ArticlePDF Available

The Mental Health Impact of Computer and Internet Training on a Multi-ethnic Sample of Community-Dwelling Older Adults: Results of a Pilot Randomised Controlled Trial


Abstract and Figures

We preliminarily explored the effects of computer and internet training in older age and attempted to address the diversity gap in the ethnogeriatric literature, given that, in our study's sample, only one-third of the participants self-identified as White. The aim of this investigation was to compare two groups - the control and the experimental conditions - regarding theme 1) computer attitudes and related self-efficacy, and theme 2) self-esteem and depressive symptomatology. Sixty non-institutionalized residents of Los Angeles County (mean age ± SD: 69.12 ± 10.37 years; age range: 51-92) were randomly assigned to either the experimental group (n=30) or the waitlist/control group (n=30). The experimental group was involved in 6 weeks of one-on-one computer and internet training for one 2-hour session per week. The same training was administered to the control participants after their post-test. Outcome measures included the four variables, organized into the two aforementioned themes. There were no significant between-group differences in either post-test computer attitudes or self-esteem. However, findings revealed that the experimental group reported greater computer self-efficacy, compared to the waitlist/control group, at post-test/follow-up [F(1,56)=28.89, p=0.001, η2 =0.01]. Additionally, at the end of the computer and internet training, there was a substantial and statistically significant decrease in depression scores among those in the experimental group when compared to the waitlist/control group [F(1,55)=9.06, p<0.004, η2 =0.02]. There were significant improvements in favour of the experimental group in computer self-efficacy and, of noteworthy clinical relevance, in depression, as evidenced by a decreased percentage of significantly depressed experimental subjects from 36.7% at baseline to 16.7% at the end of our intervention.
Content may be subject to copyright. Int J Biomed Sci Vol. 9 No. 3 September 2013 135
InternatIonal journal of BIomedIcal scIence
e Mental Health Impact of Computer and
Internet Training on a Multi-ethnic Sample of
Community-Dwelling Older Adults:
Results of a Pilot Randomised Controlled Trial
Luciana Laganá1, James J. García2
1Department of Clinical Psychology, California State University Northridge, Northridge, California, USA;
2Department of Psychology, University of North Texas, Denton, Texas, USA
Introduction: We preliminarily explored the effects of computer and internet training in older age and
attempted to address the diversity gap in the ethnogeriatric literature, given that, in our study’s sample,
only one-third of the participants self-identied as White. The aim of this investigation was to compare two
groups - the control and the experimental conditions - regarding theme 1) computer attitudes and related
self-efcacy, and theme 2) self-esteem and depressive symptomatology. Methods: Sixty non-institutionalized
residents of Los Angeles County (mean age ± SD: 69.12 ± 10.37 years; age range: 51-92) were randomly
assigned to either the experimental group (n=30) or the waitlist/control group (n=30). The experimental
group was involved in 6 weeks of one-on-one computer and internet training for one 2-hour session per
week. The same training was administered to the control participants after their post-test. Outcome mea-
sures included the four variables, organized into the two aforementioned themes. Results: There were no
signicant between-group differences in either post-test computer attitudes or self-esteem. However, nd-
ings revealed that the experimental group reported greater computer self-efcacy, compared to the wait-
list/control group, at post-test/follow-up [F(1,56)=28.89, p=0.001, η2=0.01]. Additionally, at the end of the
computer and internet training, there was a substantial and statistically signicant decrease in depression
scores among those in the experimental group when compared to the waitlist/control group [F(1,55)= 9.06 ,
p<0.004, η2=0.02]. Conclusions: There were signicant improvements in favour of the experimental group
in computer self-efcacy and, of noteworthy clinical relevance, in depression, as evidenced by a decreased
percentage of signicantly depressed experimental subjects from 36.7% at baseline to 16.7% at the end of
our intervention. (Int J Biomed Sci 2013; 9 (3): 135-147)
Keywords: Older adults; Ethnic diversity; Depression; Self-esteem; Computer attitudes; Computer self-efcacy;
Computer technology training
Corresponding author: Luciana Lag aná, Depar tment of Clinical Psychology, California State University Northridge, Northridge, California, USA.
Note: Luciana Lag an á de signed the stud y, co llected the data wit h her students, and wro te the rst dr af t of the man uscript; James J. García co nd uc ted
the data analyses, updated the liter ature, wrote the “results” section, and contributed to writing the nal version of the manuscript.
Received June 8, 2013; Accepted August 26, 2013
Copyright: ? 2013 Luciana Laganá et al. This is an open-access article distributed under the terms of the Creative Commons At tribution License
(, which permits unrestricted use, distribution, and reproduction in any medium, provided the original
author and source are credited.
Mental health iMpact of coMputer and interne t training for older adults
Septe mber 2013 Vol. 9 No. 3 Int J Biomed Sci
In the present study, we were interested in testing the
effects of a computer technology training intervention on
older adults’ computer technology comfort and well–be-
ing. It is particularly important to investigate ways to de-
crease depressive symptomatology regardless of age, as
one of the most common psychological disorders for all
Americans is depression (1). Its prevalence within senior
populations, the target of the present study, is high, with
about 5 million older adults in the U.S.A. experiencing
some form of persistent depressive symptomatology. Be-
tween 5% and 10% of older adults who visit their primary
care physicians are depressed (2, 3). Moreover, research
in this area should target non-White populations, as indi-
viduals of racial and ethnic minority backgrounds over the
age of 65 represent a rapidly growing segment of the U.S.
population, currently totaling over 13% (4). Specic to the
location of our study, in a 2008 investigation surveying
16,500 older adults residing in Los Angeles County (5),
compared to White seniors, older adults from racial/ethnic
groups reported higher rates of a variety of unmet needs
such as greater health needs, employment needs, social
isolation concerns, as well as housing, transportation, and
caregiving needs. Hispanic/Latino residents, who were
among those reporting the most unmet needs, stated that
daily activities, in particular, were a problematic issue: a
phenomenon which could certainly contribute to depres-
sive symptomatology in this population.
In addition to the troubling prevalence of depressive
symptomatology in older age, self-esteem can decrease in
older age due to social role losses, reduction of physical
beauty, decreased health, and related decits. The inclu-
sion of self-esteem in studies on interventions for geriatric
depression is methodologically appropriate and conceptu-
ally pertinent, given that there is a strong inverse asso-
ciation between self-esteem and depression in many age
groups (6, 7). Among the few geriatric studies on this
issue, one study (8) has shown a signicant inverse rela-
tionship between these two variables among older adults
with orthopedic disabilities. A more dated study on cor-
relates of self-esteem (9) revealed that seniors with low
self-esteem had signicantly more depression than those
with high self-esteem. The authors of the aforementioned
study, based on their ndings, highlighted the need to de-
velop interventions that promote enhanced self-image in
older age. Another reason for targeting self-esteem in ge-
riatric interventions is that, as it has been suggested in the
literature (10), it is best not to focus exclusively on psycho-
pathology when conducting research on seniors. Instead,
studies should include more favourable characteristics of
older adults, such as their potential for gaining higher self-
esteem following successful completion of an educational
intervention similar to the one tested herein.
Why implement a computer technology training in-
tervention to improve the well-being of older adults? This
could be a clever choice, given that, often due to fear of
stigma, older adults are usually reluctant to seek tradi-
tional mental health services, even when in need (11), as
stigma and other reasons such as lack of insurance and/or
transportation present signicant barriers to access. Con-
sequently, interventions with an “apparent” exclusively
educational focus, such as computer technology training,
could serve the well-intended purpose of bypassing se-
niors’ common resistance to pursue mental health servic-
es. For computer illiterate older adults, receiving computer
and internet training could have many benets, as use of
this technology in older age has been signicantly linked
to positive aging and seniors’ adaptation to the aging pro-
cess (12). Positive uses of the internet in older age (and
at all ages) include: communicating with loved ones, ex-
ploring hobbies/interests, accessing community resources,
and increasing socialisation (e.g., meeting people through
bulletin boards and chat rooms). There is an array of other
benets to using the internet in older age, as its utilization
affords access to health literacy and ultimately to health
action (i.e., taking control of one’s health at the community
level). Indeed, using the internet is critical for gathering
consumer health-related information on websites such as (13), as well as for geriatric healthcare
delivery and the prevention of age-related impairments
(14). Overall, online activities can provide a means to ac-
cess a variety of information as well as control interac-
tional choices (15), which could contribute to increased
well-being, as tested herein.
