Life satisfaction and problematic Internet use: Evidence for gender specific effects

Author’s Accepted Manuscript Life satisfaction and problematic Internet use: Evidence for gender specific effects Bernd Lachmann, Rayna Sariyska, Christopher Kannen, Andrew Cooper, Christian Montag www.elsevier.com/locate/psychres

PII: DOI: Reference:

S0165-1781(16)30258-X http://dx.doi.org/10.1016/j.psychres.2016.02.017 PSY9454

To appear in: Psychiatry Research Received date: 5 May 2015 Revised date: 23 December 2015 Accepted date: 12 February 2016 Cite this article as: Bernd Lachmann, Rayna Sariyska, Christopher Kannen, Andrew Cooper and Christian Montag, Life satisfaction and problematic Internet use: Evidence for gender specific effects, Psychiatry Research, http://dx.doi.org/10.1016/j.psychres.2016.02.017 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Life satisfaction and problematic Internet use: Evidence for gender specific effects Bernd Lachmanna*, Rayna Sariyskaa, Christopher Kannenb , Andrew Cooperc & Christian Montaga a

Department of Psychology, Ulm University, Ulm, Germany

b

Department of Informatics, University of Bonn, Bonn, Germany

c

Department of Psychology, Goldsmiths, University of London, London, UK

*Corresponding authors: Bernd Lachmann, Department of Psychology University of Ulm Helmholtzstrasse 8/1 D-89081 Ulm Germany Telephone: 0049 / 731 / 50 26550 Fax: 0049 / 731 / 50 32759 E-Mail:[email protected]

1.

Introduction

The Internet has revolutionized the way we communicate, conduct business and entertain ourselves. Goods can be accessed more easily, communication is faster and cheaper, and diverse forms of entertainment exist online, including TV series and computer gaming. Despite these positive aspects, a growing number of users spend an excessive amount of time on the Internet. A quickly growing body of literature is currently examining the degree to which problematic Internet use (PIU) actually reflects addictive behavior (e.g. see the review by Montag et al., 2015a). This literature has also considered if PIU can be better described by existing nosologies such as ADHD and depression (Sariyska et al., 2015). Comorbidities between PIU and several other conditions have also been examined in the literature (Han et al., 2008; Ko et al., 2012; Bozkurt et al., 2013; Carli et al., 2013). 1

Several attempts have been made to classify PIU. For example, according to Tao et al. (2010), preoccupation with the Internet and withdrawal symptoms should always be observed, in conjunction with at least one additional symptom, such as development of tolerance. Young (1998) hypothesized that five or more items should be answered positively on the Internet Addiction Test in order to consider an individual as a problematic Internet user. The items include content such as neglect of routine duties, social isolation, and being secretive about online activities. Although studies relating PIU to psychiatric disorders such as depression and ADHD are of considerable relevance, the investigation of PIU in relation to life satisfaction (well-being) and its facets is of substantial relevance too, because such a study could demonstrate that PIU might profoundly impact our well-being – even in subclinical samples (Whang et al., 2003). Therefore an investigation of the link between PIU and life satisfaction in subclinical samples warrants attention. A recent study with a rather small sample size from Turkey showed that overall life satisfaction is diminished in people with PIU (Bozoglan et al., 2013). Ko et al. (2007) also investigated life satisfaction in the context of Internet usage in Taiwan, only observing a statistical trend showing an association between lower life satisfaction and higher Internet use in N = 517 students. Pawlikowski et al. (2014) found a negative association between life satisfaction and PIU in the domain of Internet gaming, but not for Internet pornography. Adding to this, Cao et al. (2011) observed that lower scores in general life satisfaction and distinct domains such as family, friends, school, living environment and the self were associated with higher PIU scores. This study is particularly noteworthy because of its large sample size (n > 15.000). In the present study we sought to contribute to this literature by investigating whether PIU negatively affects general life satisfaction and specific facets of life satisfaction in 2

a large sample of almost 5000 participants from a different cultural background – namely from a Western society. Given robust evidence that males are more prone to show problematic Internet use than females (Morahan-Martin, 1998; Shaw et al., 2008; Lam et al., 2009), we additionally sought to investigate the potential association between life satisfaction and PIU considering gender as a possible moderator of this relationship.

2.

