Internet and Everyday Life: The Perceived Implications of Internet Use on Memory and Ability to Concentrate

Abstract

The growing role of Internet in all aspects of everyday life has led to speculations over the impacts beyond the traditional questions of access or sociability. This in mind, the main focus in this article was to examine how Finns, for majority of whom Internet use has become commonplace activity, perceive the impacts of Internet use since first adopting the technology. In this study, we examine how Internet user history and perceived computer skills, along with different sociodemographic factors, appear to reflect on the perceived impacts of Internet adoption in terms of memory and ability to concentrate. According to the results, almost one in five of the respondents reported changes concerning their memory or ability to concentrate, with skilled computer users and nonworkers, in particular, perceiving the change. Factors such as age-related differences and exposure to potential information overload at work were identified to explain the perceived change. Our data were collected in a survey-gathering information on the everyday life and well-being of Finns. The sample consisted of 2000 Finnish speakers aged 15 to 64 years. The response rate was 46 percent (N=908).

Introduction

The role of Internet in everyday life has been examined rather extensively in the past. Much of this research has focused on topics such as social ties,¹ access to information and services (digital divide),^2–5 and user purposes.⁶ However, lately, the presence of Internet, along with other new information technologies, appears more and more influential in the daily life beyond the questions of access or sociability. For instance, early findings link the increasing use of information systems at work place, along with the general information overload^7–9 with generally tiring brain.^10–12 Such activity has then been argued to contribute toward decreasing working memory and ability to concentrate already at a relatively early age.^10,13

Traditionally, studies regarding such cognitive abilities as memory and ability to concentrate tend to be associated with the field of clinical and/or cognitive psychology, as a large portion of research measuring these particular abilities takes place in laboratory settings. However, despite the increased presence of Internet and other information and communication technologies (ICTs) in most societies, the number of studies examining the relationship between Internet adoption, memory, and ability to concentrate remains limited. Much of the existing studies have focused on Internet's and ICT roles in (online) learning¹⁴ as well as new technology's role in online-based cognitive therapy^15–16 and as a support tool for those with existing brain dysfunctions.^17–19 Therefore, it is pertinent that such cognitive abilities and their relationships with Internet adoption are assessed from different perspectives.

The study by Sparrow and colleagues,²⁰ for instance, examined the effects of Google use (or Internet search engines) on memory and users' ability to access internal and external information storages. They found evidence of resulting changes in memory structures as “…people forget items they think will be available externally, and remember items they think will not be available.”²⁰ This means that people use the Internet as a way to store information in a sort of external hard drive. The authors speculated that this type of use leads to increasingly more interconnected and symbiotic relationship with the computers.²⁰ In another study, Small and colleagues²¹ examined whether active Internet use would serve to stimulate different levels of brain activity, and as a result perhaps potentially serve to delay cognitive decline often associated with aging. Their study included 12 active Internet users and 12 less-active users, all aged between 55 and 76. According to the findings, Internet search engine use appeared to increase levels of brain activity among the more-experienced and able Internet users compared to those less affluent in their skills. The authors also noted that previous studies have indicated that repetitive, routine tasks have the tendency to decrease brain activity, yet using the Internet, even after repeated use, offered continuous mental stimulation. The role of Internet use therefore offered various aspects of brain stimulation due to the fact that users had to make several decisions and choices, along with managing a variety of different information sources.²¹ In a further study, Nicholas and colleagues²² examined Internet/Web use behavior of different generations. As part of their study experiment, participants' working memory was measured. According to their findings Google generation (participants born after 1993) demonstrated much weaker working memory than the older participants.²²

All of above three studies relied on measurements conducted mostly in a laboratory setting (computer-based exercises and memory tasks, and MRI scans). In our study, however, we relied on study participants' subjective self-evaluation (survey questionnaire) of the perceived implications of Internet adoption regarding respondents' memory and ability to concentrate. Therefore, the intention was not to provide data that were directly comparable with psychological and performance measurement data, but rather a sociological perspective and a basis for future studies that could use variety of other methods for analyzing the impacts of Internet concerning similar types of cognitive abilities. Furthermore, information was provided concerning whether different sociodemographic groups differed regarding the perceived impacts. In this study, we used population-level representative data of working aged Finns, as the common feature with some of the earlier studies is that they tend to employ only a small study sample.