The rst logical step when planning a computer train-
ing intervention, prior to testing its effects on well-being,
is to test its impact on computer technology beliefs such as
computer attitudes and self-efcacy.
In the present study, we also posed the question: “Is it
possible to extend the benets of our computer technology
intervention to mental health under a controlled condi-
tion?” Indeed, if individuals, at any age, are exposed to a
positive and successful training experience in which they
feel competent and able to master a task, they may gener-
alize such a successful experience to other areas. This, in
turn, could positively impact global self-esteem and men-
tal health (16, 17).
Mental health iMpact of coMputer and interne t training for older adults Int J Biomed Sci Vol. 9 No. 3 September 2013 137
Concerning prior ndings of interventions that ad-
dressed computer attitudes and computer self-efcacy, our
laboratory obtained positive results, described in a 2008
article by the rst author (18) on a pilot study with 32 se-
niors in which computer technology training signicantly
improved computer attitudes and computer self-efcacy.
Such a theoretically- and empirically-based endeavor was
conceptualized as a Phase I trial study that did not target
mental health factors. Nonetheless, it was a critical step to-
ward ascer taining whether our i ntervention could have ben-
ecial effects on seniors; these ndings were replicated by
the rst author and her colleagues in 2011 in a randomised
controlled trial on a larger sample (19). These results are
in line with ndings from other laboratories that demon-
strated the efcacy of computer training at enhancing: a)
seniors’ computer attitudes (20-24); b) older adults’ level
of comfort with computer use after attending a SeniorNet
computer lab class (25); and c) young students’ computer
self-efcacy (26, 27). Intervention studies on older adults’
computer self-efcacy are practically non-existent, with a
few exceptions: 1) a recent study showing that computer
training and internet use by cognitively intact older adults
signicantly increased their computer-related self-efcacy
(28) and 2) the rst author’s aforementioned two studies.
Moreover, information on this outcome variable with re-
gard to ethnically diverse senior populations is scarce. If
training people in computer and internet use can improve
their computer attitudes and computer self-efcacy, then,
for the mental health-related reasons discussed above, old-
er adults – and especially ethnic minority seniors (often
neglected in research investigations) – have a clear need
for this training.
A particular focus of the present intervention was the
potential of one-on-one computer technology training to
improve mood symptomatology. Regarding the theoretical
foundation of this supposition, two related theories apply to
this discussion: 1) a resource-related mental health theory
by Hobfoll and Wells (29) and 2) a theory of conservation
of resources in older age (30) by Gatz. Briey, regarding
the rst theory, Hobfoll and Wells view resources avail-
able to seniors at both an individual level (e.g., personal
health) and at a broader level (e.g., technological resources,
internet) as having a signicant impact on mental health
outcomes. Similarly, the second theory by Gatz argues in
favour of paying careful attention to the psychosocial re-
sources of older adults to explain mental health disorders.
In our case, learning computer/internet use could be added
to seniors’ resources and consequently decrease their de-
pressive symptomatology.
However, results of geriatric interventions targeting
mood symptomatology as well as self-esteem through
computer/internet training have produced conicting re-
sults and failed to identify a well-validated intervention
that could address these clinical issues. Some research
ndings suggest that computer/internet training adminis-
tered to seniors yields a reduction in depressive symptom-
atology (31-33). However, some scholars did not identify
signicant depression-related improvements in their senior
trainees (24), while others, following 9 hours of seniors
group training over 2 weeks, only detected a trend that was
not statistically signicant at the .05 level (15). One reason
for this lack of signicance may be the group training for-
mat or the limited 2-week duration. Also, the results of a
more recent study with a sample similar to ours in terms
of health status and independent living status (34) showed
no improvements in depressive symptomatology follow-
ing two weeks during which three 4-hour training sessions
were provided to the intervention group. Moreover, com-
puter/internet training has the potential to enhance elders’
self-esteem (35), but again the scarce literature on this
topic is divisive. For instance, while the ndings of a study
showed improved self-esteem among seniors who became
computer users upon receiving weekly personal train-
ing over the course of 3 months (31), another investiga-
tion failed to demonstrate signicant self-esteem changes
among older adults who received a three-day one-on-one
computer/internet training (24). We could speculate that
maybe the short duration of the training, and/or seniors’
high level of psychological adjustment at baseline, affect-
ed the aforementioned results, but research is still needed
in order to clarify this issue. It should be noted that the out-
come of the aforementioned two studies conducted by the
rst author using computer technology training provided
no indication of whether the intervention was capable of
ameliorating seniors’ mental health, although the ndings
indicated signicant post-training improvements in com-
puter technology beliefs.
As to the present study’s hypotheses, research by the
rst author targeting the enhancement of computer atti-
tudes and self-efcacy via the same computer technology
training implemented in the current study (19) was suc-
cessful. Likewise, we expected to nd signicant post-
test improvements in trainees on both computer attitudes
and self-efcacy at the end of the intervention. Moreover,
computer technology attitudes and computer self-efca-
cy were hypothesized to be signicantly related, based
on prior ndings on this topic (36) and given that these
two variables are both dimensions of computer technol-
Mental health iMpact of coMputer and interne t training for older adults
Septe mber 2013 Vol. 9 No. 3 Int J Biomed Sci
ogy beliefs. We also chose a non-pathological variable -
self-esteem - and hypothesized signicant improvements
in this factor in the experimental trainees by the end of
our training. Specically important in the eld of ethno-
geriatric psychiatry, we chose depressive symptomatology
as an outcome variable that may be potentially amenable
to positive changes through our training, expecting sig-
nicant improvements only in experimental participants’
depressive symptomatology. We based this decision on the
available literature in support of this hypothesis, and in
particular on the encouraging results of an aforementioned
study conducted in Israel by Shapira, Barak, and Gal (33)
with Hebrew-speaking older adults (mean age=80.25 for
the experimental group and 82.60 for the control group)
using a very similar sample size (i.e., 22 experimental
older adults and 26 controls) to that of our study. Shapira
et al.s ndings showed signicant improvement in de-
pressive symptomatology following 15 weeks of computer
training and internet use. When compared to our study,
several elements were different, as their training was more
than double the length of our intervention (while most oth-
er studies implemented an even shorter training than ours)
and was presented in a group format versus a one-on-one
structure. Also, the sample was ethnically homogeneous,
as opposed to our ethnically diverse sample. Thus, testing
the current intervention adds to the available empirical lit-
erature on one-on-one computer technology training and
its effects on older ethnic minorities.
We recruited 60 community-dwelling older adults, age
51 to 92 years (mean=69.12). They volunteered to be in the
study (like the participants in the vast majority of research
in this area) and were residents of Los Angeles County,
California, United States of America (U.S.A). A variety
of sampling strategies were used by the interviewers/re-
search assistants (RAs), including purposive sampling
(i.e., using their connections in their ethnic communities)
and snowball sampling (i.e., mentioning to research par-
ticipants that we were looking for referrals to other older
adults who could participate in this research). Our goal
was to overcome some of the limitations of prior research
in this area by gathering an ethnically diverse sample and
attempting to include isolated older individuals who knew
at least one person in their community who could refer
them to us. Adopting these sampling techniques allowed
us to maximize the chances of obtaining a representative
sample of older men and women residing in Los Ange-
les County, a very ethnically diverse area. We advertised
this research project at several locations including stores,
churches, senior centres, and senior apartment complexes.
None of the respondents was recruited at a clinical facility
or via a medical referral, the signicance of which will be
addressed later in this discussion.
The following inclusion criteria were established: 1)
being at least 50 years-old; 2) being uent in English if
this was their second language (in order to avoid further
confounding our ndings with levels of acculturation); 3)
being willing and able to attend all six sessions of our one-
on-one training (even if our control/waitlist participants
rst received six weekly visits, not the training); 4) staying
in the area for the next two months, and (5) being able to
access a computer at their 顺心彩票. This last criterion was
necessary in order for all our respondents to reap the ben-
ets of our training after the study was completed (includ-
ing the training of the waitlist participants), in accordance
with the research ethics principle of offering training only
for skills that can be used feasibly in the long run. Indeed,
it has been documented that studying computer technol-
ogy techniques is not benecial to older adults unless
they practice these skills on a regular basis (37), because
computer use needs to be learned by way of both action
and practice (38). Exclusion criteria were: 1) residing in
an institutional setting; 2) being unable to grant informed
consent (i.e., not being uent in English enough to fully
comprehend the content of the consent form); and 3) hav-
ing more than ‘minor’ computer technology experience,
i.e., having turned a computer on and off or having been
exposed to other people utilizing computer technology to
write documents or use the internet. As a result of imple-
menting such criteria, all our participants were computer
and internet illiterate (i.e., non-users of the technology
taught in our training).
