304 2
1
2
2
2
2
305 1
2
2
4
2
2
306 2
2
2
3
2
2
307 2
2
2
4
2
2
308 2
2
2
4
3
2
309 2
1
2
1
1
2
310 2
1
2
1
1
2
311 2
1
2
1
1
2
312 1
1
2
2
1
2
313 1
2
2
2
1
2
314 2
1
2
1
3
1
315 2
1
2
1
3
2
316 2
1
2
2
2
2
317 2
2
2
3
2
2
318 1
2
2
4
2
2
319 2
1
2
1
1
2
320 2
1
2
1
1
2
321 1
2
2
4
3
2
322 1
2
2
4
2
2
323 1
1
2
1
2
1
324 2
1
2
1
2
2
325 2
1
2
3
2
2
326 2
2
2
4
3
2
327 1
1
2
2
2
2
328 1
2
2
4
2
1
329 2
1
2
2
2
2
330 1
1
2
1
1
1
331 2
1
2
1
1
2
332 1
1
2
2
1
2
333 1
2
2
2
1
2
334 1
2
2
2
1
2
335 2
1
2
2
1
2
336 1
1
2
2
1
1
337 2
1
2
1
1
2
338 2
1
2
1
1
1
339 2
2
2
3
1
2
340 1
2
2
2
1
2
341 2
1
2
3
1
2
342 1
1
2
2
1
2
343 1
2
2
3
2
2
344 2
2
2
1
3
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
50
Appendix A continued…
345 1
1
2
1
3
2
346 2
2
2
3
2
1
347 2
2
2
4
2
1
348 1
2
2
3
2
2
349 2
2
2
4
2
2
350 1
1
2
1
3
2
351 2
2
2
3
2
2
352 2
2
2
1
2
2
353 1
2
2
3
3
2
354 1
1
2
1
2
2
355 1
1
2
2
2
2
356 1
1
2
1
3
2
357 2
1
2
2
3
2
358 2
2
2
3
3
2
359 1
1
2
1
3
2
360 1
1
2
2
2
2
361 2
2
2
2
3
2
362 1
1
2
3
3
2
363 2
2
2
4
2
2
364 2
2
2
4
2
2
365 1
2
2
3
3
2
366 2
2
2
3
2
2
367 2
2
2
3
2
2
368 1
2
2
2
2
2
369 1
2
2
4
2
2
370 1
2
2
3
3
2
371 2
2
2
4
3
2
372 1
1
2
2
2
2
373 2
2
2
3
2
2
374 2
2
2
3
3
2
375 2
2
2
4
3
2
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
51
APPENDIX B
Sample Survey Questionnaire
Gender: ___Male ___Female
Course & Year: _______
Age: ______
On the use of Internet Laboratories
1. How often do you use the internet laboratories?
_____ Never
_____ Occasionally
_____ Always
_____ Others (Please specify):_______________________
2. Why do you use the internet facilities?
_____ for research, assignment
_____ for entertainment
_____ for communication
_____ for thesis, encoding
_____others (Please specify):_________
3. Which internet laboratory do you often go?
_____ Main Library
_____ College of Agriculture
_____ College of Arts and Sciences
_____ College of Nursing
4. Are you satisfied with the internet services at BSU?
_____ Highly Satisfied
_____ Moderately Satisfied
_____ Not Satisfied
5. Where else do you utilize internet connection outside BSU?
_____ own internet facilities
_____ computer shops
A Loglinear Analysis on the Students’ Utilization of the Internet Facilities
at Benguet State University / Nalyn T. Calias; et al. 2009
Document Outline
- A Loglinear Analysis on the Students� Utilization of the Internet
Facilities at Benguet State University
- BIBLIOGRAPHY
- ABSTRACT
- TABLE OF CONTENTS
- INTRODUCTION
- REVIEW OF LITERATURE
- THEORETICAL FRAMEWORK
- METHODOLOGY
- RESULTS AND DISCUSSIONS
- SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
- LITERATURE CITED
- APPENDIX