The following research questions were proposed:

RQ1: What kind of self-perceived implications Internet user history and perceived computer skills have on Finns in terms of memory and ability to concentrate?

RQ2: What kind of sociodemographic differences exist among the perceived implications in terms of memory and ability to concentrate?

Data and Methods

Study participants

Our data were collected in the summer 2011 as a part of a larger survey on the everyday life and well-being of Finns. The questionnaires were sent to a nationally representative random sample selected from the Central Register of Population database (through a randomized procedure by the database provider). The sample consisted of 2000 Finnish speakers aged 15 to 64 years, out of which one was deceased, and 12 were noncontacts. The survey gave a response rate of 46 percent (N=908). The sample represented relatively well Finns aged between 15 and 64 in terms of gender distribution; however, the oldest age group of 61 to 65 years was over-represented by 3 percent points. On the other hand, men aged 15 to 30 were under-represented by average 2 percent points, whereas women of the same age were slightly over-represented in the sample.²³

Outcome variables

The original question on perceived impacts of Internet adoption included seven items. These were social relationships, general knowledge, participating in cultural of sporting events, memory, ability to concentrate, purchase consumption goods, and entertainment. Under each of the seven items, we asked whether the participants perceived the changes as following: increased/improved, decreased/weakened, no change, and unable to say. Out of the seven items, we chose to focus on two items, memory and ability to concentrate, as our outcome variables.

Independent variables

As independent variables, we included the following: Internet use years (that is, how many years has the respondent been using the Internet), perceived computer skills, gender, age, education, and employment status. Gender, age, education, and employment status represent traditional sociodemographic background variables in the past Internet-related research.^3,24–28 Furthermore, Internet user years and computer skills were included, as they can be seen influential in terms of the perceived impacts of Internet since its adoption. In the questionnaire, Internet use years were inquired using an open-ended question. User years were then recoded into two categories, Less-experienced and Experienced, based on the Median, which was 10.0 (Standard deviation 4.6) user years. The self-rated computer skills were based on the scaling termed to match the traditional grading system used in the Finnish school system, thus familiar to all respondents. In the scaling system, 4 is being termed as the lowest, and 10 the highest score. In Finland, grades 8, 9, and 10 are referred as good grades; hence, they were coded as good skills, whereas grades 4 to 7 are referred from poor to average; hence, they were coded as poor/average skills. Due to the limited number of cases for both user years and computer skills (see also Table 1), we could only recode the responses into two categories to enable multivariate analysis. We also asked the respondents' year of birth and then recoded age into four categories, 15–30, 31–45, 46–60, and 60+, to provide an even age categorization.²⁴ Education was categorized as Primary education, Secondary, and BA and higher. Employment status was categorized as Work and Nonwork (the original item included the following options: Employed, Unemployed/looking for work, Retired, Student, Parental leave, and Stays home mother/father).

Table 1.