Our four outcome variables were assessed twice (pre-
and post-test) while the rest of the variables were assessed
only at baseline, to avoid burdening older adults. To quan-
tify the socio-demographic and computer-related issues
required for inclusion/exclusion purposes, we used a list of
items that covered all the inclusion and exclusion criteria;
this is the same list as the one used in the rst author’s
2011 study (19). It contains variables of interest including
age, education, household income, and ethnic background,
as well as computer ownership, access to a computer, prior
computer experience, and ability to e-mail.
Mental health iMpact of coMputer and interne t training for older adults Int J Biomed Sci Vol. 9 No. 3 September 2013 139
To calculate computer technology attitudes, we utilized
the latest version of the Older Adults’ Computer Technol-
ogy Attitudes Scale [OACTAS (18, 19)], which has achieved
strong reliability results, as reported in the aforementioned
2011 study (Cronbach’s α reliability=0.92). All its 17 items
are negatively worded, in an attempt to elicit candid re-
sponses to computer technology questions from computer
illiterate individuals. Responses are coded on a 7-point
Likert-type scale from ‘-3’ ‘strongly disagree’ to ‘+3’
‘strongly agree’; the scores are reversed before conduct-
ing data analyses, in order to have higher numbers denote
more positive computer technology attitudes.
The 30-item Computer User Self-Efcacy Scale (39)
was used to assess computer-self-efcacy; according to its
authors, the Cronbach’s α of this measure is 0.97 and the
test-retest reliability coefcient is 0.86. Items are rated on
a 6-point Likert-type scale ranging from ‘1=strongly dis-
agree’ to ‘6=strongly agree’. The only minimal adaptation
of this tool involved deleting an irrelevant introductory
item applicable only to college students.
To quantify health status (which was done for descrip-
tive purposes only), we used a very short version of a
well-validated health measure, the 12-item SF-12 Health
Survey, which is a sound measure of health status (40).
Its multi-item scale assesses 8 health concepts: physical
functioning, role limitations due to physical health prob-
lems, bodily pain, general health, vitality (energy/fatigue),
social functioning, role limitations due to emotional prob-
lems, and mental health (psychological distress and psy-
chological well-being). This measure has high test-retest
reliability scores (0.76-0.89). To minimize burdening our
older research participants, given that physical health was
not one of our outcome variables, we asked only the rst
two questions: 1) self-rated health and 2) ease of engaging
in moderate activities of daily living, including moving a
table, pushing a vacuum cleaner, bowling, or playing golf.
Self-esteem was measured using the Rosenberg Self-
Esteem Scale: a 10-item, 6-point Likert-type scale mea-
suring basic feelings of self-worth (41, 42). Its Cronbachs
α internal consistency is 0.74 among non-institutionalized
seniors (43). It has been previously utilized with older
adults (42) and recognized in the literature as an appro-
priate scale for measuring global self-esteem in older age
(44). Its utilization allowed the operationalisation of our
respondents’ global self-esteem at baseline and follow-up
We used the Beck Depression Inventory – II (BDI-II) to
assess depressive symptomatology; it contains 21 sets of 4
statements that describe varying intensities of somatic and
cognitive-affective symptoms of depression (45). Respon-
dents choose the one statement from each group that best
describes how they have been feeling for the past 2 weeks.
This tool is appropriate for use with geriatric samples (46)
and was utilized to operationalise changes in mood symp-
tomatology before and after our intervention. Based on
ndings of a study on a depressed geriatric sample (47),
the BDI-II’s internal consistency is very high (α=0.90),
and gender, ethnicity, or age are not signicantly related to
the total scores on this measure. This is ideal for our sam-
ple, as it is comprised primarily by women. An appropri-
ate BDI-II’s cut-off score for signicant depression among
geriatric populations is 10, as using this score in a study
on cognitively intact older adults led to 96.30% sensitiv-
ity in correctly identifying depressed and non-depressed
subjects (48).
Research design, procedures, and computer technol-
ogy training
In the current pilot Phase II Efcacy study, we test-
ed, for the rst time, the potential positive impact of our
one-on-one, manualized training intervention on depres-
sive symptomatology and self-esteem in older age. We
also tested changes in computer attitudes and computer
self-efcacy, to verify whether these changes were neces-
sary before improvement in well-being could occur. We
conceptualized these four variables into two themes, i.e.,
theme 1) computer attitudes and computer self-efcacy,
and theme 2) self-esteem and depressive symptomatology.
As done in the rst author’s two prior studies, the present
research was conducted following the recommendations
of the rst author’s original research model relative to the
implementation of high-quality research on community-
dwelling older adults (49). This model pays particular at-
tention to avoiding methodological challenges common to
this type of geriatric research.
We conducted a randomised, controlled 6-week inter-
vention; our research procedures were in accordance with
the ethical standards of the Institutional Review Board
of California State University Northridge concerning re-
search employing human subjects. The one-on-one com-
puter training programme imparted in this project was
designed by the rst author (18) to enhance older adults’
computer technology attitudes and self-efcacy. This is
the rst time that our laboratory has used this manual-
ized one-on-one training for well-being/mental health en-
hancement purposes. Each older adult recruited for this
study signed our consent form. Every respondent was as-
signed an RA to perform the pre- and post-tests as well as
Mental health iMpact of coMputer and interne t training for older adults
Septe mber 2013 Vol. 9 No. 3 Int J Biomed Sci
to train him/her; at baseline, RAs collected data on socio-
demographic attributes, physical health, and on the four
outcome variables. After six weeks, all participants were
re-tested on the four outcome variables, including the con-
trol subjects, who were trained following completion of
the second assessment. Control participants were visited
by their RA for one and a half hours per week without
engaging in training, in order to match the amount of at-
tention given to participants in both groups. At post-test,
our experimental subjects were asked to e-mail their RA,
in the presence of the trainer but without any assistance
(to make sure that the trainee had indeed sent the email).
All experimental subjects were able to complete this email
The rst author trained all RAs to ensure their effec-
tiveness as one-on-one computer trainers; the latter were
asked to avoid deviating from training manual instruc-
tions and, for the purpose of quality assessment, to keep a
diary of the training experience with each trainee and to
document anomalies or deviations from the instructions.
No substantial deviations were reported, as evidenced by
inspections of the diaries’ content by the second author
and several RAs. The training protocol was implemented
on a desktop computer, at locations identied by the par-
ticipants as being convenient, including the Department
of Psychology at California State University Northridge
and several libraries in the area. The rst author wrote the
training manual in order to standardise the training pro-
cedure. Its content has been described in detail elsewhere
in the literature (18-19). Generally speaking, in our train-
ing, we aimed at maximizing trainees’ active participation
in learning computer and internet use and asked RAs to
provide fast feedback to trainees on their progress during
training. After using the same training manual, the RAs in
the aforementioned 2011 study by the rst author reported
that they (as well as their trainees) found the manual and
the related training easy to follow and comprehend. Pos-
sibly due to all these procedures being in place, we did not
experience any subject loss from pre- to post-test.
Power Analyses
A-priori power analyses were run with G*Power Ver-
sion 3.1.5 (50) in order to identify the number of partici-
pants needed for each group, based on the effect sizes (ES;
η2 converted to Cohen’s ?) reported in several randomised
control trials of internet training interventions assessing
similar outcome variables to those of the current study.
According to Cohen’s (51) recommendations for adequate
power (i.e., >0.80), for theme one, the experimental and
control groups each required 12 participants for the com-
puter self-efcacy variable and 6 participants for the com-
puter attitudes variable, according to estimates provided
by Laganá (18) [EScomputer self-efcacy = 0.94; EScomputer attitudes
= 2.06]. As for theme two, the experimental and control
group each required 13 participants for the self-esteem
variable based on Billipp’s ndings (31) [ESself-es teem=0.87]
and 29 participant for depressive symptoms according to
estimates provided by Shapira, Barak, and Gal (33) [ES-
depression=0.55]. Thus, we chose the upper limit of 30 par-
ticipants per group to ensure our ability to detect these ef-
fects. Our a-priori power analyses suggest that the effects
for these interventions are subtle.