Materials and Methods

2.1 Participants Data concerning PIU and satisfaction, as well as demographic variables and education, was gathered from N = 4852 persons (n = 2343 females). The mean age of all participants was 28.96 (SD = 17.00), ranging from six (four participants at the age of six, eight participants at the age of seven) to 93 years (42 % of the participants were 17 years of age or below). 31.0% of all participants did not have a high school diploma, 5.5% had a secondary school qualification, 20.2% had a secondary school leaving certificate, 20.7% had a Baccalaureate-Diploma, and 22.6% had a university degree. The vocational distribution in the sample was as follows: 53.1% of the participants were students (45.9% school, 7.2% university), 3.1% were apprentices, 32.1% were employed, 5.0% were self-employed, 3.2% were retired, and 3,5% were unemployed. The present study was approved by the local ethics committee. Every participant gave their (electronic) consent prior to their participation in the study.

2.2 Materials The data within the study was collected from May 6th until September 28th 2014 during a public exhibition. An exhibition on a large boat with a theme of the “digital 3

society” travelled through Germany and Austria. The boat stopped in several cities for a few days before travelling to the next city. During each stop the boat opened its doors for visitors to experience the scientific installations, with no entrance fee required. Our installation offered participants the possibility to fill in questionnaires (by using a smartphone or a tablet, which were available at the stand on the exhibition). Amongst others, the questionnaires dealt with use of the Internet and life satisfaction. As an incentive, participants received individual feedback based on the responses they provided, including the responses they provided concerning life-satisfaction and PIU. For some participants (n = 3084), we also collected responses relating to individual usage of watch-based Smartphones and associated overuse of digital technologies. The results of this study – namely on `Zeitgeber´ usage and PIU – can be found in Montag et al. (2015). To assess life satisfaction we used six items retrieved from the German SocioEconomic Panel (SOEP; Siedler et al., 2009). Within the panel, a section concerning current life situations deals with several areas contributing to overall life satisfaction. For the purposes of this study, we asked for the degree of satisfaction in the following areas: health, job, income, dwelling, leisure time, and overall satisfaction with life. Following a recommendation from the SOEP, the question for overall satisfaction with life was asked at the end of the questionnaire to avoid possible interference with the questions referring to particular areas of life satisfaction. It is important to note that the questions on different facets of life satisfaction do not simply add up to the overall life satisfaction score (as mentioned above, overall life satisfaction was assessed as an independent item). The various questions on the facets of life satisfaction and overall life satisfaction are considered to be distinct, but also overlap to some extent (e.g. a person more satisfied with his income, might have a nicer apartment, and as a 4

consequence would also have a higher score on satisfaction with dwelling). All items were answered using a Likert scale ranging from 0 (“completely dissatisfied”) to 10 (“completely satisfied”). To gather information on PIU we administered a short form of the Internet addiction test (Young, 1998), the short Internet addiction test (s-IAT) from Pawlikowski et al. (2013). The short form consists of 12 items, in contrast to the original test from Young (1998) that has 20 items. The psychometric quality of the short version was considered good (Pawlikowski et al., 2013). Cronbach’s alpha was excellent in our sample (alpha = 0.91). To produce a more detailed picture on Internet use we asked the participants two additional questions concerning their time spent on the Internet: “How many hours per week do you spend on the Internet for private/business purposes?’’. Prior studies reported a strong association between hours spent on the Internet for private purposes, but not for business purposes and PIU (Sariyska et al., 2014; Montag et al., 2010).

2.3 Statistical analyses For the statistical analyses we used SPSS version 22.0 for Windows (IBM SPSS Statistics). Gender differences between the s-IAT values with respect to the participants in different groups (normal use/problematic use/internet addicts), depending on cut-off values suggested by Pawlikowski et al. (2013), were analyzed with a Chi-square test. To investigate differences in gender with respect to hours spent on the Internet, a Mann-Whitney U test was used since the variable hours spent on the Internet (business/private) was not normally distributed. In contrast, Internet addiction scores and life satisfaction variables were normally distributed and 5

could therefore be analyzed with parametric tests. The relation between facets of life satisfaction and gender was analyzed using a MANOVA. Correlations between PIU and life satisfaction were computed with Pearson’s correlation. In addition, associations between PIU and life satisfaction were compared across gender using the Fisher r-to-z transformation (Lowry, 2003). In order to get a more detailed view on the role of gender we further investigated (after applying a Bonferroni correction) the two remaining significant associations between facets of life satisfaction and PIU using moderated regression models with gender as moderator (Hayes, 2012). Finally a regression analysis was carried out to determine the most promising predictors for PIU with regard to diagnostic and interventional procedures.