Perceived Changes in Memory and Ability to Concentrate According to Internet Use Years, Perceived Computer Skills, and Sociodemographic Factors Among 15–64-Year-Old Finns in 2011

	Memory				Ability to concentrate
Independent variables	Not able to say % (n)	No change % (n)	Decreased % (n)	Increased % (n)	Not able to say % (n)	No change % (n)	Decreased % (n)	Increased % (n)
Internet use years
Experienced	15.5 (78)	59.9 (302)	15.9 (80)	8.7 (44)	14.5 (73)	67.7 (342)	12.1 (61)	5.7 (29)
Less experienced	15.9 (52)	64.9 (213)	8.8 (29)	10.4 (34)	16.5 (54)	66.1 (216)	10.1 (33)	7.3 (24)
Computer skills
Good skills	13.9 (69)	58.5 (290)	15.3 (76)	12.3 (61)	14.5 (72)	64.2 (319)	13.5 (67)	7.8 (39)
Poor/average skills	17.7 (66)	67.6 (252)	9.7 (36)	5.1 (19)	17.0 (63)	69.8 (259)	8.1 (30)	5.1 (19)
Gender
Male	16.9 (65)	61.6 (237)	11.9 (46)	9.6 (37)	18.0 (69)	64.3 (247)	10.2 (39)	7.6 (29)
Female	15.3 (75)	62.3 (306)	13.4 (66)	9.0 (44)	14.5 (71)	67.8 (333)	11.8 (58)	5.9 (29)
Age
15–30	21.1 (51)	53.7 (130)	10.3 (25)	14.9 (36)	20.7 (50)	55.4 (134)	14.5 (35)	9.5 (23)
31–45	9.7 (24)	68.1 (169)	14.9 (37)	7.3 (18)	10.5 (26)	72.2 (179)	12.5 (31)	4.8 (12)
46–60	17.4 (49)	62.1 (175)	12.8 (36)	7.8 (22)	15.3 (43)	69.0 (194)	9.3 (26)	6.4 (18)
60+	16.3 (16)	65.3 (64)	14.3 (14)	4.1 (4)	19.4 (19)	70.4 (69)	5.1 (5)	5.1 (5)
Education
BA/Higher	16.7 (89)	62.8 (334)	9.6 (51)	10.9 (58)	17.3 (92)	65.7 (349)	8.7 (46)	8.3 (44)
Secondary	12.7 (29)	62.4 (143)	17.5 (40)	7.4 (17)	12.2 (28)	69.4 (159)	14.4 (33)	3.9 (9)
Primary	19.1 (21)	59.1 (65)	16.4 (18)	5.5 (6)	17.3 (19)	64.5 (71)	14.5 (16)	3.6 (4)
Employment status
Work	14.0 (73)	66.7 (348)	13.2 (69)	6.1 (32)	13.8 (72)	71.5 (373)	10.3 (54)	4.4 (23)
Nonwork	18.6 (65)	55.1 (193)	12.3 (43)	14.0 (49)	18.9 (66)	59.0 (206)	12.3 (43)	9.7 (34)
All	16.0 (140)	62.0 (543)	12.8 (112)	9.2 (81)	16.0 (140)	66.3 (580)	11.1 (97)	6.6 (58)

Statistical analysis

The statistical analysis included basic descriptive analysis and cross tabulation, along with multinomial logistic regression analysis.²⁹ Multinomial regression analysis allowed us to describe more than just two outcomes (e.g., change–no change),³⁰ due to our interest in examining described change in both directions, either as increase or decrease. Furthermore, the questionnaire was designed to give the respondents one response option per question. For the purposes of the analysis, we merged outcome variable categories “no change” and “unable to say.” Although unable to say is a somewhat unspecific response (unlike no change), excluding those unable to say would mean significant reduction of the sample size, causing problems for the analysis. In addition, our particular interest was to examine those who perceived change compared to those who perceived no change (or were in fact unable to determine so). That is, we were interested in recording not only whether the respondent perceived change, but also (in case they did report change) what kind of change. A two-stage procedure was carried out for both outcome variables. First, we studied univariate associations between the outcome variable and Internet user years, perceived computer skills, and each one of the sociodemographic variables. Second, multivariate models were fitted to find out which of the independent variables independently predicted the outcomes. For each independent variable, the category with the highest presumed probability of no change/unable to say was taken as the reference group. The model parameters were presented as odds ratios (ORs) with statistical significances and 95% confidence intervals (CIs).