Analytic strategy
In line with the analytic strategy adopted in the 2011
randomised controlled study on only computer attitudes
and computer self-efcacy, we intended to run two sepa-
rate MANCOVAs, one per theme not correlated over 0.30,
using the Statistical Package for the Social Sciences. The
rst MANCOVA would allow the testing of post-training
changes in the two computer-focused outcome variables,
the second one in the two mental well-being variables.
However, if any of these two sets of two variables had been
correlated over 0.30, we intended to conduct Roy-Barg-
mann’s stepdown analyses instead (involving separate
ANCOVAs), thus complying with the methodological rec-
ommendations of well-respected statistical sources (52).
To avoid losing power in our analyses, given the limited
size of the sample, we did not plan to use any other covari-
ates than those dictated by the data analyses procedures,
i.e., pre-test scores when predicting post-test scores in ad-
dition to another covariate if required by Roy-Bargmann’s
procedures (in the result section below, we have detailed
this occurrence).
Concerning the frequency ndings, Table 1 illustrates
the demographic and health characteristics of the sample.
We recruited 42 women and 18 men; only about 1/3 of our
sample self-identied as White. As to perceived physi-
cal health status, 31.7 % of the sample self-rated physical
health as fair, 48.3% as good, 18.3% as very good, and
1.7% as excellent. The intercorrelation matrix, which con-
tains Pearson product-moment correlation coefcients, is
reported in Table 2. It should be noted that we did not use
the variable income in our analyses, as over one fourth of
the sample either did not know the answer or refused to
Mental health iMpact of coMputer and interne t training for older adults Int J Biomed Sci Vol. 9 No. 3 September 2013 141
answer this question. Instead, we used the variable educa-
tion as its proxy, given that education and income are often
used as proxies for socioeconomic status (53).
The means of the four outcome variables are displayed
by group in Table 3. Baseline and post-intervention mean
scores on self-esteem and depression were similar in the
two groups. Based on the typical cut-off score used in the
geriatric depression literature (48), the control group fully
qualied at baseline as being signicantly depressed, giv-
en that the mean depression scores were above the clini-
cal cut-off score of 10. The experimental group exhibited
depression scores that were almost clinically signicant,
as the mean score was over 9. However, regarding the
computer-related outcomes (i.e., computer self-efcacy
and computer attitudes), the two groups’ baseline means
appeared markedly different. To test whether these differ-
ences were statistically signicant, we compared group
means by conducting two separate t-tests. The ndings
suggested no signicant differences between the experi-
mental and control participants at baseline for computer
attitudes [t(58)=1.06, p=0.29] or computer self-efcacy
[t(58)=-1.86, p=0.07], which suggests that both groups held
statistically similar beliefs and attitudes regarding com-
puter technology at baseline. Given these results, it was
methodologically adequate to compare our two groups on
all four outcome variables.
As previously mentioned, we planned on conducting
either two MANCOVAs with two outcome variables each
- one for the computer-related theme and one for the men-
tal well-being theme - or a series of four separate ANCO-
VAs if, applying Roy-Bargmann’s suggestions (52), these
two sets of variables had correlations over 0.30. Relative
to our computer-related theme, we found a high post-test
computer attitudes and computer self-efcacy correlation
(r=-0.56, p<0.001), which required the use of two separate
A N C OVA s.
In the rst analysis/step of the Roy-Bargmann’s proce-
dure for the computer-related theme, we implemented an
ANCOVA, controlling for baseline attitudes and baseline
computer self-efcacy. This was done in order to test for
the presence of training improvements for computer atti-
tudes in the experimental participants, as this variable was
theoretically the rst computer-related factor to consider
as potentially being impacted by our intervention. Homo-
geneity of variance and regression assumptions were not
met, and there were no group differences with regard to
computer attitudes [F(1,56)=0.01, p=0.93, η2<0. 01] .
Concerning the second step of this procedure, to test
post-test group differences in computer self-efcacy due
to training, we controlled for baseline computer attitudes
and baseline computer self-efcacy as well as for post-test
computer attitudes, in line with the Roy-Bargmann’s pro-
cedure. Both the homogeneity of variance and the test of
the homogeneity of regression assumption were met. The
results of the ANCOVA showed a signicant main effect
for group at follow-up (as illustrated in Figure 1), with the
Table 1. Characteristics of the sample
Var i a b le
Age 69.12 (10.37 )
European-American 32.7
Mexican-A merican 20
Other Hispanic/Latino 7
Asian-American 15
Middle Eastern 22
American Indian/ Native American 3.3
Less than High School 33.3
Graduated from High School 35
Completed Trade School 6.7
Some college 13.3
Bachelor’s degree 5
Some graduate school 1.7
Master’s degree 1.7
Ph.D., M.D., and/or J.D. 1.7
Refused to Answer 1.7
Yearly Income
Less than $20,000 20
$20,000-$39,000 28.3
Over $40,000 25
Refused to Answer 26.3
Self-rated general health 3.10 (2 .13)
Impair ment in activities of daily living 2.13 (0.68)
Baseline computer attitudes 71.09 (23.09)
Baseline computer self-efcacy 81.60 (2 6. 53)
Baseline self-esteem 16.10 (3.8 0)
Baseline depression 10.00 (7.24)
Mental health iMpact of coMputer and interne t training for older adults
Septe mber 2013 Vol. 9 No. 3 Int J Biomed Sci
Table 2. Zero-order correlations between demographic, independent, and dependent variables
Variable Age Sex Education
in activities of
daily living
Age -0.07 - 0.11 -0.07 -0.37** -0.02 0.04
Sex -0. 31* 0.16 0.14 -0.03 0.06
Education -0.07 0.20 0.01 -0.06
Self-rated general health --0.36** 0.13 0.10
Impair ment in activities of daily living - -0.30* - 0.15
Baseline computer attitudes -0.69**
Post-test computer attitudes -
Baseline computer self-efcacy
Post-test computer self-efcacy
Baseline self-esteem
Post-test self-esteem
Baseline depression
Post-test depression
Table 2. Zero-order correlations between demographic, independent, and dependent variables (Continued)
Age -0.02 - 0.13 -0.21 -0.06 0.01 - 0.01
Sex 0.05 0.06 0.08 0.15 0.12 0.06
Education 0. 21 0.28* 0.01 0 .16 0.15 0.03
Self-rated general health -0.16 -0.05 0.33** 0.40** 0.43** 0.36**
Impair ment in activities of daily living - 0.16 -0.05 0.04 -0.27* - 0. 26* -0.30*
Baseline computer attitudes -0.56** - 0.40** 0.01 0.27* 0.18 0.25
Post-test computer attitudes - 0 . 51** -0.56** 0.01 0 .18 0.17 0.23
Baseline computer self-efcacy -0.73** -0.05 -0.13 - 0.13 - 0.23
Post-test computer self-efcacy -0.01 0.13 - 0.14 -0.30*
Baseline self-esteem -0.58** 0.08 0.04
Post-test self-esteem -0.38** 0.31*
Baseline depression -0.85**
Post-test depression -
*p<0.05; **p<0.01
Table 3. Pre- and post-intervention means of the four dependent variables by group/condition
Var i a b le Control group
baseline mean
Control group
post-intervention mean
Experimental group
baseline mean
Group post-intervention mean
Computer attitudes 74.25 73.79 67.9 3 68.65
Computer self-efcacy 75.37 75.95 87. 83 108 .18
Self-esteem 15.76 16. 46 16.44 15.66
Depression 10.96 11.40 9. 04 6.88
Mental health iMpact of coMputer and interne t training for older adults Int J Biomed Sci Vol. 9 No. 3 September 2013 143
experimental group reporting higher computer self-efca-
cy compared to the control group [F(1,56)=22.98, p=0.001,
η2=0. 01].
The mental health/well-being theme of the study was
tested via the nal two ANCOVAs, given that the corre-
lation between depression and self-esteem was over 0.30
(i.e., r=0.31, p<0.05). In step 1 of this procedure, the out-
come variable to test for training-related improvements
was self-esteem, which we conceptualized as the rst
well-being factor to possibly be impacted by our interven-
tion. Thus, in this Roy-Bargmann step-down ANCOVA,
we targeted post-test self-esteem while controlling for
baseline self-esteem and baseline depression. Like in the
rst ANCOVA, the homogeneity assumptions were not
met, and we did not detect any signicant differences con-
cerning self-esteem between the experimental and control
groups [F(1, 56)=11.07, p=0.26, η2<0.01].