3.

Results

3.1.

Data cleaning and information on the data structure

An inspection of the data revealed a total of 148 participants either choosing 0 or 99 as their age, which represents the lowest and highest available values on the age scale (data was collected via electronic tablets). Because age could be inserted using a slider, we assumed that those participants did not fill in their correct age by either not moving the slider at all or simply moving it all the way across to the other side of the scale. Since this could not be verified post hoc we decided to exclude participants with an age of 0 or 99 from any further analysis (3.05%). Additionally, we excluded one participant with a reported age of two, since this individual was unlikely to be able to read and fill in the questionnaires by himself. Participants with an age of six (four participants) and seven (eight participants) were included in the sample since writing and reading abilities are typically established by this age and therefore these participants could have answered the questions by themselves. No further 6

participants were excluded based on their responses. The data was normally distributed for all variables, except for hours spent on the Internet (private/business). We did not find any outliers on any variables. Means, standard deviations, minimum and maximum values, and skew were calculated for the variables PIU, hours spent on the Internet (private/business) per week, overall life satisfaction, as well as facets of life satisfaction, and demographic variables, and are shown in Table 1.

(Insert Table 1 about here)

3.2.

Degree of PIU in the sample in relation to gender

Three groups of participants were created according to the suggested cutoff values for Total use of Internet from Pawlikowski et al. (2013). This showed a higher percentage of females in the normal Internet use group (IAT < 31), and a higher percentage of males in the moderate (IAT 31 – 37) and problematic Internet use groups (IAT > 37) (Χ2 = 13.22, df = 2, p = 0.01). For details please refer to Table 2.

(Insert Table 2 about here)

Moreover, females showed lower mean Internet usage per week than males for private, but not for business, purposes. The means for internet use separated by gender were as follows: Private Internet use: Mfemale = 16.55, SDfemale; = 0.36, Mmale = 20.42, SDmale = 0.43; U4852) = 2524668, p < 0.01; business Internet use: Mfemale = 7.55, SDfemale; = 0.30, Mmale = 9.55, SDmale = 0.40; U(4852) = 2907678, p = 0.51).

3.3.

Gender and life satisfaction 7

Gender had an influence on several life satisfaction variables. A MANOVA revealed that males had significantly higher life satisfaction compared to females in the facets of health (F(1, 4852) = 4.44, p = 0.04; income (F(1, 4852) = 14.85, p < 0.001 and leisure (F(1, 4852) = 4.19, p = 0.04. No significant associations were observed for dwelling, job and overall life satisfaction. With respect to job satisfaction, a stand-alone ANOVA was calculated, because not everyone in the sample was employed (F(1,4543) = 0.29, p = 0.59. It should be noted that the influence of gender on income only needs to be accounted for because the remaining correlations would not hold given multiple testing (Bonferroni adjustment, dividing p = 0.05 by six tests, resulting in an alpha of 0.008 required for significance).

3.4.

Associations between life satisfaction and PIU

It can be seen in Table 3 that all measures for life satisfaction significantly and negatively correlated with the IAT score (correlations ranged between r = -0.12 to r = -0.20, p < 0.01). Therefore, high life satisfaction values were associated with low sIAT values. Further analyses revealed a significant gender effect, namely for the association between specific facets of life satisfaction health, income, dwelling, leisure and s-IAT. In females the observed associations were higher compared to males. The same was true for the association between hours spent on the Internet and s-IAT: Females showed significantly higher correlations between private use of the Internet and s-IAT values than males. Notably, hours spent on the Internet (business) and the s-IAT scores correlated significantly in the total and male sample, but not in the female subsample. The correlation between PIU and overall life satisfaction and job satisfaction showed no significant gender difference. To test for significant differences between correlations we used Fisher´s z test. After using a 8

Bonferroni correction to control for multiple testing issues, two facets of life satisfaction, health and leisure, showed significant differences between the male and female subsamples. The alpha of 0.05 was divided by six – this was the number of tests conducted (five facets of life satisfaction and general satisfaction), resulting in an alpha threshold of 0.008. A summary of the results can be found in Table 3.

