Internet Usage and Sleep Quality - Lab Report Example

Running head: INTERNET USAGE AND SLEEP QUALITY Internet Usage and Sleep Quality Jonathan Jones The of Nottingham Given current trends toward an increasingly internet-connected world, the purpose of this study was to examine the effects of internet usage on mental and physical health, particularly through the well-established connection between sleep quality and well-being. This study examines the effects of internet usage on participants’ ratings of sleep quality. Results of correlational and regression analyses show that as effort spent using the internet increases, participants experience poorer quality of sleep, as measured by reliable and valid measures of self-reported internet usage and sleep quality. However, correlations between weak social support or interactions and internet usage are not significant enough to draw conclusions. By being mindful of how internet affects sleep, internet can moderate their use to enable satisfying sleep. Internet Usage and Sleep Quality The introduction of internet technologies in the 20th century was revolutionary in the sense of connecting people and increasing the efficiency of communication tasks. By doing more in less time with computers connected to the world wide web, individuals are able to access boundless quantities of information. In the 21st century, information on the internet is created, uploaded, and shared in real time over high-speed networks, which has dramatically increased its efficacy as a production-driving and social-based tool. In addition, internet-connected mobile devices allow individuals to take the internet with them wherever they go, increasing its availability and influence over people’s daily lives. Whereas before 2009, internet use on mobile phones was rare, the trend in technology has rapidly increase since where younger people are now the most frequent user groups of mobile internet technology. Students, professionals, and casual users of internet services often have needs for information that blend the boundaries between school, work, and home life. Because of individuals’ interconnectedness with the internet and other users of the technology, there is a growing field of research on internet addiction, which is measured by a variety of scales including Young’s Internet Addiction Test (IAT), the Pathological Internet Use Questionnaire (PIUQ), and the Generalized Problematic Internet Use Scale (GPIUS). Internet addiction, also known as internet dependence, problematic internet use, or internet addiction disorder, describes excessive use of the technology in a manner that correlates to use of DSM-IV criteria, similar to previously identified forms of addiction such as drug and gambling addictions (Young, 1998). Internet addiction is largely thought of as a compulsive behaviour, and cognitions associated with it are related to notable distress in daily life. It can otherwise be thought of as an impulse-control disorder wherein people cannot help but check their computer. Young’s Internet Addition Test (IAT) has received much attention in terms of its effectiveness at measuring the concept of internet addiction, which Young and associates have developed throughout a series of papers on the subject. The IAT was developed in 1998, but it has gone through updates and modifications in order to investigate the relationships between internet addiction and other kinds of well-documented addictions (Young, 1998). Throughout these studies, the overall reliability of the IAT scale has been supported through a measure of Cronbach’s alpha, even if individuals were not addicted to the internet per se but instead only certain activities (Chang & Law, 2008). In addition to tests of reliability, Young’s Internet Addiction Test has also be subjected to systematic psychometric testing, which concluded by means of factor analysis that the IAT is high in internal consistency and concurrent validity (Widyanto & McMurran, 2004). With that in mind, the IAT can serve as a reliable instrument in evaluating the frequency and severity of an individual’s habitual internet use, evaluating whether that use is at pathological (rather than casual or professional use) levels. Similar to addictive behaviours, sleep quality is also a driver of mental and physical health, and it may be due to some personality characteristics that make it more likely for an individual to develop insomnia due to or independent of daytime dysfunction (Soehner, Kennedy, & Monk, 2007). One research question not previously given great attention has been the extent to which personality characteristics related to poor sleep quality and the onset of insomnia are also related to behaviours making people more susceptible to internet addiction and other forms of impulse-control disorders. Because poor sleep is a common problem throughout modern societies and because poor sleep is directly related to health, studies of sleep quality have focused on finding its underlying causes (Cheng, et al., 2012). Another related research question is the extent to which impairments in sleep quality, especially among younger people who are more frequent users of technology, are due to changes in modern behaviours or lifestyles or other factors such as a lack of social support (i.e. loneliness or isolation associated with increased incidence of depression and