Results

To provide an overview of the study participants and their online activity, we also examined which of the different demographic respondent groups (age, gender, education, and employment status) were the most active (using the Internet every day) leisure-time Internet users. Women (54.9 percent), the youngest age group of 15–30 years (36.6 percent), those with BA or higher degree (55.6 percent), and workers (59.9 percent) reported the highest proportions of this category.

Descriptive analysis

In Table 1, we then present the descriptive statistics, including both percentages and frequencies, concerning perceived impacts of Internet use on both memory and ability to concentrate.

Considerable differences existed in the way in which Finns perceived the impacts Internet use had in their life. Although majority of respondents stated not to perceive any changes since first adopting Internet use, the number of respondents who either felt changes or were unable to say was still relatively substantial.

Multinomial logistic regression analysis

Table 2 shows the results of the univariate multinomial logistic regression analyses for the association Internet user history and skills, as well as the sociodemographic variables, had with the outcomes. In regard of memory, OR for those with more experience in terms of Internet use years to feel decreased memory was 1.9 compared to those with less experience. Among those who perceived as having good computer skills, some respondents perceived as having decreased memory (OR 1.9), whereas others felt increase (OR 2.8) instead, when comparing with those rating as having poor/average skills. In addition, the youngest age group (18–30) felt that Internet use improved their memory significantly more often (OR 4.0) compared to the oldest age group. The difference between workers and nonworkers was also statistically significant; workers (OR 0.4) were less likely to feel increase of memory resulting from Internet adoption compared to those not in the working life.

Table 2.

Univariate Associations Between Perceived Change of Memory and Ability to Concentrate and the Independent Variables ^a

	Memory OR, p-value, 95% CI				Ability to concentrate OR, p-value, 95% CI
Main effects	Decreased		Increased		Decreased		Increased
Internet use years
Experienced	1.9^*	1.22–3.03	0.9	0.56–1.45	1.2	0.77–1.38	0.8	0.45–1.38
Less experienced	1		1		1		1
Computer skills
Good skills	1.9^**	1.22–2.86	2.8^***	1.66–4.86	1.8^**	1.17–2.90	1.7	0.96–2.98
Average/poor skills	1		1		1		1
Gender
Male	0.9	0.59–1.32	1.1	0.67–1.69	0.9	0.56–1.32	1.3	0.75–2.18
Female	1		1		1		1
Age
15–30	0.8	0.39–1.60	4.0^*	1.37–11.55	3.4^*	1.27–8.84	2.2	0.81–5.98
31–45	1.1	0.56–2.14	1.9	0.61–5.68	2.7^*	1.00–7.07	1.0	0.35–3.01
46–60	0.9	0.47–1.80	2.0	0.66–5.88	1.9	0.72–5.19	1.4	0.48–3.71
60+	1		1		1		1
Education
BA/Higher	0.6	0.32–1.03	2.0	0.82–4.70	0.6	0.32–1.08	2.3	0.79–6.41
Secondary	1.1	0.60–2.05	1.4	0.54–3.72	1.0	0.52–1.90	1.1	0.33–3.61
Primary	1		1		1		1
Employment status
Work	1.0	0.65–1.48	0.4^***	0.25–0.64	0.8	0.50–1.18	0.4^**	0.24–0.72
Nonwork	1		1		1		1

The reference category for the outcome variable: no change/unable to say.

Significance levels: ^*p<0.05; ^**p<0.01; ^***p<0.001.

OR, odds ratio; CI, confidence interval.

Contrary to memory, the impact of Internet adoption on ability to concentrate was perceived little differently among those perceiving themselves as more skilled computer users, as well as among different age groups. The OR for decreasing ability to concentrate against the less-skilled computer users was 1.8 among the more-skilled users. Further, the OR for decreasing ability to concentrate against the oldest age group was 3.4 in the age category 15–30 and 2.7 in the age category 31–45. In terms of the employment status, it appears that workers were less likely to perceive the effects on ability to concentrate as increased (OR 0.4) compared to those not in the working life.