In the nal step, we tested post-test group differences
in depressive symptomatology while controlling for base-
line depressive symptomatology and baseline self-esteem
as well as for post-test self-esteem. The homogeneity of
variance and the test of the homogeneity of regression as-
sumption were met. Results demonstrated a signicant
main effect for group at follow-up (as depicted in Figure
2), with the experimental group reporting signicantly
lower depressive symptomatology compared to the control
group [F(1, 55)=9.06, p=0.004, η2=0.02].
We also conducted pre-post reliability analyses for the
four outcome variables, in order to rule out measurement
error as a possible confound for our result, as poor reli-
ability introduces error and may decrease the ability to de-
tect an effect. Our measures were robust, given the modest
sample size, considering that a result of 0.80 and above
indicates adequate reliability (51). For computer attitudes,
the pre-test internal consistency coefcient showed excel-
lent results at 0.90 and at post-test was almost identical
at 0.89. Regarding computer self-efcacy, the pre-test in-
ternal consistency coefcient was excellent at 0.92 and at
post-test appeared somewhat improved at 0.95. For self-es-
teem, the pre-test internal consistency coefcient was 0.79
and at post-test was 0.81, indicating adequate reliability.
For BDI/ depressive symptomatology scores, the pre-test
internal consistency coefcient was .89, and at post-test
was 0.91; these results are indicative of robust clinical util-
it y.
In this study, we compared two groups of community-
dwelling seniors to investigate the effects of the rst author’s
manualized, one-on-one computer technology training for
older adults on two themes, i.e., theme 1) computer attitudes
and often-related self-efcacy and theme 2) self-esteem
and its well-known correlate, depressive symptomatology.
Alarmingly, the mean score on depression for this sample
was at the clinical cut-off score of 10; thus, our intervention
was indeed needed. As a result, the present investigation
became a randomised controlled study for a clinical sample.
Although the two variable s in each theme were signicantly
related, only one of them per theme showed to be amenable
to positive changes in older age as a function of computer
and internet training. Our ndings have psychiatric rel-
evance given that, in addition to positively impacting com-
puter self-efcacy, our computer technology training had
Computer self-efficacy scores
Control group
Intervention group
Figure 1. Visual representation of the self-efcacy results.
Control group
Intervention group
Depression Scores
Figure 2. Visual representation of the depressive symptomatol-
ogy results.
Mental health iMpact of coMputer and interne t training for older adults
Septe mber 2013 Vol. 9 No. 3 Int J Biomed Sci
benecial effects on depression in the experimental group,
pushing this group’s mean depression score well below the
clinical cut-off score. This is noteworthy, considering the
relatively small sample size of our study.
The ndings of the rst ANCOVA were somewhat
perplexing, as they contradict the results of the afore-
mentioned studies by the rst author and those of other
researchers (20-24). The results of this ANCOVA also op-
pose prior empirical evidence (55) suggesting that negative
computer attitudes in older age can stem from having lim-
ited computer technology experience, as our experimental
participants were given several weeks of experience with
this technology. More research is needed to replicate this
nding; perhaps it is indeed possible for older adults not to
improve their views of computer technology in older age
and still reap the benets of this technology.
The results of the ANCOVA on computer self-efcacy
were in line with our lab’s prior ndings (18, 19) as well as
with results from other laboratories (28). The signicantly
improved computer self-efcacy among experimental par-
ticipants may be attributable to improvements in sense of
mastery of the tasks learned during training, such as how
to surf the internet and send emails to loved ones. If liking
this technology did not improve as a result of our interven-
tion, self-assurance regarding computer usage certainly
did. Considering the already relatively high baseline mean
of computer self-efcacy for experimental participants, the
present ndings suggest that, if we provide computer and
internet training to computer-illiterate, ethnically diverse
older adults, they can still signicantly increase their com-
puter self-efcacy as a function of training regardless of
their level of pre-training condence about being able to
learn and use this technology. As part of our inquiry re-
garding a possible signicant relationship between com-
puter self-efcacy and depression, we asked this question
of the experimental group only, as this is the group in which
we detected signicant changes on these two variables as a
result of our training. However, upon conducting ancillary
analyses to examine the relationship of computer self-ef-
cacy to depression both at baseline and after the training, we
did not obtain signicant relationships at pre-test (r=- 0.16,
p=0.39) nor at post-test (r= - 0.12 , p=0.54). Furthermore,
post-hoc partial correlational analyses, controlling for the
inuence of baseline computer self-efcacy and baseline
depressive symptomatology, further revealed that post-test
computer self-efcacy among trainees was not signicantly
related to post-test depression scores (r=-0.15, p=0.46). This
result supports prior ndings on depression and internet
self-efcacy not being signicantly related among under-
graduate students (56) and extends them to an ethnically
diverse geriatric population.
The mental health/well-being theme of our study in-
cluded two variables, self-esteem and depression. Self-
esteem failed to show signicance, in line with prior lit-
erature showing that this variable does not improve with
computer training in older age (24) but in contrast with
research ndings by Billipp (31) showing that it becomes
signicantly higher in older age as a result of receiving
weekly personal training over the course of 3 months and
becoming regular user. Perhaps Billipp’s aforementioned
positive results were due to the fact that trainees became
computer users for several months, while our trainees
were not monitored in their computer use aside from par-
ticipating in our intervention. Future research is needed to
clarify this point.
Depressions scores that, prior to training, were on the
cusp of clinical signicance in our experimental group,
were very positively impacted by our training; this is the
most clinically relevant outcome of the present study. In-
deed, experimental participants started the training at levels
of depression that were statistically comparable to those of
the control subjects (who were signicantly clinically de-
pressed). Yet, as a result of training, trainees’ depression
scores were sig nicantly reduced. It is clinically notewor t hy
that the percentage of signicantly depressed experimental
subjects was reduced by 20%, i.e., from 36.7% at baseline
to only 16.7% after the intervention. Our depression result
supports some prior ndings with sample sizes similar to
ours [e.g., 22 experimental older adults and 26 controls
(33)] but conicts with results reported by other researchers
(34), which showed no signicant depression reduction as
a function of computer training in older age. Furthermore,
the total sample’s mean score of 10 on depression at baseline
suggests that many non-institutionalized, ethnically diverse
older adults are living with untreated depression: based on
this number, depression should become a very high-priority
target of community health programmes for older adults of
all ethnic backgrounds. More research is needed to corrobo-
rate our depression ndings.
Limitations of the study
Several limitations of this study must be acknowledged,
such as the previously noted modest size of its sample - al-
though some of the aforementioned geriatric studies gath-
ered comparably sized samples or even smaller samples.
Furthermore, although most of our participants were non-
White, they all resided in urban or suburban areas of Los
Angeles County, which limits generalization of our results
Mental health iMpact of coMputer and interne t training for older adults Int J Biomed Sci Vol. 9 No. 3 September 2013 145
to seniors residing in rural areas or those living outside of
the United States. Additionally, the relatively brief 6-week
intervention may not have allowed us to truly capture the
effect of internet training on our two themes over time
(i.e., longitudinally). Thus, future research directions may
include examining internet training and its effect on com-
puter attitudes and related self-efcacy as well as global
self-esteem and depressive symptoms in order to 1) quan-
tify the effect size over time and 2) identify the dose-re-
sponse of this intervention. Also, the fact that men com-
prised only 30% of our sample precluded the possibility of
conducting statistically meaningful gender comparisons.
In future studies, interested researchers should investigate
whether gender plays a signicant role in the effects of
similar interventions, as the available evidence in this area
points to older women being under-represented online and
not reporting reaping substantial benets from using the
internet (57). In order to substantiate the present results
and related explanations, there is certainly a need for more
adequately powered future investigations that should ide-
ally include the assessment of anti-depressant medication
use and the utilization of other treatment modalities for
depression. Concerning the strength of our results, in the
present study, we reported effect sizes as eta squared, as
opposed to partial eta squared; there are several advan-
tages to using eta squared, as described in the literature
(58). Our effects for both computer self-efcacy and de-
pression were not statistically trivial, according to Cohen
(51). However, as evidenced by the ndings of our power
analyses, the effects for these types of interventions tend
to be subtle, which may also reect the typical use of small
samples in this area of research.