stress). Research conducted by Do, Shin, Bautista, and Foo (2012) has examined the role of sleep duration and adolescent health outcomes, looking particularly at time spent on the internet as a moderator between the two variables. By looking at a cross-sectional online survey of 136,589 adolescents in South Korea, the authors were able to determine that shorter self-reported sleep duration was associated with a higher likelihood of self-reported depression symptoms, suicidal thoughts, and self-rated health. Sleep duration and excessive internet use are additive factors contributing toward adverse health among adolescents, which is an age-group in developed countries that uses mobile-enabled internet technology on a daily basis. By contributing to sleep deprivation, excessive internet use has both direct and indirect consequences on health (Do, Shin, Bautista, & Foo, 2013). However, by measuring sleep duration rather than sleep quality, Do, Shin, Bautista, and Foo (2012) is missing a key component of sleep’s relationship to health in terms of indicating the quality of sleep rather than its duration. Perhaps, for some individuals, a particularly those who are older, a smaller number of hours of sleep is sufficient, so long as it is of a high quality. A research question building on this line of work is whether sleep quality adds any explanatory value above and beyond simply looking at self-reported sleep duration data. Like internet use and internet addiction, scales used to measure sleep and the quality of sleep are well-established and stable on reliability measures in the clinical psychiatry literature. Sleep quality is an operational concept used to describe individual behaviour as it relates to a person’s sleepiness during the day time, which correlates negatively with subjective ratings of how deep and satisfying the person’s previous night’s sleep was. Individuals can rate poorly on standardized scales of sleep quality known as the Pittsburgh Sleep Quality Index (PSQI), which allows clinical and nonclinical researchers to assess sleep quality during the past month based on 19 self-rated questions (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). Higher scores on the PSQI represent worse sleep quality, and the scale has been used reliably to measure sleep quality among truck drivers and in randomized placebo-controlled trials (Knutson, Rathouz, Yan, Liu, & Lauderdale, 2006). For the purposes of measuring the relationship between internet use and sleep quality, Young’s Internet Addition Test and Buysse et al. (1989)’s Pittsburgh Sleep Quality Index are reliable measures. Similar to Do, Shin, Bautista, and Foo (2012), research conducted by Thomée (2012) in a thesis indicates a correlation between “intensive” internet use (defined in terms of duration of continuous use) and mental health problems in both men and women. Using computers late at night, in particular, was a cause for stress and depression that led to lost sleep and sleep disturbances. Likewise, frequent use of mobile phones (most with internet connectivity) posed a risk for sleep disturbances and depressive symptoms, which served additively to the intensive use of computers and its relationship to these problematic mental health consequences. Again, although lost sleep and sleep disturbances lead to lower sleep quality, those variables are ways of measuring sleep quality, rather than capturing the whole picture of sleep quality, which the Pittsburgh Sleep Quality Index (PSQI) attempts to measure. In addition, rather than using an established measure like Young’s IAT, the Thomée (2012) study made use of a proprietary measure of information and communication technology exposure, which encompasses forms of technology and communication that are separate from internet use. By narrowing the scope of the research down to only internet use and by utilizing scale measurements that have been widely validated in independent research, a new study can attempt to fill gaps in the literature. Despite having identified some segments of the population as being addicted to the internet, for most users of the internet, the technology enables higher levels of productivity and easier access to leisure. While the benefits of the internet are clear in the sense that complex tasks, if not already automated, can be easily processed using only an internet connection, the research literature has yet to examine in great detail the negative effects of internet use on the mental health (particularly in the sleep patterns and sleep quality) of people who utilize the internet for a wide variety of tasks, including those who are addicted to the internet as measured by a scaled approach. Since the largest groups of mobile-enabled internet and internet-utilizing people tend to be those in younger age groups, determining the effect of internet usage on sleep patterns is particularly urgent given the link between sleep patterns, sleep quality, and overall mental health among people who are still developing mentally and physically (Watkins, 2009). Internet access can improve patient outcomes in a variety of clinical settings (Afsar, 2013). However, poor sleep caused by excessive internet use is also correlated with higher anxiety, greater risk for physical health problems, and these risks are increased depending on whether individuals already have diseases or pre-existing