(Insert Table 3 about here)

We also looked at the correlations between hours spent on the Internet (private/business) and life satisfaction values. Due to the violation of the assumption of normal distribution for hours spent on the Internet (business/private), we used Spearman's Rho to calculate the correlations below. For hours spent on the Internet (private) and life satisfaction variables, the significant correlations extended from leisure ρ = -.05 (p < 0.01) to dwelling ρ = -0.13 (p < 0.01). For hours spent on the Internet (business) and life satisfaction, the significant correlations extended from ρ = -0.12 (p < 0.01) for the correlation with overall life satisfaction to ρ = -0.16 (p < 0.01)) for leisure, with the exception of income (no significant correlation).

3.5.

Association between PIU and facets of life satisfaction moderated by gender

To further analyze the role of gender as a possible moderator between PIU and facets of life satisfaction, we conducted a moderator analysis using the process script (Hayes, 2012). All facets of life satisfaction that yielded a significant gender difference after a Bonferroni correction with respect to PIU (leisure and health) were taken into account. Within the models the s-IAT values were used as the independent variable and the facets of life satisfaction leisure and health were used as the 9

dependant variables. The interaction effects between PIU and gender were significant for both leisure (R2 change = 0.0013, B = - 0.020, p = 0.01) and health (R2 change = 0.0010, B = - 0.017, p = 0.028). The moderator effect of gender on the association between PIU and leisure and health respectively can be visualized in Fig. 1. Values of the independent variable are depicted at -1 standard deviation, mean, and +1 standard deviation. As depicted in Figure 1, life satisfaction levels for leisure and health differ significantly between males and females at higher levels of the s-IAT.

(Insert Figure 1 about here)

3.6.

Predictors of PIU

In order to determine the most important predictors for PIU in the context of our assessed variables, we carried out a stepwise multiple regression with the s-IAT value as dependant variable. Facets of life satisfaction and overall life satisfaction, the demographic variables gender and age, as well as hours spent on the Internet (private/business) per week, were entered as predictors. The final model was able to explain 26.9% (R2 = 0.269) of the variance in s-IAT score (F(7,4844) = 254.42, p < .001). The independent variable hours spent on the Internet (private) explained 20.6% (R2 = 0.206) of the s-IAT score variance (β = 0.37; t = 26.74, p < 0.001), followed by age (R2 = 0.042; β = -0.24; t = -18.50, p < 0.001). The remaining variables that were accounted for in the regression model (dwelling, hours spent on the Internet (business), leisure, health, gender) were responsible for a further R2-change of only 2.0%. Please refer to Table 4 for details.

(Insert Table 4 about here) 10

4.

Discussion

The present study aimed to extend the existing research on the negative association between PIU and life satisfaction (e.g. Pawlikowski et al., 2014; Bozoglan et al., 2013; Ko et al. 2007) using a large German speaking sample. In sum, the negative association between overall life satisfaction and problematic Internet use was replicated in the present sample; all specific facets of life satisfaction correlated negatively and significantly with PIU. Although the associations are robust (which is due to the high power to detect significant effects in our study), effect sizes are rather small e.g. a correlation of -0.20 means that only four percent of the variance in life satisfaction can be explained by Internet (over)use. On that basis, our results need to be interpreted cautiously, as even small effect sizes will be statistically significant, given the very large sample size. In addition to the negative association between life satisfaction and PIU, we observed an interesting effect of gender on the associations between specific forms of life satisfaction and PIU. Nearly all correlations between specific forms of life satisfaction and PIU (with the exception of job satisfaction) differed significantly between males and females, with females showing significantly stronger associations. The associations between PIU and the specific life satisfaction facets of leisure and health are particularly noteworthy because females usually show lower scores for generalized Internet use (Li and Kirkup, 2007). This is also the case in our sample. Moreover, females showed lower Internet usage per week than males for both private and business purposes. Taking our results in to account, the facets of life satisfaction health and leisure seem to be affected at a lower level of Internet use in females. This statistical difference between males and females has been 11