In Table 3, the results of the multivariate multinomial logistic regression analyses are shown. In the models, we included only those variables that were statistically significant in the univariate analyses (Table 2) (in terms of memory, these were Internet user years, computer skills, age, and employment status. In terms of ability to concentrate, these were computer skills, age, and employment status). We then conducted the analysis by controlling for each of the significant variables in the analysis of both outcome variables. In terms of memory, only computer skills and employment status remained statistically significant. Among those with good computer skills, some felt increase of memory (OR 2.0), while some felt decrease of memory (OR 2.7) compared to those who rated themselves less skilled. Again, workers had smaller odds for increased memory against those not in the working life.

Table 3.

Multivariate Associations Between Perceived Change of Memory and Ability to Concentrate and the Independent Variables ^a

	Memory OR, p-value, 95% CI				Ability to concentrate OR, p-value, 95% CI
Main effects	Decreased		Increased		Decreased		Increased
Internet use years					na
Experienced	1.6	0.97–2.59	0.8	0.49–1.42
Less experienced	1		1
Computer skills
Good skills	2.0^**	1.21–3.20	2.7^**	1.43–5.13	1.6	0.96–2.56	1.6	0.84–2.97
Average/poor skills	1		1		1		1
Age
15–30	0.6	0.26–1.19	2.3	0.76–7.10	2.7	1.00–7.36	1.8	0.62–5.08
31–45	0.9	0.40–1.81	2.0	0.63–6.57	2.7	0.96–7.40	1.4	0.46–4.39
46–60	0.9	0.42–1.89	2.4	0.76–7.63	2.2	0.78–6.05	2.1	0.72–6.17
60+	1		1		1		1
Work for pay
Work	0.8	0.50–1.31	0.4^**	0.25–0.75	0.7	0.45–1.17	0.4^**	0.20–0.71
Nonwork	1		1		1		1

The reference category for the outcome variable: no change/unable to say.

Significance levels: ^**p<0.01; na=not apply (not significant in univariate analysis).

In terms of ability to concentrate, only employment status held a statistically significant value after controlling for the other variables; workers had smaller odds (OR 0.4) of perceiving increased ability to concentrate compared to those not in working life.

Discussion

According to the unadjusted results in terms of memory, computer skills, Internet user years, age and employment status played statistically significant roles among the respondents perceiving changes. As to computer skills, there was a clear division between the more skilled users, as on one hand there were those who perceived decrease of memory, whereas there were others who rather perceived increase of memory instead.

One explanation to the difference in the factors alleviated with the two dimensions may be found by examining some of the other independent variables in more detail. Looking at age (significant in the unadjusted model), for instance, we note that the youngest age group was more likely to perceive increased memory. As the young adults are generally the most active users³¹ (this was the case in our data as well), they are likely to perceive themselves as more skilled users, hence explaining the significant number of skilled users perceiving increasing memory. On the other hand, the more experienced users were more likely to perceive decreased memory instead, when comparing with the less-experienced users. This would perhaps also explain that older (skilled) users with longer Internet user history were therefore more likely to perceive decreased memory instead. Hence, the division among the skilled users with opposite perceptions concerning Internet adoption could perhaps be explained through age-related differences.

Regarding the question of ability to concentrate, computer skills, age, and employment status were significant in the unadjusted model. However, many of the results were different from those of memory. For instance, despite perceiving improved memory, the young were then more likely to feel that their ability to concentrate had decreased instead. The results in general concerning the decreasing ability to concentrate could be seen to reflect the increasing number of new technologies, services, and information present and available in the daily life in general. However, we must be cautious with such speculations, as the effects were quite weak, and did not remain significant in regard of computer skills and age after controlling for the other variables.