health issues (McKinley, Fien, Elliott, & Elliott, 2013). Based on research showing the effect of poor sleep on health outcomes and research establishing that internet addiction correlates with higher stress levels, I hypothesize the following: H1: Participants who rate as having poor sleep quality on the Pittsburgh Sleep Quality Index will also have high internet use (or internet addiction) ratings on the Internet Addiction Test. In a study of adolescents in India, Nalwa and Anand (2003) found a significant correlation between individuals who were addicted to the internet (i.e. pathological internet use) and who scored highly on a loneliness measure. In addition to experiencing isolation and loneliness, internet-addicted adolescents were more likely to delay other work to spend time online and felt life would be boring without the internet, leading to the conclusion that individuals were sacrificing the efficiency-gaining benefits of the internet for activities that are actually counter-productive and anti-social. Rather than increasing academic achievement, students who were internet-addicted in the study actually have internet in their lives as an impairment to education and a time-consuming resource that prevents interactions with other students in real-life through extracurricular and academic activities (Nalwa & Anand, 2003). As a result of such isolation, individuals will likely experience higher levels of stress and therefore poorer quality of sleep, and this will be reflected in their responses to questions on the IAT dealing with relationships. In alignment with other research by Cheng et al. (2012), who hypothesize that smaller social support networks is an impairment to sleep quality, I also hypothesize the following: H2: Participants who rate as having poor social support on questions related to social interactions and relationships on the IAT will have lower sleep quality as measured by the PSQI than participants who rate as having higher levels of social support on the IAT. Method Design [intentionally left blank] Participants [intentionally left blank] Materials [intentionally left blank] Procedure [intentionally left blank] Results From the 47 responses collected, the average score on the Pittsburgh Sleep Quality Index (PSQI) was a 7.66, which is within an acceptable range as defined by the authors of the scale, but anything greater than a score of 5 is associated with poor sleep quality (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). Standard deviation for the PSQI was a 3.62, which means that the bulk of the distribution was within an approximate score range of 4.04 and 11.28, which is centred on the better sleep quality end of the possible score range (which has a maximum of 21). Overall, participants in the aggregate reported average sleep quality levels that one would expect to see in a broad population. On the twenty questions asked in the Internet Addition Test (IAT), participants scored an average rating of 35.40, which is defined as “an average on-line user” that “may surf the Web a bit too long at times”, but essentially has control over his or her usage; more excessive usage, according to how the scale is defined, does not start until one scores at least a 50 on the inventory (Young, 1998). The standard deviation of scores was 18.08, which means that the bulk of the distribution was within an approximate score range of 17.30 to 53.48, representing a sample that is not significantly internet-addicted or excessive in their use of internet-connected technology. In addition, tests for normality show the distribution of scores on PSQI (Figure 1) as well as on IAT (Figure 2) tend to be normal in the sense that scores are clustered in the middle and extreme scores appear left and right of the centre cluster. A normal distribution of scores indicates the sample is a reliable representation of the population. The PSQI inventory is included in Appendix 1 along with instructions on scoring in Appendix 2 to show PSQI results are interpreted such that higher scores on the inventory equate to poorer sleep quality. Appendix 3 includes a copy of the IAT as well as instructions on scoring in Appendix 4. Higher scores on the IAT represent more internet usage. Taking these considerations together, a positive correlation means that worse sleep quality (higher PSQI) is correlated with more internet usage, which is an interpretation used to test the relationship among the variables as expressed in H1. Another way of describing that relationship is in terms of an inverse relationship between sleep quality and internet usage. A correlational analysis found a positive correlation between more internet usage and poor sleep quality that was significant (r = 0.38, p < 0.01), which affirms H1. In terms of coefficient of determination, approximately 14.61% of the variance in sleep quality can be explained in terms of internet usage by participants. The regression analysis results are shown in Table 1 for more details. The regression equation for this data set is: internet usage (IAT score) = 20.78 + 1.91 (sleep quality (PSQI score) An interpretation of this result is that for each additional +1 in the PSQI composite score (worse sleep quality), one can predict +1.91 on the participant’s score on the IAT scale (more internet usage). The resulting correlation is quite strong, and the effect of