Briey, regarding cultural considerations, to our
knowledge, this is the rst U.S.A.-based geriatric study
on the effects of computer technology training in which
about 68% of the subjects recruited are non-White. Given
that Hispanics represented 27% of our sample and Asians
15%, for a total of 42% (Whites were 32.7%), cultural is-
sues should be considered in the discussion of our depres-
sion ndings. To cite just one cultural value, due to space
limitations, familismo refers to the importance of close
family relations and intergenerational exchanges of social
support. It is a particularly strong value among Hispanic
and Asian older adults (59). In line with the concept of
familismo (not assessed here as it was beyond the scope of
this study), increased use of the internet may have created
an important avenue for increased social support and net-
working in older age, allowing older Hispanic and Asian
participants, for instance, to establish and maintain contact
(over the course of the 6 weeks of training) with grand-
children who moved away to attend college or other fam-
ily members not living nearby. This new opportunity for
strengthening family contacts via online interaction (not
assessed herein as it was not within the scope of this re-
search) might have positively affected their mental health
and well-being, and could have been a factor impacting
our depression ndings. Research in this area is needed to
experimentally test this conjecture. Also, in future stud-
ies, the recruitment of ethnically diverse older adults pre-
senting with a range of psychopathology of different kinds
would be ideal, as different mental health pathologies
could be tested for mitigation post-training. This would
allow researchers to test factors potentially affecting im-
proved psychiatric symptomatology, such as increased
online contact with loved ones, engagement in uplifting
online activities, or enhanced community interaction as a
result of acquiring community-related information or per-
sonal contacts online.
From a clinical standpoint, this investigation was a
controlled study on a clinical sample. Although we did
not intend to recruit a clinical sample, on average, our 60
community-dwelling seniors had signicant depression
at baseline, which was markedly ameliorated to the point
of no longer being clinically signicant for 20% of the
experimental group. Our ndings suggest that computer
and internet training can lead to higher levels of com-
puter self-efcacy and improved mood symptomatology
among older, computer-illiterate seniors from a variety
of ethnic backgrounds. Gains on these two outcome vari-
ables were not reported by our control subjects. If our
intervention is further conrmed as having psychiatri-
cally signicant impact, the relatively inexpensive form
of computer technology training used in this study could
become an effective yet neutral/non-pathologizing in-
tervention for the reduction of mood psychopathology
(ideally in conjunction with needed psychiatric or other
pertinent treatment). Such an intervention would be con-
sistent with the aforementioned values related to avoid-
ing the stigma of receiving mental healthcare often held
by ethnically diverse senior populations.
The authors declare that no conicting interests exist.
Mental health iMpact of coMputer and interne t training for older adults
Septe mber 2013 Vol. 9 No. 3 Int J Biomed Sci
Conception, design, manuscript drafting, and supervi-
sion of data collection: LL; Data analyses, results writing,
and critical revision of manuscript for its content: JG. All
authors read and approved the nal manuscript.
The present research was supported by NIMH grant
3 R24 MH 67851-03S1 and by NIH grant MBRS 1 SC3
GM 094075-01, Luciana Laganá, Principal Investigator.
The authors thank the rst author’s gerontology students
for their assistance with data collection and with data col-
lection and for performing the meticulous training of the
research participants.
1. American Psychological Association. Division 20 Homepage. 19 95-
2003; Chapter 4 -Priority 4.
2. Lyness JM, Caine ED, King DA, Cox C, et al. Psychiatric disorders
in older primar y care pat ients. Journal of General Internal Medicine.
1999; 14: 249-254.
3. Schulberg HC, Katon WJ, Simon GE, Rush AJ. Best clinical practice:
Guidelines for managing major depression in primary medical care.
Journal of Clinical Psychiatr y. 1999; 60 (7): 19-26.
4. U.S. Bureau of Census. Statistical Information. Los Angeles Depar t-
ment of City Planning/Demog raphic Research Unit. 2003.
5. Los Angeles County Depar tment of Commun ity and Senior Citizens
Services. L.A. County Seniors Count! Survey of the Older Adult Popu-
lation. Los Angeles, CA: Los Angeles Count y, Community and S enior
顺心彩票s. 2008.
6. Cave DG. Enacting change: A therapeutic enactment group-based pro-
gram for traumatized soldiers. Dissertation Abstracts International
Section A. 2004; 64 (12-A): 4367.
7. Repetto PB. A longitudinal study of depressive symptoms among
African-American adolescents: Their patterns of change and effects
on school outcomes and substance use. Dissertation Abstracts Inter-
national Section B. 2003; 64 (2-B): 674.
8. Zoerink DA. Exploring the relat ionship between leisure and health of
senior adults with orthopedic disabilities living in rural areas. Activi-
ties, Adaptation and Aging. 2001; 26 (2): 61-73.
9. Hunter KI, Linn MW, Harris R. Characteristics of high and low self-
esteem in the elderly. International Journal of Aging and Human
Development. 1981-1982; 14 (2): 117-126.
10. Schaie KW. Ageist language in psychological research. American Psy-
chologist. 1993; 48 (1): 49 -51.
11. Laganá L, Shanks S. Mutual biases underlying the problematic rela-
tionship between older adults and mental health providers: Any solu-
tion in sight? International Journal of Aging and Human Develop -
ment. 2002; 55 (3): 271-295.
12. Contarello A, Sarrica M, Romaioli D. Ageing in a broadband society.
An exploration on ICTs, emotional experience and social well-being
within a social representation perspective. New York, NY: Peterlang.
2011; 5: 247-258.
13. Char ness N, Holley P. The new media and older adults. American
Behavioral Scientist. 2004; 48 (4): 416- 433.
14. Charness N, Boot WR. Aging and information technology use: Poten-
tial and barriers. Current Directions in Psychological Science. 2009;
18: 253-258.
15. White H, McCon nell EM, Clipp E, Branch LG, et al. A randomized
controlled trial of the psychological i mpact of providing internet train-
ing and access to older adults. Aging and Mental Health. 2002; 6 (3):
16. Bandura A. Self-efcacy: Toward a unifying theory of behavioral
change. Psychological Re view. 1977; 8 4 (2): 191-215.
17. Bandura A. Exercise of human function ing through collective efcacy.
Current Directions in Psychological Science. 2000; 9 (3): 75-78.
18. Laganá L. Enha ncing the attitudes and self-efcacy of older adults
towards computers and the Internet: Results of a pilot st udy. Educa-
tional Gerontolog y. 2008; 34: 831-843.
19. Laganá L, Oliver T, Ai nswort h A, Edwards M. Enhancing computer
self-efcacy and attit udes in multiethnic older adults: A randomized
controlled study. Ageing & Society. 2011; 31 (6): 911-933.
20. Charness N, Schuma nn C, Boritz G. Training older adults in word
processi ng: effects of age, training technique, and computer anxiety.
International Journal of Technology and Aging. 1992; 5 (1): 79-105.
21. Jay GM, Willis SL. Inuence of direct computer experience on older
adults’ attitudes towards computers. Journal of Gerontology: Psycho-
logical Sciences. 1992; 47: 250–257.
22. Dyck JL, Smither JA. Age differences in computer anxiety: The role
of computer experience, gender and education. Journal of Educat ional
Computing Research. 1994; 10: 231-240.
23. Morrell RW, Etcht KV. Instr uctional design for older computer users:
The inuence of cognitive factors. In W. A. Rogers, A. D. Fisk, et al.
(Eds.). Aging and skilled perfor mance: Advances in theory and appli-
cations. Hillsdale, NJ: Lawrence Erlbaum Associates. 1996; p 241-26 5.
24. Kelley CL, Mor rell RW, Park DC, Mayhorn CB. Predictors of elec-
tronic bu lletin board system use in older adults. Educational Gerontol-
ogy. 1999; 25 (1): 19-35.
25. Segrist K A. Attitudes of older adults toward a computer training pro-
gram. Educational Gerontology. 2004; 30 (7 ): 563-571.
26. Ch ristian CAS. The different ial effects of training on computer self-
efcacy. Dissertation Abstracts International Section A. 2000; 61
(5-A): 180 6.