demonstrated first by a simple test searching for significant differences between the observed correlations, and additionally by a more elaborate moderator analysis. In contrast, one would expect, given our data, that males need to pass a higher threshold of Internet use to report being affected with respect to health and leisure satisfaction. As men tend to engage in more unhealthy coping behavior in general (Berrigan et al., 2003), and digital consumption forms an important aspect of leisure activities for males (Griffiths and Hunt, 1995; Comber et al., 1997), it might take longer for males to recognize their own PIU and its effect on life satisfaction. This gender specific difference in coping behavior could be one explanation for the findings of this study. Brand et al. (2014) proposed a model on the development and maintenance of generalized and specific Internet addiction where coping plays a role as a mediator of the link between personality and Internet addiction. Although life satisfaction is not mentioned explicitly, it could be a predisposing factor. Brand et al. (2014) hypothesize that specific predispositions could increase the probability that a person feels rewarded when using the Internet. Consequently this person would increase their use of the Internet for reward until this behavior leads finally to PIU. Following this rationale, and taking the findings of the present study in to account, we argue that gender could be an important variable which influences the association between life satisfaction and PIU. In sum, the present study further underlines the link between life satisfaction and PIU. Moreover, we extend findings in the literature by showing an interesting effect of gender on associations between specific life satisfaction and PIU. This study also has some limitations worth noting. The present study only includes cross-sectional data, so we cannot make inferences about the causal nature of the relationship between the included variables. Furthermore, the observed correlations are only 12

small to moderate in size, and due to the high statistical power (because of the high sample size), even small correlations, such as that between hours spent on the Internet (business) and PIU (rho = -0.05), became significant. Lastly, only participants familiar with Smartphones or tablets could take part in the study, which may mean that the sample in the present study is not fully representative of the general population. On the other hand, the use of Smartphones and tablets is pervasive in modern Western societies, so we anticipate that this bias would be negligible. Given the equal gender distribution and large standard deviations for age, we believe that our findings are of interest for researchers and clinicians interested in the degree to which problematic internet usage affects life satisfaction. The current study also extends the existing research, given that many previous studies have used samples that predominantly feature university students. Future studies should investigate the relations between life satisfaction and PIU in more detail by using longitudinal research designs that will allow for stronger causal inferences to be made between these variables.

Acknowledgement The present study was funded by the German Research Foundation (MO 2363/6-1). Moreover, the position of CM is funded by a Heisenberg grant awarded to him by the German Research Foundation (MO 2363/3-1).

Contributors. Author Contributions BL, RS, and CM designed the study. BL and CM prepared the exhibition for data collection. CK programed the online platform and preprocessed the data. BL, AC and RS performed statistical analyses. BL and CM

13

wrote the manuscript. AC did check the manuscript for usage errors. All authors contributed substantially to the final version of the paper.

Conflict of interest. There are no actual or potential conflict of interest.

14

References

Berrigan D., Dodd K., Troiano R.P., Krebs-Smith S.M., Barbash, R.B., 2003. Patterns of health behavior in US adults. Preventive Medicine 36, 615-623.

Bozkurt, H., Coskun, M., Ayaydin, H., Adak, I., Zoroglu, S.S., 2013. Prevalence and patterns of psychiatric disorders in referred adolescents with Internet addiction. Psychiatry Clin Neurosci 67, 352-359.

Bozoglan, B., Demirer, V., Sahin, I., 2013. Loneliness, self‐esteem, and life satisfaction as predictors of Internet addiction: A cross‐sectional study among Turkish university students. Scandinavian journal of psychology 54, 313-319.

Brand, M., Young, K. S., & Laier, C. 2014. Prefrontal control and Internet addiction:

a

theoretical

model

and

review

of

neuropsychological

and

neuroimaging findings. Frontiers in human neuroscience, 8.

Cao, H., Sun, Y., Wan, Y., Hao, J., Tao, F, 2011. Problematic Internet use in Chinese adolescents and its relation to psychosomatic symptoms and life satisfaction. BMC Public Health 11, 80-82.

Carli, V., Durkee, T., Wasserman, D., Hadlaczky, G., Despalins, R., Kramarz, E., Wassermann, C., Sarchiapone, M., Hoven, C.W., Brunner, R., Kaess, M., 2013. The

association

between

pathological

internet

use

and

comorbid

psychopathology: a systematic review. Psychopathology 46, 1-13.

Comber, C., Colley, A., Hargreaves, D.J., Dorn, L., 1997. The effects of age, gender and computer experience upon computer attitudes. Educational Research 39, 123-133.