marginal internet use on someone’s sleep quality probably depends on how many hours of sleep have already been given up to internet usage. To set a sense for how the results compare in terms of relationships on a scatterplot, see Figure 3, which presents scaled responses to PSQI and IAT along with a trend curve depicting the line of best fit. Scores tend to be clustered toward lower end on both scales, with some extreme responses on the right and upper parts of the chart. With all 47 responses represented, the chart shows the overall trend in the dataset. In order to determine the validity of H2, which hypothesized that participants who rate as having poorer social support on the IAT will have lower sleep quality as measured by the PSQI, a subscale on the IAT was created based on the 6 items on the IAT having to deal with relationships and social interactions related to internet usage. These 6 items were questions 3, 4, 5, 9, 13, and 19, and they all had some indication of how using the internet affects relationships both on- and off-line, with higher scores indicating that internet usage negatively affects relationships and lower scores indicating that internet usage has less or no effect on relationships. Participants’ scores were summed on all 6 items, and they were evaluated on a 30-point scale. The average of the sub-scores was 7.87 with a standard deviation of 5.47. Like the total IAT scores, social sub-scores tended to be distributed normally, but with a slight skew to the right (as seen in Figure 4). Ultimately, a correlational analysis found a slightly positive but not statistically significant result (r = 0.22, p = 0.15) indicating a correlation between internet usage’s effect on relationships and its effect on sleep quality (see Table 2). Like the relationship between regular IAT scores and PSQI scores, the trend is slightly positive (see Figure 5), but not significant like the aggregated IAT scores. Discussion Statistical analysis revealed that an inverse relationship between internet usage and sleep quality, which is a result that mirrors what Thomée (2012) discovered for lost sleep and sleep disturbances and what Do, Shin, Bautista, and Foo (2012) discovered for sleep duration in adolescents. However, by expanding the results of their research into the area of sleep quality by employing the reliable PSQI and IAT scales, this study has confirmed the hypothesized relationship. Instead of using proprietary or unestablished measures of sleep quality and internet usage, the results here are based on scales that have been in the research literature for more than a decade and have been subjected to rigorous validity testing. Doing this adds validity to the research demonstrating a link between using the internet excessively and losing out on high quality sleep. It also draws attention to the negative social effects of excessive internet usage, which is growing at an accelerated rate with the popularity of internet-ready mobile devices. Adolescents, for whom internet (along with other technologies such as SMS) is the most ever-present factor, are most likely going to be the first generation where the effects of high computer use and poor sleep quality will be seen in mental and physical health outcomes. More in-depth epidemiological studies similar to the one conducted by Do, Shin, Bautista, and Foo (2012) will be important in measuring these effects across very large samples of participants, hopefully using established measures such as PSIQ and IAT. This study was unable to confirm conclusions offered by Nalwa and Anand (2003) and Cheng et al. (2012) explaining social isolation as a moderating variable in the relationship between internet usage and sleep quality. Because this study only looked at the social dimensions (or effects) of excessive internet usage by looking at a subset of IAT questions looking at social relationships, it is not possible to use this study as evidence that conclusions in Nalwa and Anand (2003) and Cheng et al. (2012) are inaccurate. While excessive internet usage is associated loosely with depression (Do, Shin, Bautista, & Foo, 2013), social isolation is likely a result of depression rather than its cause—therefore creating an indirect relationship between excessive internet usage and depression (a common cause of sleep loss). Future research should explore the links between these variables (including mental health generally) in a way that sheds light on how internet usage interacts with depressive symptoms. The goal of that line of research should be not only exploring the linkage between the variables, but also in demonstrating how different behaviours related to the internet can produce better mental and physical health outcomes. Having established that a correlation exists, additional studies may need to be conducted in order to find out the underlying cognitive mechanism that exists behind poorer sleep quality in those who are either addicted to or who are excessively using internet-connected devices. Neuroimaging studies would offer valuable insights on how the brain works in receiving stimulation from the internet and how that stimulation affects sleep mechanisms necessary for restful, satisfying sleep (see Steriade and McCarley (2010) for more detail on waking and sleeping from a biological perspective). Using research conclusions like the findings presented in