27. Torkzadeh R, Pughoeft K, Hall L. Computer self-efcacy, training
effectiveness and user attit udes: An empirical study. Behaviour and
Informat ion Technolog y. 1999; 18 (4): 299-309.
28. Wild KV, Mattek NC, Maxwell SA, Dodge HH, et al. Computer-
related self-efcacy and an xiety in older adults wit h and wit hout mild
cognitive impair ment. Alzheimer’s & Dement ia: The Journal of the
Alzheimer’s Association. 2012; 8 (6 ): 544-552.
29. Hobfoll SE, Wells JD. Conservation of resources, stress, and aging:
Why do some slide and some spring? In J. Lomranz (Ed.), Handbook
of aging and mental health: An integrated approach. New York: Ple-
num. 1998; p121-134.
30. Gatz M. Toward a developmentally infor med theory of mental disorder
in older adults. In J. Lomranz (Ed.), Handbook of aging and mental
health: An integrated approach. New York: Plen um. 1998; p101-120.
31. Billipp SH. T he psychosocial impact of interactive computer use
within a vulnerable elderly population: A report on a randomized pro-
spective trial in 顺心彩票 health care setting. Public Health Nursing. 2001;
18 (2): 138 -145.
32. McConatha D, Tahmaseb McConatha J, Dermigny R. The use of inter-
active computer services to enhance the quality of life for long-term
care residents. Gerontologist. 199 4; 34 (4): 553 -556.
33. Shapira N, Bara k A, Gal I. Promoting older adults’ well-being th rough
Inter net training and use. Aging & Mental Health . 20 0 7; 11 (5): 477-
Mental health iMpact of coMputer and interne t training for older adults Int J Biomed Sci Vol. 9 No. 3 September 2013 1 47
34. Slegers K, van Boxtel M, Jolles J. Effects of computer training and
Inter net usage on the well-being and quality of life of older adults: A
randomized, controlled study. Journal of Gerontolog y: Psychological
Sciences. 2008; 63 (3): 176-184.
35. Eilers ML. Older adults and computer education: ‘Not to have the
world a closed door’. International Journal of Technology and Aging.
1989; 2 (1): 56-76.
36. Wu YT, Tsai CC. University students’ Internet attitudes and Internet
self-efcacy: a study at three Universities in Taiwan. CyberPsychol-
ogy and Behavior. 2006; 9 (4): 441-450.
37. Buse CE. E-scaping the ageing body? Computer technologies and
embodiment in later life. Ageing and Societ y. 2010; 30: 987-1009.
38. Wacquant L. Body and Soul. Oxford: Oxford Universit y Press. 2004.
39. Cassidy S, Eachus P. Development of the Computer User Self-Ef cacy
(CUSE) Scale: Investigat ing the relationship between computer self-
efcacy, gender and experience with computer s. Journal of Educa-
tional Computing Research . 2002; 26 (2): 169-189.
40. Ware JE, Kosinski M, Keller SD. A 12-Item Short-For m Health Survey
(SF-12): Construction of scales and preliminary tests of reliabilit y and
validity. Medical Care. 1996; 32 (3): 220-233.
41. Rosenberg M. Societ y and the adolescent self-image. Princeton, NJ:
Princeton University Press. 1965.
42. Ranzijn R, Keeves J, Luszcz M, Feather NT. The role of self-perceived
usefulness and competence in the self-esteem of elderly adults: Con-
rmatory factor analyses of the Bachman revision of Ronseberg’s
Self-Esteem Scale. Journal of Gerontology B Psychology Science and
Social Sciences. 1998; 53 (2): 96 -104.
43. Ward RA. The impact of subjective age and stig ma on older persons.
Journal of Gerontology. 1977; 32: 227-232 .
44. Breytspraa k LM, George LK. Self-concept and self-esteem. In D.J.
Mangen & W.A. Peterson (Eds.), Research instruments in social ger-
ontology clinical and social psychology. Minneapolis, MN: University
of Minnesota Press. 1982; p241-302.
45. Beck AT, Steer RA. Manual for the revised Beck Depression Inven-
tory. San Antonio, TX: Psychological Corporation. 19 87.
46. Olin JT, Schneider LS, Eaton EM, Zemansky MF. The Geriatr ic
Depression Scale and the Beck Depression Inventory as screening
instruments in an older adult out patient population. Psychological
Assessment. 1992; 4: 190-192.
47. Steer RA, Rissmiller DJ, Beck AT. Use of the Beck Depression Inven-
tory-II with depressed ger iatric inpatients. Behaviour Research and
The rap y. 2000; 38 (3): 311-318.
48. Laprise R, Vezina J. Diagnostic per formance of the Geriatric Depres-
sion Scale and the Beck Depression Inventory with nursing 顺心彩票 resi-
dents. Canadian Journal on Aging. 1998; 17 (4): 401-413.
49. Laganá L, Rubin M. Methodological challenges and potential solu-
tions for the i ncorporation of sound community-based research into
Service-Learning. In A. Furco & S. H. Billig (Eds.), Advances in Ser-
vice-Learning Research - Volume I: 顺心彩票 Learni ng: The essence
of the pedagogy. Greenwich, CT: Information Age Publishing. 2002;
50. Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analy-
ses using G*Power 3.1: Tests for cor relation and regression analyses.
Behavior Research Methods. 2009; 41 (4): 1149-1160.
51. Cohen J. Statistical power analysis for t he behavioral sciences (2nd Edi-
tio n). Hillsdale, NJ: Lawrence Erlbaum. 1988.
52. Tabachnick BG, Fidell LS. Using Multiva riate Statistics. Fifth edition,
Pearson /Allyn and Bacon, Boston, Massachusetts. 2007.
53. Cable G. Income, race, and preventable hospitalizations: a small area
analysis in New Jersey. Journal of Health Care for the Poor and
Underserved. 2002; 13 (1): 66-80.
54. Groth-Marnat G. Handbook of psychological assessment (5th ed.).
New York, NY: Wiley. 200 9.
55. Hawthorn DD. Interface design and engagement with older people.
Behaviour & Information Technolog y. 2007; 26: 333-341.
56. Eastin MS, LaRose R. Inter net self-efcacy and the psychology of the
digital d ivide. Journal of Computer-Mediated Communication. 2000;
6 (1):
57. Dur ndell A, Haag Z. Computer self-efcacy, computer anxiety, atti-
tudes towards the I nternet and reported experience with the Inter net,
by gender, in an East European sample. Computers in Human Behav-
ior. 2002; 18: 521-535.
58. Levine TR, Hullett CR. Et a squared, part ial eta squared, and misre-
porting of effect size in communication research. Human Communica-
tion Research. 2002; 28 (4): 612-625.
59. Ruiz ME. Familismo and lial piety among Latino and Asian elders:
Reevaluating family and social suppor t. Hispanic Health Care Inter-
national. 2007; 5 (2): 81-89.
... This finding implies that investing in digital-media support can have beneficial outcomes. Past work shows that formal computer training can enhance the computer and technical self-efficacy of older adults (Czaja et al. 2018;Laganá and García 2013;Seo et al. 2019). Exploring whether informal support sources may also benefit self-efficacy, and whether informal and formal support differs in the effect on self-efficacy, may offer guidance in determining what kinds of support may be most useful. ...
Full-text available
Given that older adults constitute a highly heterogeneous group that engages with digital media in varying ways, there is likely to be large variation in technology support needs, something heretofore unaddressed in the literature. Drawing on in-depth qualitative interviews with a multinational sample of older adults, the authors explore the support needs of older adults for using digital media, including their perceptions of whether the support they receive meets their needs. Participants obtained assistance from both informal (e.g., adult children) and formal (e.g., computer classes) sources. However, the support given can lack immediacy, leaving older adults dependent on others’ availability to provide it. Educational approaches that emphasize individualized support and wide availability might allow an improved experience for a population that is increasingly online with an interest in a wide range of activities.
... When con- sidering the effects of studies evidencing high or moder- ate methodological quality, the majority being RCT's, three of these studies evidenced no statistically signifi- cant effects (Mountain et al., 2014;Slegers et al., 2007Slegers et al., , 2008White et al., 2002). The remaining three studies ( Lagana and Garcia, 2013;Shapira et al., 2007 ...