Griffiths, M.D., Hunt, N., 1995. Computer game playing in adolescence: Prevalence and demographic indicators. Journal of Community & Applied Social Psychology 5, 189-193. 15

Han, D.H., Lee, Y.S., Na, C., Ahn, J.Y., Chung, U.S., Daniels, M.A., Haws, C.A., Renshaw, P.F., 2008. The effect of methylphenidate on Internet video game play in children with attention-deficit/hyperactivity disorder. Compr Psychiatry 50, 251256.

Hayes, A. F. 2012. PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. 2012.

Ko, C.-H., Yen, J.-Y., Yen, C.-F., Lin, H.-C., Yang, M.-J., 2007. Factors predictive for incidence and remission of internet addiction in young adolescents: a prospective study. CyberPsychol Behavior 10, 545-551.

Ko, C.-H., Yen, J.-Y., Yen, C.-F., Chen, C.-S., Chen, C.-C., 2012. The association between Internet addiction and psychiatric disorder: A review of the literature. European Psychiatry 27, 1–8.

Lam, L.-T., Peng, Z.-W., Mai, J.-C., Jing, J. 2009. Factors associated with Internet addiction among adolescents. CyberPsychol Behavior 12, 551-5555.

Li, N., Kirkup, G., 2007. Gender and cultural differences in Internet use: A study of China and the UK. Computers & Education 48, 301-307.

Lowry, R. 2003. Significance of the difference between two correlation coefficients. VassarStats. http://faculty. vassar. edu/lowry/rdiff. html. (retrieved November 23, 2015).

Montag. C., Jurkiewicz, M., Reuter, M., 2010. Low self-directedness is a better predictor for problematic internet use than high neuroticism. Computers in Human Behavior 26, 1531-1535.

16

Montag, C., Kannen, C., Lachmann, B., Sariyska, R., Duke, É., Reuter, M., Markowetz, A., 2015. The importance of analogue zeitgebers to reduce digital addictive tendencies in the 21st century. Addictive Behaviors Reports 2, 23-27.

Montag, C, Duke, É., Reuter, M. 2015a. A Short Summary of Neuroscientific Findings on Internet Addiction. In: Montag, C, Reuter, M. (Eds.), Internet Addicition: Neuroscientific Approaches and Therapeutical Interventions. Springer, New York, pp. 131-139.

Morahan-Martin, J., 1998. The gender gap in Internet use: Why men use the Internet more than women—a literature review. CyberPsychol Behavior 1, 3-10.

Pawlikowski, M., Altstötter-Gleich, C., Brand, M., 2013. Validation and psychometric properties of a short version of Young’s Internet Addiction Test. Computers in Human Behavior 29, 1212-1223.

Pawlikowski, M., Nader, I. W., Burger, C., Stieger, S., & Brand, M. 2014. Pathological Internet use–It is a multidimensional and not a unidimensional construct. Addiction Research & Theory, 22(2), 166-175.

Sariyska, R., Reuter, M., Bey, K., Sha, P., Li, M., Chen, Y.-F., Liu, W.-Y., Zhu, Y.K., Li, C.-B., Suárez-Rivillas, A., Feldmann, M., Hellmann, M., Keiper, J., Markett, S., Young, K.S., Montag, C., 2014. Self-esteem, personality and Internet Addiction: A cross-cultural comparison study. Personality and Individual Differences 61, 28-33.

Sariyska, R., Reuter, M., Lachmann, B., & Montag, C. 2015. Attention deficit/hyperactivity disorder is a better predictor for problematic Internet use than depression: evidence from Germany. J Addict Res Ther, 6(209), 2.

Shaw, M., Black, D.W., 2008. Internet addiction. CNS drugs 22, 353-365.

17

Siedler, T., Schupp, J., Spiess, C.K., Wagner, G.G., 2009. The german socioeconomic panel (soep) as reference data set. Schmollers Jahrbuch: Journal of Applied

Social

Science

Studies/Zeitschrift

für

Wirtschafts-

und

Sozialwissenschaften 129, 367-374.

Tao, R., Huang, X., Wang, J., Zhang, H., Zhang, Y., Li, M. 2010. Proposed diagnostic criteria for internet addiction. Addiction 105, 556-564.

Whang, L.S.M., Lee, S., Chang, G., 2003. Internet over-users' psychological profiles: a behavior sampling analysis on internet addiction. CyberPsychol Behavior 6, 143-150.