The only variable that remained statistically significant even after controlling for the other variables in regard of both memory and ability to concentrate was employment status. In both cases, workers were less likely to perceive increased memory and ability to concentrate compared to those not working. This might perhaps be explained by the fact that since employed, they were more likely to use the different information technologies and systems often required in many of the present day jobs (in our data, workers were also more likely to use the Internet during their leisure time compared to the nonworkers), and therefore be more likely to be subjected to potential information overload.^7,32 Instead, nonworkers were less likely to be subjected to these demands.

The intention in this study was to provide an overview of how the impacts of Internet adoption are perceived by users, particularly in terms of their memory and ability to concentrate. Based on the results, it appears that people tend to differ significantly in terms of the way in which they perceive the impacts of Internet adoption. Although most of the respondents perceiving no change in regard of their memory or ability to concentrate since first adopting the Internet, there were still almost 40 percent of users who either felt changes, one way or another, or were unsure about the possible implications. Further, there were also respondents with a similar sociodemographic background that could report very different perceptions concerning the impacts of Internet. This also serves to indicate that for the same reasons, the perceived impacts are vastly multileveled, thus signaling that there is indeed a need for a more careful examination of the potential impacts of adopting Internet, but also acquiring deeper level knowledge of the respondents' background variables (e.g., place of residence, family situation, and leisure activities).

The methodology in the earlier related studies^20–22 were largely based on measurements conducted in a laboratory setting, as they were designed to measure particular abilities and activities. However, our study was conducted largely from a sociological perspective; hence, the intention was to examine the participants' self-perceptions. Yet, using self-reported data offers various limitations for analysis, particularly concerning questions on cognitive abilities. For deeper understanding of such abilities, research in a more controlled environment (laboratory settings) is needed, preferably with different types of reference groups to learn to what extent the users' own perceptions appear valid.

It would also be interesting to examine, for instance, how long-term use reflects on the likes of memory and ability to concentrate by examining potential differences between periods of Internet use with period of nonuse, along with adopting qualitative methods such as interviews to aid data collection. Internet users also be compared against those who have never used the Internet (preferably of the same age), and also more detailed data could be collected concerning user purposes, thus to study potential implications among different user types.

In this study we relied on data from one county only. It would be valuable to gain access for similar data in other countries as well, and thus enable comparison of countries with high penetration rates and average user hours with those less inclined for this type of activities.

Footnotes

Acknowledgments

This article was supported in part by the grants from the Finnish Cultural Foundation, The Varsinais-Suomi Regional Fund, and Turku University Foundation.

Author Disclosure Statement

No competing financial interests exist

References

Kraut

, Patterson

, Lundmark

et al.

Internet paradox: a social technology that reduces social involvement and psychological well-being?

American Psychologist, 1998; 53:1017–1031.

Hargittai

. Second-level digital divide: differences in people's online skillsFirst Monday 2002; 7www.firstmonday.org/issues/issue7_4/hargittai/. 2012 Jan 18.

Hargittai

, Walejko

. The participation divide: content creation and sharing in the digital age. Information, Communication and Society, 2008; 11:239–256.

van Dijk

JAGM

. Digital divide research, achievements and shortcomings. Poetics, 2006; 34:221–235.

Räsänen

. Information society for all? Structural characteristics of Internet use in 15 European Countries. European Societies, 2006; 8:59–81.

Wellman

, Haythornthwaite

. 2002. The internet in everyday life. Oxford: Blackwell.

Bawden

, Robinson

. The dark side of information: overload, anxiety and other pathologies. Journal of Information Science, 2009; 35:180–191.

Chen

C-Y

, Pedersen

, Murphy

. The influence of perceived information overload on student participation and knowledge construction in computer-mediated communication. Instructional Science, 2011[Epub ahead of print]10.1007/s11251-011-9179-0.