this study, further research explorations can be justified in looking at those brain structures and interactions in light of 21st century trends toward greater dependence on internet technologies for communication and social linkage. The problems created by poor sleep not only extend to worsened mental and physical health, but it also has the potential to create divisions between members of particular demographic segments. The research presented here was limited in the sense that it did not collect data on demographic information, such as gender and age. While most studies of this kind collect such data to explore gender or age differences within the sample, those findings were not available for analysis. Future studies should look to investigate what role, if any, demographic factors play in the relationship between the variables. Since usage of internet-connected devices is not as common older people (Loges & Jung, 2001), for example, perhaps a sample of older participants would score low on the IAT. Based on the findings of this study, low scores on the IAT measure for an older sample would predict low scores on the PSQI, indicating better sleep quality. The digital divide between gender is also well-documented, with females consistently being slower to use both computers and the Internet (Tapscott, 1998). Like the situation described with older demographics, it is very much possible that since females use internet less and therefore, from the conclusions of this research, have better quality sleep that some of the variance in job or academic performance based on gender may be related to these factors. References Afsar, B. (2013). The relation between Internet and social media use and the demographic and clinical parameters, quality of life, depression, cognitive function and sleep quality in hemodialysis patients. General Hospital Psychiatry, 35, 625-630. Buysse, D., Reynolds, C., Monk, T., Berman, S., & Kupfer, D. (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28, 193-213. Chang, M., & Law, S. (2008). Factor structure for Young’s Internet Addiction Test: A confirmatory study. Computers in Human Behavior, 24, 2597-2619. Cheng, S., Shih, C., Lee, H., Hou, Y., Chen, K., Chen, K., . . . Yang, C. (2012). A study on the sleep quality of incoming university students. Psychiatry Research, 197, 270-274. Do, Y., Shin, E., Bautista, M., & Foo, K. (2013). The associations between self-reported sleep duration and adolescent health outcomes: What is the role of time spent on Internet use? Sleep Medicine, 14, 195-200. Knutson, K., Rathouz, P., Yan, L., Liu, K., & Lauderdale, D. (2006). Stability of the Pittsburgh Sleep Quality Index and the Epworth Sleepiness questionnaires over 1 year in early middle-aged adults: The CARDIA Study. Sleep, 29, 1503-1506. Loges, W., & Jung, J. (2001). Exploring the digital divide: Internet connectedness and age. Communication Research, 28, 536-562. McKinley, S., Fien, M., Elliott, R., & Elliott, D. (2013). Sleep and psychological health during early recovery from critical illness: An observational study. Journal of Psychosomatic Research, 75, 539-545. Nalwa, K., & Anand, A. (2003). Internet addiction in students: A cause of concern. CyberPsychology & Behavior, 6, 653-656. Soehner, A., Kennedy, K., & Monk, T. (2007). Personality correlates with sleep-wake variables. Chronobiology International, 24, 889-903. Steriade, M., & McCarley, R. (2010). Brain control of wakefulness and sleep. New York: Springer. Tapscott, D. (1998). Growing up digital: the rise of the Net generation. New York: McGraw-Hill. Thomée, S. (2012). ICT use and mental health in young adults. Gothenburg: Sahlgrenska Academy. Watkins, S. (2009). The young and the digital: What the migration to social-network sites, games, and anytime, anywhere media means for our future. New York: Beacon Press. Widyanto, L., & McMurran, M. (2004). The psychometric properties of the Internet Addiction Test. CyberPsychology & Behavior, 7, 449-456. Young, K. (1998). Internet addiction: The emergence of a new clinical disorder. CyberPsychology & Behavior, 1, 237-244. Appendix 1 [insert PSQI here] Appendix 2 [insert PSQI scoring instructions here] Appendix 3 [insert IAT here] Appendix 4 [insert IAT scoring instructions here] Table 1. Regression Statistics Multiple R 0.382252 R Square 0.146116 Adjusted R Square 0.127141 Standard Error 16.88965 Observations 47 ANOVA   df SS MS F Significance F Regression 1 2196.614 2196.614 7.70039 0.008012 Residual 45 12836.71 285.2601 Total 46 15033.32         Coefficients Standard Error t Stat P-value Intercept 20.77966 5.817596 3.571864 0.000858 X Variable 1 1.909322 0.688054 2.774958 0.008012 Table 2. Regression Statistics Multiple R 0.215266 R Square 0.04634 Adjusted R Square 0.025147 Standard Error 5.398573 Observations 47 ANOVA   df SS MS F Significance F Regression 1 63.72769 63.72769 2.186605 0.146182 Residual 45 1311.506 29.14459 Total 46 1375.234         Coefficients Standard Error t Stat P-value Intercept 5.381356 1.859525 2.893942 0.005847 X Variable 1 0.325212 0.219928 1.478717 0.146182 Figure Captions Figure 1. PSQI Score Frequency Distribution. Figure 2. IAT Score Frequency Distribution. Figure 3. Sleep Quality vs. Internet Usage. Figure 4. IAT Social Scale Score Frequency Distribution Figure 5. Sleep Quality vs. IAT Social Scale Score Figure 1. Figure 2. Figure 3. Figure 4. Read More