Young, K.S., 1998. Internet addiction: The emergence of a new clinical disorder. CyberPsychol Behavior 1, 237-244.

life satisfaction: leisure

Gender specific association between s-IAT and leisure 7.8 7.6 7.4 7.2 7 6.8 6.6 6.4

Male Female -1 SD

Mean s-IAT

+1 SD

life satisfaction: health

Gender specific association between s-IAT and health 7.8 7.6 7.4 7.2 7 6.8 6.6 6.4

Male Female -1 SD

Mean s-IAT

+1 SD

18

Fig. 1. Association between PIU (s-IAT values) and the facets of life satisfaction leisure and health moderated by gender. The s-IAT Values are given as variables at -1 standard deviation, mean, and +1 standard deviation. Table 1. Means and Standard Deviation, Minimum, Maximum, and Skew of all used variables. (N = 4852) Variables

Mean

SD

Min

Max

Skew

Age

28.97

16.99

2

93

0.68

s-IAT

25.31

8.75

12

60

1.23

Internet (private), hours per week, (HIP)

18.55

20.71

0

99

5.34

Internet (business), hours per week, (HIB)

8.58

16.42

0

99

14.40

Health (H)

7.30

2.32

0

10

-1.11

Job (J)

7.05

2.49

0

10

-0.96

Income (I)

6.15

3.09

0

10

-0.56

Dwelling (D)

7.65

2.56

0

10

-1.33

Leisure (L)

7.30

2.37

0

10

-0.99

Overall life satisfaction (OLS)

7.74

2.16

0

10

-1.41

Short Internet Addiction Test (s-IAT).

Table 2. Distribution of internet use in percent distinguished by gender. s-IAT < 31

s-IAT 31 - 37

s-IAT > 37

Total sample (N = 4852) s-IAT

74.1%

17.8%

8.1%

Male (N = 2509) s-IAT

72.0%

19.0%

9.0%

Female (N = 2343) s-IAT

76.4%

16.5%

7.1%

Short Internet Addiction Test (s-IAT). Cutoff values suggested by Pawlikowski et al. (2013).

Table 3. Correlations between short Internet Addiction Test (s-IAT), quantitative variables, demographic variables, and life satisfaction. HIP Total sample (N = 4852) s-IAT Male (N = 2509) s-IAT Female (N = 2343) s-IAT

HIB

Age

H

I

D

L

OLS

J *

0.40

*1

-0.05

*1

-0.27

*

-0.16

*

-0.12

*

-0.18

*

-0.14

*

-0.20

*

-0.17 (N = 4456)

0.36

*1

-0.06

*1

-0.24

-0.13

*

-0.10

*

-0.16

*

-0.11

*

-0.19

*

-0.15 (N = 2284)

0.43

*1

-0.03

1

-0.30

-0.20

*

-0.16

*

-0.22

*

-0.18

*

-0.22

*

-0.18 (N = 2172)

*

*

p= p= p= p= p = 0.02 p = 0.02 p = ns 0.01 p = ns p = ns 0.002 0.006 0.006 z = 2.12 z = 2.17 z = -2.9 z = 2.2 z = 2.5 z = 2.49 Pearson correlations between s-IAT and age, partial correlations (correcting for age) between s-IAT, age, facets of life satisfaction (health (H), income (I), dwelling (D), leisure (L), job (J)), overall life satisfaction (OLS), and test Fisher´s z

19

for significance between male and female correlations with Fisher´s z test. Spearman correlations for hours spend * 1 on the Internet (private; HIP/business; HIB) and s-IAT; p < 0.01; Spearman correlation. Table 4. Stepwise regression for problematic Internet use (PIU) (N = 4852). 2

R

β

p

Hours spend on the Internet (private)

0.206

0.373

<0.01

Age

0.249

-0.240

<0.01

Dwelling

0.260

-0.065

<0.01

Hours spend on the Internet (business)

0.264

0.068

<0.01

Leisure

0.267

-0.050

<0.01

Health

0.268

-0.038

<0.01

Gender

0.269

-0.032

0.01

Independent variables

Dependant variable: s-IAT.

Highlights 

We investigate the relation of problematic Internet use (PIU) and life satisfaction.



Facets of life satisfaction health and leisure were negatively correlated with PIU.



Male reported a higher total amount of PIU than females.



Association between life satisfaction and PIU was significantly higher for females.



We assume gender specific thresholds concerning life satisfaction and PIU.

20