Full-text available
This systematic review explored the effectiveness of technology-based interventions in promoting the mental health and wellbeing of people aged 65 and over. Data were collected as part of a wider review commissioned by the National Institute for Health and Care Excellence (NICE) in England on the effectiveness of different actions to promote the mental wellbeing and independence of older people. All studies identified through this review were subject to a detailed critical appraisal of quality, looking at internal and external validity. Twenty-one papers covering evaluations of technological interventions were identified. They examined the psychosocial effects of technologies for education, exposure to, and/or training to use, computers and the internet, telephone/internet communication and computer gaming. Few studies took the form of randomized controlled trials, with little comparability in outcome measures, resulting in an inconsistent evidence base with moderate strength and quality. However, three out of six studies with high or moderate quality ratings (all focused on computer/internet training) reported statistically significant positive effects on psychosocial outcomes, including increased life satisfaction and experienced social support, as well as reduced depression levels among intervention recipients. The review results highlight the need for more methodologically rigorous studies evaluating the effects of technology-based interventions on mental wellbeing. Well-performed technology-based interventions to promote various aspects of mental wellbeing, as identified in this review, can serve as best practice examples in this emerging field.
... When considering the effects of studies evidencing high or moderate methodological quality, the majority being RCT's, three of these studies evidenced no statistically significant effects (Mountain et al., 2014;Slegers et al., 2007Slegers et al., , 2008White et al., 2002). The remaining three studies (Lagana and Garcia, 2013;Shapira et al., 2007et al., 2008) did however report statistically significant positive effects among the intervention participants. ...
Full-text available
Given a changing demographic landscape, the promotion of older adults’ wellbeing and independence is a public health issue. In recent years, the potential of technology-based resources for the promotion of wellbeing in later life has been highlighted. This systematic review analysed the effectiveness of technology-based interventions for the promotion of mental wellbeing among adults aged 65 and over without extensive health or social care needs. The data originates from an evidence review project commissioned by the National Institute for Health and Care Excellence (NICE) in the UK on the effectiveness of different actions to promote the mental wellbeing of older people. Systematic searches were performed in 8 bibliographic databases. Publications from the period 2003 onwards were considered. From the original review data material, 25 intervention studies were selected for this review, covering technology use for educational purposes, computer/internet exposure or training, telephone/internet communication, or computer gaming. The number of studies employing an RCT design and looking at comparable outcomes was low, resulting in the strength of the evidence being moderate and somewhat inconsistent. When considering the six studies with higher quality ratings, four of them - all focused on computer/internet training - reported significant favorable effects on one or several outcomes among intervention recipients (e.g. increased life satisfaction, experienced social support). While the review results highlight a lack of methodologically rigorous studies evaluating the effects of technology-based interventions for optimal ageing, they also present promising examples of effective interventions that can serve as best practice examples in this emerging field.
As the world population ages and older adults comprise a growing proportion of current and potential Internet users, understanding the state of Internet use among older adults as well as the ways their use has evolved may clarify how best to support digital media use within this population. This article synthesizes the quantitative literature on Internet use among older adults, including trends in access, skills, and types of use, while exploring social inequalities in relation to each domain. We also review work on the relationship between health and Internet use, particularly relevant for older adults. We close with specific recommendations for future work, including a call for studies better representing the diversity of older adulthood and greater standardization of question design.
Currently, there is no reliable cure for mental disorder such as depression. However, there are a few strategies which can help in the treatment of their symptoms. These comprise both pharmacological and non-pharmacological approaches. The purpose of this article is to discuss the role of the Internet and computer-based programs as an appropriate intervention tool for older adults with depression. This is done by conducting a literature search in the databases Web of Science, Scopus, MEDLINE and Springer, and consequently by evaluating the findings of the relevant studies. Based on the findings, computer-based programs targeted at older people with depression may be beneficial in several ways: they are non-invasive treatments, they can be tailored-made to older people’s needs, they are cost-effective and can be made widely available, and they appear to be an effective intervention tool, especially as far as the short-term effects are concerned. Nevertheless, it is important to pay close attention to the methodological standards in future clinical studies, as well as to the efficacy of these computer-based programs aimed at older individuals with depression.
Full-text available
Regression methods were used to select and score 12 items from the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) to reproduce the Physical Component Summary and Mental Component Summary scales in the general US population (n = 2,333). The resulting 12-item short-form (SF-12) achieved multiple R squares of 0.911 and 0.918 in predictions of the SF-36 Physical Component Summary and SF-36 Mental Component Summary scores, respectively. Scoring algorithms from the general population used to score 12-item versions of the two components (Physical Component Summary and Mental Component Summary) achieved R squares of 0.905 with the SF-36 Physical Component Summary and 0.938 with the SF-36 Mental Component Summary when cross-validated in the Medical Outcomes Study. Test-retest (2-week) correlations of 0.89 and 0.76 were observed for the 12-item Physical Component Summary and the 12-item Mental Component Summary, respectively, in the general US population (n = 232). Twenty cross-sectional and longitudinal tests of empirical validity previously published for the 36-item short-form scales and summary measures were replicated for the 12-item Physical Component Summary and the 12-item Mental Component Summary, including comparisons between patient groups known to differ or to change in terms of the presence and seriousness of physical and mental conditions, acute symptoms, age and aging, self-reported 1-year changes in health, and recovery from depression. In 14 validity tests involving physical criteria, relative validity estimates for the 12-item Physical Component Summary ranged from 0.43 to 0.93 (median = 0.67) in comparison with the best 36-item short-form scale. Relative validity estimates for the 12-item Mental Component Summary in 6 tests involving mental criteria ranged from 0.60 to 1.07 (median = 0.97) in relation to the best 36-item short-form scale. Average scores for the 2 summary measures, and those for most scales in the 8-scale profile based on the 12-item short-form, closely mirrored those for the 36-item short-form, although standard errors were nearly always larger for the 12-item short-form.
Presents an integrative theoretical framework to explain and to predict psychological changes achieved by different modes of treatment. This theory states that psychological procedures, whatever their form, alter the level and strength of self-efficacy. It is hypothesized that expectations of personal efficacy determine whether coping behavior will be initiated, how much effort will be expended, and how long it will be sustained in the face of obstacles and aversive experiences. Persistence in activities that are subjectively threatening but in fact relatively safe produces, through experiences of mastery, further enhancement of self-efficacy and corresponding reductions in defensive behavior. In the proposed model, expectations of personal efficacy are derived from 4 principal sources of information: performance accomplishments, vicarious experience, verbal persuasion, and physiological states. Factors influencing the cognitive processing of efficacy information arise from enactive, vicarious, exhortative, and emotive sources. The differential power of diverse therapeutic procedures is analyzed in terms of the postulated cognitive mechanism of operation. Findings are reported from microanalyses of enactive, vicarious, and emotive modes of treatment that support the hypothesized relationship between perceived self-efficacy and behavioral changes. (21/2 p ref)
This study investigated the impact of shifts in age identification by older people from 'middle aged' to 'elderly' within the context of the labeling theory of deviance. From 323 interviews of older people (over 60), it was found that age identification was unrelated to attitudes toward old people and that, contrary to predictions from labeling theory, the label 'elderly' did not affect self esteem through any 'gate keeping' process. Attachment of negative stereotypes, or stigma, to growing old was, however, strongly related to self derogation. The implications of these findings are discussed concerning (1) the usefulness of a 'deviance' perspective on aging and (2) the applicability and sensitivity of labeling conceptualizations.
The abstract for this document is available on CSA Illumina.To view the Abstract, click the Abstract button above the document title.
This study compared the diagnostic performance of the Beck Depression Inventory (BDI) and the Geriatric Depression Scale (GDS) in correctly identifying depressed and nondepressed older people who are nursing 顺心彩票 residents without cognitive disorders. At the usual cut-off scores of 10 and 11, sensitivity was 96.30 per cent for the BDI and 88.89 per cent for the GDS, while their specificity rates were 46.15 and 56.41 per cent respectively. Using Receiver Operating Characteristics (ROC) curves, the results obtained using these scales were compared with the diagnosis of psychiatric disorder according to the DSM-III-R. Contrary to the hypothesis, no difference was found between the area under the ROC for the BDI (Az = .87; SD = .04) and for the GDS (Az = .85; SD = .05). The exclusion of the somatic items, or the somatic factor, did not change the diagnostic performance of the BDI. Indices of temporal stability, concurrent validity and agreement with clinical diagnoses also confirmed the reliability and validity of these two scales for older people living in nursing 顺心彩票s.