Beaudoin

. Explaining the relationship between Internet use and interpersonal trust: taking into account motivation and information overload. Journal of Computer-Mediated Communication, 2008; 13:550–568.

10.

Müller

. 2010. Aivot pitävät kutituksesta vaan eivät kurituksesta. Brain@Work - Yhteisvastuu aivoista–seminar presentation September 7th 2010 2011 Oct 12.

11.

Klingberg

. Limitations in information processing in the human brain: neuroimaging of dual-task performance and working memory tasks. Progress in Brain Research, 2000; 126:95–102.

12.

Klingberg

. 2009. The overflowing brain: information overload and the limits of working memory. New York: Oxford University Press.

13.

Kauppinen

, Hanhela

, Kandolin

et al. 2010. Työ ja terveys Suomessa 2009. Helsinki, Finland: Työterveyslaitos.

14.

Saadé

, Bouchaib

. The impact of cognitive absorption on perceived usefulness and perceived ease of use in on-line learning: an extension of the technology acceptance model. Information and Management, 2005; 42:317–327.

15.

Berger

, Hohl

, Caspar

. Internet-based treatment for social phobia: a randomized controlled trial. Journal of Clinical Psychology, 2009; 65:1021–1035.

16.

Anderson

, Jacobs

, Rothbaum

. Computer-supported cognitive behavioral treatment of anxiety disorders. Journal of Clinical Psychology, 2004; 60:253–267.

17.

Egan

. The role of the Internet in the lives of people with traumatic brain injury PhD thesis. School of social work and human services, The University of Queensland, 2010.

18.

Karim

, Hinterberger

, Richter

et al. Neural internet: Web surfing with brain potentials for the completely paralyzed. Neurorehabilitation and Neural Repair, 2006; 20:508–515.

19.

Bergquist

, Gehl

, Mandrekar

et al. The effect of internet-based cognitive rehabilitation in persons with memory impairments after severe traumatic brain injury. Journal of Computer-Mediated Communication, 2009; 13:550–568.

20.

Sparrow

, Liu

, Wegner

. Google effects on memory: cognitive consequences of having information at our fingertips. Science, 2011; 333:776–778.

21.

Small

, Moody

, Siddarth

et al. Your brain on Google: patterns of cerebral activation during internet searching. American Journal of Geriatric Psychiatry, 2009; 17:116–126.

22.

Nicholas

, Rowlands

, Clark

et al. Google Generation II: web behaviour experiments with the BBC. Aslib Proceedings: New Information Perspectives, 2011; 63:28–45.

23.

Statistics Finland. 2010. Statistical yearbook of Finland 2010. Helsinki, Finland: Tilastokeskus.

24.

Räsänen

. The aftermath of the ICT revolution? Media and communication technology preferences in Finland in 1999 and 2004. New Media and Society, 2008; 10:225–246.

25.

Tsatsou

. Why Internet use? A quantitative examination of the role of everyday life and Internet policy and regulation. Technology in Society, 2011; 33:73–83.

26.

Gershuny

. Web use and net nerds: a neofunctionalist analysis of the impact of information technology at home. Social Forces, 2003; 82:141–168.

27.

Robinson

, DiMaggio

, Hargittai

. New social survey perspectives on the digital divide. IT and Society, 2003; 1:1–22.

28.

Katz

, Rice

, Aspden

. The Internet, 1995–2000. American Behavioral Scientist, 2001; 45:405–419.

29.

Hosmer

, Lemeshow

. 2000. Applied logistic regression. New York: Wiley.

30.

Greene

. 1993. Econometric analysis, fifth. New Jersey: Prentice Hall.

31.

Statistics Finland. 2011. Use of information and communications technology [e-publication] Helsinki: Statistics Finland www.stat.fi/til/sutivi/2011/sutivi_2011_2011-11-02_tie_001_en.html. 2012 Aug 14.

32.

Brennan

. The scientific management of information overload. Journal of Business and Management, 2011; 17:121–134.